Quick start

Trim Galore reads FASTQ files (plain or gzip-compressed), trims adapters and low-quality bases, and writes trimmed FASTQ plus a per-file trimming report. The defaults work for most Illumina libraries.

Single-end

trim_galore input.fastq.gz

Outputs:

• input_trimmed.fq.gz (trimmed reads)
• input.fastq.gz_trimming_report.txt (text report)
• input.fastq.gz_trimming_report.json (structured report for MultiQC)

Paired-end

trim_galore --paired sample_R1.fastq.gz sample_R2.fastq.gz

Outputs:

• sample_R1_val_1.fq.gz and sample_R2_val_2.fq.gz (validated paired reads)
• A trimming report per input file

Parallel processing

Speedup is near-linear up to about 16 cores. Beyond about 20 the run is usually I/O-bound, so adding cores helps less.

trim_galore --cores 8 --paired sample_R1.fastq.gz sample_R2.fastq.gz

RRBS libraries

trim_galore --rrbs --paired sample_R1.fastq.gz sample_R2.fastq.gz

For non-directional libraries, add --non_directional. See the Bisulfite & RRBS guide for the biology behind these modes.

Run FastQC alongside

trim_galore --fastqc input.fastq.gz

FastQC is built in via the bundled fastqc-rust library: no Java or external fastqc install needed. Outputs are FastQC 0.12.1-compatible HTML + ZIP files.

Common combinations

# Small RNA: auto-lowers --length to 18 bp
trim_galore --small_rna input.fastq.gz

# 2-colour aware quality trimming (NextSeq, NovaSeq): replaces -q
trim_galore --2colour 20 input.fastq.gz

# Force-disable poly-G trimming (it is auto-enabled on 2-colour data)
trim_galore --no_poly_g input.fastq.gz

# Trim a poly-A tail (mRNA-seq libraries)
trim_galore --poly_a input.fastq.gz

# Multiple adapters, up to 3 occurrences per read
trim_galore -a AGCTCCCG -a TTTCATTAT -a TTTATTCGGAT -n 3 input.fastq.gz

# Adapters from a FASTA file
trim_galore -a "file:./adapters.fa" input.fastq.gz

Get the full flag list

trim_galore --help

For context on individual flags and how they interact, see the user guide.