nf-core pipeline · clip, clip-seq, peak-calling
nf-core/clipseq
A Nextflow pipeline for quality control, mapping, deduplication and peak calling on CLIP sequencing data.
What nf-core/clipseq is
nf-core/clipseq is a bioinformatics analysis pipeline designed for processing cross-linking and immunoprecipitation sequencing (CLIP-seq) data. The workflow enables researchers to investigate RNA-protein interactions by turning raw sequencing reads into mapped genomic sites, identified crosslink events, and binding peaks.
Originally written by Charlotte West and Anob Chakrabarti from the Luscombe Lab at The Francis Crick Institute, the pipeline originated during an April 2020 hackathon. It automates standard data-handling challenges common to CLIP experiments, such as filtering abundant non-coding RNAs, resolving PCR duplicates through unique molecular identifiers (UMIs), and comparing alternative peak-calling algorithms.
What the pipeline does
The pipeline starts with read quality assessment and adapter trimming using Cutadapt and FastQC. Reads are then pre-mapped against reference sequences such as rRNA and tRNA using Bowtie 2 to remove abundant contamination before genome alignment with STAR. Mapped reads are deduplicated based on unique molecular identifiers using UMI-tools.
Following deduplication, BEDTools identifies crosslink sites and produces bedGraph coverage tracks. For binding site discovery, the pipeline offers several peak-calling tools: iCount, Paraclu, PureCLIP, and Piranha. DREME is then used for motif detection on the identified sites. Quality metrics across the workflow—including library complexity from Preseq and regional genomic distribution from RSeQC—are collected and summarised in a MultiQC report.
Key facts
- Workflow framework
- Nextflow (DSL2)
- Original developers
- Charlotte West and Anob Chakrabarti (Luscombe Lab, The Francis Crick Institute)
- Latest release
- 1.0.0 (2021-04-27)
- Trimming tool
- Cutadapt
- Pre-mapping tool
- Bowtie 2
- Genome aligner
- STAR
- Deduplication tool
- UMI-tools
- Peak callers
- iCount, Paraclu, PureCLIP, Piranha
- Motif discovery tool
- DREME
- Reporting
- MultiQC
What you can use it for
- RNA-binding protein target identification
- Determine transcriptome-wide RNA interaction sites and binding peaks for RNA-binding proteins using CLIP-seq datasets.
- Peak-caller benchmarking
- Compare peak detection sensitivity and consistency across multiple algorithms (iCount, Paraclu, PureCLIP, and Piranha) within a single run.
- Binding motif discovery
- Identify sequence motifs enriched at protein-RNA crosslink sites using the integrated DREME step.
Run it here
The research agent runs nf-core/clipseq 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
- Provides multiple established peak-calling algorithms (iCount, Paraclu, PureCLIP, and Piranha) to suit different CLIP variants and experimental protocols.
- Filters out rRNA and tRNA contaminants via a dedicated Bowtie 2 pre-mapping step prior to genome alignment.
- Handles unique molecular identifiers (UMIs) via UMI-tools to remove PCR duplication bias accurately.
- Aggregates quality metrics from Preseq, RSeQC, and FastQC into a unified MultiQC report that includes peak-calling comparisons.
Watch out for
- Requires reference sequences for pre-mapping (such as rRNA and tRNA) and an annotated reference genome.
- Built specifically for CLIP-seq protocols and is not configured for general RNA-seq expression quantification.
Questions
Can I run nf-core/clipseq without my own cluster?
The research agent runs nf-core/clipseq 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/clipseq under?
MIT, like every nf-core pipeline.
Which peak callers are supported by nf-core/clipseq?
The pipeline supports four peak-calling tools: iCount, Paraclu, PureCLIP, and Piranha.
How does the pipeline handle PCR duplicates?
It performs deduplication based on unique molecular identifiers (UMIs) using UMI-tools.
What quality control tools are included?
The workflow runs FastQC for read quality, Preseq for library complexity estimation, RSeQC for regional distribution, and compiles the metrics with MultiQC.
How does the workflow remove rRNA and tRNA reads?
It includes a pre-mapping step using Bowtie 2 to filter reads aligning to abundant non-coding RNAs before full genome mapping with STAR.
How to cite it
10.5281/zenodo.4723016
Written from nf-core/clipseq's own README. Statements citing a figure that does not appear there are removed before publication. Licence and running details come from our registry.