Mercurial > repos > rliterman > csp2

comparison CSP2/CSP2_env/env-d9b9114564458d9d-741b3de822f2aaca6c6caa4325c4afce/opt/bbmap-39.01-1/taxonomy.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: Prints the full taxonomy of a string.
9 String may be a gi number, NCBI taxID, or Latin name.
10 An NCBI identifier should just be a number or ncbi|number.
11 A gi number should be gi|number.
12 Please read bbmap/docs/guides/TaxonomyGuide.txt for more information.
13 Not: It is more convenient to use taxonomy.jgi-psf.org.
14
15 Usage: taxonomy.sh tree=<tree file> <identifier>
16 Alternate usage: taxonomy.sh tree=<tree file> in=<file>
17
18 Usage examples:
19 taxonomy.sh tree=tree.taxtree.gz homo_sapiens canis_lupus 9606
20 taxonomy.sh tree=tree.taxtree.gz gi=gitable.int1.d.gz in=refseq.fasta
21
22 Processing parameters:
23 in=<file> A file containing named sequences, or just the names.
24 out=<file> Output file. If blank, use stdout.
25 tree=<file> Specify a TaxTree file like tree.taxtree.gz.
26 On Genepool, use 'auto'.
27 gi=<file> Specify a gitable file like gitable.int1d.gz. Only needed
28 if gi numbers will be used. On Genepool, use 'auto'.
29 accession= Specify one or more comma-delimited NCBI accession to taxid
30 files. Only needed if accesions will be used; requires ~45GB
31 of memory. On Genepool, use 'auto'.
32 level=null Set to a taxonomic level like phylum to just print that level.
33 minlevel=-1 For multi-level printing, do not print levels below this.
34 maxlevel=life For multi-level printing, do not print levels above this.
35 silva=f Parse headers using Silva or semicolon-delimited syntax.
36 taxpath=auto Set the path to taxonomy files; auto only works at NERSC.
37
38 Parameters without an '=' symbol will be considered organism identifiers.
39
40 * Note *
41 Tree and table files are in /global/projectb/sandbox/gaag/bbtools/tax
42 For non-Genepool users, or to make new ones, use taxtree.sh and gitable.sh
43
44 Java Parameters:
45 -Xmx This will set Java's memory usage,
46 overriding autodetection.
47 -Xmx20g will specify 20 gigs of RAM, and -Xmx200m will specify
48 200 megs. The max is typically 85% of physical memory.
49 -eoom This flag will cause the process to exit if an out-of-memory
50 exception occurs. Requires Java 8u92+.
51 -da Disable assertions.
52
53 Please contact Brian Bushnell at bbushnell@lbl.gov if you encounter any problems.
54 "
55 }
56
57 #This block allows symlinked shellscripts to correctly set classpath.
58 pushd . > /dev/null
59 DIR="${BASH_SOURCE[0]}"
60 while [ -h "$DIR" ]; do
61 cd "$(dirname "$DIR")"
62 DIR="$(readlink "$(basename "$DIR")")"
63 done
64 cd "$(dirname "$DIR")"
65 DIR="$(pwd)/"
66 popd > /dev/null
67
68 #DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )/"
69 CP="$DIR""current/"
70
71 z="-Xmx8g"
72 z2="-Xms8g"
73 set=0
74
75 if [ -z "$1" ] || [[ $1 == -h ]] || [[ $1 == --help ]]; then
76 usage
77 exit
78 fi
79
80 calcXmx () {
81 source "$DIR""/calcmem.sh"
82 setEnvironment
83 parseXmx "$@"
84 if [[ $set == 1 ]]; then
85 return
86 fi
87 freeRam 2000m 84
88 z="-Xmx${RAM}m"
89 z2="-Xms${RAM}m"
90 }
91 calcXmx "$@"
92
93 taxonomy() {
94 local CMD="java $EA $EOOM $z -cp $CP tax.PrintTaxonomy $@"
95 echo $CMD >&2
96 eval $CMD
97 }
98
99 taxonomy "$@"