3a68b9598a
Filter reads from a FASTQ file using a list of identifiers.
Each entry in the input FASTQ file (or files) is checked against all
entries in the identifier list. Matches are included by default, or
excluded if the --invert flag is supplied. Paired-end files are kept
consistent (in order).
This is almost certainly not the most efficient way to implement this
filtering procedure. I tested a few different strategies and this one
seemed the fastest. Current timing with 16 processes is about 10
minutes per 1M paired reads with gzip'd input and output, depending on
the length of the identifier list to filter by.
usage: filter_fastq.py [-h] [-i INPUT] [-1 READ1] [-2 READ2] [-p NUM_THREADS]
[-o OUTPUT] [-f FILTER_FILE] [-v] [--gzip]
32 lines
895 B
Makefile
32 lines
895 B
Makefile
# $NetBSD: Makefile,v 1.1 2021/05/27 17:11:42 brook Exp $
|
|
|
|
PKGNAME= filter-fastq-0.0.0.20210527
|
|
GITHUB_PROJECT= filter-fastq
|
|
GITHUB_TAG= d2c9218
|
|
DISTNAME= filter-fastq
|
|
CATEGORIES= biology
|
|
MASTER_SITES= ${MASTER_SITE_GITHUB:=stephenfloor/}
|
|
EXTRACT_SUFX= .zip
|
|
DIST_SUBDIR= ${GITHUB_PROJECT}
|
|
|
|
MAINTAINER= pkgsrc-users@NetBSD.org
|
|
HOMEPAGE= https://github.com/stephenfloor/filter-fastq/
|
|
COMMENT= Filter reads from a FASTQ file
|
|
LICENSE= mit
|
|
|
|
WRKSRC= ${WRKDIR}/filter-fastq-d2c92182674a6d5aa257fb63eb60ac24ddb8b4a0
|
|
USE_LANGUAGES= # none
|
|
NO_BUILD= yes
|
|
|
|
PYTHON_VERSIONS_ACCEPTED= 27
|
|
|
|
REPLACE_PYTHON+= filter_fastq.py
|
|
|
|
INSTALLATION_DIRS+= bin share/doc/filter_fastq
|
|
|
|
do-install:
|
|
${INSTALL_SCRIPT} ${WRKSRC}/filter_fastq.py ${DESTDIR}${PREFIX}/bin
|
|
${INSTALL_DATA} ${WRKSRC}/README.md ${DESTDIR}${PREFIX}/share/doc/filter_fastq
|
|
|
|
.include "../../lang/python/application.mk"
|
|
.include "../../mk/bsd.pkg.mk"
|