Mercurial > repos > rliterman > csp2

comparison CSP2/CSP2_env/env-d9b9114564458d9d-741b3de822f2aaca6c6caa4325c4afce/opt/bbmap-39.01-1/callgenes.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 February 10, 2020
7
8 Description: Finds orfs and calls genes in unspliced prokaryotes.
9 This includes bacteria, archaea, viruses, and mitochondria.
10 Can also predict 16S, 23S, 5S, and tRNAs.
11
12 Usage: callgenes.sh in=contigs.fa out=calls.gff outa=aminos.faa out16S=16S.fa
13
14 File parameters:
15 in=<file> A fasta file; the only required parameter.
16 out=<file> Output gff file.
17 outa=<file> Amino acid output.
18 out16s=<file> 16S output.
19 model=<file> A pgm file or comma-delimited list.
20 If unspecified a default model will be used.
21 stats=stderr Stats output (may be stderr, stdin, a file, or null).
22 hist=null Gene length histogram.
23 compareto= Optional reference gff file to compare with the gene calls.
24 'auto' will name it based on the input file name.
25
26 Formatting parameters:
27 json=false Print stats in JSON.
28 binlen=20 Histogram bin length.
29 bins=2000 Maximum histogram bins.
30 pz=f (printzero) Print histogram lines with zero count.
31
32
33
34 Other parameters:
35 minlen=60 Don't call genes shorter than this.
36 trd=f (trimreaddescription) Set to true to trim read headers after
37 the first whitespace. Necessary for IGV.
38 merge=f For paired reads, merge before calling.
39 detranslate=f Output canonical nucleotide sequences instead of amino acids.
40 recode=f Re-encode nucleotide sequences over called genes, leaving
41 non-coding regions unchanged.
42
43 Please contact Brian Bushnell at bbushnell@lbl.gov if you encounter any problems.
44 "
45 }
46
47 #This block allows symlinked shellscripts to correctly set classpath.
48 pushd . > /dev/null
49 DIR="${BASH_SOURCE[0]}"
50 while [ -h "$DIR" ]; do
51 cd "$(dirname "$DIR")"
52 DIR="$(readlink "$(basename "$DIR")")"
53 done
54 cd "$(dirname "$DIR")"
55 DIR="$(pwd)/"
56 popd > /dev/null
57
58 #DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )/"
59 CP="$DIR""current/"
60
61 z="-Xmx6g"
62 z2="-Xms6g"
63 set=0
64
65 if [ -z "$1" ] || [[ $1 == -h ]] || [[ $1 == --help ]]; then
66 usage
67 exit
68 fi
69
70 calcXmx () {
71 source "$DIR""/calcmem.sh"
72 setEnvironment
73 parseXmx "$@"
74 }
75 calcXmx "$@"
76
77 function callgenes() {
78 local CMD="java $EA $EOOM $z $z2 -cp $CP prok.CallGenes $@"
79 #Too long to echo sometimes since wildcards can be expanded
80 #echo $CMD >&2
81 eval $CMD
82 }
83
84 callgenes "$@"