fastp Workflow - Usage Guide

Overview

fastp is a modern, all-in-one FASTQ preprocessor that does adapter trimming, quality/length filtering, 2-color poly-G removal, base correction, and QC reporting in a single fast C++ pass, replacing separate Cutadapt, Trimmomatic, and FastQC steps for bulk Illumina data. Its defining feature is paired-end adapter detection by overlap analysis (no adapter sequence needed). It is the default for bulk preprocessing, but not a substitute for cutadapt's precision on small-RNA/amplicon adapters or for proper UMI-based molecule counting.

Prerequisites

conda install -c bioconda fastp

Quick Start

Tell your AI agent what you want to do: - "Process my paired-end FASTQ files with fastp" - "Clean up my NovaSeq data with poly-G trimming" - "Run fastp with Q20 filtering and a minimum length of 36 bp" - "Aggregate all my fastp reports with MultiQC"

Example Prompts

Basic processing

"Run fastp on my paired-end FASTQ files and generate HTML and JSON reports"

"Process all samples in my data directory with fastp"

Custom parameters

"Use fastp with Q20 filtering, a minimum length of 36 bp, and 3' sliding-window trimming"

"Enable overlap base correction for my paired-end reads"

Platform-specific

"Process my NovaSeq data with poly-G trimming"

"Merge overlapping paired-end reads from my short-insert library"

Batch processing

"Run fastp on all my samples and aggregate the JSON reports with MultiQC"

What the Agent Will Do

Run fastp with overlap-based PE adapter trimming (no adapter sequence needed)
Apply per-read quality filtering and optional light window trimming with a length gate
Handle platform-specific issues (auto poly-G for 2-color instruments)
Generate HTML and JSON reports for MultiQC aggregation
Optionally merge overlapping reads or extract UMIs (dedup happens later, post-alignment)

Comparison with the traditional pipeline

Task	Traditional	fastp
QC report	FastQC	Built-in HTML/JSON
Adapter trim	Cutadapt	Overlap analysis (no sequence needed)
Quality trim/filter	Trimmomatic	Built-in
Poly-G	Manual	Auto for 2-color
Speed	3 passes	1 pass

Tips

For paired-end data fastp needs no adapter sequence; it finds the read overlap and trims read-through automatically. Provide --adapter_sequence for single-end.
--correction uses the overlap to fix a low-quality base from its high-quality mate (paired-end only).
Poly-G trimming auto-enables for NextSeq/NovaSeq from the instrument ID; leave it on.
Do NOT use --dedup for RNA-seq, amplicon, or any assay where identical reads are real signal; it is sequence-identity dedup and removes biological duplicates. Use UMIs or coordinate dedup instead.
--umi only extracts the UMI; molecule-accurate dedup/consensus happens after alignment (umi_tools/fgbio).
For light RNA-seq trimming, skip aggressive quality cutting; the aligner soft-clips tails.
JSON reports feed MultiQC directly for cohort-level review.

Resources

read-qc/adapter-trimming - Precise adapter/primer control for small-RNA and amplicon read-qc/quality-filtering - Detailed quality/length filtering and the trim-light evidence base read-qc/quality-reports - Aggregate fastp JSON across samples with MultiQC read-qc/umi-processing - Molecule-accurate UMI dedup and consensus after alignment alignment-files/duplicate-handling - Coordinate-based duplicate marking for DNA variant calling