👤🧬🚫 Remove human reads from a sequencing run 👤🧬️🚫
nohuman removes human reads from sequencing reads by classifying them with kraken2 against a custom database
built from all of the genomes in the Human Pangenome Reference Consortium's (
HPRC) second release. It can take any type of
sequencing technology. Read more about the development of this method here.
$ conda install -c bioconda nohuman
Important
You will need to install kraken2 yourself using this install method.
curl -sSL nohuman.mbh.sh | sh
# or with wget
wget -nv -O - nohuman.mbh.sh | shYou can also pass options to the script like so
$ curl -sSL nohuman.mbh.sh | sh -s -- --help
install.sh [option]
Fetch and install the latest version of nohuman, if nohuman is already
installed it will be updated to the latest version.
Options
-V, --verbose
Enable verbose output for the installer
-f, -y, --force, --yes
Skip the confirmation prompt during installation
-p, --platform
Override the platform identified by the installer [default: apple-darwin]
-b, --bin-dir
Override the bin installation directory [default: /usr/local/bin]
-a, --arch
Override the architecture identified by the installer [default: x86_64]
-B, --base-url
Override the base URL used for downloading releases [default: https://github.com/mbhall88/nohuman/releases]
-h, --help
Display this help message
Important
You will need to install kraken2 yourself using this install method.
$ cargo install nohumanDocker images are hosted on the GitHub Container registry.
Prerequisite: apptainer (previously singularity)
$ URI="docker://ghcr.io/mbhall88/nohuman:latest"
$ apptainer exec "$URI" nohuman --helpThe above will use the latest version. If you want to specify a version then use a tag like so.
$ VERSION="0.2.1"
$ URI="docker://ghcr.io/mbhall88/nohuman:${VERSION}"Prerequisite: docker
$ docker pull ghcr.io/mbhall88/nohuman:latest
$ docker run ghcr.io/mbhall88/nohuman:latest nohuman --helpYou can find all the available tags here.
Important
You will need to install kraken2 yourself using this install method.
$ git clone https://github.com/mbhall88/nohuman.git
$ cd nohuman
$ cargo build --release
$ target/release/nohuman -hnohuman now keeps a manifest of the available Kraken2 databases so you can install as many versions as you want.
List the available versions (the default is always the most recent dataset, currently HPRC.r2 that includes the
latest Human Pangenome Reference genomes):
$ nohuman --list-db-versions
Download the default (latest) database:
$ nohuman --download
Download a specific version or fetch every available release:
$ nohuman --download --db-version HPRC.r1
$ nohuman --download --db-version all
By default, databases are cached under $HOME/.nohuman/db/<version>. When you run nohuman without any additional
options it will automatically choose the newest database you have installed. Use --db-version to pin a specific
version, or --db to point at a directory that already contains a Kraken2 database (for example, a shared install):
$ nohuman --db-version HPRC.r1 -t 4 in.fq
$ nohuman --db /data/my_kraken_db -t 4 in.fq
Tip
Set the NOHUMAN_DB environment variable to override the default database location for every command without having to
pass --db each time.
$ nohuman -c
[2023-12-14T04:10:46Z INFO ] All dependencies are available
$ nohuman -t 4 in.fq
this will pass 4 threads to kraken2 and output the clean reads as in.nohuman.fq.
You can specify where to write the output file with -o
$ nohuman -t 4 -o clean.fq in.fq
If you have paired-end Illumina reads
$ nohuman -t 4 in_1.fq in_2.fq
or to specify a different path for the output
$ nohuman -t 4 --out1 clean_1.fq --out2 clean_2.fq in_1.fq in_2.fq
Set a minimum confidence score for kraken2 classifications
$ nohuman --conf 0.5 in.fq
or write the kraken2 read classification output to a file
$ nohuman -k kraken.out in.fq
or write the kraken2 sample report to file
$ nohuman -r kraken.report in.fq
Tip
Compressed output will be inferred from the specified output path(s). If no output path is provided, the same
compression as the input will be used. To override the output compression format, use the --output-type option.
Supported compression formats are gzip (.gz), zstandard (zst), bzip2 (.bz2), and xz (.xz). If multiple threads are provided, these
will be used for compression of the output (where possible).
You can invert the functionality of nohuman to keep only the human reads by using the --human/-H flag.
$ nohuman -h
Remove human reads from a sequencing run
Usage: nohuman [OPTIONS] [INPUT]...
Arguments:
[INPUT]... Input file(s) to remove human reads from
Options:
-o, --out1 <OUTPUT_1> First output file.
-O, --out2 <OUTPUT_2> Second output file.
-c, --check Check that all required dependencies are available and exit
-d, --download Download the database
-D, --db <PATH> Path to the database [default: /home/michael/.nohuman/db]
--db-version <VERSION> Name of a downloaded database version to use (use `all` with
`--download` to fetch every version)
--list-db-versions List available database versions and exit
-F, --output-type <FORMAT> Output compression format. u: uncompressed; b: Bzip2; g: Gzip; x: Xz (Lzma); z: Zstd
-t, --threads <INT> Number of threads to use in kraken2 and optional output compression. Cannot be 0 [default: 1]
-H, --human Output human reads instead of removing them
-C, --conf <[0, 1]> Kraken2 minimum confidence score [default: 0.0]
-k, --kraken-output <FILE> Write the Kraken2 read classification output to a file
-r, --kraken-report <FILE> Write the Kraken2 report with aggregate counts/clade to file
-v, --verbose Set the logging level to verbose
-h, --help Print help (see more with '--help')
-V, --version Print version
$ nohuman --help
Remove human reads from a sequencing run
Usage: nohuman [OPTIONS] [INPUT]...
Arguments:
[INPUT]...
Input file(s) to remove human reads from
Options:
-o, --out1 <OUTPUT_1>
First output file.
Defaults to the name of the first input file with the suffix "nohuman" appended.
e.g. "input_1.fastq" -> "input_1.nohuman.fq".
Compression of the output file is determined by the file extension of the output file name.
Or by using the `--output-type` option. If no output path is given, the same compression
as the input file will be used.
-O, --out2 <OUTPUT_2>
Second output file.
Defaults to the name of the first input file with the suffix "nohuman" appended.
e.g. "input_2.fastq" -> "input_2.nohuman.fq".
Compression of the output file is determined by the file extension of the output file name.
Or by using the `--output-type` option. If no output path is given, the same compression
as the input file will be used.
-c, --check
Check that all required dependencies are available and exit
-d, --download
Download the database
-D, --db <PATH>
Path to the database
[default: ~/.nohuman/db]
--db-version <VERSION>
Name of a downloaded database version to use (use `all` with `--download` to fetch every version)
--list-db-versions
List available database versions and exit
-F, --output-type <FORMAT>
Output compression format. u: uncompressed; b: Bzip2; g: Gzip; x: Xz (Lzma); z: Zstd
If not provided, the format will be inferred from the given output file name(s), or the
format of the input file(s) if no output file name(s) are given.
-t, --threads <INT>
Number of threads to use in kraken2 and optional output compression. Cannot be 0
[default: 1]
-H, --human
Output human reads instead of removing them
-C, --conf <[0, 1]>
Kraken2 minimum confidence score
[default: 0.0]
-k, --kraken-output <FILE>
Write the Kraken2 read classification output to a file
-r, --kraken-report <FILE>
Write the Kraken2 report with aggregate counts/clade to file
-v, --verbose
Set the logging level to verbose
-h, --help
Print help (see a summary with '-h')
-V, --version
Print version
Hostile is an alignment-based approach that performs well. It take longer and uses
more memory than the nohuman kraken approach, but has slightly better accuracy for Illumina data. See the paper for
more details and for other alternate approaches.
Hall, Michael B., and Lachlan J. M. Coin. “Pangenome databases improve host removal and mycobacteria classification from clinical metagenomic data” GigaScience, April 4, 2024. https://doi.org/10.1093/gigascience/giae010
@article{hall_pangenome_2024,
title = {Pangenome databases improve host removal and mycobacteria classification from clinical metagenomic data},
volume = {13},
issn = {2047-217X},
url = {https://doi.org/10.1093/gigascience/giae010},
doi = {10.1093/gigascience/giae010},
urldate = {2024-04-07},
journal = {GigaScience},
author = {Hall, Michael B and Coin, Lachlan J M},
month = jan,
year = {2024},
pages = {giae010},
}