TrimGalore

In this section:

Description
Available Versions
Loading the Module
Job Submission
Job Array for Multiple Samples
Job Array for Paired-end (Multiple Samples)
Main Options and Features
Output Files
Optimization Tips
References

Description 

According to the documentation of TrimGalore, this is a tool for consistent adapter trimming and removal of low-quality bases from next-generation sequencing (NGS) data. It has special handling for RRBS (Reduced Representation Bisulfite Sequencing) libraries.

It automates adapter detection, offers Phred-based quality trimming, support for paired-end data, and parallelization for efficient processing.

Available Versions 

trim_galore/2.2.0 (default)

Note

To check the available versions:

module avail trim_galore

Loading the Module 

# Load TrimGalore
module load trim_galore/2.2.0

# Verify installation
trim_galore --version

# Check help
trim_galore --help

Job Submission 

Below are example scripts to use TrimGalore on GridUnesp. Always remember to define INPUT and OUTPUT for job-nanny.

Example 1: Single-end Processing 

submit_trimgalore_se.sh

#!/bin/bash
#SBATCH -J trim_se
#SBATCH -N 1
#SBATCH -n 1
#SBATCH -t 12:00:00
#SBATCH --mem=8G

export INPUT="sample.fastq.gz"
export OUTPUT="sample_trimmed.fq.gz"

module load trim_galore/2.2.0

job-nanny trim_galore --fastqc sample.fastq.gz

Example 2: Paired-end Processing 

submit_trimgalore_pe.sh

#!/bin/bash
#SBATCH -J trim_pe
#SBATCH -N 1
#SBATCH -c 8
#SBATCH -t 24:00:00
#SBATCH --mem=16G

export INPUT="sample_R1.fastq.gz sample_R2.fastq.gz"
export OUTPUT="sample_R1_val_1.fq.gz sample_R2_val_2.fq.gz"

module load trim_galore/2.2.0

job-nanny trim_galore --paired --cores $SLURM_CPUS_PER_TASK \
                      --fastqc sample_R1.fastq.gz sample_R2.fastq.gz

Example 3: Paired-end with Uncompressed Files 

submit_trimgalore_pe_uncompressed.sh

#!/bin/bash
#SBATCH -J trim_pe_uncomp
#SBATCH -N 1
#SBATCH -c 8
#SBATCH -t 24:00:00
#SBATCH --mem=16G

export INPUT="sample_R1.fastq sample_R2.fastq"
export OUTPUT="sample_R1_val_1.fq sample_R2_val_2.fq"

module load trim_galore/2.2.0

job-nanny trim_galore --paired --cores $SLURM_CPUS_PER_TASK \
                      sample_R1.fastq sample_R2.fastq

Note

By default, output compression mirrors input compression: - Input .fastq.gz → output .fq.gz - Input .fastq → output .fq

Example 4: RRBS Mode (Bisulfite-seq)

submit_trimgalore_rrbs.sh

#!/bin/bash
#SBATCH -J trim_rrbs
#SBATCH -N 1
#SBATCH -c 4
#SBATCH -t 24:00:00
#SBATCH --mem=16G

export INPUT="bisulfite_R1.fastq.gz bisulfite_R2.fastq.gz"
export OUTPUT="bisulfite_R1_val_1.fq.gz bisulfite_R2_val_2.fq.gz"

module load trim_galore/2.2.0

job-nanny trim_galore --paired --rrbs --cores $SLURM_CPUS_PER_TASK \
                      bisulfite_R1.fastq.gz bisulfite_R2.fastq.gz

Example 5: Size Optimization with `--clumpify`

For data like ATAC-seq or amplicons that benefit from reordering for better compression:

submit_trimgalore_clumpify.sh

#!/bin/bash
#SBATCH -J trim_clump
#SBATCH -N 1
#SBATCH -c 8
#SBATCH -t 12:00:00
#SBATCH --mem=16G

export INPUT="amplicon_data.fastq.gz"
export OUTPUT="amplicon_trimmed.fq.gz"

module load trim_galore/2.2.0

# For intermediate files: --clumpify
# For storage: add --compression 6
job-nanny trim_galore --clumpify --cores $SLURM_CPUS_PER_TASK \
                      --compression 6 amplicon_data.fastq.gz

Example 6: Poly-A Removal 

For RNA-seq libraries enriched with poly-A:

submit_trimgalore_polya.sh

#!/bin/bash
#SBATCH -J trim_polya
#SBATCH -N 1
#SBATCH -c 8
#SBATCH -t 12:00:00
#SBATCH --mem=16G

export INPUT="rna_sample_R1.fastq.gz rna_sample_R2.fastq.gz"
export OUTPUT="rna_sample_R1_val_1.fq.gz rna_sample_R2_val_2.fq.gz"

module load trim_galore/2.2.0

job-nanny trim_galore --paired --polyA --cores $SLURM_CPUS_PER_TASK \
                      --fastqc rna_sample_R1.fastq.gz rna_sample_R2.fastq.gz

Example 7: Force Uncompressed Output 

submit_trimgalore_dont_gzip.sh

#!/bin/bash
#SBATCH -J trim_nogzip
#SBATCH -N 1
#SBATCH -c 4
#SBATCH -t 12:00:00
#SBATCH --mem=8G

export INPUT="sample.fastq.gz"
export OUTPUT="sample_trimmed.fq"

module load trim_galore/2.2.0

job-nanny trim_galore --dont_gzip --cores $SLURM_CPUS_PER_TASK \
                      sample.fastq.gz

Job Array for Multiple Samples 

submit_trimgalore_array.sh

#!/bin/bash
#SBATCH -J trim_array
#SBATCH --array=1-10
#SBATCH -N 1
#SBATCH -c 4
#SBATCH -t 12:00:00
#SBATCH --mem=8G

SAMPLES=(
    "sample1"
    "sample2"
    "sample3"
    "sample4"
    "sample5"
    "sample6"
    "sample7"
    "sample8"
    "sample9"
    "sample10"
)

SAMPLE=${SAMPLES[$SLURM_ARRAY_TASK_ID-1]}
export INPUT="${SAMPLE}.fastq.gz"
export OUTPUT="${SAMPLE}_trimmed.fq.gz"

module load trim_galore/2.2.0

job-nanny trim_galore --fastqc ${SAMPLE}.fastq.gz

Job Array for Paired-end (Multiple Samples)

submit_trimgalore_array_pe.sh

#!/bin/bash
#SBATCH -J trim_array_pe
#SBATCH --array=1-10
#SBATCH -N 1
#SBATCH -c 8
#SBATCH -t 24:00:00
#SBATCH --mem=16G

SAMPLES=(
    "sample1"
    "sample2"
    "sample3"
    "sample4"
    "sample5"
    "sample6"
    "sample7"
    "sample8"
    "sample9"
    "sample10"
)

SAMPLE=${SAMPLES[$SLURM_ARRAY_TASK_ID-1]}
export INPUT="${SAMPLE}_R1.fastq.gz ${SAMPLE}_R2.fastq.gz"
export OUTPUT="${SAMPLE}_R1_val_1.fq.gz ${SAMPLE}_R2_val_2.fq.gz"

module load trim_galore/2.2.0

job-nanny trim_galore --paired --cores $SLURM_CPUS_PER_TASK \
                      --fastqc ${SAMPLE}_R1.fastq.gz ${SAMPLE}_R2.fastq.gz

Main Options and Features 

TrimGalore Options
Option	Abbrev./Value	Description
`--paired`		Paired-end mode
`--cores N`	`-j N`	Number of cores for parallel processing (up to ~8)
`--fastqc`		Runs integrated FastQC on trimmed files
`--rrbs`		Special mode for RRBS libraries
`--polyA`		Removes poly-A tails (recommended for RNA-seq)
`--nextseq N`		Quality trimming specific to 2-color instruments
`--clumpify`		Reorders reads for better compression (requires `--cores >= 2`)
`--compression N`		gzip compression level (1 to 9). Default: 1 (fastest)
`--retain_unpaired`		Keeps unpaired reads in separate files
`--dont_gzip`		Outputs plain text (uncompressed)
`--length N`		Discards reads shorter than N (default: 20)
`--quality N`		Phred quality for trimming (default: 20)

Tip

For paired-end data, consider using --retain_unpaired to avoid losing reads that lost their pair during trimming.

Output Files 

Files generated by TrimGalore
Mode	Output files
Single-end	`_trimmed.fq.gz`, `_trimming_report.txt`, `*_trimming_report.json`
Paired-end	`_val_1.fq.gz`, `_val_2.fq.gz` (valid pairs)
Paired-end (with `--retain_unpaired`)	`_unpaired_1.fq.gz`, `_unpaired_2.fq.gz` (unpaired reads)
FastQC	HTML report and ZIP file with post-trimming quality

Optimization Tips 

Parallelism: Use --cores up to 8 for near-linear speed gains
Compression: --compression 1 (default) is faster; --compression 6-9 reduces file size
Clumpify: Benefits ATAC-seq, amplicons, RNA-seq; requires --cores >= 2
Memory: TrimGalore is lightweight, but integrated FastQC may consume more memory

References 

Official documentation: https://www.trimgalore.com/
GitHub repository: https://github.com/FelixKrueger/TrimGalore
User Guide: https://www.trimgalore.com/user-guide/
RRBS Guide: https://www.trimgalore.com/rrbs-guide/
v2.0 migration notes: https://www.trimgalore.com/v2-migration-notes/