# Corollary Labs

> Corollary is a home for scientific AI: run open biology and chemistry AI models online with no GPU and no installation, explore scientific datasets in the browser, and use a research agent that runs bioinformatics software, queries scientific databases and runs nf-core pipelines for you. Corollary is a non-commercial research demonstrator. It is provided free of charge, without warranty or support, and must not be used for commercial purposes or relied upon for clinical, diagnostic, regulatory or safety-critical decisions.

- 301 AI models (145 runnable online from their page), 248 datasets, 42 bioinformatics programs, 63 scientific databases and 156 nf-core pipelines.
- Every model page states the licence and whether commercial use is permitted, and each page's description is written from the model's or tool's own documentation.
- Run a model: open its page at https://corollary-labs.com/models/<id> and press Run, or ask the research agent at https://corollary-labs.com/agent.
- Developer API: https://corollary-labs.com/developers
- MCP server for AI agents (no key, read-only): https://corollary-labs.com/mcp — setup at https://corollary-labs.com/developers/docs/mcp

## What you can do here (one page per task)

- [Check genome assembly completeness online](https://corollary-labs.com/for/genome-assembly-completeness): Assess the completeness of your genome assembly, transcriptome, or gene set with BUSCO 6.1.0 using curated OrthoDB 12 single-copy orthologues, without installing local tools or lineage datasets.
- [Assemble a genome from sequencing reads](https://corollary-labs.com/for/de-novo-genome-assembly): Assemble genomes de novo using MaSuRCA from short paired-end reads alone or combined with long reads, then evaluate assembly contiguity and lineage-specific completeness with BUSCO.
- [Recover genomes from a metagenome](https://corollary-labs.com/for/metagenome-binning): Recover metagenome-assembled genomes from your contigs and read cohort using Bowtie 2, samtools, and MetaCAT, with lineage completeness and contamination checked by BUSCO.
- [Predict genes in a bacterial genome](https://corollary-labs.com/for/bacterial-gene-prediction): Predict protein-coding genes in bacterial or archaeal assemblies with Prodigal, then annotate functions using DIAMOND or MMseqs2 homology searches against reference databases.
- [Annotate a phage genome online](https://corollary-labs.com/for/phage-genome-annotation): Annotate bacteriophage contigs using Pharokka to identify protein-coding genes, PHROG functional categories, tRNAs, CRISPR arrays, and AMR or virulence markers.
- [Find viruses and prophages in a genome or metagenome](https://corollary-labs.com/for/find-viruses-in-sequence-data): Identify viral contigs and integrated proviruses in assembled genomes or metagenomes using geNomad, then determine their completeness and contamination with CheckV.
- [Check viral genome completeness and contamination](https://corollary-labs.com/for/viral-genome-quality): Assess viral genome completeness, host contamination and MIUViG quality tiers with CheckV. Establish whether your sequences represent complete genomes or fragments before downstream annotation.
- [Classify a phage to ICTV genus and species](https://corollary-labs.com/for/phage-taxonomy): Classify bacteriophage genomes to ICTV genus and species using taxmyPHAGE and the INPHARED database, applying the official 70% and 95% nucleotide identity thresholds.
- [Find RNA viruses in metatranscriptome assemblies](https://corollary-labs.com/for/rna-virus-discovery): Identify RNA viruses in metatranscriptome assemblies by screening for RdRp domains across eight profile HMM collections, with MMseqs2 taxonomy against RefSeq RNA viruses.
- [Find CRISPR arrays and Cas genes in a genome](https://corollary-labs.com/for/crispr-array-detection): Identify CRISPR arrays, predict Cas operons, and classify subtypes across bacterial, archaeal, or metagenomic assemblies without local software installation.
- [Analyse a pooled CRISPR screen](https://corollary-labs.com/for/crispr-screen-analysis): Analyse pooled CRISPR knockout screens from sgRNA counts using MAGeCK, with gene ranking and honest false discovery rate correction across screening targets.
- [Differential expression from RNA-seq counts](https://corollary-labs.com/for/rna-seq-differential-expression): Estimate log fold changes and false discovery rates directly from unnormalised RNA-seq count matrices using negative binomial models with empirical Bayes shrinkage, batch adjustment, and stable gene identifier mapping.
- [Run an nf-core pipeline without a cluster](https://corollary-labs.com/for/run-nf-core-pipelines): Run standard nf-core pipelines locally using Nextflow. Validate your samplesheet against the pinned pipeline schema, execute on a dedicated single-machine job, and inspect MultiQC reports without managing cluster queues.
- [Call and filter variants from aligned reads](https://corollary-labs.com/for/variant-calling-and-filtering): Call, normalise, and filter genetic variants from aligned sequencing reads across explicit genome builds, generating standardised VCF records with calibrated population and functional effect annotations.
- [Call peaks in ATAC-seq and ChIP-seq data](https://corollary-labs.com/for/atac-and-chip-peak-calling): Call peaks from ATAC-seq and ChIP-seq alignments using MACS3, annotate genomic features with HOMER and bedtools, and query transcription-factor motifs against JASPAR.
- [Quality-control and trim sequencing reads](https://corollary-labs.com/for/read-quality-control): Quality-control, filter, and trim raw FASTQ files using fastp or Trimmomatic. Inspect quality reports, assess base scores and adapter contamination, and prepare clean reads for downstream analysis.
- [Deconvolve bulk RNA-seq into cell types](https://corollary-labs.com/for/bulk-rna-seq-deconvolution): Estimate cell-type proportions in bulk RNA-seq or Visium spots from an annotated single-cell reference using Rectangle, correcting for cell size differences and isolating uncharacterised signal as Unknown.
- [Analyse single-cell RNA-seq data](https://corollary-labs.com/for/single-cell-rna-seq-analysis): Process single-cell RNA-seq from raw counts through adaptive quality control, clustering, and cell-type annotation to pseudobulk differential expression across biological replicates.
- [Analyse spatial transcriptomics data](https://corollary-labs.com/for/spatial-transcriptomics-analysis): Analyse spatial transcriptomics data across sequencing and imaging platforms. Inspect QC, calculate spatially variable genes, map neighbourhoods, and deconvolve mixed spots using reference atlases.
- [Quantify multiplexed tissue images](https://corollary-labs.com/for/multiplexed-tissue-imaging): Quantify multiplexed tissue sections into single-cell marker intensities, cell phenotypes, and spatial neighbourhood architectures using spatialproteomics without local software installation.
- [Predict single-cell responses to unseen perturbations](https://corollary-labs.com/for/perturbation-response-prediction): Predict single-cell transcriptomic shifts under unmeasured perturbations, doses, or contexts using LabCompass flow matching models, evaluated directly against mean-shift and no-change baselines.
- [Predict a protein structure from its sequence](https://corollary-labs.com/for/protein-structure-prediction): Predict or retrieve three-dimensional protein structures from sequence, evaluate per-residue pLDDT and inter-domain PAE confidence metrics, and identify disordered regions or experimental caveats.
- [Dock a ligand into a protein online](https://corollary-labs.com/for/protein-ligand-docking): Dock small molecules into prepared protein structures using AutoDock Vina, SMINA, DiffDock, or Boltz-2, with automatic control redocking and geometric sanity checks.
- [Analyse a protein sequence](https://corollary-labs.com/for/protein-sequence-analysis): Clean and characterise protein sequences, calculate physicochemical properties accurately, and identify domains and homologs using deterministic statistics, database queries, and sequence embeddings.
- [Align sequences and build a phylogenetic tree](https://corollary-labs.com/for/multiple-sequence-alignment-phylogeny): Clean sequence sets, align with MAFFT or MUSCLE, and infer maximum-likelihood phylogenetic trees using IQ-TREE or FastTree, with model selection and honest alignment metrics.
- [Number, model and analyse antibodies](https://corollary-labs.com/for/antibody-structure-and-numbering): Number antibody variable domains, annotate CDRs, generate 3D coordinates, and screen developability attributes using ANARCI, ImmuneBuilder, and sequence-based analysis tools without local installation.
- [Calculate molecular properties from SMILES](https://corollary-labs.com/for/molecule-property-calculation): Calculate molecular descriptors and assess drug-likeness rules from SMILES. Receive standardised structures, physicochemical properties, and rule violations without local installations.
- [Predict ADMET properties of a compound](https://corollary-labs.com/for/admet-prediction): Standardise small-molecule structures and predict 52 ADMET and toxicity endpoints using ADMET-AI and Chemprop D-MPNN ensembles, benchmarked against reference distributions of approved drugs.
- [Screen a compound library against a target](https://corollary-labs.com/for/virtual-screening): Screen compound libraries against protein targets using AutoDock Vina, RDKit, and enrichment validation without local installation.
- [Build a PLS calibration from spectra](https://corollary-labs.com/for/spectra-calibration): Build partial least squares calibrations from spectral matrices, assess predictive performance with robust cross-validation schemes, inspect VIP profiles, and resolve mixture spectra using MCR-ALS.
- [Preprocess an untargeted LC-MS feature table](https://corollary-labs.com/for/metabolomics-preprocessing): Convert an untargeted LC-MS feature table into an analysis-ready matrix using label-free diagnostics, blank and detection filtering, LOESS drift correction evaluated on held-out QCs, and matrix-specific normalisation.
- [Process accelerometer data for sleep and activity](https://corollary-labs.com/for/accelerometer-sleep-and-activity): Extract physical activity, rest-activity rhythms, and sleep estimates from raw wrist or hip accelerometer files using GGIR, configured for your study cohort.
- [Segment organs and read findings in abdominal CT](https://corollary-labs.com/for/abdominal-ct-organ-segmentation): Research reads of contrast-enhanced abdominal CT: 36 structures, organ volumes and finding scores, with the clinical line drawn.
- [Fit dose-response and kinetics curves](https://corollary-labs.com/for/lab-assay-curve-fitting): Fit enzyme kinetics and dose-response curves using lmfit and scipy. Obtain Km, Vmax, or IC50 with confidence intervals and residual plots from assay data without installing any software.
- [Analyse flow cytometry data](https://corollary-labs.com/for/flow-cytometry-analysis): Parse FCS files, apply spillover compensation matrices, transform fluorescence scales, execute sequential gating from controls, and calculate population counts and parent frequencies.
- [Quantify microscopy images and gels](https://corollary-labs.com/for/microscopy-image-quantification): Quantify raw intensity data from microscopy images, western blots, and colony plates without manual installation. Run background subtraction, segmentation, densitometry, and colocalisation using validated Python tools.
- [Run a literature review with citations you can check](https://corollary-labs.com/for/literature-review): Search live scientific indexes, inspect methods in open-access full text, and build an evidence synthesis where every claim links directly to a retrieved citation.
- [Design an experiment with the statistics settled first](https://corollary-labs.com/for/experiment-design): Turn an open question into a staged, falsifiable protocol with sample size, randomisation, specific controls, and statistical analysis fixed before running an assay.
- [Choose and check a statistical test](https://corollary-labs.com/for/statistical-analysis-review): Choose the correct statistical test directly from your experimental design, evaluate assumptions, and report effect sizes with confidence intervals rather than isolated p-values.
- [Generate individual-level synthetic data from a summary table](https://corollary-labs.com/for/synthetic-data-generation): Generate an individual-level synthetic cohort matching summary statistics from a baseline table or dataset. You get a downloadable CSV and an itemised moment-by-moment verification report.
- [Reproduce a published figure or benchmark](https://corollary-labs.com/for/reproduce-a-published-result): Test a specific published benchmark or quantitative figure by setting explicit tolerances beforehand, running the canonical code on isolated compute, and reporting an evidence-based verdict.
- [Query scientific databases without an API key](https://corollary-labs.com/for/query-scientific-databases): Query public scientific databases and datasets directly through an automated research agent. Retrieve verified records, explore schemas, and attribute every returned figure to its exact source without personal API keys.

## Compare by task

- [Protein structure prediction models you can run online](https://corollary-labs.com/run/protein-structure-prediction): These tools predict three-dimensional coordinates for proteins, complexes, and nucleic acids from sequence data or retrieve solved structures, serving researchers in structural biology and molecular design.
- [Protein design models you can run online](https://corollary-labs.com/run/protein-design): Computational models that generate novel protein backbones, fit sequences to structural scaffolds, and design targeted binders for structural biology and protein engineering applications.
- [Protein language models you can run online](https://corollary-labs.com/run/protein-language-models): Protein language models extract biological representations, score mutations, and design protein sequences or structures from primary amino acid chains and structural tokens.
- [Molecular docking and binding affinity models you can run online](https://corollary-labs.com/run/molecular-docking): Tools in this category predict three-dimensional binding poses and calculate protein–ligand binding affinities, helping researchers evaluate candidate molecules and understand macromolecular interactions.
- [Antibody design and analysis models you can run online](https://corollary-labs.com/run/antibody-design): Computational antibody tools number, model, humanise, and design immunoglobulin sequences and structural coordinates, supporting researchers across therapeutic antibody engineering and repertoire analysis.
- [DNA and genomics models you can run online](https://corollary-labs.com/run/dna-genomics-models): DNA and genomic foundation models process raw nucleotide sequences to predict regulatory elements, assess variant effects, and generate biological sequences across single loci or million-base-pair spans.
- [Single-cell foundation models you can run online](https://corollary-labs.com/run/single-cell-models): Foundation and predictive models for single-cell transcriptomics, enabling cell type annotation, latent embedding generation, gene regulatory network reconstruction, and in silico perturbation forecasting.
- [Chemistry and ADMET prediction models you can run online](https://corollary-labs.com/run/chemistry-admet): Computational tools and models in this category evaluate small-molecule ADMET profiles, predict physicochemical properties and binding affinities, generate novel structures, and model chemical reactions.
- [Molecular simulation and materials models you can run online](https://corollary-labs.com/run/molecular-simulation): Computational engines and machine learning potentials for atomistic simulation, molecular dynamics, crystal property prediction, and generative materials discovery across chemistry and materials science.
- [Pathology and medical imaging models you can run online](https://corollary-labs.com/run/pathology-medical-imaging): Pathology and medical imaging models analyse histology slides and clinical scans for cancer subtyping, tissue segmentation, multimodal clinical question answering, and spatial molecular profiling.
- [Biomedical language models you can run online](https://corollary-labs.com/run/biomedical-language-models): Biomedical language models extract entities from clinical text, query scientific publications, and reason over biological data such as proteins, molecules, and medical imaging for researchers and clinicians.
- [Weather, climate and Earth observation models you can run online](https://corollary-labs.com/run/earth-climate-models): These models forecast global weather, emulate multi-year climate variability, analyse remote sensing imagery, and predict solar events for Earth and space scientists.
- [Genome assembly and quality tools you can run online](https://corollary-labs.com/run/genome-assembly-tools): These tools construct genome sequences from raw reads, bin metagenomic contigs into microbial genomes, and evaluate assembly completeness using conserved orthologues.
- [Phage and virus genomics tools you can run online](https://corollary-labs.com/run/phage-virus-tools): Computational pipelines and classifiers for detecting viral and phage sequences, estimating assembly completeness, annotating phage genes, and classifying phage lifestyles and taxonomy.
- [CRISPR analysis tools you can run online](https://corollary-labs.com/run/crispr-tools): These tools detect natural CRISPR arrays and Cas genes in genomic sequences, design human SpCas9 guide RNAs, and identify essential genes from genome-scale knockout screens.
- [Sequence search, alignment and phylogenetics tools you can run online](https://corollary-labs.com/run/sequence-alignment-tools): These tools process biological sequences, search databases, remove redundancy, build multiple sequence alignments, and infer evolutionary trees for nucleotide, protein, and structural data.
- [Sequencing read analysis tools you can run online](https://corollary-labs.com/run/ngs-analysis-tools): Tools across the primary sequencing analysis lifecycle, spanning raw read quality control, reference alignment, genomic arithmetic, variant and peak calling, and downstream transcriptomic quantification.

## AI models

- [ESMFold](https://corollary-labs.com/models/esmfold2): Predicts all-atom 3D protein structures directly from single amino acid sequences without alignments in seconds. (MIT, commercial use permitted; runs online)
- [Boltz-2](https://corollary-labs.com/models/boltz-2): Predicts 3D complex structures and binding affinities for protein-ligand pairs in a single forward pass. (MIT, commercial use permitted; runs online)
- [AlphaFold2](https://corollary-labs.com/models/alphafold2): Predicts all-atom 3D protein structures from sequence data using multiple sequence alignments. (Apache 2.0 (code), check licence before commercial use; runs online)
- [AlphaFold DB Lookup](https://corollary-labs.com/models/alphafold-db): Retrieve pre-computed AlphaFold protein structures and confidence scores across 247 million UniProt entries. (CC BY 4.0, commercial use permitted; runs online)
- [PDB Structure Fetch](https://corollary-labs.com/models/pdb-fetch): Retrieves experimentally determined protein coordinates and metadata for any PDB entry. (CC0, commercial use permitted; runs online)
- [ESM2-3B](https://corollary-labs.com/models/esm2-3b): A 3-billion-parameter protein language model generating 1,280-dimensional sequence embeddings. (MIT, commercial use permitted; runs online)
- [ESM2-650M](https://corollary-labs.com/models/esm2-650m): Generates 1,280-dimensional embeddings and masked residue predictions from protein sequences. (MIT, commercial use permitted; runs online)
- [ColabFold](https://corollary-labs.com/models/colabfold): MMseqs2-accelerated AlphaFold2 protein monomer and complex structure prediction pipeline. (MIT, commercial use permitted; listed, not yet runnable)
- [OpenFold](https://corollary-labs.com/models/openfold): Predicts three-dimensional protein structures from amino acid sequences using a retrained PyTorch implementation. (Apache-2.0, commercial use permitted; listed, not yet runnable)
- [Chai-1](https://corollary-labs.com/models/chai-1): Predicts 3D structures of multimodal biochemical complexes, including proteins, nucleic acids, and small molecules. (Chai Discovery Community License, check licence before commercial use; runs online)
- [RoseTTAFold All-Atom](https://corollary-labs.com/models/rosettafold-aa): Full-atom structure prediction for proteins, non-canonical residues, glycans, metals, and small-molecule cofactors. (MIT code / non-commercial weights, non-commercial use only; listed, not yet runnable)
- [OmegaFold](https://corollary-labs.com/models/omegafold): Predicts protein structures from single sequences without multiple sequence alignments, suited for orphan proteins. (Apache-2.0, commercial use permitted; listed, not yet runnable)
- [AlphaFlow / BioEmu](https://corollary-labs.com/models/alphaflow): Generates protein conformational ensembles directly from sequence for dynamic and disordered structures. (MIT, commercial use permitted; listed, not yet runnable)
- [RFDiffusion](https://corollary-labs.com/models/rfdiffusion): Generates de novo protein backbones and targeted binders using SE(3) denoising diffusion. (Non-commercial (BSD-derived), non-commercial use only; runs online)
- [ProteinMPNN](https://corollary-labs.com/models/proteinmpnn): Generates amino acid sequences predicted to fold into specified target protein backbone structures. (MIT, commercial use permitted; runs online)
- [BindCraft](https://corollary-labs.com/models/bindcraft): An automated pipeline designing target-specific protein binders via AlphaFold2 co-folding hallucination and MPNN. (MIT, commercial use permitted; listed, not yet runnable)
- [ESM-IF1](https://corollary-labs.com/models/esm-if1): Generates protein sequences for specified backbone structures using ESM2 language model representations. (MIT, commercial use permitted; runs online)
- [ProGen2](https://corollary-labs.com/models/progen2): Generates protein and antibody sequences conditionally using autoregressive language modelling. (BSD-3-Clause, commercial use permitted; runs online)
- [DPLM-2](https://corollary-labs.com/models/dplm-2): Generates protein sequences and structures simultaneously using a unified diffusion process. (Apache-2.0, commercial use permitted; listed, not yet runnable)
- [SaProt](https://corollary-labs.com/models/saprot): A structure-aware protein language model predicting variant effects and generating embeddings using 3Di tokens. (MIT, commercial use permitted; runs online)
- [EVOLVEpro](https://corollary-labs.com/models/evolvepro): Active learning model for directed evolution that proposes mutant sequences from small experimental batches. (MIT, commercial use permitted; listed, not yet runnable)
- [DiffDock](https://corollary-labs.com/models/diffdock): Diffusion-based generative model predicting 3D poses of small-molecule ligands bound to proteins. (MIT, commercial use permitted; runs online)
- [AutoDock Vina](https://corollary-labs.com/models/autodock-vina): Predicts small-molecule binding poses and affinity scores against protein structures. (Apache-2.0, commercial use permitted; runs online)
- [DiffDock-PP](https://corollary-labs.com/models/diffdock-pp): A diffusion-based model for docking protein–protein interfaces. (MIT, commercial use permitted; listed, not yet runnable)
- [Gnina](https://corollary-labs.com/models/gnina): Rescores protein–ligand docked poses using convolutional neural networks to improve virtual screening enrichment. (Apache-2.0, commercial use permitted; listed, not yet runnable)
- [SMINA](https://corollary-labs.com/models/smina): A molecular docking tool optimised for fragment-sized ligands and custom scoring functions. (Apache-2.0, commercial use permitted; runs online)
- [MM/GBSA](https://corollary-labs.com/models/mmgbsa): Estimates binding free energies to rank congeneric small-molecule series against protein targets. (GPL-3.0, check licence before commercial use; listed, not yet runnable)
- [OpenFE / Lomap](https://corollary-labs.com/models/openfe): Plans and executes alchemical free energy perturbation networks to calculate relative ligand binding affinities. (MIT, commercial use permitted; listed, not yet runnable)
- [AbLang2](https://corollary-labs.com/models/ablang2): An antibody language model for generating paired-chain sequence embeddings and guiding CDR design. (BSD-3-Clause, commercial use permitted; runs online)
- [AntiFold](https://corollary-labs.com/models/antifold): An antibody inverse folding model that designs CDR sequences tailored to antibody backbone frameworks. (BSD-3-Clause, commercial use permitted; runs online)
- [IgFold](https://corollary-labs.com/models/igfold): Predicts antibody 3D structures from heavy and light chains in sub-second runtimes. (JHU Academic License, non-commercial use only; listed, not yet runnable)
- [ImmuneBuilder](https://corollary-labs.com/models/immunebuilder): Fast structure prediction for antibodies, nanobodies, and T-cell receptors across full repertoires. (BSD-3-Clause, commercial use permitted; runs online)
- [ANARCI](https://corollary-labs.com/models/anarci): Numbers antibody sequences and annotates framework and CDR regions across multiple numbering schemes. (BSD-3-Clause, commercial use permitted; runs online)
- [Paragraph](https://corollary-labs.com/models/paragraph): Predicts antibody binding interface residues from structure to support humanisation and immunogenicity studies. (BSD-3-Clause, commercial use permitted; listed, not yet runnable)
- [Sapiens](https://corollary-labs.com/models/sapiens): Germline-guided antibody humanisation tool proposing mutations that increase humanness while preserving affinity. (BSD-3-Clause, commercial use permitted; runs online)
- [MAMMAL solubility](https://corollary-labs.com/models/proteinsol): Binary protein solubility classification directly from amino-acid sequences. (Apache-2.0, commercial use permitted; runs online)
- [Evo 2](https://corollary-labs.com/models/evo2): A genome-scale foundation model for nucleotide sequence generation, variant effect scoring, and genome modelling. (Arc Institute License, check licence before commercial use; runs online)
- [Nucleotide Transformer v2](https://corollary-labs.com/models/nucleotide-transformer): A DNA foundation model that generates nucleotide sequence representations for regulatory element and variant analysis. (CC-BY-NC-SA-4.0, non-commercial use only; runs online)
- [scGPT](https://corollary-labs.com/models/scgpt): A foundation model trained on 33 million human cells for single-cell annotation and perturbation prediction. (MIT, commercial use permitted; listed, not yet runnable)
- [Geneformer](https://corollary-labs.com/models/geneformer): A transformer model for gene network inference and target prioritisation from single-cell transcriptomes. (Apache-2.0, commercial use permitted; listed, not yet runnable)
- [CellFM](https://corollary-labs.com/models/cellfm): A 100-million-cell foundation model for pan-tissue single-cell annotation and cross-dataset integration. (Apache-2.0, commercial use permitted; listed, not yet runnable)
- [TimesFM 2.5](https://corollary-labs.com/models/timesfm): A decoder-only foundation model for zero-shot numerical time-series forecasting without task-specific training. (Apache-2.0, commercial use permitted; runs online)
- [Enformer](https://corollary-labs.com/models/enformer): Predicts gene expression and regulatory tracks across long genomic windows to assess variant effects. (Apache-2.0, commercial use permitted; runs online)
- [DNABERT-2](https://corollary-labs.com/models/dnabert-2): A byte-pair encoded transformer model generating nucleotide embeddings for genomic sequence classification. (Apache-2.0, commercial use permitted; runs online)
- [SCENIC+ / pySCENIC](https://corollary-labs.com/models/pyscenic): Infers enhancer-driven gene regulatory networks from paired single-cell ATAC-seq and RNA-seq data. (GPL-3.0, check licence before commercial use; listed, not yet runnable)
- [Molecular Property Calculator](https://corollary-labs.com/models/molecule-properties): Computes physicochemical descriptors and druglikeness rule compliance for small molecules from SMILES strings. (MIT, commercial use permitted; runs online)
- [MolFormer](https://corollary-labs.com/models/molformer): A chemical language transformer that extracts molecular embeddings from SMILES strings for property prediction. (Apache-2.0, commercial use permitted; runs online)
- [GROVER](https://corollary-labs.com/models/grover): A molecular graph transformer predicting ADMET endpoints from small-molecule SMILES strings. (MIT, commercial use permitted; listed, not yet runnable)
- [ADMET-AI](https://corollary-labs.com/models/admetlab3): Predicts 52 small-molecule ADMET endpoints simultaneously using Chemprop D-MPNN ensembles. (MIT, commercial use permitted; runs online)
- [Chemprop D-MPNN — toxicity panel](https://corollary-labs.com/models/dmpnn): Predicts small-molecule toxicity endpoints and structural alerts from chemical structures. (MIT, commercial use permitted; runs online)
- [REINVENT 4](https://corollary-labs.com/models/reinvent4): Generates and optimises small molecules against multi-objective property profiles using reinforcement learning. (Apache-2.0, commercial use permitted; listed, not yet runnable)
- [MolMIM](https://corollary-labs.com/models/molmim): MolMIM generates and optimises small molecules toward target properties using a smooth latent space. (Apache-2.0, commercial use permitted; runs online)
- [MegaMolBART](https://corollary-labs.com/models/megamolbart): Generates SMILES strings and predicts forward reaction products for chemical design and synthesis planning. (Apache-2.0, commercial use permitted; listed, not yet runnable)
- [Reaction prediction and fingerprints](https://corollary-labs.com/models/rxnfp): Predicts forward reactions and retrosynthetic routes for small molecules, and generates reaction fingerprints. (MIT, commercial use permitted; runs online)
- [MAMMAL](https://corollary-labs.com/models/mammal): A biomedical foundation model predicting binding affinity, blood–brain barrier permeability, and solubility. (Apache-2.0, check licence before commercial use; runs online)
- [BioVERSE](https://corollary-labs.com/models/bioverse): A multimodal model answering natural-language questions about proteins and small molecules from their embeddings. (Apache-2.0, check licence before commercial use; listed, not yet runnable)
- [BioMedGPT-R1](https://corollary-labs.com/models/biomedgpt-r1): A biomedical language model providing chain-of-thought reasoning over text, small molecules, and protein sequences. (Apache-2.0, commercial use permitted; runs online)
- [Galactica](https://corollary-labs.com/models/galactica): A scientific language model designed to study scientific tasks and serve as a baseline for domain knowledge. (CC-BY-NC-4.0, non-commercial use only; runs online)
- [QAMAS](https://corollary-labs.com/models/qamas): Simulate mitochondrial ATP synthesis, oxygen consumption and membrane potential. (MIT, commercial use permitted; runs online)
- [OpenMM](https://corollary-labs.com/models/openmm): GPU-native molecular dynamics simulation engine with a scriptable Python API for biomolecular systems. (MIT, commercial use permitted; runs online)
- [GROMACS](https://corollary-labs.com/models/gromacs): Performs molecular dynamics simulations of large biomolecular systems and membrane proteins. (LGPL-2.1, commercial use permitted; listed, not yet runnable)
- [AmberTools](https://corollary-labs.com/models/amber): A molecular simulation suite providing parameterisation, solvation, and free-energy evaluation for molecular systems. (GPL-3.0, check licence before commercial use; listed, not yet runnable)
- [AutoDock-GPU](https://corollary-labs.com/models/autodock-gpu): GPU-accelerated molecular docking tool for high-throughput virtual screening of compound libraries. (LGPL-2.1, commercial use permitted; listed, not yet runnable)
- [PaperQA2 (Crow)](https://corollary-labs.com/models/paperqa2): Grounded retrieval-augmented generation agent that answers scientific queries using cited literature. (Apache-2.0, commercial use permitted; runs online)
- [Sequence Analytics](https://corollary-labs.com/models/sequence-analysis): Calculates standard physicochemical properties and amino acid composition deterministically from protein sequences. (MIT, commercial use permitted; runs online)
- [Zero-To-CAD Qwen3-VL 2B](https://corollary-labs.com/models/adskailab-zero-to-cad-qwen3-vl-2b): Vision-language model generating executable CadQuery CAD programs from multi-view rendered images. (See model card, check licence before commercial use; runs online)
- [Kimina-Prover Distill 1.7B](https://corollary-labs.com/models/ai-mo-kimina-prover-distill-1-7b): A 1.7-billion-parameter language model that generates Lean 4 proofs for competition-level mathematical theorems. (See model card, check licence before commercial use; runs online)
- [Kimina-Prover Distill 8B](https://corollary-labs.com/models/ai-mo-kimina-prover-distill-8b): A distilled 8-billion-parameter language model for automated Lean 4 theorem proving in mathematics. (See model card, check licence before commercial use; runs online)
- [Kimina-Prover Preview Distill 7B](https://corollary-labs.com/models/ai-mo-kimina-prover-preview-distill-7b): A distilled 7B language model that generates formal mathematical proofs in Lean 4. (See model card, check licence before commercial use; runs online)
- [GENA-LM BERT base (T2T)](https://corollary-labs.com/models/airi-institute-gena-lm-bert-base-t2t): A BERT-base genomic foundation model trained on T2T human assemblies for masked DNA sequence understanding. (See model card, check licence before commercial use; runs online)
- [GENA-LM BERT large (T2T)](https://corollary-labs.com/models/airi-institute-gena-lm-bert-large-t2t): Genomic foundation model based on BERT large, trained on the T2T human genome assembly for sequence analysis. (See model card, check licence before commercial use; runs online)
- [ModernGENA base](https://corollary-labs.com/models/airi-institute-moderngena-base): A 377M-parameter ModernBERT DNA foundation model for genomic sequence modelling and downstream analysis tasks. (See model card, check licence before commercial use; runs online)
- [ModernGENA large](https://corollary-labs.com/models/airi-institute-moderngena-large): A 377M-parameter ModernBERT foundation model for vertebrate DNA sequence modelling and masked sequence prediction. (See model card, check licence before commercial use; runs online)
- [ESMC 300M](https://corollary-labs.com/models/biohub-esmc-300m): A 300M-parameter protein language model for sequence embedding, masked language modelling, and classification. (See model card, check licence before commercial use; runs online)
- [ESMC 600M](https://corollary-labs.com/models/biohub-esmc-600m): A 600M-parameter protein language model providing sequence embeddings and masked language modelling. (See model card, check licence before commercial use; runs online)
- [ESMC 6B](https://corollary-labs.com/models/biohub-esmc-6b): A 6-billion-parameter protein language model producing sequence representations and masked residue predictions. (See model card, check licence before commercial use; runs online)
- [ESMFold2](https://corollary-labs.com/models/biohub-esmfold2): Predicts all-atom 3D structures for proteins, nucleic acids, and ligands from sequence or sequence alignments. (See model card, check licence before commercial use; runs online)
- [ESMFold2-Experimental](https://corollary-labs.com/models/biohub-esmfold2-experimental): Predicts all-atom protein structures and confidence metrics from sequences for experimental evaluation. (See model card, check licence before commercial use; runs online)
- [ESMFold2-Experimental-Fast](https://corollary-labs.com/models/biohub-esmfold2-experimental-fast): A fast, experimental protein structure prediction model generating all-atom coordinates from amino acid sequences. (See model card, check licence before commercial use; runs online)
- [ESMFold2-Fast](https://corollary-labs.com/models/biohub-esmfold2-fast): Predicts all-atom 3D protein structures rapidly from single amino acid sequences for high-throughput screening. (See model card, check licence before commercial use; runs online)
- [DoctoBERT-fr-base](https://corollary-labs.com/models/doctolib-lab-doctobert-fr-base): A 111M-parameter French RoBERTa encoder pretrained for biomedical and clinical text processing. (See model card, check licence before commercial use; runs online)
- [DoctoModernBERT-fr-base](https://corollary-labs.com/models/doctolib-lab-doctomodernbert-fr-base): A 149M-parameter French biomedical encoder with an 8,192-token context window for clinical NLP and token classification. (See model card, check licence before commercial use; runs online)
- [AIFS Single 1.0](https://corollary-labs.com/models/ecmwf-aifs-single-1-0): A data-driven weather forecasting model by ECMWF for deterministic global predictions. (See model card, check licence before commercial use; runs online)
- [HuatuoGPT-Vision 7B](https://corollary-labs.com/models/freedomintelligence-huatuogpt-vision-7b): A multimodal medical model that answers clinical questions across radiology, pathology, and dermatology imagery. (See model card, check licence before commercial use; runs online)
- [ether0](https://corollary-labs.com/models/futurehouse-ether0): A 24-billion parameter language model that reasons in English to design and modify molecular structures as SMILES. (See model card, check licence before commercial use; runs online)
- [Carbon-3B](https://corollary-labs.com/models/huggingfacebio-carbon-3b): A 3-billion-parameter autoregressive DNA foundation model for sequence generation and variant-effect prediction. (See model card, check licence before commercial use; runs online)
- [Carbon-500M](https://corollary-labs.com/models/huggingfacebio-carbon-500m): A 500M-parameter autoregressive DNA language model designed primarily as a draft model for speculative decoding. (See model card, check licence before commercial use; runs online)
- [Carbon-8B](https://corollary-labs.com/models/huggingfacebio-carbon-8b): An 8B-parameter autoregressive DNA model for genomic prediction, generation, and long-context sequence retrieval. (See model card, check licence before commercial use; runs online)
- [BioCLIP 2](https://corollary-labs.com/models/imageomics-bioclip-2): Vision foundation model for zero-shot and few-shot species classification across the tree of life. (See model card, check licence before commercial use; runs online)
- [ChatNT](https://corollary-labs.com/models/instadeepai-chatnt): Conversational model answering natural-language queries for DNA, RNA, and protein sequence tasks. (See model card, check licence before commercial use; runs online)
- [Nucleotide Transformer 2.5B](https://corollary-labs.com/models/instadeepai-nucleotide-transformer-2-5b-multi-species): A 2.5B-parameter DNA language model trained across 850 species genomes for molecular phenotype prediction. (See model card, check licence before commercial use; runs online)
- [AIMNet2 (B97-3c, 2025)](https://corollary-labs.com/models/isayevlab-aimnet2-2025): A neural network interatomic potential calculating molecular energies, forces, and charges at the B97-3c level. (See model card, check licence before commercial use; runs online)
- [AIMNet2-NSE](https://corollary-labs.com/models/isayevlab-aimnet2-nse): A neural network interatomic potential for open-shell chemistry, radicals and non-singlet molecular states. (See model card, check licence before commercial use; runs online)
- [AIMNet2-Pd](https://corollary-labs.com/models/isayevlab-aimnet2-pd): A neural network interatomic potential predicting energies, forces, and charges for palladium-containing systems. (See model card, check licence before commercial use; runs online)
- [AIMNet2-rxn](https://corollary-labs.com/models/isayevlab-aimnet2-rxn): A neural-network interatomic potential for predicting energies, forces, and pathways in organic chemical reactions. (See model card, check licence before commercial use; runs online)
- [AIMNet2 ωB97M-D3](https://corollary-labs.com/models/isayevlab-aimnet2-wb97m-d3): Neural network interatomic potential trained at the ωB97M-D3 level of theory for fast molecular simulations. (See model card, check licence before commercial use; runs online)
- [MACE-MP-0](https://corollary-labs.com/models/mace-foundations-mace-mp-0): An equivariant message-passing potential trained on the Materials Project for inorganic crystal simulation. (See model card, check licence before commercial use; runs online)
- [MACE-MPA-0](https://corollary-labs.com/models/mace-foundations-mace-mpa-0): MACE foundation model predicting energy, forces, and charges for inorganic materials and molecules. (See model card, check licence before commercial use; runs online)
- [MarinDNA m5.1 1B base model](https://corollary-labs.com/models/marin-dna-marin-dna-exp135-m5-1): A 1.12-billion-parameter causal language model for DNA sequence generation and variant effect prediction. (See model card, check licence before commercial use; runs online)
- [MIST 1.8B · G298](https://corollary-labs.com/models/mist-models-mist-1-8b-09sntn03-g298): Predicts molecular Gibbs free energy at 298.15 K from SMILES strings using a 1.8-billion-parameter transformer. (See model card, check licence before commercial use; runs online)
- [MIST 1.8B · HIV](https://corollary-labs.com/models/mist-models-mist-1-8b-1a4puhg2-hiv): A 1.8-billion-parameter molecular foundation model predicting HIV-1 replication inhibition from SMILES strings. (See model card, check licence before commercial use; runs online)
- [MIST 1.8B · H298](https://corollary-labs.com/models/mist-models-mist-1-8b-3fbbz4is-h298): A 1.8-billion parameter transformer predicting molecular enthalpy at 298.15 K from Kekulized SMILES strings. (See model card, check licence before commercial use; runs online)
- [MIST 1.8B · ZPVE](https://corollary-labs.com/models/mist-models-mist-1-8b-6nmcwyrp-zpve): A 1.8-billion-parameter molecular transformer predicting zero-point vibrational energy from SMILES strings. (See model card, check licence before commercial use; runs online)
- [MIST 1.8B · U298](https://corollary-labs.com/models/mist-models-mist-1-8b-85f24xkj-u298): Predicts molecular internal energy at 298.15 K from SMILES representations using a 1.8-billion-parameter transformer. (See model card, check licence before commercial use; runs online)
- [MIST 1.8B · QM8](https://corollary-labs.com/models/mist-models-mist-1-8b-8nd1ot5j-qm8): A 1.8-billion-parameter molecular foundation model fine-tuned to predict electronic spectra properties from SMILES strings. (See model card, check licence before commercial use; runs online)
- [MIST 1.8B · U0](https://corollary-labs.com/models/mist-models-mist-1-8b-a7akimjj-u0): A 1.8-billion-parameter transformer predicting molecular internal energy at 0 K from SMILES strings. (See model card, check licence before commercial use; runs online)
- [MIST 1.8B base](https://corollary-labs.com/models/mist-models-mist-1-8b-dh61satt): A 1.8-billion-parameter foundation model using masked language modelling on molecular SMILES for chemistry prediction. (See model card, check licence before commercial use; runs online)
- [MIST 1.8B · μ (dipole)](https://corollary-labs.com/models/mist-models-mist-1-8b-ez05expv-mu): A 1.8-billion-parameter foundation model predicting molecular dipole moments directly from SMILES strings. (See model card, check licence before commercial use; runs online)
- [MIST 1.8B · BBBP](https://corollary-labs.com/models/mist-models-mist-1-8b-fbdn8e35-bbbp): A molecular foundation model predicting blood-brain barrier penetration from SMILES strings. (See model card, check licence before commercial use; runs online)
- [MIST 1.8B · ESOL](https://corollary-labs.com/models/mist-models-mist-1-8b-hxiygjsm-esol): A molecular foundation model fine-tuned to predict aqueous solubility from chemical SMILES strings. (See model card, check licence before commercial use; runs online)
- [MIST 1.8B · FreeSolv](https://corollary-labs.com/models/mist-models-mist-1-8b-iwqj2cld-freesolv): A molecular foundation model predicting hydration free energy for neutral molecules from SMILES representations. (See model card, check licence before commercial use; runs online)
- [MIST 1.8B · Cv](https://corollary-labs.com/models/mist-models-mist-1-8b-j356b3nf-cv): Predicts constant-volume heat capacity from molecular SMILES strings for thermodynamic and chemical screening. (See model card, check licence before commercial use; runs online)
- [MIST 1.8B · HOMO](https://corollary-labs.com/models/mist-models-mist-1-8b-jmjosq12-homo): A molecular foundation model predicting highest occupied molecular orbital energy directly from chemical SMILES. (See model card, check licence before commercial use; runs online)
- [MIST 1.8B · Lipo](https://corollary-labs.com/models/mist-models-mist-1-8b-jvt4azpz-lipo): Predicts octanol-water distribution coefficients (logD) from molecular SMILES strings using a 1.8B transformer. (See model card, check licence before commercial use; runs online)
- [MIST 1.8B · HOMO-LUMO gap](https://corollary-labs.com/models/mist-models-mist-1-8b-kayun6v3-gap): A 1.8-billion-parameter molecular transformer predicting the HOMO-LUMO gap from Kekulized SMILES strings. (See model card, check licence before commercial use; runs online)
- [MIST 1.8B · SIDER](https://corollary-labs.com/models/mist-models-mist-1-8b-l1wfo7oa-sider): Predicts 27 MedDRA adverse drug reactions from molecular SMILES strings using a 1.8-billion-parameter encoder. (See model card, check licence before commercial use; runs online)
- [MIST 1.8B · ClinTox](https://corollary-labs.com/models/mist-models-mist-1-8b-lu1l5ieh-clintox): Predicts clinical trial toxicity and FDA approval status of chemical compounds from SMILES strings. (See model card, check licence before commercial use; runs online)
- [MIST 1.8B · BACE](https://corollary-labs.com/models/mist-models-mist-1-8b-m50jgolp-bace): A 1.8-billion-parameter model predicting small-molecule inhibition of Alzheimer's target BACE-1 from SMILES strings. (See model card, check licence before commercial use; runs online)
- [MIST 1.8B · LUMO](https://corollary-labs.com/models/mist-models-mist-1-8b-n14wshc9-lumo): A 1.8-billion-parameter model predicting molecular LUMO energy from SMILES strings for chemical property discovery. (See model card, check licence before commercial use; runs online)
- [MIST 1.8B · α (polarizability)](https://corollary-labs.com/models/mist-models-mist-1-8b-rcwary93-alpha): A 1.8-billion parameter molecular transformer predicting isotropic polarizability directly from chemical SMILES. (See model card, check licence before commercial use; runs online)
- [MIST 1.8B · Tox21](https://corollary-labs.com/models/mist-models-mist-1-8b-uop1z0dc-tox21): A 1.8-billion-parameter model predicting 12 toxicity endpoints for small molecules from SMILES strings. (See model card, check licence before commercial use; runs online)
- [MIST 1.8B · ⟨R²⟩](https://corollary-labs.com/models/mist-models-mist-1-8b-xxe7t35e-r2): Predicts molecular electronic spatial extent from SMILES strings using a 1.8-billion-parameter transformer encoder. (See model card, check licence before commercial use; runs online)
- [MIST 26.9M · odour](https://corollary-labs.com/models/mist-models-mist-26-9m-48kpooqf-odour): A molecular transformer predicting 135 scent descriptors from chemical SMILES strings. (See model card, check licence before commercial use; runs online)
- [MIST 26.9M · dn](https://corollary-labs.com/models/mist-models-mist-26-9m-6hk5coof-dn): A molecular language model fine-tuned to predict Lewis basicity and boron trifluoride affinity from SMILES strings. (See model card, check licence before commercial use; runs online)
- [MIST 26.9M · boiling point](https://corollary-labs.com/models/mist-models-mist-26-9m-b302p09x-bp): Predicts molecular normal boiling points in degrees Celsius directly from SMILES strings. (See model card, check licence before commercial use; runs online)
- [MIST 26.9M · flash point](https://corollary-labs.com/models/mist-models-mist-26-9m-cyuo2xb6-fp): A molecular transformer predicting flash-point temperature from Kekulized SMILES strings. (See model card, check licence before commercial use; runs online)
- [MIST 26.9M · melting point](https://corollary-labs.com/models/mist-models-mist-26-9m-y3ge5pf9-mp): Predicts molecular melting points in degrees Celsius from SMILES representations using a 26.9M-parameter encoder. (See model card, check licence before commercial use; runs online)
- [MIST 27.1M · ETN](https://corollary-labs.com/models/mist-models-mist-27-1m-1gcxtg8y-etn): A molecular transformer predicting the normalized Dimroth-Reichardt solvent polarity parameter from SMILES strings. (See model card, check licence before commercial use; runs online)
- [MIST 28M · FreeSolv](https://corollary-labs.com/models/mist-models-mist-28m-0uiq7o7m-freesolv): A 28-million-parameter transformer predicting hydration free energies of small neutral molecules in water. (See model card, check licence before commercial use; runs online)
- [MIST 28M · BBBP](https://corollary-labs.com/models/mist-models-mist-28m-3xpfhv48-bbbp): A molecular transformer predicting blood-brain barrier permeability from SMILES strings for CNS drug screening. (See model card, check licence before commercial use; runs online)
- [MIST 28M · pKa](https://corollary-labs.com/models/mist-models-mist-28m-6zlgl2qn-pka): Predicts acid dissociation constants (pKa) in DMSO for small molecules from Kekulized SMILES strings. (See model card, check licence before commercial use; runs online)
- [MIST 28M · HIV](https://corollary-labs.com/models/mist-models-mist-28m-8fh43gke-hiv): A molecular transformer predicting HIV-1 replication inhibition in T-cell lines from SMILES strings. (See model card, check licence before commercial use; runs online)
- [MIST 28M · BACE](https://corollary-labs.com/models/mist-models-mist-28m-8loj3bab-bace): Predicts beta-secretase 1 inhibition from molecular SMILES strings for Alzheimer's disease drug discovery. (See model card, check licence before commercial use; runs online)
- [MIST 28M · ClinTox](https://corollary-labs.com/models/mist-models-mist-28m-97vfcykk-clintox): Predicts clinical toxicity and FDA approval status from molecular SMILES strings for drug candidates. (See model card, check licence before commercial use; runs online)
- [MIST 28M · tmQM](https://corollary-labs.com/models/mist-models-mist-28m-ggd8iisr-tmqm): A molecular transformer predicting quantum-mechanical properties of transition-metal complexes from SMILES strings. (See model card, check licence before commercial use; runs online)
- [MIST 28M · QM8](https://corollary-labs.com/models/mist-models-mist-28m-gzwqzpcr-qm8): Predicts electronic spectra properties for small organic molecules from SMILES strings using a 28M transformer. (See model card, check licence before commercial use; runs online)
- [MIST 28M · ESOL](https://corollary-labs.com/models/mist-models-mist-28m-kcwb9le5-esol): A 28M-parameter transformer predicting aqueous solubility (log S) of small molecules from SMILES strings. (See model card, check licence before commercial use; runs online)
- [MIST 28M · Tox21](https://corollary-labs.com/models/mist-models-mist-28m-kw4ks27p-tox21): A 28-million parameter transformer predicting 12 toxicity endpoints from molecular SMILES strings. (See model card, check licence before commercial use; runs online)
- [MIST 28M · solvent properties](https://corollary-labs.com/models/mist-models-mist-28m-solvent-properties): A 28-million-parameter transformer predicting physical and solvatochromic descriptors of organic solvents from SMILES. (See model card, check licence before commercial use; runs online)
- [MIST 28M base](https://corollary-labs.com/models/mist-models-mist-28m-ti624ev1): A 28M-parameter molecular language model pretrained on SMILES strings for molecular property prediction. (See model card, check licence before commercial use; runs online)
- [MIST 28M · ToxCast](https://corollary-labs.com/models/mist-models-mist-28m-ttqcvt6fs-toxcast): A 28-million parameter transformer predicting 617 toxicity assay endpoints directly from chemical SMILES strings. (See model card, check licence before commercial use; runs online)
- [MIST 28M · Lipo](https://corollary-labs.com/models/mist-models-mist-28m-xzr5ulva-lipo): Predicts the octanol-water distribution coefficient (logD at pH 7.4) of small molecules from SMILES strings. (See model card, check licence before commercial use; runs online)
- [MIST 28M · MUV](https://corollary-labs.com/models/mist-models-mist-28m-yr1urd2c-muv): Predicts 17 bioassay activity endpoints from SMILES strings for unbiased virtual screening. (See model card, check licence before commercial use; runs online)
- [MIST 28M · SIDER](https://corollary-labs.com/models/mist-models-mist-28m-z8qo16uy-sider): A molecular transformer predicting 27 adverse drug reaction classes from SMILES strings. (See model card, check licence before commercial use; runs online)
- [MIST 27.0M · conductivity](https://corollary-labs.com/models/mist-models-mist-conductivity-27-0m-2mpg8dcd): A molecular transformer that predicts ionic conductivity and thermodynamic parameters for liquid electrolyte mixtures. (See model card, check licence before commercial use; runs online)
- [MIST mixtures](https://corollary-labs.com/models/mist-models-mist-mixtures-zffffbex): Predicts thermodynamic excess properties and mixing behaviour of binary chemical mixtures from molecular SMILES. (See model card, check licence before commercial use; runs online)
- [Indus SDE v0.2](https://corollary-labs.com/models/nasa-impact-indus-sde-v0-2): A 125M-parameter masked language model for extracting scientific concepts from Earth and atmospheric science texts. (See model card, check licence before commercial use; runs online)
- [NASA-SMD-IBM](https://corollary-labs.com/models/nasa-impact-nasa-smd-ibm-v0-1): RoBERTa-based encoder model domain-adapted for NASA Science Mission Directorate earth and space science literature. (See model card, check licence before commercial use; runs online)
- [OpenMed AnatomyDetect](https://corollary-labs.com/models/openmed-openmed-ner-anatomydetect-electramed-109m): A 109M-parameter transformer for detecting anatomical structures, body parts, and tissues in biomedical text. (See model card, check licence before commercial use; runs online)
- [OpenMed BloodCancerDetect](https://corollary-labs.com/models/openmed-openmed-ner-bloodcancerdetect-tinymed-65m): A compact 65M-parameter model that identifies chronic lymphocytic leukaemia clinical entities in biomedical text. (See model card, check licence before commercial use; runs online)
- [OpenMed ChemicalDetect](https://corollary-labs.com/models/openmed-openmed-ner-chemicaldetect-modernmed-149m): A 149M-parameter transformer for detecting chemical compounds and drug names in biomedical and clinical text. (See model card, check licence before commercial use; runs online)
- [OpenMed DiseaseDetect](https://corollary-labs.com/models/openmed-openmed-ner-diseasedetect-biomed-335m): A biomedical transformer for detecting disease names and medical conditions in scientific and clinical text. (See model card, check licence before commercial use; runs online)
- [OpenMed DNADetect](https://corollary-labs.com/models/openmed-openmed-ner-dnadetect-supermedical-125m): A 125M-parameter RoBERTa model for extracting DNA, RNA, protein, cell line, and cell type entities from biomedical text. (See model card, check licence before commercial use; runs online)
- [OpenMed GenomeDetect](https://corollary-labs.com/models/openmed-openmed-ner-genomedetect-modernmed-149m): A 149M-parameter ModernBERT model for extracting gene and protein entity mentions from English biomedical text. (See model card, check licence before commercial use; runs online)
- [OpenMed GenomicDetect](https://corollary-labs.com/models/openmed-openmed-ner-genomicdetect-pubmed-335m): Identifies and extracts genomic and cell-line entity mentions from biomedical literature and clinical texts. (See model card, check licence before commercial use; runs online)
- [OpenMed OncologyDetect](https://corollary-labs.com/models/openmed-openmed-ner-oncologydetect-multimed-568m): A biomedical transformer for named entity recognition of cancer genetics and anatomical entities in text. (See model card, check licence before commercial use; runs online)
- [OpenMed OrganismDetect](https://corollary-labs.com/models/openmed-openmed-ner-organismdetect-tinymed-82m): A compact biomedical transformer model for detecting species and organism mentions in scientific literature. (See model card, check licence before commercial use; runs online)
- [OpenMed PathologyDetect](https://corollary-labs.com/models/openmed-openmed-ner-pathologydetect-tinymed-135m): A DistilBERT model for extracting disease entities from clinical text and biomedical literature. (See model card, check licence before commercial use; runs online)
- [OpenMed PharmaDetect](https://corollary-labs.com/models/openmed-openmed-ner-pharmadetect-superclinical-434m): A transformer model for identifying and extracting chemical and drug entities from biomedical and clinical text. (See model card, check licence before commercial use; runs online)
- [OpenMed ProteinDetect](https://corollary-labs.com/models/openmed-openmed-ner-proteindetect-superclinical-141m): A 141M-parameter DeBERTa model that identifies proteins, complexes, and variants in biomedical literature and clinical text. (See model card, check licence before commercial use; runs online)
- [OpenMed SpeciesDetect](https://corollary-labs.com/models/openmed-openmed-ner-speciesdetect-electramed-109m): A 109M-parameter transformer for detecting species and organism names in biomedical literature and clinical texts. (See model card, check licence before commercial use; runs online)
- [p-IgGen](https://corollary-labs.com/models/opig-p-iggen): An autoregressive protein language model for designing diverse paired antibody heavy and light chain sequences. (See model card, check licence before commercial use; runs online)
- [OpenPhenom](https://corollary-labs.com/models/recursionpharma-openphenom): Generates phenomic embeddings from multichannel cell microscopy images for drug discovery and perturbation analysis. (See model card, check licence before commercial use; runs online)
- [FlashPPI](https://corollary-labs.com/models/tattabio-flashppi): A dense retrieval model for rapid, proteome-wide microbial protein-protein interaction screening. (See model card, check licence before commercial use; runs online)
- [gLM2 650M](https://corollary-labs.com/models/tattabio-glm2-650m): A 650M-parameter genomic foundation model encoding contigs with both amino acid and nucleotide tokens. (See model card, check licence before commercial use; runs online)
- [AstroLLaMA](https://corollary-labs.com/models/universetbd-astrollama): A specialized 7-billion-parameter language model for astronomy and astrophysics literature text generation. (See model card, check licence before commercial use; runs online)
- [BioReason-Pro RL](https://corollary-labs.com/models/wanglab-bioreason-pro-rl): A reinforcement-learning-tuned language model for biological reasoning and protein function prediction. (See model card, check licence before commercial use; runs online)
- [BioReason-Pro SFT](https://corollary-labs.com/models/wanglab-bioreason-pro-sft): Multimodal reasoning model for protein function prediction with structured chain-of-thought rationales. (See model card, check licence before commercial use; runs online)
- [Genos 1.2B](https://corollary-labs.com/models/zhejianglab-genos-1-2b): A 1.2-billion-parameter genomic foundation model for high-efficiency sequence analysis and prediction tasks. (See model card, check licence before commercial use; runs online)
- [OneGenome-Rice](https://corollary-labs.com/models/zhejianglab-onegenome-rice): A genomic foundation model analysing long DNA sequences for rice functional genomics and precision breeding. (See model card, check licence before commercial use; runs online)
- [Make-A-Shape · multi-view 20M](https://corollary-labs.com/models/ADSKAILab/Make-A-Shape-multi-view-20m): Generates detailed 3D geometric shapes from four multi-view images for engineering and shape modelling research. (listed, not yet runnable)
- [Make-A-Shape · point-cloud 20M](https://corollary-labs.com/models/ADSKAILab/Make-A-Shape-point-cloud-20m): Generates detailed 3D geometric shapes from point cloud inputs for engineering and computational research. (listed, not yet runnable)
- [Make-A-Shape · single-view 20M](https://corollary-labs.com/models/ADSKAILab/Make-A-Shape-single-view-20m): Generates detailed 3D geometric shapes from single 2D images in two seconds for engineering research. (listed, not yet runnable)
- [Make-A-Shape · voxel 16³](https://corollary-labs.com/models/ADSKAILab/Make-A-Shape-voxel-16res-20m): Generates detailed 3D geometries from coarse 16³ voxel inputs for engineering and materials prototyping. (listed, not yet runnable)
- [Make-A-Shape · voxel 32³](https://corollary-labs.com/models/ADSKAILab/Make-A-Shape-voxel-32res-20m): Generates detailed 3D shapes from 32³ voxel input representations for scientific and engineering research. (listed, not yet runnable)
- [WaLa DM1 1B](https://corollary-labs.com/models/ADSKAILab/WaLa-DM1-1B): Generates detailed 3D shapes from a single-view depth map in four seconds using wavelet latent diffusion. (listed, not yet runnable)
- [WaLa DM4 1B](https://corollary-labs.com/models/ADSKAILab/WaLa-DM4-1B): Generates 3D shapes from four-view depth maps in four seconds using wavelet latent diffusion. (listed, not yet runnable)
- [WaLa DM6 1B](https://corollary-labs.com/models/ADSKAILab/WaLa-DM6-1B): Generates 3D geometric shapes from six-view depth maps for engineering and materials research. (listed, not yet runnable)
- [WaLa MVDream DM6](https://corollary-labs.com/models/ADSKAILab/WaLa-MVDream-DM6): Generates six-view depth maps from text prompts as an intermediate representation for 3D shape generation. (listed, not yet runnable)
- [WaLa MVDream RGB4](https://corollary-labs.com/models/ADSKAILab/WaLa-MVDream-RGB4): Text-to-multiview diffusion model generating four consistent RGB views from text prompts for 3D shape synthesis. (listed, not yet runnable)
- [WaLa PC 1B](https://corollary-labs.com/models/ADSKAILab/WaLa-PC-1B): Generates complete 3D shapes from point cloud inputs using wavelet-based latent diffusion in 2 to 4 seconds. (listed, not yet runnable)
- [WaLa RGB4 1B](https://corollary-labs.com/models/ADSKAILab/WaLa-RGB4-1B): A 1-billion-parameter diffusion model that generates 3D shapes from four-view RGB images in 2 to 4 seconds. (listed, not yet runnable)
- [WaLa SK 1B](https://corollary-labs.com/models/ADSKAILab/WaLa-SK-1B): A billion-parameter 3D generative model that synthesises detailed 3D shapes from 2D sketch inputs. (listed, not yet runnable)
- [WaLa SV 1B](https://corollary-labs.com/models/ADSKAILab/WaLa-SV-1B): A 1-billion-parameter model generating detailed 3D shapes from single-view images using wavelet latent representations. (listed, not yet runnable)
- [WaLa UN 1B](https://corollary-labs.com/models/ADSKAILab/WaLa-UN-1B): A billion-parameter 3D diffusion model generating detailed shapes unconditionally from noise for design exploration. (listed, not yet runnable)
- [WaLa VX16 1B](https://corollary-labs.com/models/ADSKAILab/WaLa-VX16-1B): A 1-billion-parameter 3D generative model that reconstructs detailed shapes from 16³ voxel grids in 2–4 seconds. (listed, not yet runnable)
- [NexaMass V3 Struct](https://corollary-labs.com/models/AethronPhantom/NexaMass-V3-Struct): A compact tandem mass spectrometry encoder that predicts molecular fingerprints to narrow candidate structures. (listed, not yet runnable)
- [RudolfV 2](https://corollary-labs.com/models/Aignostics/rudolfv-2): A 1.1-billion-parameter pathology foundation model for image feature extraction in digital histopathology. (listed, not yet runnable)
- [ACE2-ERA5](https://corollary-labs.com/models/allenai/ACE2-ERA5): A global atmospheric emulator for subseasonal to decadal climate variability and multi-year projections. (listed, not yet runnable)
- [HiRO-ACE](https://corollary-labs.com/models/allenai/HiRO-ACE): Climate simulation framework generating 3 km precipitation fields from 100 km inputs using diffusion downscaling. (listed, not yet runnable)
- [OlmoEarth v1.1 Base](https://corollary-labs.com/models/allenai/OlmoEarth-v1_1-Base): A remote sensing foundation model for analysing Sentinel-1, Sentinel-2, and Landsat satellite imagery. (listed, not yet runnable)
- [OlmoEarth v1.1 Nano](https://corollary-labs.com/models/allenai/OlmoEarth-v1_1-Nano): A 5.5-million-parameter vision transformer for remote sensing analysis on satellite imagery and time series. (listed, not yet runnable)
- [OlmoEarth v1.1 Tiny](https://corollary-labs.com/models/allenai/OlmoEarth-v1_1-Tiny): A 12M-parameter foundation model for remote-sensing tasks across Sentinel-1, Sentinel-2, and Landsat imagery. (listed, not yet runnable)
- [OlmoEarth v1.2 Base](https://corollary-labs.com/models/allenai/OlmoEarth-v1_2-Base): An Earth-observation foundation model for analysing Sentinel-1, Sentinel-2, and Landsat satellite imagery. (listed, not yet runnable)
- [OlmoEarth v1.2 Nano](https://corollary-labs.com/models/allenai/OlmoEarth-v1_2-Nano): A compact remote-sensing foundation model extracting features from Sentinel and Landsat satellite imagery. (listed, not yet runnable)
- [OlmoEarth v1.2 Small](https://corollary-labs.com/models/allenai/OlmoEarth-v1_2-Small): A 36M-parameter vision transformer extracting representations from satellite imagery for Earth and climate science. (listed, not yet runnable)
- [OlmoEarth v1.2 Tiny](https://corollary-labs.com/models/allenai/OlmoEarth-v1_2-Tiny): A compact 12-million-parameter vision model for Earth observation and remote-sensing analysis. (listed, not yet runnable)
- [OlmoEarth v1 Base](https://corollary-labs.com/models/allenai/OlmoEarth-v1-Base): A remote sensing foundation model for analysing Sentinel-1, Sentinel-2, and Landsat imagery and time series. (listed, not yet runnable)
- [OlmoEarth v1 FT — AWF (Base)](https://corollary-labs.com/models/allenai/OlmoEarth-v1-FT-AWF-Base): A fine-tuned Earth-observation model mapping land use and land cover in southern Kenya for conservation. (listed, not yet runnable)
- [OlmoEarth v1 FT — Ecosystem Type Mapping (Base)](https://corollary-labs.com/models/allenai/OlmoEarth-v1-FT-EcosystemTypeMapping-Base): Predicts ecosystem categories from satellite imagery across North Africa for biodiversity and conservation workflows. (listed, not yet runnable)
- [OlmoEarth v1 FT — Forest Loss Driver (Base)](https://corollary-labs.com/models/allenai/OlmoEarth-v1-FT-ForestLossDriver-Base): Identifies causes of satellite-detected forest loss, such as agriculture, logging, and fire, for Earth scientists. (listed, not yet runnable)
- [OlmoEarth v1 FT — LFMC (Base)](https://corollary-labs.com/models/allenai/OlmoEarth-v1-FT-LFMC-Base): Predicts live fuel moisture content in woody vegetation to support wildfire-risk monitoring from satellite imagery. (listed, not yet runnable)
- [OlmoEarth v1 FT — Mangrove (Base)](https://corollary-labs.com/models/allenai/OlmoEarth-v1-FT-Mangrove-Base): An Earth observation model fine-tuned to map mangrove extent for coastal conservation and habitat monitoring. (listed, not yet runnable)
- [OlmoEarth v1 Large](https://corollary-labs.com/models/allenai/OlmoEarth-v1-Large): A 308M-parameter Vision Transformer foundation model for satellite imagery and Earth observation tasks. (listed, not yet runnable)
- [OlmoEarth v1 Nano](https://corollary-labs.com/models/allenai/OlmoEarth-v1-Nano): A lightweight vision transformer for remote sensing analysis across Sentinel and Landsat satellite imagery. (listed, not yet runnable)
- [OlmoEarth v1 Tiny](https://corollary-labs.com/models/allenai/OlmoEarth-v1-Tiny): A 6.2M-parameter vision transformer foundation model for analysing satellite imagery and image time series. (listed, not yet runnable)
- [SamudrACE-E3SMv3](https://corollary-labs.com/models/allenai/SamudrACE-E3SMv3): A stochastic coupled atmosphere–ocean emulator for simulating long-term preindustrial climate variability. (listed, not yet runnable)
- [Mosaic](https://corollary-labs.com/models/AMLab-UvA/mosaic): A 214M-parameter transformer for fast, probabilistic global weather forecasting with preserved spectral fidelity. (listed, not yet runnable)
- [SE-600M](https://corollary-labs.com/models/arcinstitute/SE-600M): Generates embeddings from human single-cell RNA profiles to aid cell-state and perturbation analysis. (listed, not yet runnable)
- [STACK Large](https://corollary-labs.com/models/arcinstitute/Stack-Large): A single-cell transcriptomics model for in-context learning across cell types and biological perturbations. (listed, not yet runnable)
- [Stack-Large Aligned](https://corollary-labs.com/models/arcinstitute/Stack-Large-Aligned): A 217M-parameter single-cell foundation model aligned for cross-batch consistency and perturbation analysis. (listed, not yet runnable)
- [OpenDDE](https://corollary-labs.com/models/aurekaresearch/OpenDDE): An all-atom biomolecular foundation model using co-folding for structure prediction and drug discovery. (listed, not yet runnable)
- [ESMC 300M (Dec 2024)](https://corollary-labs.com/models/biohub/esmc-300m-2024-12): A 300-million-parameter protein language model for generating protein embeddings in biological workflows. (listed, not yet runnable)
- [ESMC 600M (Dec 2024)](https://corollary-labs.com/models/biohub/esmc-600m-2024-12): A 600-million-parameter protein language model checkpoint for protein engineering and variant-effect analysis. (listed, not yet runnable)
- [Promera](https://corollary-labs.com/models/bjing-mit/promera): Generative model for biomolecular co-folding, binder filtering, and miniprotein and nanobody design. (listed, not yet runnable)
- [PeptiVerse](https://corollary-labs.com/models/ChatterjeeLab/PeptiVerse): A suite of machine learning models predicting biophysical and therapeutic properties of natural and modified peptides. (listed, not yet runnable)
- [AVES2-BEATs](https://corollary-labs.com/models/EarthSpeciesProject/esp-aves2-sl-beats-all): A bioacoustic audio encoder that produces transferable representations for wildlife monitoring and vocal analysis. (listed, not yet runnable)
- [NatureLM-audio](https://corollary-labs.com/models/EarthSpeciesProject/NatureLM-audio): Audio-language foundation model for species identification, vocalisation detection, and bioacoustic captioning. (listed, not yet runnable)
- [AIFS ENS 1.0](https://corollary-labs.com/models/ecmwf/aifs-ens-1.0): A probabilistic data-driven weather forecasting model for upper-air, surface variables and cyclone tracks. (listed, not yet runnable)
- [AIFS ENS 2.0](https://corollary-labs.com/models/ecmwf/aifs-ens-2.0): A data-driven medium-range probabilistic weather forecasting model generating global ensemble forecasts. (listed, not yet runnable)
- [AIFS Single 0.2.1](https://corollary-labs.com/models/ecmwf/aifs-single-0.2.1): A data-driven medium-range global weather forecasting model developed by ECMWF. (listed, not yet runnable)
- [AIFS Single 1.1](https://corollary-labs.com/models/ecmwf/aifs-single-1.1): A data-driven medium-range weather forecasting model producing 6-hour interval atmospheric and surface predictions. (listed, not yet runnable)
- [AIFS Single 2.0](https://corollary-labs.com/models/ecmwf/aifs-single-2.0): A data-driven medium-range weather model by ECMWF providing deterministic 15-day global atmospheric and wave forecasts. (listed, not yet runnable)
- [OC25](https://corollary-labs.com/models/facebook/OC25): Catalysis model simulating explicit-solvent and catalyst systems for electrochemistry and energy applications. (listed, not yet runnable)
- [OMat24](https://corollary-labs.com/models/facebook/OMAT24): Inorganic material property prediction models using EquiformerV2 and eSEN architectures. (listed, not yet runnable)
- [OMC25](https://corollary-labs.com/models/facebook/OMC25): Predicts properties of organic molecular crystals for chemistry and materials science researchers. (listed, not yet runnable)
- [OMol25](https://corollary-labs.com/models/facebook/OMol25): Predicts molecular properties and energies for chemistry and materials science researchers. (listed, not yet runnable)
- [UMA](https://corollary-labs.com/models/facebook/UMA): A family of universal atomic graph neural networks for chemistry and materials science simulations. (listed, not yet runnable)
- [AlphaGenome](https://corollary-labs.com/models/google/alphagenome-all-folds): Predicts DNA regulatory features, such as gene expression and TF binding, at single-nucleotide resolution. (listed, not yet runnable)
- [MedASR](https://corollary-labs.com/models/google/medasr): A 105-million-parameter speech recognition model fine-tuned on clinical dictation for medical transcription. (listed, not yet runnable)
- [MedGemma 1.5 4B](https://corollary-labs.com/models/google/medgemma-1.5-4b-it): A 4-billion-parameter multimodal model interpreting medical imaging and clinical text for biomedical researchers. (listed, not yet runnable)
- [MedGemma 27B](https://corollary-labs.com/models/google/medgemma-27b-it): An instruction-tuned 27B medical AI for clinical question answering, health records, and medical image interpretation. (listed, not yet runnable)
- [MedSigLIP](https://corollary-labs.com/models/google/medsiglip-448): A dual-encoder vision-language model mapping medical images and text into a shared embedding space for healthcare AI. (listed, not yet runnable)
- [Path Foundation](https://corollary-labs.com/models/google/path-foundation): Vision transformer developed by Google to extract feature embeddings from histopathology image patches. (listed, not yet runnable)
- [TxGemma 27B Chat](https://corollary-labs.com/models/google/txgemma-27b-chat): A 27-billion-parameter conversational language model for therapeutic discovery and drug property prediction. (listed, not yet runnable)
- [TxGemma 27B Predict](https://corollary-labs.com/models/google/txgemma-27b-predict): A 27B language model predicting drug, target, and molecular properties across diverse therapeutic modalities. (listed, not yet runnable)
- [TxGemma 2B](https://corollary-labs.com/models/google/txgemma-2b-predict): A 2-billion-parameter language model fine-tuned for therapeutic property prediction across drug discovery tasks. (listed, not yet runnable)
- [TxGemma 9B Chat](https://corollary-labs.com/models/google/txgemma-9b-chat): A conversational 9B-parameter language model for drug discovery and therapeutic property prediction with reasoning. (listed, not yet runnable)
- [TxGemma 9B Predict](https://corollary-labs.com/models/google/txgemma-9b-predict): A 9B-parameter language model fine-tuned on Therapeutics Data Commons for therapeutic property prediction. (listed, not yet runnable)
- [NTv3 650M](https://corollary-labs.com/models/InstaDeepAI/NTv3_650M_post): A multi-species genomics foundation model predicting functional tracks and genomic elements across 1 Mb contexts. (listed, not yet runnable)
- [ML4Floods](https://corollary-labs.com/models/isp-uv-es/ml4floods): Optical satellite image segmentation model for global flood extent mapping from Sentinel-2 and Landsat data. (listed, not yet runnable)
- [StarCOP](https://corollary-labs.com/models/isp-uv-es/starcop): A segmentation model for detecting methane plumes in airborne and satellite hyperspectral remote sensing imagery. (listed, not yet runnable)
- [SuperIX](https://corollary-labs.com/models/isp-uv-es/superIX): Enhances Sentinel-2 satellite imagery to finer resolutions and evaluates super-resolution algorithms for researchers. (listed, not yet runnable)
- [Protenix-RNA](https://corollary-labs.com/models/LiteFold/protenix-rna): A Protenix-lineage PyTorch model fine-tuned for RNA three-dimensional structure prediction. (listed, not yet runnable)
- [Botanic0-L](https://corollary-labs.com/models/living-models/Botanic0-L): Plant genomic foundation model for regulatory annotation, expression analysis, and variant effect prediction. (listed, not yet runnable)
- [Botanic0-M](https://corollary-labs.com/models/living-models/Botanic0-M): A mid-sized plant genomic foundation model for regulatory annotation, expression inference, and variant effect analysis. (listed, not yet runnable)
- [Botanic0-S](https://corollary-labs.com/models/living-models/Botanic0-S): A compact 4-layer plant genomic foundation model pre-trained on nuclear DNA across 43 diverse plant genomes. (listed, not yet runnable)
- [Botanic1-L](https://corollary-labs.com/models/living-models/Botanic1-L): Plant genomic foundation model for sequence embedding, zero-shot variant effect scoring, and downstream fine-tuning. (listed, not yet runnable)
- [Botanic1-M](https://corollary-labs.com/models/living-models/Botanic1-M): A 688M-parameter bidirectional state-space plant genomic foundation model for sequence modelling. (listed, not yet runnable)
- [Botanic1-M SAE (k=64, codebook 8192)](https://corollary-labs.com/models/living-models/Botanic1-M-sae-k64-codebook8192): Sparse autoencoders decomposing Botanic1-M activations into 8,192 features for plant genomic interpretability. (listed, not yet runnable)
- [Botanic1-S](https://corollary-labs.com/models/living-models/Botanic1-S): A 318M-parameter bidirectional Mamba-2 foundation model for analysing plant genomes at single-nucleotide resolution. (listed, not yet runnable)
- [ABMIL (CONCH v1.5 features, pc108-24k)](https://corollary-labs.com/models/MahmoodLab/abmil.base.conch_v15.pc108-24k): Attention-based MIL classifier for slide-level cancer subtyping using CONCH v1.5 features. (listed, not yet runnable)
- [ABMIL (UNI2 features, pc108-24k)](https://corollary-labs.com/models/MahmoodLab/abmil.base.uni_v2.pc108-24k): Attention-based multiple-instance learning classifier for whole-slide histology inference using UNI2 patch features. (listed, not yet runnable)
- [ABMIL (UNI features, pc108-24k)](https://corollary-labs.com/models/MahmoodLab/abmil.base.uni.pc108-24k): Slide-level attention-based classifier aggregating UNI patch features across pan-cancer histology images. (listed, not yet runnable)
- [CONCH](https://corollary-labs.com/models/MahmoodLab/CONCH): A pathology vision-language model for zero-shot classification, retrieval, and analysis on H&E whole-slide images. (listed, not yet runnable)
- [CONCH v1.5](https://corollary-labs.com/models/MahmoodLab/conchv1_5): A pathology vision-language model for image feature extraction and zero-shot or few-shot tasks. (listed, not yet runnable)
- [HEST Tissue Segmentation](https://corollary-labs.com/models/MahmoodLab/hest-tissue-seg): Segments tissue from background in H&E and IHC histology images for spatial transcriptomics and slide processing. (listed, not yet runnable)
- [KRONOS](https://corollary-labs.com/models/MahmoodLab/KRONOS): A self-supervised vision model for feature extraction from multiplexed tissue images in spatial proteomics. (listed, not yet runnable)
- [Madeleine](https://corollary-labs.com/models/MahmoodLab/madeleine): A companion pathology model developed by Mahmood Lab for open histology evaluation pipelines. (listed, not yet runnable)
- [SEAL](https://corollary-labs.com/models/MahmoodLab/SEAL): A pathology foundation model for whole-slide image representation, cancer classification, and biomarker tasks. (listed, not yet runnable)
- [TITAN](https://corollary-labs.com/models/MahmoodLab/TITAN): Slide-level pathology foundation model generating whole-slide embeddings for oncology and biomarker prediction. (listed, not yet runnable)
- [UNI](https://corollary-labs.com/models/MahmoodLab/UNI): A self-supervised Vision Transformer patch encoder for computational pathology and histology workflows. (listed, not yet runnable)
- [UNI2-h](https://corollary-labs.com/models/MahmoodLab/UNI2-h): Vision transformer generating 1,536-dimensional embeddings from histology patches for pathology analysis. (listed, not yet runnable)
- [MMPT-FM](https://corollary-labs.com/models/Merck/MMPT-FM): Generates matched molecular pair transformations and chemical analogues for medicinal chemistry design. (listed, not yet runnable)
- [TEDDY](https://corollary-labs.com/models/Merck/TEDDY): Single-cell foundation models trained on 116 million cells for cross-donor and cross-disease representations. (listed, not yet runnable)
- [Aurora](https://corollary-labs.com/models/microsoft/aurora): An Earth system foundation model for global weather forecasting, atmospheric chemistry, and cyclone tracking. (listed, not yet runnable)
- [BioEmu](https://corollary-labs.com/models/microsoft/bioemu): Generative model predicting protein equilibrium structural ensembles and thermodynamic stabilities from sequence. (listed, not yet runnable)
- [CoLiPRI](https://corollary-labs.com/models/microsoft/colipri): A 3D vision–language encoder that jointly represents chest CT scans and radiology reports for medical imaging research. (listed, not yet runnable)
- [MatterGen](https://corollary-labs.com/models/microsoft/mattergen): A diffusion model that generates novel, stable inorganic crystal structures with user-defined target properties. (listed, not yet runnable)
- [MatterSim](https://corollary-labs.com/models/microsoft/mattersim): An atomistic foundation model for predicting material properties across varied temperatures and pressures. (listed, not yet runnable)
- [Skala 1.0](https://corollary-labs.com/models/microsoft/skala-1.0): A neural exchange-correlation functional for DFT delivering hybrid accuracy at semi-local computational cost. (listed, not yet runnable)
- [Skala 1.1](https://corollary-labs.com/models/microsoft/skala-1.1): A deep-learning exchange-correlation functional for DFT combining semi-local speed with hybrid accuracy. (listed, not yet runnable)
- [Equiformer v3](https://corollary-labs.com/models/mirror-physics/equiformer_v3): An SE(3)-equivariant graph transformer predicting energies and forces for molecules and crystalline materials. (listed, not yet runnable)
- [MIST 28M · QM9](https://corollary-labs.com/models/mist-models/mist-28M-kkgx0omx-qm9): Predicts quantum-mechanical properties of small organic molecules using a 28-million parameter architecture. (listed, not yet runnable)
- [Surya Solar Flares](https://corollary-labs.com/models/nasa-ibm-ai4science/solar_flares_surya): A solar physics model that predicts solar flares from full-disk magnetogram and EUV time series. (listed, not yet runnable)
- [Surya Solar Wind](https://corollary-labs.com/models/nasa-ibm-ai4science/solar_wind_surya): Forecasts solar wind plasma and interplanetary magnetic fields at the L1 point for solar physicists. (listed, not yet runnable)
- [Surya 1.0](https://corollary-labs.com/models/nasa-ibm-ai4science/Surya-1.0): A 366M-parameter foundation model trained on solar observatory data for space weather forecasting. (listed, not yet runnable)
- [cBottle](https://corollary-labs.com/models/nvidia/cbottle): A cascaded diffusion model for generating kilometer-scale Earth atmospheric states on HEALPix geometry. (listed, not yet runnable)
- [FourCastNet 3](https://corollary-labs.com/models/nvidia/fourcastnet3): A geometric deep learning model for rapid global probabilistic weather and subseasonal-to-seasonal forecasting. (listed, not yet runnable)
- [StormCast V1](https://corollary-labs.com/models/nvidia/stormcast-v1-era5-hrrr): A mesoscale neural model forecasting 99 atmospheric state variables at kilometre scale using generative diffusion. (listed, not yet runnable)
- [LULA-1.1](https://corollary-labs.com/models/omtx/lula-1.1): A sequence-only model predicting protein-ligand binding scores from amino-acid sequences and SMILES strings. (listed, not yet runnable)
- [CYP Inhibition Model](https://corollary-labs.com/models/openadmet/cyp1a2-cyp2d6-cyp3a4-cyp3c9-chemeleon-baseline): A multi-task baseline model predicting pIC50 inhibition values across four cytochrome P450 isoforms. (listed, not yet runnable)
- [PXR Activation Model](https://corollary-labs.com/models/openadmet/pxr-chemeleon-baseline): A baseline CheMeleon model predicting pregnane X receptor activation potency to assess drug-drug interaction risks. (listed, not yet runnable)
- [OpenFold3](https://corollary-labs.com/models/OpenFold/OpenFold3): Predicts 3D structures of proteins, nucleic acids, ligands, and complexes for structural biology and drug discovery. (listed, not yet runnable)
- [OrbMol](https://corollary-labs.com/models/Orbital-Materials/OrbMol): A molecular foundation-model interatomic potential predicting energies, forces, and properties across chemical systems. (listed, not yet runnable)
- [Aion Base](https://corollary-labs.com/models/polymathic-ai/aion-base): An omnimodal foundation model predicting across 39 astronomical survey data types using masked modelling. (listed, not yet runnable)
- [AstroCLIP](https://corollary-labs.com/models/polymathic-ai/astroclip): A multimodal astronomy model aligning galaxy spectra and images into a shared embedding space. (listed, not yet runnable)
- [WALRUS](https://corollary-labs.com/models/polymathic-ai/walrus): A 1.3-billion-parameter space–time Transformer surrogate for multi-domain continuum dynamics simulations. (listed, not yet runnable)
- [DeepSpot-M](https://corollary-labs.com/models/ratschlab/DeepSpotM): Multimodal deep-learning model that predicts spatially resolved gene expression directly from H&E whole-slide images — a LoRA adapter over the Midnight pathology foundation model, covering 19k+ genes. Used to generate the TCGA and HEST virtual spatial transcriptomics atlases for computational pathology and oncology research. (listed, not yet runnable)
- [Nesso](https://corollary-labs.com/models/recursionpharma/nesso): Predicts protein–ligand binding affinities and binder status from protein sequences and small molecules. (listed, not yet runnable)
- [AQAffinity](https://corollary-labs.com/models/SandboxAQ/AQAffinity): An open-source protein–ligand binding affinity prediction head built on OpenFold3 for drug discovery. (listed, not yet runnable)
- [BOA — Basis Overlap Architecture](https://corollary-labs.com/models/sciai-lab/boa): An equivariant graph neural network predicting molecular ground-state electron densities using basis overlap matrices. (listed, not yet runnable)
- [eva-rna](https://corollary-labs.com/models/ScientaLab/eva-rna): A transformer foundation model generating sample- and gene-level embeddings from human and mouse transcriptomic profiles. (listed, not yet runnable)
- [PhenoSeq](https://corollary-labs.com/models/Sentinal4D/PhenoSeq): Predicts single-cell scGPT RNA-seq embeddings from multi-channel fluorescence microscopy imaging features. (listed, not yet runnable)
- [Rhaister](https://corollary-labs.com/models/tahoebio/Rhaister): A statistical model predicting cellular perturbation responses across drugs, knockdowns, and cytokines. (listed, not yet runnable)
- [Tahoe-100M-SCVI](https://corollary-labs.com/models/tahoebio/Tahoe-100M-SCVI-v1): A single-cell variational autoencoder mapping drug perturbations across cancer cell lines into a latent space. (listed, not yet runnable)
- [Tahoe-x1](https://corollary-labs.com/models/tahoebio/Tahoe-x1): Perturbation-trained transformer foundation models for single-cell cancer research and drug discovery. (listed, not yet runnable)
- [Clinical Camel 70B](https://corollary-labs.com/models/wanglab/ClinicalCamel-70B): A 70-billion-parameter clinical language model fine-tuned for medical dialogue and clinical decision support. (listed, not yet runnable)
- [GO-GPT](https://corollary-labs.com/models/wanglab/gogpt): Generative model predicting Gene Ontology terms across all three ontology aspects directly from protein sequences. (listed, not yet runnable)
- [MedSAM](https://corollary-labs.com/models/wanglab/medsam-vit-base): A promptable medical image segmentation model fine-tuned on CT, MRI, X-ray, ultrasound, and histology data. (listed, not yet runnable)
- [X-Cell](https://corollary-labs.com/models/Xaira-Therapeutics/X-Cell): A diffusion model predicting single-cell transcriptomic responses to CRISPRi perturbations across cellular contexts. (listed, not yet runnable)

## Bioinformatics programs (run by the research agent)

- [BUSCO](https://corollary-labs.com/tools/busco): Assesses completeness of genome assemblies, gene sets, and transcriptomes using universal single-copy orthologs.
- [MaSuRCA](https://corollary-labs.com/tools/masurca): Assembles genomes from short Illumina reads, alone or combined with long reads, using super-reads.
- [MetaCAT](https://corollary-labs.com/tools/metacat): Bins metagenome assemblies into genomes using coverage and composition, and links them with host traits.
- [mitochondria-simulator](https://corollary-labs.com/tools/mitochondria-simulator): Simulates Illumina or PacBio reads from a mitochondrial genome with chloroplast and nuclear contamination, then assembles them, to find the depth and assembler that recover it.
- [geNomad](https://corollary-labs.com/tools/genomad): Identifies plasmids and viruses in sequencing data, predicts proviruses, and assigns viral taxonomy.
- [CheckV](https://corollary-labs.com/tools/checkv): A command-line pipeline for assessing the completeness, contamination, and quality of viral genomes from metagenomes.
- [Pharokka](https://corollary-labs.com/tools/pharokka): Rapid, standardised genome annotation pipeline for bacteriophage isolates and metaviromes.
- [BACPHLIP](https://corollary-labs.com/tools/bacphlip): Predicts whether a bacteriophage genome represents a temperate or virulent lifestyle using conserved protein domains.
- [taxmyPHAGE](https://corollary-labs.com/tools/taxmyphage): Classifies complete dsDNA bacteriophage genomes to genus and species following official ICTV guidelines.
- [RdRpCATCH](https://corollary-labs.com/tools/rdrpcatch): Identifies RNA virus RNA-dependent RNA polymerase sequences using aggregated profile hidden Markov models.
- [MinCED](https://corollary-labs.com/tools/minced): Identifies CRISPR arrays in complete genomes and assembled metagenomic contigs.
- [CRISPRCasFinder](https://corollary-labs.com/tools/crisprcasfinder): Identifies CRISPR arrays and neighbouring Cas genes in genome sequences for researchers.
- [DeepCRISPR](https://corollary-labs.com/tools/deepcrispr): Predicts SpCas9 sgRNA on-target knockout efficacy and off-target cleavage profiles in human genomic sequences.
- [Nextflow](https://corollary-labs.com/tools/nextflow): A workflow system for creating scalable, portable, and reproducible scientific data-analysis pipelines.
- [RADAR CT](https://corollary-labs.com/tools/radar-ct): Generalist vision-language model that segments 36 anatomical structures and scores 145 findings on abdominal CT.
- [MitoMimics](https://corollary-labs.com/tools/mitomimics): Simulates mitochondrial movement, fission and fusion in a cell and renders it as synthetic widefield microscopy timelapses with ground-truth masks and tracks.
- [GGIR](https://corollary-labs.com/tools/ggir): Processes raw wrist and hip accelerometer data into physical activity, sleep, and circadian metrics.
- [Rectangle](https://corollary-labs.com/tools/rectangle): Estimates cell-type proportions in bulk and spatial transcriptomic data using single-cell RNA sequencing references.
- [samtools](https://corollary-labs.com/tools/samtools): A command-line toolsuite for reading, writing, sorting, indexing, and viewing SAM, BAM, and CRAM alignment files.
- [bcftools](https://corollary-labs.com/tools/bcftools): A toolkit to call, filter, merge, query, and summarise genetic variants in VCF, gVCF, and BCF2 files.
- [Bowtie 2](https://corollary-labs.com/tools/bowtie2): Aligns sequencing reads to long reference genomes using memory-efficient FM indexing.
- [BWA](https://corollary-labs.com/tools/bwa): A sequence alignment package that maps DNA sequencing reads against large reference genomes using FM-indexing.
- [bedtools](https://corollary-labs.com/tools/bedtools): A command-line suite for intersecting, merging, and analysing genomic intervals across standard bioinformatics formats.
- [MACS3](https://corollary-labs.com/tools/macs3): A peak caller identifying enriched genomic regions from bulk and single-cell ChIP-seq, ATAC-seq, and DNA assays.
- [HOMER](https://corollary-labs.com/tools/homer): Discovers sequence motifs and analyses functional genomics sequencing data such as ChIP-seq and RNA-seq.
- [fastp](https://corollary-labs.com/tools/fastp): Quality control and preprocessing software that trims, filters and profiles short-read FASTQ sequencing data in one pass.
- [Trimmomatic](https://corollary-labs.com/tools/trimmomatic): A Java tool for trimming sequencing adapters and low-quality bases from single-end and paired-end Illumina NGS reads.
- [featureCounts](https://corollary-labs.com/tools/featurecounts): Assigns mapped sequencing reads to genomic features such as genes, exons, and promoters for expression analysis.
- [SeqKit](https://corollary-labs.com/tools/seqkit): A cross-platform toolkit for fast inspection, filtering, conversion, and analysis of FASTA and FASTQ files.
- [BLAST+](https://corollary-labs.com/tools/blast): Finds regions of local similarity and calculates statistical significance across nucleotide and protein sequences.
- [DIAMOND](https://corollary-labs.com/tools/diamond): Accelerated sequence aligner for comparing protein and translated DNA sequences against protein databases.
- [MMseqs2](https://corollary-labs.com/tools/mmseqs2): Search and cluster massive protein and nucleotide sequence sets across CPUs and GPUs.
- [CD-HIT](https://corollary-labs.com/tools/cd-hit): Clusters protein and nucleotide sequences by sequence identity to eliminate redundancy.
- [MAFFT](https://corollary-labs.com/tools/mafft): A multiple sequence alignment program for amino acid or nucleotide sequences.
- [MUSCLE](https://corollary-labs.com/tools/muscle): Generates multiple sequence alignments for proteins and nucleotides, alongside protein structure alignments.
- [FastTree](https://corollary-labs.com/tools/fasttree): Infers approximately-maximum-likelihood phylogenetic trees from large alignments.
- [IQ-TREE](https://corollary-labs.com/tools/iqtree): Infers maximum-likelihood phylogenetic trees with automated model selection and ultrafast branch support.
- [Prodigal](https://corollary-labs.com/tools/prodigal): Protein-coding gene prediction software for bacterial, archaeal, and metagenomic DNA sequences.
- [ViennaRNA (RNAfold)](https://corollary-labs.com/tools/viennarna): Predicts and compares RNA secondary structures and their thermodynamics for computational and molecular biologists.
- [PLINK](https://corollary-labs.com/tools/plink): A statistical genetics toolset for whole-genome association studies and large-scale population genetic analyses.
- [GCTA](https://corollary-labs.com/tools/gcta): Tool for estimating heritability, genetic correlation, and running association tests from genome-wide data.
- [MAGeCK](https://corollary-labs.com/tools/mageck): Identifies functionally important genes from genome-scale CRISPR-Cas9 knockout screens using sgRNA read counts.

## Scientific databases (queried by the research agent, no API key)

- [Reactome](https://corollary-labs.com/databases/reactome): A curated, peer-reviewed knowledgebase of human biological pathways and molecular reactions for life scientists.
- [Harvard Dataverse](https://corollary-labs.com/databases/dataverse): A multidisciplinary data repository for sharing, preserving, and citing research data regardless of affiliation.
- [Zenodo](https://corollary-labs.com/databases/zenodo): A general-purpose open repository operated by CERN for sharing and preserving research datasets, software and papers.
- [Figshare](https://corollary-labs.com/databases/figshare): A research repository hosting institutional deposits, publisher records, and supplementary data behind figures.
- [Gene Expression Omnibus](https://corollary-labs.com/databases/geo): A public functional genomics archive for gene expression and DNA methylation profiling datasets.
- [DANDI Archive](https://corollary-labs.com/databases/dandi): A repository of BRAIN Initiative neurophysiology and neuroimaging data standardised in NWB and BIDS formats.
- [ICPSR](https://corollary-labs.com/databases/icpsr): A repository of social and behavioural science studies, including specialised topical archives.
- [NACDA](https://corollary-labs.com/databases/nacda): A repository of longitudinal research data focusing on ageing, retirement, and late-life health.
- [UniProtKB](https://corollary-labs.com/databases/uniprot): A reference database providing curated protein sequences and functional annotations for biological research.
- [PDB (RCSB)](https://corollary-labs.com/databases/pdb): A repository of experimentally determined 3D macromolecular structures with tools for search, analysis, and visualization.
- [AlphaFold DB](https://corollary-labs.com/databases/alphafolddb): A database by Google DeepMind and EMBL-EBI providing over 200 million pre-computed protein structure predictions.
- [ChEMBL](https://corollary-labs.com/databases/chembl): A curated database of bioactive molecules and protein targets for drug discovery and chemogenomics research.
- [Europe PMC](https://corollary-labs.com/databases/europepmc): A life sciences literature repository providing access to millions of full-text articles, citations and preprints.
- [UniRef90 / UniRef50](https://corollary-labs.com/databases/uniref): Clustered UniProt protein sequences at 90% and 50% identity, supporting MSA-based folding. (catalogued, not queried live)
- [ESM Atlas](https://corollary-labs.com/databases/esmatlas): A database of 59.7 million metagenomic protein structure predictions covering proteins absent from AlphaFold DB.
- [STRING](https://corollary-labs.com/databases/string): Database of known and predicted protein-protein interactions and functional association networks across organisms.
- [CATH](https://corollary-labs.com/databases/cath): A structural hierarchy classifying protein domains and their evolutionary relationships. (catalogued, not queried live)
- [Pfam](https://corollary-labs.com/databases/pfam): A database of protein families, domains and functional annotations built using profile hidden Markov models. (catalogued, not queried live)
- [InterPro](https://corollary-labs.com/databases/interpro): Integrates predictive signatures from 13 databases to classify protein sequences into families, domains and sites.
- [PDB-CCD](https://corollary-labs.com/databases/pdb-ccd): Reference dictionary defining all residues and small molecule components observed in Protein Data Bank structures. (catalogued, not queried live)
- [IEDB](https://corollary-labs.com/databases/iedb): A database and prediction resource containing 7.45 million B-cell and T-cell immune epitopes for immunology workflows. (catalogued, not queried live)
- [SKEMPI 2](https://corollary-labs.com/databases/skempi2): A database of protein–protein binding free energy and kinetic changes upon mutation for solved complexes. (catalogued, not queried live)
- [Gene Ontology](https://corollary-labs.com/databases/go): A standardised functional genomics ontology classifying gene and gene product attributes across species. (catalogued, not queried live)
- [FireProtDB](https://corollary-labs.com/databases/fireprotdb): A curated database of protein stability measurements for single-point mutants to train and test stability models. (catalogued, not queried live)
- [CZ CELLxGENE](https://corollary-labs.com/databases/cellxgene): A single-cell platform to explore, download, and analyse standardised human and mouse transcriptomic atlases.
- [BindingDB](https://corollary-labs.com/databases/bindingdb): A public database of experimentally measured binding affinities between proteins and drug-like small molecules.
- [MGnify Human Gut (UHGG)](https://corollary-labs.com/databases/mgnify-human-gut): Metagenome-assembled and isolate genomes from the human gastrointestinal tract, clustered into species representatives with a matching protein catalogue.
- [MetaboAnalyst](https://corollary-labs.com/databases/metaboanalyst): A web platform and compound dictionary for metabolomics data processing, statistical analysis, and pathway annotation.
- [MetaCyc](https://corollary-labs.com/databases/metacyc): Curated database of experimentally elucidated metabolic pathways, reactions, and enzymes across all domains of life. (catalogued, not queried live)
- [eggNOG](https://corollary-labs.com/databases/eggnog): A database providing hierarchical orthologous groups and functional annotations for comparative genomics.
- [NIAGADS Open Access](https://corollary-labs.com/databases/niagads): Open-access Alzheimer's disease genetics, variant annotations, and GWAS summary statistics for researchers.
- [FlyWire](https://corollary-labs.com/databases/flywire): A whole-brain connectome resource providing proofread neurons, synapses, and annotations for adult Drosophila.
- [ADNI](https://corollary-labs.com/databases/adni): A longitudinal repository of neuroimaging, biomarker, and clinical data across Alzheimer's disease progression. (catalogued, not queried live)
- [Ensembl](https://corollary-labs.com/databases/ensembl): A centralised genome browser and database providing automated annotation for human and other vertebrate genomes.
- [UCSC Genome Browser](https://corollary-labs.com/databases/ucsc): A web-based graphical viewer and tool suite for exploring reference genomes and aligned biological annotation tracks.
- [ClinVar](https://corollary-labs.com/databases/clinvar): A public archive of human genomic variants and their clinical significance assertions across diseases and drug responses.
- [dbSNP](https://corollary-labs.com/databases/dbsnp): A public repository of human single nucleotide variations and short genetic variants for biomedical research.
- [gnomAD](https://corollary-labs.com/databases/gnomad): Population allele frequencies and gene constraint scores derived from exomes and genomes.
- [GWAS Catalog](https://corollary-labs.com/databases/gwas-catalog): A curated database of human genome-wide association studies and variant–trait associations for biomedical research.
- [Open Targets Platform](https://corollary-labs.com/databases/opentargets): A genomics database providing target–disease association evidence, drug mechanisms, and fine-mapped credible sets.
- [GTEx](https://corollary-labs.com/databases/gtex): A genomics database providing gene expression data and single-tissue eQTLs across 54 human tissues.
- [ENCODE](https://corollary-labs.com/databases/encode): A functional genomics database cataloguing regulatory elements, transcription and chromatin states in human and mouse.
- [RegulomeDB](https://corollary-labs.com/databases/regulomedb): A genomics resource providing regulatory evidence and prioritisation scores for non-coding variants.
- [JASPAR](https://corollary-labs.com/databases/jaspar): A curated database of non-redundant transcription factor binding profiles across eukaryotic taxa.
- [ReMap 2022](https://corollary-labs.com/databases/remap): A catalogue of regulatory regions derived from integrated public ChIP-seq data across four model organisms.
- [cBioPortal](https://corollary-labs.com/databases/cbioportal): A cancer genomics database providing mutation, copy number, expression, and clinical data from public studies.
- [EMDB](https://corollary-labs.com/databases/emdb): A public repository of cryo-EM density maps and tomograms of macromolecular complexes and subcellular structures.
- [PRIDE Archive](https://corollary-labs.com/databases/pride): A repository of mass-spectrometry proteomics projects and their associated files.
- [MetaboLights](https://corollary-labs.com/databases/metabolights): A repository and reference database for cross-species, cross-platform metabolomics experiments and metabolite data.
- [Human Protein Atlas](https://corollary-labs.com/databases/hpa): A Swedish open-access resource mapping spatial protein and RNA expression across human tissues, cells, and organs.
- [g:Profiler](https://corollary-labs.com/databases/gprofiler): A web server for functional enrichment analysis and gene identifier conversion across biological databases.
- [Guide to PHARMACOLOGY](https://corollary-labs.com/databases/gtopdb): An expert-curated database of pharmacological targets, ligands, and their quantitative interactions.
- [Mouse Phenome Database](https://corollary-labs.com/databases/mpd): A database of strain-level phenotype and genotype measurements across diverse laboratory mouse populations.
- [Paleobiology Database](https://corollary-labs.com/databases/pbdb): A scientific database of fossil occurrences, taxa, and collections with geographic and stratigraphic context.
- [WoRMS](https://corollary-labs.com/databases/worms): Authoritative register and classification of marine species names, synonyms, and distributions.
- [NCBI BLAST](https://corollary-labs.com/databases/blast): NCBI BLAST finds regions of local similarity between biological sequences to infer functional and evolutionary links.
- [NCBI (E-utilities and Datasets)](https://corollary-labs.com/databases/ncbi): Programmatic access to NCBI biomedical, genomic, sequence, and literature databases via APIs.
- [European Nucleotide Archive](https://corollary-labs.com/databases/ena): EMBL-EBI's repository of raw sequencing reads, genome assemblies, and annotated nucleotide sequences.
- [ArrayExpress](https://corollary-labs.com/databases/arrayexpress): A repository of functional genomics experiments, including microarray and sequencing studies, hosted by EMBL-EBI.
- [CIViC](https://corollary-labs.com/databases/civic): An open, community-curated knowledgebase documenting the clinical relevance of cancer variants.
- [openFDA](https://corollary-labs.com/databases/openfda): An open-access platform providing indexed US Food and Drug Administration datasets and APIs for researchers.
- [PGS Catalog](https://corollary-labs.com/databases/pgs-catalog): An open database of published polygenic scores and evaluation metadata for genomic research.
- [SigCom LINCS](https://corollary-labs.com/databases/sigcom-lincs): A search platform for LINCS L1000 gene-expression signatures of chemical and genetic perturbations.

## nf-core pipelines

- [nf-core/abotyper](https://corollary-labs.com/pipelines/abotyper): A bioinformatics pipeline to determine ABO blood group phenotypes from long-read sequencing data.
- [nf-core/airrflow](https://corollary-labs.com/pipelines/airrflow): Analyse bulk and single-cell BCR and TCR sequencing data using the Immcantation framework.
- [nf-core/alleleexpression](https://corollary-labs.com/pipelines/alleleexpression): A Nextflow pipeline for allele-specific expression analysis using STAR-WASP, UMI deduplication, and phASER.
- [nf-core/ampliseq](https://corollary-labs.com/pipelines/ampliseq): A bioinformatics analysis pipeline for amplicon sequencing denoising, taxonomic assignment and diversity profiling.
- [nf-core/atacseq](https://corollary-labs.com/pipelines/atacseq): A Nextflow pipeline for quality control, peak calling, and differential accessibility analysis of ATAC-seq data.
- [nf-core/bacass](https://corollary-labs.com/pipelines/bacass): Assembles and annotates bacterial genomes from short, long, or hybrid sequencing reads.
- [nf-core/bacmodel](https://corollary-labs.com/pipelines/bacmodel): A bioinformatics pipeline for functional annotation and metabolic reconstruction of bacterial genomes.
- [nf-core/bactmap](https://corollary-labs.com/pipelines/bactmap): A reference-mapping pipeline that calls variants and builds pseudogenome alignments from bacterial genome reads.
- [nf-core/bamtofastq](https://corollary-labs.com/pipelines/bamtofastq): Converts mapped and unmapped BAM or CRAM files into FASTQ reads with integrated quality control.
- [nf-core/cageseq](https://corollary-labs.com/pipelines/cageseq): Processes CAGE-seq data through adapter trimming, alignment, and tag clustering for gene expression analysis.
- [nf-core/callingcards](https://corollary-labs.com/pipelines/callingcards): Processes calling cards sequencing data from yeast and mammalian models to identify and quantify genomic insertions.
- [nf-core/cellpainting](https://corollary-labs.com/pipelines/cellpainting): Processes Cell Painting microscopy assays through illumination correction, feature extraction, and quality control.
- [nf-core/chipseq](https://corollary-labs.com/pipelines/chipseq): A Nextflow pipeline for ChIP-seq quality control, alignment, peak calling, and differential binding analysis.
- [nf-core/circdna](https://corollary-labs.com/pipelines/circdna): A Nextflow pipeline identifying extrachromosomal circular DNA from Circle-seq, WGS, and ATAC-seq data.
- [nf-core/circrna](https://corollary-labs.com/pipelines/circrna): Analyses total RNA sequencing data to identify, quantify, and predict miRNA targets for circular RNAs.
- [nf-core/clipseq](https://corollary-labs.com/pipelines/clipseq): A Nextflow pipeline for quality control, mapping, deduplication and peak calling on CLIP sequencing data.
- [nf-core/coproid](https://corollary-labs.com/pipelines/coproid): Identifies the biological host of ancient and modern faeces using endogenous DNA and microbiome machine learning.
- [nf-core/createpanelrefs](https://corollary-labs.com/pipelines/createpanelrefs): Builds panels of normals and reference models for copy number and variant calling tools from multiple samples.
- [nf-core/createtaxdb](https://corollary-labs.com/pipelines/createtaxdb): Automated and parallelised generation of reference databases for multiple taxonomic profilers and classifiers.
- [nf-core/crisprseq](https://corollary-labs.com/pipelines/crisprseq): A Nextflow pipeline for evaluating targeted gene editing quality and identifying hits from pooled CRISPR screens.
- [nf-core/cutandrun](https://corollary-labs.com/pipelines/cutandrun): Pipeline for processing CUT&RUN, CUT&Tag, and TIPseq data from raw reads to peak calling and quality reports.
- [nf-core/daamicrobiome](https://corollary-labs.com/pipelines/daamicrobiome): Microbiome differential abundance pipeline combining multiple tools via k-intersection or simulation consensus.
- [nf-core/dartseq](https://corollary-labs.com/pipelines/dartseq): A bioinformatics pipeline for m6A RNA modification detection and quality control in DART-seq sequencing data.
- [nf-core/datasync](https://corollary-labs.com/pipelines/datasync): A Nextflow pipeline for transferring files and directories across storage systems while validating their integrity.
- [nf-core/deepmodeloptim](https://corollary-labs.com/pipelines/deepmodeloptim): Augments biological data towards optimal task-specific training sets for PyTorch deep learning models.
- [nf-core/deepmutscan](https://corollary-labs.com/pipelines/deepmutscan): A bioinformatics pipeline for analysing deep mutational scanning data generated by shotgun DNA sequencing.
- [nf-core/demo](https://corollary-labs.com/pipelines/demo): A minimal Nextflow pipeline for bioinformatics workshops, demonstrations, and training sessions.
- [nf-core/demultiplex](https://corollary-labs.com/pipelines/demultiplex): Nextflow pipeline for demultiplexing raw data from various sequencing platforms into FASTQ files.
- [nf-core/denovotranscript](https://corollary-labs.com/pipelines/denovotranscript): A de novo transcriptome assembly pipeline for paired-end bulk RNA-seq data without a reference genome.
- [nf-core/detaxizer](https://corollary-labs.com/pipelines/detaxizer): Identifies and optionally filters specific taxonomic sequences, such as human DNA, from FASTQ sequencing data.
- [nf-core/diaproteomics](https://corollary-labs.com/pipelines/diaproteomics): Automated quantitative analysis of data-independent acquisition (DIA) proteomics mass spectrometry measurements.
- [nf-core/differentialabundance](https://corollary-labs.com/pipelines/differentialabundance): Performs differential abundance and gene set analyses on observation matrices from RNA-seq and microarrays.
- [nf-core/diseasemodulediscovery](https://corollary-labs.com/pipelines/diseasemodulediscovery): A Nextflow pipeline for identifying and evaluating network-based disease modules from molecular interaction data.
- [nf-core/drop](https://corollary-labs.com/pipelines/drop): Detects aberrant expression, splicing, and mono-allelic expression in RNA-seq data for rare disorder diagnosis.
- [nf-core/drugresponseeval](https://corollary-labs.com/pipelines/drugresponseeval): Evaluate and benchmark computational drug response prediction models using standardised cross-validation protocols.
- [nf-core/dualrnaseq](https://corollary-labs.com/pipelines/dualrnaseq): Simultaneous analysis of host and pathogen gene expression from dual RNA-seq sequencing data.
- [nf-core/eager](https://corollary-labs.com/pipelines/eager): A reproducible Nextflow pipeline for processing, quality-controlling and analysing ancient genomic sequencing data.
- [nf-core/epigenomesegmentation](https://corollary-labs.com/pipelines/epigenomesegmentation): A chromatin segmentation pipeline annotating functional genomic states from histone, chromatin, and methylation data.
- [nf-core/epitopeprediction](https://corollary-labs.com/pipelines/epitopeprediction): A Nextflow pipeline that predicts peptide binding to MHC class I and II molecules from variants, proteins, or peptides.
- [nf-core/evexplorer](https://corollary-labs.com/pipelines/evexplorer): A Nextflow pipeline for quality control, normalisation, and differential expression analysis of extracellular vesicle RNA.
- [nf-core/fastqrepair](https://corollary-labs.com/pipelines/fastqrepair): A Nextflow pipeline to recover corrupted FASTQ.gz files, fix uncompliant reads, and restore read pairing.
- [nf-core/fastquorum](https://corollary-labs.com/pipelines/fastquorum): A Nextflow pipeline to generate consensus reads from sequencing data using unique molecular identifiers.
- [nf-core/fetchngs](https://corollary-labs.com/pipelines/fetchngs): Pipeline to fetch metadata and raw FastQ files from public repositories for downstream bioinformatics analyses.
- [nf-core/funcprofiler](https://corollary-labs.com/pipelines/funcprofiler): Read-based functional profiling of microbiome short-read sequencing data across multiple profiling tools.
- [nf-core/funcscan](https://corollary-labs.com/pipelines/funcscan): Screens assembled nucleotide contigs and metagenomes for antimicrobial peptides, resistance genes, and biosynthetic clusters.
- [nf-core/genephylomodeler](https://corollary-labs.com/pipelines/genephylomodeler): A bioinformatics pipeline that fits evolutionary models and detects natural selection in protein-coding gene alignments.
- [nf-core/genomeannotator](https://corollary-labs.com/pipelines/genomeannotator): Identifies coding gene structures in eukaryotic draft genomes using external evidence and ab initio models.
- [nf-core/genomeassembler](https://corollary-labs.com/pipelines/genomeassembler): Assembles, polishes, and scaffolds haploid or unphased genomes from Oxford Nanopore or PacBio HiFi sequencing reads.
- [nf-core/genomeqc](https://corollary-labs.com/pipelines/genomeqc): A bioinformatics pipeline for evaluating and comparing the quality of multiple genome assemblies and annotations.
- [nf-core/genomeskim](https://corollary-labs.com/pipelines/genomeskim): Performs quality control, filtering, and organelle assembly or genome analysis from genome skimming data.
- [nf-core/genomicrelatedness](https://corollary-labs.com/pipelines/genomicrelatedness): Infers genetic relatedness from low-coverage whole-genome sequencing data.
- [nf-core/gwas](https://corollary-labs.com/pipelines/gwas): A Nextflow pipeline under development for conducting genome-wide association studies.
- [nf-core/hadge](https://corollary-labs.com/pipelines/hadge): Demultiplexes single-cell sequencing samples using combined hashing and genotype deconvolution methods.
- [nf-core/hgtseq](https://corollary-labs.com/pipelines/hgtseq): A Nextflow pipeline to identify horizontal gene transfer and host integration sites from sequencing data.
- [nf-core/hic](https://corollary-labs.com/pipelines/hic): A Nextflow pipeline for processing, normalising, and analysing chromosome conformation capture (Hi-C) sequencing data.
- [nf-core/hicar](https://corollary-labs.com/pipelines/hicar): A Nextflow pipeline for simultaneous profiling of chromatin accessibility and cis-regulatory chromatin interactions.
- [nf-core/hlatyping](https://corollary-labs.com/pipelines/hlatyping): Precision human leukocyte antigen typing from next-generation sequencing data for DNA and RNA samples.
- [nf-core/imcyto](https://corollary-labs.com/pipelines/imcyto): A Nextflow pipeline for image segmentation and single-cell expression extraction from Imaging Mass Cytometry data.
- [nf-core/isoseq](https://corollary-labs.com/pipelines/isoseq): A Nextflow pipeline for Iso-Seq genome annotation, generating FLNC reads and gene models using uLTRA and TAMA.
- [nf-core/kmermaid](https://corollary-labs.com/pipelines/kmermaid): A bioinformatics pipeline for comparative analysis of omic sequences using k-mer based similarity methods.
- [nf-core/lncpipe](https://corollary-labs.com/pipelines/lncpipe): A bioinformatics pipeline for analysing long non-coding RNAs from RNA sequencing datasets.
- [nf-core/longraredisease](https://corollary-labs.com/pipelines/longraredisease): A long-read sequencing pipeline to detect and interpret structural variants in rare diseases and neurodevelopment.
- [nf-core/lsmquant](https://corollary-labs.com/pipelines/lsmquant): Preprocesses and analyses light-sheet microscopy images of tissue-cleared samples.
- [nf-core/mag](https://corollary-labs.com/pipelines/mag): A bioinformatics pipeline for quality control, assembly, binning, and annotation of metagenomic sequencing data.
- [nf-core/magmap](https://corollary-labs.com/pipelines/magmap): Maps metagenomic and metatranscriptomic reads to large collections of genomes for gene quantification.
- [nf-core/marsseq](https://corollary-labs.com/pipelines/marsseq): Preprocessing pipeline for MARS-seq v2 single-cell RNA sequencing data with RNA velocity estimation via StarSolo.
- [nf-core/mcmicro](https://corollary-labs.com/pipelines/mcmicro): An end-to-end pipeline that transforms multi-channel whole-slide microscopy images into single-cell data.
- [nf-core/meerpipe](https://corollary-labs.com/pipelines/meerpipe): Processes MeerKAT pulsar data to generate calibration products, pulse arrival times and timing residuals.
- [nf-core/metaboigniter](https://corollary-labs.com/pipelines/metaboigniter): Pre-processes mass spectrometry metabolomics data with MS1 quantification and MS2 identification.
- [nf-core/metapep](https://corollary-labs.com/pipelines/metapep): Predicts peptide MHC and HLA binding affinity directly from metagenomic sequences, taxa, and proteins.
- [nf-core/metatdenovo](https://corollary-labs.com/pipelines/metatdenovo): Assemble and annotate metatranscriptomic and metagenomic sequencing data for prokaryotes, eukaryotes and viruses.
- [nf-core/methylarray](https://corollary-labs.com/pipelines/methylarray): An end-to-end Nextflow pipeline for processing, filtering, and analysing Illumina DNA methylation array data.
- [nf-core/methylong](https://corollary-labs.com/pipelines/methylong): Extract DNA methylation calls, phase variants, and identify differentially methylated regions from long reads.
- [nf-core/methylseq](https://corollary-labs.com/pipelines/methylseq): A Nextflow pipeline for quality control, alignment and methylation extraction from bisulfite and TAPS sequencing data.
- [nf-core/mhcquant](https://corollary-labs.com/pipelines/mhcquant): Identifies and quantifies MHC-eluted peptides from data-dependent acquisition mass spectrometry datasets.
- [nf-core/mitodetect](https://corollary-labs.com/pipelines/mitodetect): A Nextflow pipeline for end-to-end analysis of mitochondrial next-generation sequencing data.
- [nf-core/mnaseseq](https://corollary-labs.com/pipelines/mnaseseq): Pipeline for processing micrococcal nuclease sequencing data and mapping genome-wide nucleosome positions.
- [nf-core/molkart](https://corollary-labs.com/pipelines/molkart): An automated pipeline for processing Resolve Biosciences Molecular Cartography combinatorial FISH data.
- [nf-core/mspepid](https://corollary-labs.com/pipelines/mspepid): Identifies peptides and rescores spectra from proteomics and peptidomics mass spectrometry data.
- [nf-core/multiplesequencealign](https://corollary-labs.com/pipelines/multiplesequencealign): A Nextflow pipeline to run and systematically benchmark multiple sequence alignment methods and protein structures.
- [nf-core/nanoseq](https://corollary-labs.com/pipelines/nanoseq): A Nextflow pipeline for basecalling, demultiplexing, alignment and analysis of Oxford Nanopore DNA and RNA data.
- [nf-core/nanostring](https://corollary-labs.com/pipelines/nanostring): A bioinformatics pipeline for quality control and normalisation of NanoString nCounter gene expression data.
- [nf-core/nascent](https://corollary-labs.com/pipelines/nascent): A Nextflow pipeline for processing nascent transcript and transcriptional start site sequencing assays.
- [nf-core/ncrnannotator](https://corollary-labs.com/pipelines/ncrnannotator): Annotates non-coding RNA loci across genome assemblies using Rfam covariance models and Infernal.
- [nf-core/omicsgenetraitassociation](https://corollary-labs.com/pipelines/omicsgenetraitassociation): Integrates multi-omic datasets to perform correlated meta-analysis and functional enrichment for trait associations.
- [nf-core/oncoanalyser](https://corollary-labs.com/pipelines/oncoanalyser): A Nextflow pipeline for analysing and reporting cancer DNA and RNA sequencing data using the WiGiTS toolkit.
- [nf-core/pacsomatic](https://corollary-labs.com/pipelines/pacsomatic): A Nextflow pipeline for somatic variant and methylation analysis using paired PacBio HiFi tumour and normal reads.
- [nf-core/pacvar](https://corollary-labs.com/pipelines/pacvar): Processes PacBio long-read sequencing data for whole-genome variant discovery and tandem repeat characterisation.
- [nf-core/pairgenomealign](https://corollary-labs.com/pipelines/pairgenomealign): Pairwise whole-genome alignment of query genomes against a target genome using the LAST toolchain.
- [nf-core/pangenome](https://corollary-labs.com/pipelines/pangenome): Constructs locally directed, acyclic pangenome graphs from collections of sequences using Nextflow.
- [nf-core/panoramaseq](https://corollary-labs.com/pipelines/panoramaseq): A Nextflow pipeline for processing sequencing-based spatial transcriptomics data from in-situ arrays.
- [nf-core/pathogenepidemiology](https://corollary-labs.com/pipelines/pathogenepidemiology): A profile-based Nextflow pipeline for genomic epidemiological analysis of predefined pathogens.
- [nf-core/pathogensurveillance](https://corollary-labs.com/pipelines/pathogensurveillance): A population genomics pipeline for pathogen identification, variant detection, and phylogenetic biosurveillance.
- [nf-core/phageannotator](https://corollary-labs.com/pipelines/phageannotator): Identifies, annotates, and quantifies bacteriophage sequences from genomic and metagenomic assemblies.
- [nf-core/phaseimpute](https://corollary-labs.com/pipelines/phaseimpute): Phases and imputes genetic data across low-pass sequencing and reference panel datasets.
- [nf-core/phyloplace](https://corollary-labs.com/pipelines/phyloplace): A Nextflow pipeline for phylogenetic placement and taxonomic classification of query sequences using EPA-NG and GAPPA.
- [nf-core/pixelator](https://corollary-labs.com/pipelines/pixelator): Processes Proximity Network Assay sequencing data into single-cell protein abundance and interactomics graphs.
- [nf-core/plasmodiumdrugres](https://corollary-labs.com/pipelines/plasmodiumdrugres): Analyses drug resistance markers from microhaplotype data to estimate allele frequencies and prevalences.
- [nf-core/proteinannotator](https://corollary-labs.com/pipelines/proteinannotator): Generates sequence-level functional, domain, and secondary structure annotations for protein FASTA sequences.
- [nf-core/proteinfamilies](https://corollary-labs.com/pipelines/proteinfamilies): Generates de novo protein families and updates existing models from amino acid sequence data.
- [nf-core/proteinfold](https://corollary-labs.com/pipelines/proteinfold): Predicts three-dimensional structures of proteins, nucleic acids and complexes using multiple folding engines.
- [nf-core/proteogenomicsdb](https://corollary-labs.com/pipelines/proteogenomicsdb): Builds customised proteogenomics databases from reference proteomes, non-canonical sequences, and genomic variants.
- [nf-core/provenancereport](https://corollary-labs.com/pipelines/provenancereport): A Nextflow pipeline that renders reproducible Quarto reports with full data, runtime and environment provenance.
- [nf-core/radseq](https://corollary-labs.com/pipelines/radseq): A Nextflow variant-calling pipeline for restriction site-associated DNA sequencing (RADseq) data.
- [nf-core/rangeland](https://corollary-labs.com/pipelines/rangeland): Processes satellite imagery and auxiliary data to analyse land-cover change and vegetation trends.
- [nf-core/raredisease](https://corollary-labs.com/pipelines/raredisease): A bioinformatics pipeline for calling and scoring genetic variants from rare disease WGS and WES data.
- [nf-core/rarevariantburden](https://corollary-labs.com/pipelines/rarevariantburden): Performs summary-count rare variant burden tests comparing sequenced cases against public gnomAD control cohorts.
- [nf-core/readsimulator](https://corollary-labs.com/pipelines/readsimulator): Nextflow pipeline for simulating amplicon, target capture, metagenome, and whole-genome sequencing reads.
- [nf-core/references](https://corollary-labs.com/pipelines/references): A bioinformatics pipeline for generating genomic indices and reference files for common organisms.
- [nf-core/reportho](https://corollary-labs.com/pipelines/reportho): Comparative analysis and consensus aggregation of orthologue predictions for protein queries.
- [nf-core/ribomsqc](https://corollary-labs.com/pipelines/ribomsqc): A quality control pipeline that monitors mass spectrometer performance during ribonucleoside analysis.
- [nf-core/riboseq](https://corollary-labs.com/pipelines/riboseq): A Nextflow pipeline for processing ribosome profiling data, predicting translated ORFs, and evaluating translational efficiency.
- [nf-core/rnadnavar](https://corollary-labs.com/pipelines/rnadnavar): Somatic mutation detection pipeline integrating RNA and DNA sequencing data using a consensus calling approach.
- [nf-core/rnafusion](https://corollary-labs.com/pipelines/rnafusion): A Nextflow bioinformatics pipeline for detecting and visualising gene fusions and splicing events from RNA-seq data.
- [nf-core/rnaseq](https://corollary-labs.com/pipelines/rnaseq): A bioinformatics pipeline for quality control, alignment, and quantification of reference-based RNA sequencing data.
- [nf-core/rnasplice](https://corollary-labs.com/pipelines/rnasplice): An RNA sequencing pipeline for analysing alternative splicing, differential transcript usage, and exon usage.
- [nf-core/rnastructurome](https://corollary-labs.com/pipelines/rnastructurome): Analyses SHAPE and DMS chemical RNA structure-probing data to quantify reactivity and predict secondary structures.
- [nf-core/rnavar](https://corollary-labs.com/pipelines/rnavar): A Nextflow pipeline for RNA-seq variant discovery following GATK4 best practice guidelines.
- [nf-core/sammyseq](https://corollary-labs.com/pipelines/sammyseq): Processes SAMMY-seq sequencing data to investigate chromatin accessibility states and genome compartmentalisation.
- [nf-core/sarek](https://corollary-labs.com/pipelines/sarek): Nextflow pipeline for detecting germline and somatic variants from whole-genome and targeted sequencing data.
- [nf-core/scdownstream](https://corollary-labs.com/pipelines/scdownstream): A Nextflow pipeline for quality control, batch integration, clustering, and analysis of single-cell RNA-seq data.
- [nf-core/scnanoseq](https://corollary-labs.com/pipelines/scnanoseq): A Nextflow pipeline for processing 10X Genomics single-cell and single-nuclei Oxford Nanopore RNA sequencing data.
- [nf-core/scrnaseq](https://corollary-labs.com/pipelines/scrnaseq): Processes single-cell RNA sequencing data across multiple aligners into count matrices for transcriptomic research.
- [nf-core/seqinspector](https://corollary-labs.com/pipelines/seqinspector): A dedicated quality control pipeline for sequencing data, producing sample and group MultiQC reports.
- [nf-core/seqsubmit](https://corollary-labs.com/pipelines/seqsubmit): Automates the submission of raw sequencing reads, metagenomic assemblies, MAGs, and bins to the ENA.
- [nf-core/slamseq](https://corollary-labs.com/pipelines/slamseq): A Nextflow pipeline for processing SLAMSeq sequencing data, quantifying conversions, and identifying target genes.
- [nf-core/smrnaseq](https://corollary-labs.com/pipelines/smrnaseq): A Nextflow pipeline for quality control, quantification, and discovery in small RNA sequencing data.
- [nf-core/sopa](https://corollary-labs.com/pipelines/sopa): Nextflow pipeline for processing and analysing single-cell spatial transcriptomics and multiplex imaging data.
- [nf-core/spatialaxe](https://corollary-labs.com/pipelines/spatialaxe): Pipeline for segmentation, transcript assignment and quality control of 10x Genomics Xenium and Atera spatial data.
- [nf-core/spatialvi](https://corollary-labs.com/pipelines/spatialvi): A Nextflow pipeline for processing and analysing 10x Genomics Visium spatial transcriptomics data.
- [nf-core/spinningjenny](https://corollary-labs.com/pipelines/spinningjenny): Simulates the first industrial revolution using agent-based models for economic and historical research.
- [nf-core/stableexpression](https://corollary-labs.com/pipelines/stableexpression): Identifies stably expressed genes across RNA-seq and microarray datasets to select RT-qPCR reference genes.
- [nf-core/taxprofiler](https://corollary-labs.com/pipelines/taxprofiler): Classifies and profiles shotgun metagenomic reads across multiple tools and reference databases simultaneously.
- [nf-core/tbanalyzer](https://corollary-labs.com/pipelines/tbanalyzer): Bioinformatics pipeline for analysing members of the Mycobacterium tuberculosis complex.
- [nf-core/tfactivity](https://corollary-labs.com/pipelines/tfactivity): Identifies differentially active transcription factors across biological conditions using expression and chromatin data.
- [nf-core/troughgraph](https://corollary-labs.com/pipelines/troughgraph): Quantitative assessment of permafrost landscapes and regional thaw levels built using Nextflow.
- [nf-core/tumourevo](https://corollary-labs.com/pipelines/tumourevo): Models tumour clonal evolution and subclonal architecture from matched tumour-normal whole-genome sequencing data.
- [nf-core/variantbenchmarking](https://corollary-labs.com/pipelines/variantbenchmarking): Benchmarks germline and somatic small variants, structural variants and copy number alterations against truth sets.
- [nf-core/variantcatalogue](https://corollary-labs.com/pipelines/variantcatalogue): Generates population variant catalogues and allele frequencies from whole genome sequence data.
- [nf-core/variantprioritization](https://corollary-labs.com/pipelines/variantprioritization): Functional annotation and clinical interpretation of somatic and germline cancer variants using PCGR and CPSR.
- [nf-core/viralintegration](https://corollary-labs.com/pipelines/viralintegration): Identifies viral integration sites in host genomes from sequencing reads using a chimeric read alignment strategy.
- [nf-core/viralmetagenome](https://corollary-labs.com/pipelines/viralmetagenome): Reconstructs viral consensus genomes and detects intra-host single-nucleotide variants from metagenomic sequencing data.
- [nf-core/viralrecon](https://corollary-labs.com/pipelines/viralrecon): Assembly and intra-host variant calling for viral sequencing data from Illumina and Oxford Nanopore platforms.

## Datasets (browse rows, schema and statistics in the browser)

- [ABC-1M](https://corollary-labs.com/datasets/ADSKAILab/ABC-1M): A corpus of one million CAD-quality 3D shapes from the ABC dataset for engineering generative models.
- [codeparrot_megatron](https://corollary-labs.com/datasets/ADSKAILab/codeparrot_megatron): A Megatron-formatted release of the CodeParrot dataset for large-scale code language-model pretraining.
- [LLM-narrative-planning-taskset](https://corollary-labs.com/datasets/ADSKAILab/LLM-narrative-planning-taskset): A benchmark evaluating large language model planning and reasoning across multi-step engineering scenarios.
- [Zero-To-CAD-100k](https://corollary-labs.com/datasets/ADSKAILab/Zero-To-CAD-100k): A curated 100,000-sample dataset of diverse parametric CAD programs and multi-view renders for 3D model generation.
- [Zero-To-CAD-1m](https://corollary-labs.com/datasets/ADSKAILab/Zero-To-CAD-1m): Executable CAD construction sequences and multi-view renders for training parametric 3D computer-aided design models.
- [MedSynth](https://corollary-labs.com/datasets/Ahmad0067/MedSynth): Synthetic medical dialogue and SOAP note pairs to train clinical summarisation models without patient privacy risks.
- [aimo-validation-aime](https://corollary-labs.com/datasets/AI-MO/aimo-validation-aime): A validation benchmark of 90 recent AIME competition mathematics problems with integer solutions.
- [aimo-validation-amc](https://corollary-labs.com/datasets/AI-MO/aimo-validation-amc): A validation benchmark of 83 reformatted AMC 12 competition problems with integer answers for mathematical reasoning.
- [aimo-validation-math-level-4](https://corollary-labs.com/datasets/AI-MO/aimo-validation-math-level-4): Level-4 pre-calculus MATH benchmark problems with integer outputs for AIMO challenge validation.
- [aimo-validation-math-level-5](https://corollary-labs.com/datasets/AI-MO/aimo-validation-math-level-5): A dataset of highest-difficulty level-5 MATH problems with integer answers for AIMO challenge validation.
- [aops_raw](https://corollary-labs.com/datasets/AI-MO/aops_raw): A collection of 80,661 contest mathematics problems and solutions from the Art of Problem Solving forums.
- [CombiBench](https://corollary-labs.com/datasets/AI-MO/CombiBench): A Lean 4 benchmark of 100 formalised combinatorial mathematics problems for automated theorem proving systems.
- [GeometryLeanBench](https://corollary-labs.com/datasets/AI-MO/GeometryLeanBench): A Lean 4 geometry theorem proving benchmark for automated reasoning and mathematics research.
- [Kimina-Prover-Promptset](https://corollary-labs.com/datasets/AI-MO/Kimina-Prover-Promptset): A curated dataset of challenging mathematical problems in Lean 4 for reinforcement learning theorem proving.
- [minif2f_test](https://corollary-labs.com/datasets/AI-MO/minif2f_test): A corrected Lean 4 test set for the miniF2F formal mathematics benchmark.
- [NuminaMath-1.5](https://corollary-labs.com/datasets/AI-MO/NuminaMath-1.5): 860K+ competition math problems from 17 sources with verified solutions — the training backbone of the gold-medal solution at the 2024 AI Mathematical Olympiad.
- [NuminaMath-CoT](https://corollary-labs.com/datasets/AI-MO/NuminaMath-CoT): NuminaMath with Chain-of-Thought reasoning annotations.
- [NuminaMath-LEAN](https://corollary-labs.com/datasets/AI-MO/NuminaMath-LEAN): Formal Lean 4 dataset of 100K mathematical competition problems for training and testing automated theorem provers.
- [NuminaMath-TIR](https://corollary-labs.com/datasets/AI-MO/NuminaMath-TIR): A dataset of mathematical problems paired with tool-integrated reasoning paths and code execution outputs.
- [olympiads](https://corollary-labs.com/datasets/AI-MO/olympiads): Olympiad-level mathematical competition problems and solutions structured for mathematical reasoning models.
- [olympiads-ref](https://corollary-labs.com/datasets/AI-MO/olympiads-ref): A structured dataset of competition mathematics problems and verified solutions for reasoning models.
- [olympiads-ref-base](https://corollary-labs.com/datasets/AI-MO/olympiads-ref-base): A reference dataset of mathematical olympiad problems used as a foundation for NuminaMath training splits.
- [OpenTME](https://corollary-labs.com/datasets/Aignostics/OpenTME): Pre-analysed TCGA H&E whole-slide images with tumour-microenvironment spatial features from Aignostics.
- [olmoearth-paper-embeddings](https://corollary-labs.com/datasets/allenai/olmoearth-paper-embeddings): Precomputed Earth-observation foundation model embeddings across 24 downstream evaluation tasks.
- [peS2o](https://corollary-labs.com/datasets/allenai/peS2o): A multi-domain open-access scientific pretraining corpus of up to 40 million cleaned papers for language models.
- [ATB](https://corollary-labs.com/datasets/AllTheBacteria/ATB): A standardised dataset of bacterial coding and intergenic sequences from approximately two million genomes.
- [Bac-Corpus-dna-intergenic-sequences-high-diversity](https://corollary-labs.com/datasets/AllTheBacteria/Bac-Corpus-dna-intergenic-sequences-high-diversity): A clustered collection of 1.45 billion bacterial intergenic DNA sequences for pretraining genomic language models.
- [Bac-Corpus-protein-sequences-high-diversity](https://corollary-labs.com/datasets/AllTheBacteria/Bac-Corpus-protein-sequences-high-diversity): A corpus of approximately 940 million clustered bacterial protein sequences for pretraining language models.
- [SPIRE](https://corollary-labs.com/datasets/AllTheBacteria/SPIRE): A planetary-scale metagenomics resource aggregating global environmental microbiome samples for researchers.
- [BioMysteryBench-full](https://corollary-labs.com/datasets/Anthropic/BioMysteryBench-full): A biology evaluation suite designed to test expert-level scientific reasoning in frontier models.
- [BioMysteryBench-preview](https://corollary-labs.com/datasets/Anthropic/BioMysteryBench-preview): A preview benchmark evaluating AI scientific reasoning through anonymised bioinformatics analysis tasks.
- [opengenome2](https://corollary-labs.com/datasets/arcinstitute/opengenome2): A curated genomic database of 8.8 trillion base pairs across all domains of life for foundation model training.
- [Perturb-Sapiens](https://corollary-labs.com/datasets/arcinstitute/Perturb-Sapiens): A whole-organism atlas of predicted single-cell perturbation responses across 28 human tissues.
- [Replogle-Nadig-Preprint](https://corollary-labs.com/datasets/arcinstitute/Replogle-Nadig-Preprint): Perturb-seq single-cell perturbation dataset for perturbation-response modelling.
- [SE-167M-Human](https://corollary-labs.com/datasets/arcinstitute/SE-167M-Human): A dataset of 167 million human single-cell RNA expression profiles across diverse tissues and cell types.
- [Stack-CellxGene45M](https://corollary-labs.com/datasets/arcinstitute/Stack-CellxGene45M): A curated collection of 45 million single-cell transcriptomic profiles for model alignment and cross-study analysis.
- [State-Tahoe-Filtered](https://corollary-labs.com/datasets/arcinstitute/State-Tahoe-Filtered): A curated single-cell perturbation dataset for training and benchmarking cross-study cell-state models.
- [ThousandWorlds](https://corollary-labs.com/datasets/AstroAutomata/ThousandWorlds): A multi-simulator benchmark dataset of exoplanet general circulation models for spatial climate emulation.
- [DynaCLR-data](https://corollary-labs.com/datasets/biohub/DynaCLR-data): Live-cell microscopy time-series dataset with infection labels for dynamic contrastive cell embedding learning.
- [ESMC-SAE-Features](https://corollary-labs.com/datasets/biohub/ESMC-SAE-Features): A feature table annotating 16,384 latent features from a sparse autoencoder trained on the ESMC-6B protein model.
- [popsicle](https://corollary-labs.com/datasets/biohub/popsicle): A cryo-electron tomography benchmark for macromolecular localization and 3D cellular structure segmentation.
- [finemed-fr](https://corollary-labs.com/datasets/doctolib-lab/finemed-fr): A 19.2B-word French medical pretraining corpus drawn from web text, PDFs, and Wikipedia with multi-axis quality labels.
- [finemed-rephrased-fr](https://corollary-labs.com/datasets/doctolib-lab/finemed-rephrased-fr): A 13.6-million-document French medical corpus rewritten by large language models to amplify clinical entity density.
- [PATHOS-PLM-EMBEDDINGS](https://corollary-labs.com/datasets/DSIMB/PATHOS-PLM-EMBEDDINGS): Precomputed protein language model embeddings for wild-type and mutant human SwissProt proteins.
- [BEANS-Zero](https://corollary-labs.com/datasets/EarthSpeciesProject/BEANS-Zero): A bioacoustics benchmark evaluating zero-shot audio-language models on animal vocalisation tasks.
- [drug-target-activity](https://corollary-labs.com/datasets/eve-bio/drug-target-activity): Interaction and potency measurements for 1,397 mostly FDA-approved small molecule drugs across three target classes.
- [principia-bench](https://corollary-labs.com/datasets/facebook/principia-bench): A curated benchmark evaluating language models on deriving mathematical objects from STEM problems.
- [principia-collection](https://corollary-labs.com/datasets/facebook/principia-collection): A STEM reasoning dataset for training language models to derive mathematical objects and numerical answers.
- [atomvison_atomistic_stm_images_2d_materials_unique_chemical_compositions_structure_v1-1](https://corollary-labs.com/datasets/foundry-ml/atomvison_atomistic_stm_images_2d_materials_unique_chemical_compositions_structure_v1-1): Simulated scanning tunnelling microscopy images of 2D materials for machine learning on atomic-scale imaging.
- [atomvison_simulated_atomistic_stem_images_2d_materials_unique_chemical_compositions_structure_ba](https://corollary-labs.com/datasets/foundry-ml/atomvison_simulated_atomistic_stem_images_2d_materials_unique_chemical_compositions_structure_ba): Simulated STEM images of 2D materials paired with structure metadata for training machine learning models.
- [dataset_concrete_compressive_strength](https://corollary-labs.com/datasets/foundry-ml/dataset_concrete_compressive_strength): Concrete mix-design and compressive strength data for civil engineering and materials science machine learning.
- [dataset_debyet_aflow](https://corollary-labs.com/datasets/foundry-ml/dataset_debyet_aflow): A curated dataset of calculated Debye temperatures for 4,896 materials, designed for machine learning property prediction.
- [dataset_exfoliatione](https://corollary-labs.com/datasets/foundry-ml/dataset_exfoliatione): A tabular dataset of calculated exfoliation energies for 636 materials to support discovery of 2D layered compounds.
- [dataset_li_conductivity](https://corollary-labs.com/datasets/foundry-ml/dataset_li_conductivity): Experimental lithium-ion solid electrolyte conductivity data for 372 materials to support machine learning discovery.
- [dataset_metallicglass_dmax](https://corollary-labs.com/datasets/foundry-ml/dataset_metallicglass_dmax): Experimental maximum casting diameters and thermal properties for 998 metallic glasses to assess casting feasibility.
- [dataset_metallicglass_rc](https://corollary-labs.com/datasets/foundry-ml/dataset_metallicglass_rc): Experimental critical cooling rate measurements for 2,125 metallic glasses to support glass-forming ability modelling.
- [dataset_metallicglass_rc_llm](https://corollary-labs.com/datasets/foundry-ml/dataset_metallicglass_rc_llm): LLM-extracted critical cooling rates for 297 metallic glasses across research literature.
- [dataset_mg_alloy](https://corollary-labs.com/datasets/foundry-ml/dataset_mg_alloy): A tabular dataset of 365 biomedical magnesium alloys linking composition and processing to mechanical properties.
- [dataset_perovskite_asr](https://corollary-labs.com/datasets/foundry-ml/dataset_perovskite_asr): Perovskite oxide area-specific resistance data across four temperatures for solid-oxide fuel cell machine learning.
- [dataset_perovskite_conductivity](https://corollary-labs.com/datasets/foundry-ml/dataset_perovskite_conductivity): Perovskite conductivity measurements supporting machine learning screening for solid-oxide fuel cell electrolytes.
- [dataset_perovskite_formatione](https://corollary-labs.com/datasets/foundry-ml/dataset_perovskite_formatione): Formation energy data for 9,646 perovskite compounds to support machine learning screening for material stability.
- [dataset_perovskite_habs](https://corollary-labs.com/datasets/foundry-ml/dataset_perovskite_habs): Perovskite hydrogen absorption measurements for machine-learning screening of protonic solid oxide cell materials.
- [dataset_perovskite_stability_updated](https://corollary-labs.com/datasets/foundry-ml/dataset_perovskite_stability_updated): A curated DFT dataset for benchmarking machine learning models predicting perovskite stability and durability.
- [dataset_perovskite_tec](https://corollary-labs.com/datasets/foundry-ml/dataset_perovskite_tec): A curated dataset of 137 experimental thermal expansion coefficients for perovskite oxides across temperature ranges.
- [dataset_rpv_tts](https://corollary-labs.com/datasets/foundry-ml/dataset_rpv_tts): A tabular dataset of reactor pressure vessel steel measurements for predicting irradiation-induced embrittlement.
- [dataset_thermalcond_aflow](https://corollary-labs.com/datasets/foundry-ml/dataset_thermalcond_aflow): Thermal conductivity values and material compositions from AFLOW for machine learning screening of thermal materials.
- [dataset_thermalexp_aflow](https://corollary-labs.com/datasets/foundry-ml/dataset_thermalexp_aflow): A benchmark tabular dataset of 4,886 AFLOW thermal expansion coefficients for inorganic materials modelling.
- [dielectric_constant_v1-1](https://corollary-labs.com/datasets/foundry-ml/dielectric_constant_v1-1): DFT-calculated dielectric and optical properties of 1,056 inorganic compounds for machine learning material screening.
- [diffusion_v1-4](https://corollary-labs.com/datasets/foundry-ml/diffusion_v1-4): Tabular dataset of DFT-calculated solute migration barriers for 408 host-impurity pairs in dilute alloys.
- [direct_electron_detectorceleritas_xs_simulated_readout_images_electron_counting_model_v1-1](https://corollary-labs.com/datasets/foundry-ml/direct_electron_detectorceleritas_xs_simulated_readout_images_electron_counting_model_v1-1): Simulated Celeritas XS readout images for training electron-counting models in cryo-EM and STEM.
- [double_perovskite_bandgap_v1-1](https://corollary-labs.com/datasets/foundry-ml/double_perovskite_bandgap_v1-1): Computed band gaps of 1,306 double perovskite oxide compounds for machine learning screening in optoelectronics.
- [elastic_tensor_v1-1](https://corollary-labs.com/datasets/foundry-ml/elastic_tensor_v1-1): A tabular dataset of density functional theory elastic tensors and moduli for 1,181 inorganic crystalline compounds.
- [electromigration_v1-1](https://corollary-labs.com/datasets/foundry-ml/electromigration_v1-1): Electromigration dataset of effective charge values for 49 metal host-impurity pairs in interconnect materials.
- [elwood_md_v1-2](https://corollary-labs.com/datasets/foundry-ml/elwood_md_v1-2): Molecular-dynamics simulation dataset mapping monomer SMILES strings to physical and thermodynamic properties.
- [foundry_aflow_band_gaps_v1-1](https://corollary-labs.com/datasets/foundry-ml/foundry_aflow_band_gaps_v1-1): A curated dataset of AFLOW band gaps formatted for training and evaluating machine learning models.
- [foundry_g4mp2_solvation_v1-2](https://corollary-labs.com/datasets/foundry-ml/foundry_g4mp2_solvation_v1-2): High-accuracy G4MP2 thermochemical and solvation energy dataset across multiple solvents for machine learning.
- [foundry_moses_v1-1](https://corollary-labs.com/datasets/foundry-ml/foundry_moses_v1-1): A molecular dataset benchmark for training and evaluating generative models producing drug-like chemical structures.
- [foundry_mp_band_gaps_v1-1](https://corollary-labs.com/datasets/foundry-ml/foundry_mp_band_gaps_v1-1): A curated dataset of Materials Project band gaps for machine-learning benchmarking in inorganic electronic structure.
- [foundry_oqmd_band_gaps_v1-1](https://corollary-labs.com/datasets/foundry-ml/foundry_oqmd_band_gaps_v1-1): A dataset of Open Quantum Materials Database band gaps for machine learning on inorganic crystal electronic structure.
- [foundry_osdb_v1-1](https://corollary-labs.com/datasets/foundry-ml/foundry_osdb_v1-1): Tabular dataset of zeolite-guest poses and binding energies for machine learning property prediction and screening.
- [foundry_qmc_ml_v1-1](https://corollary-labs.com/datasets/foundry-ml/foundry_qmc_ml_v1-1): Quantum Monte Carlo and electronic structure benchmark dataset for 1,175 small molecules.
- [foundry_stan_segmentation_v1-1](https://corollary-labs.com/datasets/foundry-ml/foundry_stan_segmentation_v1-1): A materials electron-microscopy image segmentation dataset for machine learning feature extraction.
- [heusler_magnetization_v1-1](https://corollary-labs.com/datasets/foundry-ml/heusler_magnetization_v1-1): Saturation magnetisation and structural properties of 1,153 Heusler compounds for materials science discovery.
- [mask_rcnn_defect_detection_v1-1](https://corollary-labs.com/datasets/foundry-ml/mask_rcnn_defect_detection_v1-1): Microscopy dataset annotated for instance-segmentation defect detection in materials inspection.
- [perovskite_opbandcenter_v1-1](https://corollary-labs.com/datasets/foundry-ml/perovskite_opbandcenter_v1-1): DFT-calculated oxygen p-band centre energies across 2,912 perovskite oxides for catalytic property modelling.
- [perovskite_stability_v1-1](https://corollary-labs.com/datasets/foundry-ml/perovskite_stability_v1-1): A benchmark dataset of DFT-calculated thermodynamic stability values for 1,929 perovskite oxide compositions.
- [piezoelectric_tensor_v1-1](https://corollary-labs.com/datasets/foundry-ml/piezoelectric_tensor_v1-1): A density functional theory dataset of piezoelectric tensors for 941 inorganic materials for machine learning.
- [semiconductor_defectlevels_v1-1](https://corollary-labs.com/datasets/foundry-ml/semiconductor_defectlevels_v1-1): Computed defect charge state transition levels for 2,910 semiconductor-impurity pairs to predict doping and trap states.
- [steel_strength_v1-1](https://corollary-labs.com/datasets/foundry-ml/steel_strength_v1-1): A tabular dataset of elemental compositions and mechanical properties for 312 steel alloys to support alloy design.
- [superconductivity_v1-1](https://corollary-labs.com/datasets/foundry-ml/superconductivity_v1-1): A tabular dataset of experimentally measured superconducting critical temperatures for 16,414 material compositions.
- [training_locating_atoms_stem_images_v1-2](https://corollary-labs.com/datasets/foundry-ml/training_locating_atoms_stem_images_v1-2): A training dataset of scanning transmission electron microscopy images for localising atomic positions.
- [wolverton_oxides_v1-1](https://corollary-labs.com/datasets/foundry-ml/wolverton_oxides_v1-1): DFT-computed thermodynamic and electronic properties of 4,914 ABO3 perovskite oxides for materials machine learning.
- [medical-o1-reasoning-SFT](https://corollary-labs.com/datasets/FreedomIntelligence/medical-o1-reasoning-SFT): A medical chain-of-thought reasoning dataset for supervised fine-tuning of language models.
- [medical-o1-verifiable-problem](https://corollary-labs.com/datasets/FreedomIntelligence/medical-o1-verifiable-problem): Verifiable medical reasoning problems with ground-truth answers for aligning and training medical language models.
- [BixBench](https://corollary-labs.com/datasets/futurehouse/BixBench): A benchmark of 205 reproducible scientific questions paired with data capsules to evaluate AI systems.
- [ether0-benchmark](https://corollary-labs.com/datasets/futurehouse/ether0-benchmark): A chemistry reasoning benchmark for language models featuring tasks where all answers are molecular structures.
- [lab-bench](https://corollary-labs.com/datasets/futurehouse/lab-bench): A biology benchmark for evaluating AI systems on scientific research tasks from protocols to molecular cloning.
- [crosslab-nmc](https://corollary-labs.com/datasets/GenData-Research/crosslab-nmc): A benchmark and evidence corpus for evaluating scientific claims about battery cathodes against literature evidence.
- [GDPa1](https://corollary-labs.com/datasets/ginkgo-datapoints/GDPa1): A biophysical developability benchmark providing experimental assay measurements for 242 antibodies.
- [GDPx1](https://corollary-labs.com/datasets/ginkgo-datapoints/GDPx1): A functional genomics DRUG-seq dataset measuring chemical perturbations in A549 cells.
- [GDPx2](https://corollary-labs.com/datasets/ginkgo-datapoints/GDPx2): DRUG-seq transcriptomic profiling across four primary human cell types treated with 85 compounds.
- [GDPx3](https://corollary-labs.com/datasets/ginkgo-datapoints/GDPx3): A high-content Cell Painting imaging dataset for training machine learning models in drug discovery.
- [GDPx4](https://corollary-labs.com/datasets/ginkgo-datapoints/GDPx4): DRUG-seq transcriptomic dataset for studying gene-drug interactions in engineered HEK293 cells.
- [spiqa](https://corollary-labs.com/datasets/google/spiqa): A multimodal benchmark for evaluating question answering across scientific paper figures, tables, and text.
- [carbon-pretraining-corpus](https://corollary-labs.com/datasets/HuggingFaceBio/carbon-pretraining-corpus): A 1.1-trillion-nucleotide DNA and RNA corpus spanning eukaryotes and prokaryotes for genomic foundation models.
- [clinvar-vep](https://corollary-labs.com/datasets/HuggingFaceBio/clinvar-vep): A ClinVar variant effect prediction benchmark for evaluating genomic language models on variant pathogenicity.
- [genomic-niah](https://corollary-labs.com/datasets/HuggingFaceBio/genomic-niah): A long-context retrieval benchmark testing genomic language models on DNA sequences from 24 to 786 kilobase pairs.
- [perturbation-bench](https://corollary-labs.com/datasets/HuggingFaceBio/perturbation-bench): Evaluates DNA foundation models on pairwise sequence discrimination with codon and CAG repeat perturbations.
- [traitgym](https://corollary-labs.com/datasets/HuggingFaceBio/traitgym): Evaluates genomic foundation models predicting causal non-coding variant effects across diverse human phenotypes.
- [TreeOfLife-200M](https://corollary-labs.com/datasets/imageomics/TreeOfLife-200M): A 233-million-image biodiversity dataset across 933,798 taxa for training biological foundation vision models.
- [multi_species_genomes](https://corollary-labs.com/datasets/InstaDeepAI/multi_species_genomes): A 174-billion-nucleotide multi-species genomic dataset for pre-training DNA foundation models across diverse taxa.
- [NTv3_benchmark_dataset](https://corollary-labs.com/datasets/InstaDeepAI/NTv3_benchmark_dataset): A multi-species genomic benchmark dataset providing functional tracks and structural genome annotations.
- [nucleotide_transformer_downstream_tasks](https://corollary-labs.com/datasets/InstaDeepAI/nucleotide_transformer_downstream_tasks): A collection of 18 genomic classification benchmark tasks covering promoters, enhancers, splice sites, and histone marks.
- [plant-genomic-benchmark](https://corollary-labs.com/datasets/InstaDeepAI/plant-genomic-benchmark): A multi-species plant genomics benchmark evaluating sequence models across eight regulatory and expression tasks.
- [true-cds-protein-tasks](https://corollary-labs.com/datasets/InstaDeepAI/true-cds-protein-tasks): A genomic benchmark evaluating machine learning models on five protein tasks paired with true coding sequences.
- [winnow-ms-datasets](https://corollary-labs.com/datasets/InstaDeepAI/winnow-ms-datasets): Tandem mass spectrometry datasets and peptide predictions for calibration and FDR estimation in de novo sequencing.
- [CloudSEN12Plus](https://corollary-labs.com/datasets/isp-uv-es/CloudSEN12Plus): A benchmark dataset of over 49,000 Sentinel-2 patches with expert annotations for cloud and shadow segmentation.
- [opensr-test](https://corollary-labs.com/datasets/isp-uv-es/opensr-test): A benchmark dataset and evaluation suite for testing real-world Sentinel-2 satellite imagery super-resolution.
- [rtm_emulation](https://corollary-labs.com/datasets/isp-uv-es/rtm_emulation): A benchmark dataset for training fast machine learning emulators to replace atmospheric radiative transfer models.
- [WorldFloodsv2](https://corollary-labs.com/datasets/isp-uv-es/WorldFloodsv2): A global remote sensing dataset pairing optical satellite imagery with flood extent segmentation masks.
- [ChemBench](https://corollary-labs.com/datasets/jablonkagroup/ChemBench): A curated evaluation benchmark assessing the chemistry and materials science reasoning of large language models.
- [chempile-caption](https://corollary-labs.com/datasets/jablonkagroup/chempile-caption): An image-to-text dataset of chemistry figures and captions sourced from LibreTexts for multimodal models.
- [chempile-code](https://corollary-labs.com/datasets/jablonkagroup/chempile-code): A curated Python code corpus for training models on computational chemistry, biology, and materials workflows.
- [chempile-education](https://corollary-labs.com/datasets/jablonkagroup/chempile-education): A chemistry education text dataset covering textbooks, university lectures, and competition problems for language models.
- [chempile-instruction](https://corollary-labs.com/datasets/jablonkagroup/chempile-instruction): A multi-turn dialogue dataset for instruction tuning chemistry-focused large language models across multiple subdomains.
- [chempile-lift](https://corollary-labs.com/datasets/jablonkagroup/chempile-lift): A 35.5-billion-token dataset of natural language question-answer pairs for chemical property prediction using LLMs.
- [chempile-mlift](https://corollary-labs.com/datasets/jablonkagroup/chempile-mlift): A multimodal chemistry dataset of 61.5 million molecule images and text representations for property prediction.
- [chempile-paper](https://corollary-labs.com/datasets/jablonkagroup/chempile-paper): A corpus of 13.9 billion tokens from chemistry papers and preprints for training and evaluating language models.
- [chempile-reasoning](https://corollary-labs.com/datasets/jablonkagroup/chempile-reasoning): Multi-step chemistry reasoning corpus — open-domain QA, NLI, and multiple-choice items with chains of reasoning for training and evaluating chemical reasoning models.
- [MaCBench](https://corollary-labs.com/datasets/jablonkagroup/MaCBench): A multimodal benchmark assessing vision large language models across chemistry and materials science.
- [Atompack](https://corollary-labs.com/datasets/LeMaterial/Atompack): Public atomistic machine learning datasets distributed in the Atompack format for scalable multi-worker training.
- [LeMat-Bulk](https://corollary-labs.com/datasets/LeMaterial/LeMat-Bulk): Primary bulk materials database aggregating 1M+ crystal structures with DFT-computed formation energies, band gaps, and elastic properties for materials discovery.
- [LeMat-Bulk-DFT-Hull](https://corollary-labs.com/datasets/LeMaterial/LeMat-Bulk-DFT-Hull): Convex hull reference data for materials stability analysis across DFT and machine learning interatomic potentials.
- [LeMat-Bulk-DFT-Hull-All](https://corollary-labs.com/datasets/LeMaterial/LeMat-Bulk-DFT-Hull-All): A bulk materials DFT convex hull dataset for computational materials science and stability discovery.
- [LeMat-Bulk-MLIP-Formation-Energies](https://corollary-labs.com/datasets/LeMaterial/LeMat-Bulk-MLIP-Formation-Energies): Formation energies computed across multiple MLIPs for bulk materials stability and benchmarking.
- [LeMat-Bulk-MLIP-Hull](https://corollary-labs.com/datasets/LeMaterial/LeMat-Bulk-MLIP-Hull): Convex hull stability data for bulk materials computed with machine-learning interatomic potentials and DFT.
- [LeMat-BulkUnique](https://corollary-labs.com/datasets/LeMaterial/LeMat-BulkUnique): A deduplicated dataset of over five million inorganic crystal structures curated for generative materials modelling.
- [LeMat-GenBench-embeddings](https://corollary-labs.com/datasets/LeMaterial/LeMat-GenBench-embeddings): Precomputed embeddings for evaluating generative materials models on crystal-structure prediction tasks.
- [LeMat-Rho](https://corollary-labs.com/datasets/LeMaterial/LeMat-Rho): Electronic charge densities, Bader charges, and r2SCAN DFT properties for 69,117 inorganic crystal structures.
- [LeMat-Synth](https://corollary-labs.com/datasets/LeMaterial/LeMat-Synth): A curated materials synthesis dataset for synthesizability prediction and synthesis-route generation.
- [LeMat-Synth-Papers](https://corollary-labs.com/datasets/LeMaterial/LeMat-Synth-Papers): A curated corpus of materials literature for training and evaluating synthesis procedure extraction models.
- [LeMat-Traj](https://corollary-labs.com/datasets/LeMaterial/LeMat-Traj): A large-scale dataset of bulk material molecular dynamics and relaxation trajectories for machine learning potentials.
- [LeMat-Traj-subset](https://corollary-labs.com/datasets/LeMaterial/LeMat-Traj-subset): Curated molecular-dynamics trajectory subset for prototyping and small-scale interatomic potential training.
- [AlphaFoldDB](https://corollary-labs.com/datasets/LiteFold/AlphaFoldDB): AlphaFoldDB prediction index — open database of predicted protein 3D structures with confidence scores, providing structural coverage for known protein sequences at scale.
- [BFD](https://corollary-labs.com/datasets/LiteFold/BFD): An indexed distribution of the BFD clustered protein sequence database for homology search and alignment.
- [CATH](https://corollary-labs.com/datasets/LiteFold/CATH): Hierarchical protein domain classification dataset structured into train and test splits partitioned by S35 clusters.
- [CycPepMPDB](https://corollary-labs.com/datasets/LiteFold/CycPepMPDB): Curated experimental membrane permeability measurements and structural data for cyclic peptides.
- [DisProt](https://corollary-labs.com/datasets/LiteFold/DisProt): Curated database of intrinsically disordered proteins and regions with experimental evidence and functional annotations.
- [ESMAtlas](https://corollary-labs.com/datasets/LiteFold/ESMAtlas): Predicted metagenomic protein structures from MGnify generated with ESMFold and packaged as WebDataset shards.
- [Evolutionary](https://corollary-labs.com/datasets/LiteFold/Evolutionary): Precomputed MMseqs2-style evolutionary multiple sequence alignment data for protein language models and folding.
- [FireProtDB](https://corollary-labs.com/datasets/LiteFold/FireProtDB): FireProtDB — curated thermostability mutation data for proteins, supporting protein-engineering and stability-prediction modelling.
- [FLIP2](https://corollary-labs.com/datasets/LiteFold/FLIP2): A protein engineering benchmark for evaluating fitness prediction models under realistic distribution shifts.
- [GO](https://corollary-labs.com/datasets/LiteFold/GO): A structured tabular Parquet version of Gene Ontology terms for functional annotation across organisms.
- [GOA](https://corollary-labs.com/datasets/LiteFold/GOA): Evidence-coded Gene Ontology annotations for UniProtKB proteins, RNAs, and protein complexes.
- [HumanProteinAtlas](https://corollary-labs.com/datasets/LiteFold/HumanProteinAtlas): Tabular gene and protein expression annotations across human tissues, pathology, and subcellular compartments.
- [IEDB](https://corollary-labs.com/datasets/LiteFold/IEDB): A tabular export of experimentally characterised immune epitopes and assays across diverse immunological contexts.
- [InterPro](https://corollary-labs.com/datasets/LiteFold/InterPro): Integrated protein family, domain, and functional-site signatures for sequence classification and annotation.
- [MegaScale-Tsuboyama2023](https://corollary-labs.com/datasets/LiteFold/MegaScale-Tsuboyama2023): Experimental thermodynamic protein folding stabilities across mutations for training protein stability predictors.
- [Mgnify](https://corollary-labs.com/datasets/LiteFold/Mgnify): A multi-terabyte index and distribution of microbiome protein sequences and tables from EMBL-EBI MGnify.
- [NCBI](https://corollary-labs.com/datasets/LiteFold/NCBI): A Parquet shard index of NCBI RefSeq curated protein sequences and metadata for biological research.
- [OpenProteinSet](https://corollary-labs.com/datasets/LiteFold/OpenProteinSet): A massive open-source training corpus of protein sequence alignments and template structures for structure prediction.
- [PDB](https://corollary-labs.com/datasets/LiteFold/PDB): Tabular index and mirror of 98,824 Protein Data Bank mmCIF macromolecular structures for structural machine learning.
- [PDB-CCD](https://corollary-labs.com/datasets/LiteFold/PDB-CCD): A Parquet dataset of the Protein Data Bank Chemical Component Dictionary for structural machine learning.
- [Pfam](https://corollary-labs.com/datasets/LiteFold/Pfam): A JSONL mirror of Pfam-A protein domain annotations and clan mappings with row-level provenance.
- [ProteinGym](https://corollary-labs.com/datasets/LiteFold/ProteinGym): A standardized benchmark suite evaluating protein fitness and variant-effect predictors across millions of mutations.
- [protenix-data](https://corollary-labs.com/datasets/LiteFold/protenix-data): Preprocessed biomolecular dataset for training AlphaFold3-style all-atom structure prediction models.
- [SKEMPI2](https://corollary-labs.com/datasets/LiteFold/SKEMPI2): Experimental benchmark of protein-protein binding affinity, kinetics, and thermodynamics changes upon mutation.
- [STRING](https://corollary-labs.com/datasets/LiteFold/STRING): A database of physical and functional protein-protein interactions with sequence and organism metadata.
- [SwissSidechain](https://corollary-labs.com/datasets/LiteFold/SwissSidechain): Database of non-natural amino-acid sidechains providing 3D coordinates, topologies, and rotamer libraries.
- [UniProtKB](https://corollary-labs.com/datasets/LiteFold/UniProtKB): Processed UniProtKB release — comprehensive, high-quality protein sequence and functional annotation knowledgebase (Swiss-Prot + TrEMBL) reformatted for ML pipelines.
- [UniRef50](https://corollary-labs.com/datasets/LiteFold/UniRef50): UniProt reference cluster representatives at 50% sequence identity for protein machine learning and sequence searches.
- [UniRef90](https://corollary-labs.com/datasets/LiteFold/UniRef90): A shard index of UniProt protein sequences clustered at 90% identity for sequence search and language model training.
- [Botanic1-pretraining](https://corollary-labs.com/datasets/living-models/Botanic1-pretraining): A genomic pre-training dataset containing 7.2 million 8,192-base-pair windows across 320 land plant species.
- [CCA](https://corollary-labs.com/datasets/luminary-shift/CCA): A dataset of compressor and turbomachinery simulations for machine learning-assisted aerospace design.
- [Pump](https://corollary-labs.com/datasets/luminary-shift/Pump): Centrifugal pump CFD simulations across operating conditions for training turbomachinery surrogate models.
- [SHIFT-Crash](https://corollary-labs.com/datasets/luminary-shift/SHIFT-Crash): A vehicle crash-simulation dataset capturing structural deformation under impact for safety and physics modelling.
- [Submarine](https://corollary-labs.com/datasets/luminary-shift/Submarine): Submarine hydrodynamics CFD dataset providing submerged-body flow simulations for marine engineering.
- [SUV](https://corollary-labs.com/datasets/luminary-shift/SUV): A CFD dataset of SUV-class vehicles for training machine learning models on automotive aerodynamics.
- [WING](https://corollary-labs.com/datasets/luminary-shift/WING): Wing-flow computational fluid dynamics dataset for machine learning-based aerodynamic surrogate modelling.
- [hest](https://corollary-labs.com/datasets/MahmoodLab/hest): A paired histology and spatial transcriptomics dataset linking 1,276 profiles to high-resolution whole slide images.
- [hest-bench](https://corollary-labs.com/datasets/MahmoodLab/hest-bench): A pathology benchmark evaluating spatial-transcriptomics and pathology models on aligned histology and expression tasks.
- [Patho-Bench](https://corollary-labs.com/datasets/MahmoodLab/Patho-Bench): A benchmark dataset of standardised splits and labels for evaluating whole-slide image foundation models.
- [TCGA-UniformTumor-8K](https://corollary-labs.com/datasets/MahmoodLab/TCGA-UniformTumor-8K): A pan-cancer pathology benchmark of 25,495 high-resolution image regions for evaluating slide encoders.
- [UNI2-h-features](https://corollary-labs.com/datasets/MahmoodLab/UNI2-h-features): Precomputed UNI2-h patch features for TCGA, CPTAC, and PANDA slides to support downstream pathology models.
- [viper](https://corollary-labs.com/datasets/MahmoodLab/viper): A benchmark for assessing vision-language models on rat histopathology in veterinary and toxicologic pathology.
- [B3DB](https://corollary-labs.com/datasets/maomlab/B3DB): Curated blood-brain barrier permeability measurements and chemical descriptors for central nervous system drug discovery.
- [ChAFF](https://corollary-labs.com/datasets/maomlab/ChAFF): Curated chemistry dataset for machine learning benchmarking on molecular properties and bioassays.
- [CryptoCEN](https://corollary-labs.com/datasets/maomlab/CryptoCEN): A co-expression network for Cryptococcus neoformans to predict gene function and prioritise drug targets.
- [Molecule3D](https://corollary-labs.com/datasets/maomlab/Molecule3D): A dataset of roughly four million DFT-derived 3D molecular structures and quantum properties for geometric learning.
- [TDC](https://corollary-labs.com/datasets/maomlab/TDC): Curated datasets spanning 22 single-instance prediction tasks for machine learning in therapeutic discovery.
- [MetaMathQA](https://corollary-labs.com/datasets/meta-math/MetaMathQA): A mathematical question-answering dataset augmented from GSM8K and MATH to train and evaluate mathematical reasoning.
- [msr-acc-tae25](https://corollary-labs.com/datasets/microsoft/msr-acc-tae25): A dataset of 73,040 coupled-cluster total atomization energies for evaluating and training quantum chemistry models.
- [miriad-4.4M](https://corollary-labs.com/datasets/miriad/miriad-4.4M): A 4.4-million-pair medical question-answering corpus distilled from peer-reviewed literature for LLM workflows.
- [miriad-5.8M](https://corollary-labs.com/datasets/miriad/miriad-5.8M): A dataset of 5.8 million biomedical question-answer pairs for training and evaluating healthcare language models.
- [excess-properties](https://corollary-labs.com/datasets/mist-models/excess-properties): Curated dataset of temperature- and composition-dependent excess properties for chemical mixtures.
- [core-sdo](https://corollary-labs.com/datasets/nasa-ibm-ai4science/core-sdo): Multi-modal Solar Dynamics Observatory dataset for solar foundation models and space weather applications.
- [surya-bench-flare-forecasting](https://corollary-labs.com/datasets/nasa-ibm-ai4science/surya-bench-flare-forecasting): A full-disk solar flare forecasting benchmark dataset based on NOAA GOES X-ray observations from 2010 to 2024.
- [EO-via-NLP](https://corollary-labs.com/datasets/nasa-impact/EO-via-NLP): A dataset pairing Earth observation data and text to support natural language interfaces and entity recognition.
- [WxC-Bench](https://corollary-labs.com/datasets/nasa-impact/WxC-Bench): A standardised benchmark for evaluating AI models across six atmospheric and earth science downstream tasks.
- [drivaerml](https://corollary-labs.com/datasets/neashton/drivaerml): High-fidelity scale-resolving automotive CFD dataset of morphed DrivAer cars for machine learning.
- [SARLO-80](https://corollary-labs.com/datasets/ONERA/SARLO-80): 119K paired SAR/optical images with text captions at 80cm resolution for multimodal learning.
- [Octant_CYP_inhibition_reactivity_blog_release](https://corollary-labs.com/datasets/openadmet/Octant_CYP_inhibition_reactivity_blog_release): Cytochrome P450 inhibition and metabolic reactivity screening data across diverse small molecules for ADMET modelling.
- [openadmet-expansionrx-challenge-data](https://corollary-labs.com/datasets/openadmet/openadmet-expansionrx-challenge-data): ADMET profiling data for RNA-targeted small molecules across multiple pharmacokinetic endpoints.
- [openadmet-expansionrx-challenge-train-data](https://corollary-labs.com/datasets/openadmet/openadmet-expansionrx-challenge-train-data): Training dataset containing real-world ADMET endpoints from Expansion Therapeutics drug discovery campaigns.
- [frontierscience](https://corollary-labs.com/datasets/openai/frontierscience): A benchmark assessing AI models on expert-level reasoning tasks across the natural sciences.
- [healthbench](https://corollary-labs.com/datasets/openai/healthbench): A medical benchmark evaluating AI assistants on multi-turn conversations using physician-written rubrics.
- [healthbench-professional](https://corollary-labs.com/datasets/openai/healthbench-professional): A benchmark of physician-graded clinical dialogues for evaluating language models on expert-level health tasks.
- [MedDialog](https://corollary-labs.com/datasets/OpenMed/MedDialog): A dataset of 251,731 patient-doctor dialogue pairs for training and evaluating clinical conversation models.
- [Medical-Reasoning-SFT-Mega](https://corollary-labs.com/datasets/OpenMed/Medical-Reasoning-SFT-Mega): A supervised fine-tuning dataset of 1.79 million medical chain-of-thought reasoning conversations across seven models.
- [synthvision-annotated-kimi](https://corollary-labs.com/datasets/OpenMed/synthvision-annotated-kimi): A synthetic medical imaging dataset annotated by Kimi K2.5 for visual question answering and diagnostic evaluation.
- [synthvision-annotated-qwen](https://corollary-labs.com/datasets/OpenMed/synthvision-annotated-qwen): A synthetic medical imaging dataset annotated by Qwen 3.5 for training multimodal clinical models.
- [synthvision-seeds](https://corollary-labs.com/datasets/OpenMed/synthvision-seeds): A curated collection of 119,137 medical image records serving as seeds for the SynthVision data generation pipeline.
- [OAS](https://corollary-labs.com/datasets/opig/OAS): A curated database of antibody sequences from immune repertoire studies for antibody machine learning.
- [MofasaDB](https://corollary-labs.com/datasets/Orbital-Materials/MofasaDB): Metal-organic framework dataset from Orbital — large-scale curated MOF structures for materials-discovery ML and synthetic chemistry workflows.
- [camelyon16-features](https://corollary-labs.com/datasets/owkin/camelyon16-features): Pre-extracted Phikon features from CAMELYON16 whole slide images for benchmarking cancer classification methods.
- [her2-challenge-2026](https://corollary-labs.com/datasets/owkin/her2-challenge-2026): A computational pathology dataset for predicting breast cancer HER2 status from whole-slide histology images.
- [nct-crc-he](https://corollary-labs.com/datasets/owkin/nct-crc-he): Colorectal cancer histopathology patch dataset across nine tissue classes for image classification benchmarking.
- [plism-dataset-tiles](https://corollary-labs.com/datasets/owkin/plism-dataset-tiles): Histology tile dataset across 7 scanners and 13 staining conditions for benchmarking foundation model robustness.
- [AirfRANS_original](https://corollary-labs.com/datasets/PLAID-datasets/AirfRANS_original): Graph-structured CFD dataset of RANS simulations over NACA airfoils for physics learning and graph neural networks.
- [active_matter](https://corollary-labs.com/datasets/polymathic-ai/active_matter): High-fidelity simulations of biological active matter for benchmarking neural PDE solvers and collective dynamics.
- [MHD_64](https://corollary-labs.com/datasets/polymathic-ai/MHD_64): A 3D compressible magnetohydrodynamics turbulence dataset at 64³ resolution for neural operator benchmarking.
- [planetswe](https://corollary-labs.com/datasets/polymathic-ai/planetswe): Spherical rotating shallow-water simulations with Earth-like topography and periodic forcings for atmospheric surrogate models.
- [rayleigh_benard](https://corollary-labs.com/datasets/polymathic-ai/rayleigh_benard): A 2D thermal convection simulation dataset for benchmarking surrogate models across varying flow regimes.
- [supernova_explosion_64](https://corollary-labs.com/datasets/polymathic-ai/supernova_explosion_64): Hydrodynamic simulation dataset of supernova blastwaves propagating through turbulent interstellar gas clouds.
- [constellaration](https://corollary-labs.com/datasets/proxima-fusion/constellaration): A dataset of quasi-isodynamic stellarator plasma boundaries and MHD equilibria for fusion energy optimization.
- [TCGA_virtual_spatial_transcriptomics_atlas](https://corollary-labs.com/datasets/ratschlab/TCGA_virtual_spatial_transcriptomics_atlas): A virtual spatial transcriptomics atlas across 32 cancer types from 28,664 whole-slide images in TCGA.
- [rxrx3](https://corollary-labs.com/datasets/recursionpharma/rxrx3): High-content microscopy dataset covering cellular perturbations in human cells for drug discovery researchers.
- [rxrx3-core](https://corollary-labs.com/datasets/recursionpharma/rxrx3-core): Curated phenomics dataset of Cell Painting images and embeddings for benchmarking drug-target interactions.
- [equational-theories-benchmark](https://corollary-labs.com/datasets/SAIRfoundation/equational-theories-benchmark): A mathematical reasoning benchmark evaluating whether language models can determine equational implications over magmas.
- [equational-theories-selected-problems](https://corollary-labs.com/datasets/SAIRfoundation/equational-theories-selected-problems): Curated mathematical implication problems for benchmarking automated theorem proving and language model reasoning.
- [aqcat25-dataset](https://corollary-labs.com/datasets/SandboxAQ/aqcat25-dataset): 13.5M DFT calculation trajectories for heterogeneous catalysis and machine learning potential training.
- [SAIR](https://corollary-labs.com/datasets/SandboxAQ/SAIR): Over one million protein-ligand pairs with cofolded 3D structures and binding affinities for drug discovery.
- [EmeraldBay](https://corollary-labs.com/datasets/tahoebio/EmeraldBay): Single-cell perturbation dataset pairing transcriptomics and drug sensitivity across 52 cell lines and 91 treatments.
- [parse-de-rhaister](https://corollary-labs.com/datasets/tahoebio/parse-de-rhaister): Differential-expression statistics for 12 donors, 18 immune cell types, and 90 cytokines from a Parse PBMC screen.
- [replogle-nadig-de-rhaister](https://corollary-labs.com/datasets/tahoebio/replogle-nadig-de-rhaister): Differential expression statistics across four cell lines and ~2,000 CRISPR knockdowns used to train Rhaister.
- [Tahoe-100M](https://corollary-labs.com/datasets/tahoebio/Tahoe-100M): Single-cell transcriptomic atlas of over 100 million profiles across 50 cancer cell lines and 1,100 small molecules.
- [tahoe-de-rhaister](https://corollary-labs.com/datasets/tahoebio/tahoe-de-rhaister): Differential-expression summary statistics across 50 cancer cell lines and 384 drug perturbations.
- [Tahoe-x1-embeddings](https://corollary-labs.com/datasets/tahoebio/Tahoe-x1-embeddings): Pre-computed single-cell embeddings from the Tahoe-x1-3B foundation model across cancer cell lines and small molecules.
- [OG](https://corollary-labs.com/datasets/tattabio/OG): A curated prokaryotic and viral genomic corpus formatted with mixed amino acid and nucleotide sequences.
- [OMG](https://corollary-labs.com/datasets/tattabio/OMG): A 3.1-trillion base pair mixed-modality metagenomic corpus for pretraining genomic language models.
- [arxiv-abstracts-large](https://corollary-labs.com/datasets/UniverseTBD/arxiv-abstracts-large): A corpus of 1.7 million arXiv paper abstracts for training scientific and astronomy language models.
- [AstroLLaVA_convos](https://corollary-labs.com/datasets/UniverseTBD/AstroLLaVA_convos): Astronomical images with captions and synthetic dialogues for training multimodal vision-language models.
- [bioreason-pro-sft-reasoning-data](https://corollary-labs.com/datasets/wanglab/bioreason-pro-sft-reasoning-data): A training corpus of biological reasoning traces and annotations for supervised fine-tuning of protein models.
- [CT_DeepLesion-MedSAM2](https://corollary-labs.com/datasets/wanglab/CT_DeepLesion-MedSAM2): CT lesion segmentation mask annotations generated using MedSAM2 for model training and evaluation.
- [img_virus_plasmid](https://corollary-labs.com/datasets/wanglab/img_virus_plasmid): A catalogue combining uncultivated viral genomes from IMG/VR v4 and plasmid sequences from IMG/PR with metadata.
- [kegg](https://corollary-labs.com/datasets/wanglab/kegg): KEGG pathway entries paired with genomic variants and reasoning traces to evaluate mechanistic disease prediction.
- [X-Atlas-Orion](https://corollary-labs.com/datasets/Xaira-Therapeutics/X-Atlas-Orion): A genome-wide Perturb-seq atlas of 8 million human cells across 18,903 protein-coding gene knockdowns.
- [X-Atlas-Pisces](https://corollary-labs.com/datasets/Xaira-Therapeutics/X-Atlas-Pisces): A single-cell Perturb-seq atlas pairing CRISPRi knockdowns with transcriptomes across 16 biological contexts.

## Optional

- [Full text of every page](https://corollary-labs.com/llms-full.txt): Every write-up above in full: overview, how it works, uses, strengths, limits and questions.
- [Science AI Radar](https://corollary-labs.com/radar): Daily, source-checked notes on new AI releases for biology, chemistry and materials science.