Mercurial > repos > rliterman > csp2

comparison CSP2/CSP2_env/env-d9b9114564458d9d-741b3de822f2aaca6c6caa4325c4afce/opt/bbmap-39.01-1/splitbytaxa.sh @ 69:33d812a61356

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
planemo upload commit 2e9511a184a1ca667c7be0c6321a36dc4e3d116d
author jpayne
date Tue, 18 Mar 2025 17:55:14 -0400
parents
children
comparison
equal deleted inserted replaced
67:0e9998148a16 69:33d812a61356
1 #!/bin/bash
2
3 usage(){
4 echo "
5 Written by Brian Bushnell
6 Last modified Jan 7, 2020
7
8 Description: Splits sequences according to their taxonomy,
9 as determined by the sequence name. Sequences should
10 be labeled with a gi number, NCBI taxID, or species name.
11
12 Usage: splitbytaxa.sh in=<input file> out=<output pattern> tree=<tree file> table=<table file> level=<name or number>
13
14 Input may be fasta or fastq, compressed or uncompressed.
15
16
17 Standard parameters:
18 in=<file> Primary input.
19 out=<file> Output pattern; must contain % symbol.
20 overwrite=f (ow) Set to false to force the program to abort rather than
21 overwrite an existing file.
22 showspeed=t (ss) Set to 'f' to suppress display of processing speed.
23 ziplevel=2 (zl) Set to 1 (lowest) through 9 (max) to change compression
24 level; lower compression is faster.
25
26 Processing parameters:
27 level=phylum Taxonomic level, such as phylum. Filtering will operate on
28 sequences within the same taxonomic level as specified ids.
29 tree= A taxonomic tree made by TaxTree, such as tree.taxtree.gz.
30 table= A table translating gi numbers to NCBI taxIDs.
31 Only needed if gi numbers will be used.
32 On Genepool, use 'tree=auto table=auto'.
33 * Note *
34 Tree and table files are in /global/projectb/sandbox/gaag/bbtools/tax
35 For non-Genepool users, or to make new ones, use taxtree.sh and gitable.sh
36
37 Java Parameters:
38 -Xmx This will set Java's memory usage, overriding automatic
39 memory detection.
40 -Xmx20g will specify 20 gigs of RAM, and -Xmx200m will specify
41 200 megs. The max is typically 85% of physical memory.
42 -eoom This flag will cause the process to exit if an out-of-memory
43 exception occurs. Requires Java 8u92+.
44 -da Disable assertions.
45
46 Please contact Brian Bushnell at bbushnell@lbl.gov if you encounter any problems.
47 "
48 }
49
50 #This block allows symlinked shellscripts to correctly set classpath.
51 pushd . > /dev/null
52 DIR="${BASH_SOURCE[0]}"
53 while [ -h "$DIR" ]; do
54 cd "$(dirname "$DIR")"
55 DIR="$(readlink "$(basename "$DIR")")"
56 done
57 cd "$(dirname "$DIR")"
58 DIR="$(pwd)/"
59 popd > /dev/null
60
61 #DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )/"
62 CP="$DIR""current/"
63
64 z="-Xmx4g"
65 z2="-Xms4g"
66 set=0
67
68 if [ -z "$1" ] || [[ $1 == -h ]] || [[ $1 == --help ]]; then
69 usage
70 exit
71 fi
72
73 calcXmx () {
74 source "$DIR""/calcmem.sh"
75 setEnvironment
76 parseXmx "$@"
77 if [[ $set == 1 ]]; then
78 return
79 fi
80 freeRam 1000m 84
81 z="-Xmx${RAM}m"
82 z2="-Xms${RAM}m"
83 }
84 calcXmx "$@"
85
86 splitbytaxa() {
87 local CMD="java $EA $EOOM $z -cp $CP tax.SplitByTaxa $@"
88 echo $CMD >&2
89 eval $CMD
90 }
91
92 splitbytaxa "$@"