Single-Cell Biology · Xaira-Therapeutics
X-Atlas-Pisces
A single-cell Perturb-seq atlas pairing CRISPRi knockdowns with transcriptomes across 16 biological contexts.
What X-Atlas-Pisces is
X-Atlas-Pisces is a single-cell Perturb-seq dataset developed by Xaira Therapeutics. It pairs targeted genetic knockdowns via CRISPR interference (CRISPRi) with single-cell gene expression readouts, covering 25.6 million perturbed transcriptomes across 16 biologically diverse contexts.
The dataset addresses the need for large-scale, paired perturbation and expression profiles across varied cellular backgrounds. It serves as an experimental resource for training and evaluating the X-Cell model in single-cell biology, medicine, and genomics.
How it was made
The atlas compiles CRISPRi perturbation data from diverse biological systems, including standard cell lines (HCT116, HEK293T, and HepG2), induced pluripotent stem cells (iPSCs), multi-lineage differentiating iPSCs, and Jurkat T lymphoma cells in both resting and CD3/CD28-activated states.
Perturbation experiments use dual-guide single-cell RNA sequencing. Subsets within the atlas include cells from X-Atlas/Orion covering HCT116 and HEK293T screens, cells with dual guides targeting a 200-gene validation set or non-targeting controls from HepG2 and iPSC screens, and dual-guide-containing cells from resting and active Jurkat screens.
Key facts
- Dataset size
- 25.6 million single-cell transcriptomes
- Biological contexts
- 16 diverse contexts
- Perturbation technology
- CRISPRi dual-guide knockdowns
- Context types
- HCT116, HEK293T, HepG2, iPSCs, differentiating iPSCs, resting and activated Jurkat T cells
- Intended application
- Training and evaluating the X-Cell model
What you can use it for
- Machine learning model training
- Supplying paired perturbation and single-cell gene expression data to train and benchmark predictive models such as X-Cell.
- Context-specific perturbation mapping
- Comparing transcriptomic responses to genetic knockdowns across different cell lineages, including stem cells and transformed lines.
- Immune activation profiling
- Examining the impact of genetic perturbations in resting versus CD3/CD28-activated Jurkat T lymphoma cells.
Explore the data
No queryable preview
This dataset is not available in a queryable form — usually an unusual file format or a repository too large to convert. The original source may still be readable.Strengths and limits
Good at
- Encompasses 25.6 million single-cell transcriptomes across 16 diverse cellular contexts.
- Captures functional genomic responses in both basal cell lines and dynamic states, such as differentiating stem cells and activated T cells.
- Employs dual-guide CRISPRi to link defined gene knockdowns with single-cell transcriptomic shifts.
Watch out for
- Perturbations in HepG2 and iPSC screens are confined to a 200-gene validation set and non-targeting controls.
- Portions of the screens are designated for scheduled addition rather than immediate inclusion.
Questions
Can I look inside X-Atlas-Pisces without downloading it?
Not here: it is not available in a queryable form. Its description and statistics are still shown.
What licence is X-Atlas-Pisces under?
Its authors declare cc-by-nc-sa-4.0. Read the terms on the original page before reusing the data.
What is the primary method used to induce perturbations in this atlas?
The atlas uses CRISPR interference (CRISPRi) implemented with dual-guide pairs to silence specific target genes.
What cell types are present in X-Atlas-Pisces?
It includes HCT116, HEK293T, HepG2, induced pluripotent stem cells (iPSCs), multi-lineage differentiating iPSCs, and both resting and CD3/CD28-activated Jurkat T cells.
What model was this dataset designed to evaluate?
The dataset was generated to train and evaluate the X-Cell model.
Written from X-Atlas-Pisces's own dataset card. Statements citing a figure that does not appear there are removed before publication. Licence and running details come from our registry.