|
jpayne@69
|
1 #!/bin/bash
|
|
jpayne@69
|
2
|
|
jpayne@69
|
3 usage(){
|
|
jpayne@69
|
4 echo "
|
|
jpayne@69
|
5 Written by Brian Bushnell
|
|
jpayne@69
|
6 Last modified December 22, 2021
|
|
jpayne@69
|
7 This script requires at least 52GB RAM.
|
|
jpayne@69
|
8 It is designed for NERSC and uses hard-coded paths.
|
|
jpayne@69
|
9
|
|
jpayne@69
|
10 Description: Removes all reads that map to the cat, dog, mouse, or human genome with at least 95% identity after quality trimming.
|
|
jpayne@69
|
11 Removes approximately 98.6% of human 2x150bp reads, with zero false-positives to non-animals.
|
|
jpayne@69
|
12 NOTE! This program uses hard-coded paths and will only run on Nersc systems.
|
|
jpayne@69
|
13
|
|
jpayne@69
|
14 Usage: removecatdogmousehuman.sh in=<input file> outu=<clean output file>
|
|
jpayne@69
|
15
|
|
jpayne@69
|
16 Input may be fasta or fastq, compressed or uncompressed.
|
|
jpayne@69
|
17
|
|
jpayne@69
|
18 Parameters:
|
|
jpayne@69
|
19 threads=auto (t) Set number of threads to use; default is number of logical processors.
|
|
jpayne@69
|
20 overwrite=t (ow) Set to false to force the program to abort rather than overwrite an existing file.
|
|
jpayne@69
|
21 interleaved=auto (int) If true, forces fastq input to be paired and interleaved.
|
|
jpayne@69
|
22 trim=t Trim read ends to remove bases with quality below minq.
|
|
jpayne@69
|
23 Values: t (trim both ends), f (neither end), r (right end only), l (left end only).
|
|
jpayne@69
|
24 untrim=t Undo the trimming after mapping.
|
|
jpayne@69
|
25 minq=4 Trim quality threshold.
|
|
jpayne@69
|
26 ziplevel=2 (zl) Set to 1 (lowest) through 9 (max) to change compression level; lower compression is faster.
|
|
jpayne@69
|
27 outm=<file> File to output the reads that mapped to human.
|
|
jpayne@69
|
28
|
|
jpayne@69
|
29 ***** All BBMap parameters can be used; run bbmap.sh for more details. *****
|
|
jpayne@69
|
30
|
|
jpayne@69
|
31 Please contact Brian Bushnell at bbushnell@lbl.gov if you encounter any problems.
|
|
jpayne@69
|
32 "
|
|
jpayne@69
|
33 }
|
|
jpayne@69
|
34
|
|
jpayne@69
|
35 #This block allows symlinked shellscripts to correctly set classpath.
|
|
jpayne@69
|
36 pushd . > /dev/null
|
|
jpayne@69
|
37 DIR="${BASH_SOURCE[0]}"
|
|
jpayne@69
|
38 while [ -h "$DIR" ]; do
|
|
jpayne@69
|
39 cd "$(dirname "$DIR")"
|
|
jpayne@69
|
40 DIR="$(readlink "$(basename "$DIR")")"
|
|
jpayne@69
|
41 done
|
|
jpayne@69
|
42 cd "$(dirname "$DIR")"
|
|
jpayne@69
|
43 DIR="$(pwd)/"
|
|
jpayne@69
|
44 popd > /dev/null
|
|
jpayne@69
|
45
|
|
jpayne@69
|
46 #DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )/"
|
|
jpayne@69
|
47 CP="$DIR""current/"
|
|
jpayne@69
|
48 JNI="-Djava.library.path=""$DIR""jni/"
|
|
jpayne@69
|
49 JNI=""
|
|
jpayne@69
|
50
|
|
jpayne@69
|
51 z="-Xmx50g"
|
|
jpayne@69
|
52 z2="-Xms50g"
|
|
jpayne@69
|
53 set=0
|
|
jpayne@69
|
54
|
|
jpayne@69
|
55 if [ -z "$1" ] || [[ $1 == -h ]] || [[ $1 == --help ]]; then
|
|
jpayne@69
|
56 usage
|
|
jpayne@69
|
57 exit
|
|
jpayne@69
|
58 fi
|
|
jpayne@69
|
59
|
|
jpayne@69
|
60 calcXmx () {
|
|
jpayne@69
|
61 source "$DIR""/calcmem.sh"
|
|
jpayne@69
|
62 setEnvironment
|
|
jpayne@69
|
63 parseXmx "$@"
|
|
jpayne@69
|
64 if [[ $set == 1 ]]; then
|
|
jpayne@69
|
65 return
|
|
jpayne@69
|
66 fi
|
|
jpayne@69
|
67 }
|
|
jpayne@69
|
68 calcXmx "$@"
|
|
jpayne@69
|
69
|
|
jpayne@69
|
70 function removecatdogmousehuman() {
|
|
jpayne@69
|
71 local CMD="java $EA $EOOM $z $z2 $JNI -cp $CP align2.BBMap minratio=0.9 maxindel=3 bwr=0.16 bw=12 quickmatch fast minhits=2 path=/global/cfs/cdirs/bbtools/mousecatdoghuman/ pigz unpigz zl=6 qtrim=r trimq=10 untrim idtag usemodulo printunmappedcount ztd=2 kfilter=25 maxsites=1 k=14 bloomfilter $@"
|
|
jpayne@69
|
72 echo $CMD >&2
|
|
jpayne@69
|
73 eval $CMD
|
|
jpayne@69
|
74 }
|
|
jpayne@69
|
75
|
|
jpayne@69
|
76 removecatdogmousehuman "$@"
|
