krakenXtract

Extract reads from a FASTQ file based on taxonomic classification via Kraken2.

Written in Rust.

Background

I recently wanted to extract reads from a medium-ish sized (6GB) FASTQ file (~5.5 million reads), based on taxonomic classifications. For that I used the great KrakenTools. This however took a while both parse the Kraken2 output file and extract/write the matching reads. Having been wanting to experiment with Rust for a while, this inspired me to re-implement the extract_kraken_reads.py script in Rust as a learning exercise.

This is currently an early implementation (and my first Rust programme!), with plans to expand functionality.

Current features

• Extract all reads from a fastq file based on a taxonomic id
• Supports interleaved fastq files
• Extract parents or the children of the specified taxon id
• Supports both uncompressed or gzip inputs.
• Multithreaded
• ~4.4x speed up compared to KrakenTools

Benchmarks (rough)

For more detail see benchmarks

Installation

Download the latest release.

Alternatively, build from source:

Clone the repository:

bash git clone https://github.com/Sam-Sims/krakenxtract

Install rust/cargo:

To install please refer to the rust documentation: docs

Build and add to path:

bash cd kraken-extract cargo build --release export PATH=$PATH:$(pwd)/target/release

All executables will be in the directory kraken-extract/target/release.

Usage

bash kraken-extract --kraken <kraken_output> --fastq <fastq_file> --taxid <taxonomic_id> --output <output_file>

Arguments

-k, --kraken <KRAKEN_OUTPUT> -t, --taxid <TAXID> -r, --report <REPORT_OUTPUT> -f, --fastq <FASTQ_FILE> -o, --output <OUTPUT_LOCATION> --compression-mode <COMPRESSION> [default: fast] --parents --children --no-compress --exclude -h, --help Print help -V, --version Print version

--parents: This will extract all the reads classified at all taxons between the root and the specified --taxid

--children: This will extract all the reads classified as decendents or subtaxa of --taxid (Including the taxid)

--compression_mode: This defines the compression mode of the output fastq.gz file - fast / default / best

--no-compress: This will output a plaintext fastq file

--exclude: This will output every read except those matching the taxid. Works with --parents and --children

Future plans

• [x] Support unzipped fastq files
• [ ] Support paired end FASTQ files
• [x] --include-parents and --include-children arguments
• [ ] Supply multiple taxonomic IDs to extract
• [x] Exclude taxonomic IDs
• [ ] --append
• [x] --compression-mode <fast/default/best>
• [x] More verbose output
• [ ] Proper benchmarks
• [ ] Output fasta format (for blast??)
• [x] Output non gz
• [ ] Tests

Version

• 0.2.3

Changelog

0.2.3

• Code optimisations

0.2.2

• Increased verbosity of outputs to user
• --no-compress flag to output a standard, plaintext fastq file
• --exclude to exclude specified reads. Works with --children and --parents
• Docstrings under the hood

0.2.1

• Fixes to reduce memory usage

0.2.0

• Detect and handle gz files or plain files
• --compression arg to select compression type
• zlib-ng to speed up gzip handling
• --children and --parents to save children and parents based on kraken report

0.1.0

• First release