nf-core pipeline · amp, amr, antibiotic-resistance
nf-core/funcscan
Screens assembled nucleotide contigs and metagenomes for antimicrobial peptides, resistance genes, and biosynthetic clusters.
What nf-core/funcscan is
nf-core/funcscan is a bioinformatics analysis pipeline designed for screening nucleotide sequences, such as assembled genomic and metagenomic contigs, for functional genes. It specialises in detecting antimicrobial peptides (AMPs), antibiotic resistance genes (ARGs), biosynthetic gene clusters (BGCs), and carbohydrate-active enzymes (CAZymes).
Developed by Jasmin Frangenberg, Anan Ibrahim, Louisa Perelo, Moritz E. Beber, and James A. Fellows Yates, the pipeline resolves the challenge of coordinating disparate functional annotation tools. By automating multiple predictive models and standardising outputs into harmonised ontologies and unified reports, it provides an end-to-end framework for functional genomics and natural product discovery.
What the pipeline does
The workflow accepts assembled nucleotide sequences in FASTA format. It executes initial quality assessment with SeqKit, performs taxonomic classification of prokaryotic contigs with MMseqs2, and annotates coding sequences using Prodigal, Pyrodigal, Prokka, or Bakta. General protein domains and families across the predicted coding sequences are annotated with InterProScan.
Contigs are then screened through specialised functional discovery modules. AMP candidates are identified using ampir, Macrel, AMPlify, and HMMER. ARGs are detected using ABRicate, AMRFinderPlus, fARGene, RGI, and DeepARG, with argNorm mapping results to the Antibiotic Resistance Ontology. BGCs are identified using antiSMASH, BiG-SLiCE, DeepBGC, GECCO, and HMMER, while CAZymes and their gene clusters are identified via run_dbcan. Results across samples are compiled using AMPcombi, hAMRonization, comBGC, and MultiQC.
Key facts
- Pipeline engine
- Nextflow DSL2
- Input format
- FASTA or multi-FASTA contig files via CSV samplesheet
- AMP detection tools
- ampir, Macrel, HMMER, AMPlify
- ARG detection tools
- ABRicate, AMRFinderPlus, fARGene, RGI, DeepARG
- BGC detection tools
- antiSMASH, BiG-SLiCE, DeepBGC, GECCO, HMMER
- CAZyme detection tool
- run_dbcan
- Aggregation tools
- AMPcombi, hAMRonization, comBGC, MultiQC
- Ontology mapping
- Antibiotic Resistance Ontology (ARO) via argNorm
What you can use it for
- Metagenomic antimicrobial resistance profiling
- Screen assembled metagenomic contigs for known and novel resistance determinants using multiple predictors standardised against the Antibiotic Resistance Ontology.
- Natural product discovery
- Mine prokaryotic genomes or environmental assemblies for secondary metabolite biosynthetic gene clusters using rule-based and deep-learning predictors.
- Peptide antibiotic screening
- Identify putative antimicrobial peptide sequences from contigs using ensemble predictions across ampir, Macrel, AMPlify, and profile hidden Markov models.
- Carbohydrate metabolism annotation
- Identify carbohydrate-active enzymes, CAZyme gene clusters, and associated substrate predictions in complex microbial assemblies.
Run it here
The research agent runs nf-core/funcscan with Nextflow on its own machine — up to 32 cores, 256 GiB of memory and 24 hours per run, which covers the test profile and real single-sample runs. It builds the samplesheet, checks it against the pipeline's schema and reads the results back. When a run is too big for one machine, it gives you the exact command for your own cluster instead.
Strengths and limits
Good at
- Integrates multiple independent discovery tools for AMPs, ARGs, and BGCs into a single automated pipeline.
- Harmonises heterogeneous ARG tool outputs to the Antibiotic Resistance Ontology using argNorm and hAMRonization.
- Generates consolidated summary tables across multiple samples using AMPcombi, comBGC, and MultiQC.
- Uses containerised DSL2 modules to run tasks reproducibly across various compute infrastructures.
Watch out for
- Operates on pre-assembled nucleotide contigs and cannot take raw, unassembled sequencing reads directly.
- Taxonomic classification and initial gene annotation steps are focused strictly on contigs of prokaryotic origin.
- Pipeline execution parameters cannot be passed via Nextflow custom configuration files and must be supplied via CLI flags or parameter files.
Questions
Can I run nf-core/funcscan without my own cluster?
The research agent runs nf-core/funcscan with Nextflow on its own machine — up to 32 cores, 256 GiB of memory and 24 hours per run, which covers the test profile and real single-sample runs. It builds the samplesheet, checks it against the pipeline's schema and reads the results back. When a run is too big for one machine, it gives you the exact command for your own cluster instead.
What licence is nf-core/funcscan under?
MIT, like every nf-core pipeline.
What input data does the pipeline require?
Input is supplied via a CSV samplesheet specifying sample identifiers alongside paths to FASTA files containing assembled nucleotide contigs.
Can screening modules be executed independently?
Yes, screening categories can be individually selected using flags such as --run_amp_screening, --run_arg_screening, and --run_bgc_screening.
How does the pipeline handle differences in gene naming across ARG tools?
It applies argNorm to map outputs from DeepARG, AMRFinderPlus, and ABRicate to standard Antibiotic Resistance Ontology terms, then collates them with hAMRonization.
How to cite it
10.5281/zenodo.7643099
Written from nf-core/funcscan's own README. Statements citing a figure that does not appear there are removed before publication. Licence and running details come from our registry.