ITSxpress: Software to rapidly trim the internal transcribed spacer (ITS) region of FASTQ files

Documentation Status CI/CD Pipeline (Test & Publish Docker) Anaconda-Server Badge GitHub release (latest by date) https://zenodo.org/badge/DOI/10.5281/zenodo.1304348.svg

Authors

  • Adam R. Rivers, US Department of Agriculture, Agricultural Research Service, Gainesville, FL, USA

  • Sveinn V. Einarsson, Queens University, Charlotte, NC, USA

Citations

Einarsson SV, Rivers AR. 2024. ITSxpress version 2: software to rapidly trim internal transcribed spacer sequences with quality scores for amplicon sequencing. Microbiol Spectr 0:e00601-24. (doi: 10.1128/spectrum.00601-24)

Rivers AR, Weber KC, Gardner TG, Liu, S, Armstrong, SD. ITSxpress: Software to rapidly trim internally transcribed spacer sequences with quality scores for marker gene analysis [version 1]. F1000Research 2018, 7:1418 (doi: 10.12688/f1000research.15704.1)

Introduction

The internally transcribed spacer region is a region between highly conserved the small subunit (SSU) of rRNA and the large subunit (LSU) of the rRNA. In Eukaryotes it contains the 5.8s genes and two variable length spacer regions. In amplicon sequencing studies it is common practice to trim off the conserved (SSU, 5,8S or LSU) regions. Bengtsson-Palme et al. (2013) published software the software package ITSx to do this.

ITSxpress is designed to support the calling of exact amplicon sequence variants (ASVs) rather than OTUs. This newer method of sequence error-correction requires quality score data from each sequence, so each input sequence must be trimmed. ITSXpress makes this possible by taking FASTQ data, de-replicating the sequences with Vsearch then identifying the start and stop sites using HMMER hmmsearch. Results are parsed and the trimmed files are returned. The ITS1, ITS2 or the entire ITS region including the 5.8s rRNA gene can be selected. ITSxpress uses the hmm models from ITSx so results are comparable.

ITSxpress is also a QIIME2 plugin. Starting from version 2.0.0 of ITSxpress, the QIIME2 plugin is included with the command line version of ITSxpress. The installation method will be slightly different depending on whether QIIME2 is being used.

Installing ITSxpress for use as a QIIME2 plugin

To install ITSxpress as a plugin for QIIME 2, first install QIIME 2 Amplicon Distribution as a separate Conda/Mamba environemnt using their instructions https://library.qiime2.org/quickstart/qiime2 then add ITSxpress to the QIIME 2 Conda environment. The examples below are for QIIME2 version 2026.7 an so please update the commands if you want a newer release.

For Linux:

conda env create \
  --name rachis-qiime2-2026.7 \
  --file https://raw.githubusercontent.com/qiime2/distributions/refs/heads/dev/2026.7/qiime2/released/rachis-qiime2-linux-64-conda.yml
conda activate rachis-qiime2-2026.7
conda install -c bioconda -c conda-forge itsxpress
qiime dev refresh-cache

For maxOS (Intel) and OS X:

conda env create \
    --name rachis-qiime2-2026.7 \
    --file https://raw.githubusercontent.com/qiime2/distributions/refs/heads/dev/2026.7/qiime2/released/rachis-qiime2-osx-64-conda.yml
conda activate rachis-qiime2-2026.7
conda install -c bioconda -c conda-forge itsxpress
qiime dev refresh-cache

For macOS (Arm / Apple Silicon):

CONDA_SUBDIR=osx-64 conda env create \
 --name rachis-qiime2-2026.7 \
 --file https://raw.githubusercontent.com/qiime2/distributions/refs/heads/dev/2026.7/qiime2/released/rachis-qiime2-osx-64-conda.yml
conda activate rachis-qiime2-2026.7
conda config --env --set subdir osx-64
CONDA_SUBDIR=osx-64 conda install -c bioconda -c conda-forge itsxpress
qiime dev refresh-cache

Installing ITSxpress for standalone use

For Linux, maxOS (Intel), and OS X:

conda create -n itsxpressenv -c bioconda -c conda-forge itsxpress
conda activate itsxpressenv

For macOS (Arm/Apple Silicon):

CONDA_SUBDIR=osx-64 conda create -n itsxpressenv -c bioconda -c conda-forge itsxpress
conda activate itsxpressenv
conda  config --env --set subdir osx-64

Running ITSxpress as a Docker container

docker pull ghcr.io/usda-ars-gbru/itsxpress
docker run [Options...] itsxpress

Dependencies

The software requires Vsearch, Hmmer and Biopython. Bioconda takes care of this for you so it is the preferred installation method.

Usage

Option

Description

-h, –help

Show this help message and exit.
–fastq
A .fastq, .fq, .fastq.gz or .fq.gz file.
Required.
–single_end
A flag to specify that the fastq file is single-ended (not
paired). Default is false.
–fastq2
A .fastq, .fq, .fastq.gz or .fq.gz file
representing read 2 if present, optional.
–outfile
The trimmed FASTQ file, if it ends in gz it will be
gzipped.
–outfile2
The trimmed FASTQ read 2 file, if it ends in gz it will
be gzipped. If used, reads will be retuned as unmerged pairs
rather than than merged.

–tempdir

Specify the temp file directory. Default is None.

–keeptemp

Should intermediate files be kept? Default is false.

–region

Options : {ITS2, ITS1, ALL}
–taxa
Select the taxonomic group sequenced: {Alveolata, Bryophyta,
Bacillariophyta, Amoebozoa, Euglenozoa, Fungi, Chlorophyta,
Rhodophyta, Phaeophyceae, Marchantiophyta, Metazoa, Oomycota,
Haptophyceae, Raphidophyceae, Rhizaria, Synurophyceae,
Tracheophyta, Eustigmatophyceae, Parabasalia, All}.
Default Fungi. For QIIME single letter codes are used, see help menu for list.
–cluster_id
The percent identity for clustering reads range [0.99-1.0],
set to 1 for exact de-replication. Default 1.0.

–log

Log file. Default is ITSxpress.log.

–threads

Number of processor threads to use. Default is 1. There is not much benefit over 4.
–reversed_primers
Primers are in reverse orientation as in Taylor et al. 2016,
DOI:10.1128/AEM.02576-16. If selected ITSxpress returns
trimmed reads flipped to the forward orientation
–allow_staggered_reads
Allow merging staggered reads with –fastq_allowmergestagger
for Vsearch –fastq_mergepairs. See Vsearch documentation.
(Optional) Default is true.
–trim-ccs
PacBio CCS reads mode. Stitch fake primers (GACAGGTACAAGAAGGA
and adapter ACTGGAGACTGGGTTAA) with uniform quality scores.

QIIME2 plugin Examples

Example 1: trimming fungal paired-end Illumina reads (outputting unmerged reads)

When processing Illumina paired-end fungal data intended for downstream analysis with DADA2, you must maintain separate forward and reverse read structures. The trim-pair-output-unmerged command trims the ITS region from both reads but keeps them unmerged.

qiime itsxpress trim-pair-output-unmerged \
  --i-per-sample-sequences paired-end-demux.qza \
  --p-region ITS1 \
  --p-taxa F \
  --p-threads 4 \
  --o-trimmed trimmed-paired-unmerged.qza \
  --verbose

Key Parameters:

  • --p-region ITS1: Targets the ITS1 sub-region specifically.

  • --p-taxa F: Restricts HMM searching to Fungi (selected by default).

  • --o-trimmed: Generates an output artifact of type SampleData[PairedEndSequencesWithQuality].

Example 2: trimming fungal single-end reads

If your dataset consists of single-end reads use the trim-single command. Using this for merged paired reads is not recommended by DADA2. It uses the information from both reads in its error modeling.

qiime itsxpress trim-single \
  --i-per-sample-sequences single-end-demux.qza \
  --p-region ITS2 \
  --p-taxa F \
  --p-threads 4 \
  --o-trimmed trimmed-single.qza \
  --verbose

Key Parameters:

  • --i-per-sample-sequences: Accepts a SampleData[SequencesWithQuality] artifact containing single-end data.

  • --p-region ITS2: Targets the ITS2 sub-region.

Example 3: trimming PacBio CCS long reads (single-end)

PacBio Circular Consensus Sequencing (CCS) generates long, high-fidelity reads that can span the entire ITS region (ITS1, 5.8S, and ITS2). To handle these long reads correctly, you must pass the --p-trim-ccs flag to the trim-single command. This feature orients input sequence reads against a universal reference database with Vsearch. It outputs trimmed reads with synthetic dummy primers on the ends (forward: GACAGGTACAAGAAGGA, reverse complement of reverse: ACTGGAGACTGGGTTAA) with uniform Phred quality scores of 93 (~) to satisfy downstream QIMME2 DADA2 plugin requirements. At completion, ITSxpress provides a log message displaying the example qiime dada2 denoise-ccs arguments for denoising.

qiime itsxpress trim-single \
  --i-per-sample-sequences pacbio-ccs-demux.qza \
  --p-region ALL \
  --p-taxa F \
  --p-trim-ccs \
  --p-threads 4 \
  --o-trimmed trimmed-pacbio-ccs.qza \
  --verbose

Key Parameters:

  • --p-region ALL: Instructs the plugin to extract the full ITS composite sequence (ITS1 + 5.8S + ITS2).

  • --p-trim-ccs: Explicitly enables the specialized trimming mode optimized for PacBio CCS reads.

Stand Alone Command Line Examples

Use case 1: Trimming the ITS2 region from a fungal amplicon sequencing dataset with forward and reverse gzipped FASTQ files using two cpu threads. Return a single merged file for use in Deblur.

itsxpress --fastq r1.fastq.gz --fastq2 r2.fastq.gz --region ITS2 \
--taxa Fungi --log logfile.txt --outfile trimmed_reads.fastq.gz --threads 2

ITSxpress can take uncompressed, gzipped or zstd compressed FASTQ files and it can write uncompressed, gzipped or zstd compressed FASTQ files. It expects FASTQ files to end in: .fq, .fastq, .fq.gz, fastq.gz, .fq.zst or fastq.zst.

Use case 2: Trimming the ITS2 region from a fungal amplicon sequencing dataset with forward and reverse gzipped FASTQ files using two cpu threads. Return forward and reverse read files for use in Dada2.

itsxpress --fastq r1.fastq.gz --fastq2 r2.fastq.gz --region ITS2 \
--taxa Fungi --log logfile.txt --outfile trimmed_reads_r1.fastq.gz --outfile2 trimmed_reads_r2.fastq.gz --threads 2

ITSxpress can take uncompressed, gzipped or zstd compressed FASTQ files and it can write uncompressed, gzipped or zstd compressed FASTQ files. It expects FASTQ files to end in: .fq, .fastq, .fq.gz, fastq.gz, .fq.zst or fastq.zst.

Use case 3: Trimming the ITS2 region from a fungal amplicon sequencing dataset with an single-ended gzipped FASTQ files using two cpu threads.

itsxpress --fastq single-end.fastq.gz --single_end --region ITS2 --taxa Fungi \
--log logfile.txt --outfile trimmed_reads.fastq.gz --threads 2

Single ended data is less common and may come from a platform like Oxford Nanopore or PacBio with long single-end reads. Pre-merging paired-end reads is not recommended because the next step for most people is DADA2 which prefers paired-end reads.

ITS HMM Models

The HMM models for identifying ITS regions were built by Henrik Nilsson at the University of Gothenburg. Originally, they were built for the ITSx software package; Dr. Nilsson has continued to update the models, but not on the same schedule as ITSx. To provide clarity about the reference data used, I have created a separate, versioned repository for the ITS sequence models: ITS_HMMs on GitHub and assigned it a DOI.

ITSXpress currently uses ITS_HMMS version 2.0.0.

API Documentation

The API documentation is available at https://itsxpress.readthedocs.io

License information

This software is a work of the United States Department of Agriculture, Agricultural Research Service. Under 17 U.S. Code § 105 the work is not copyrightable. It is released under a Creative Commons CC0 public domain attribution.