Mercurial > repos > rliterman > csp2

comparison CSP2/CSP2_env/env-d9b9114564458d9d-741b3de822f2aaca6c6caa4325c4afce/opt/bbmap-39.01-1/stats.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 March 3, 2020
7
8 Description: Generates basic assembly statistics such as scaffold count,
9 N50, L50, GC content, gap percent, etc. For multiple files, please use
10 statswrapper.sh. Works with fasta and fastq only (gzipped is fine).
11 Please read bbmap/docs/guides/StatsGuide.txt for more information.
12
13 Usage: stats.sh in=<file>
14
15 Parameters:
16 in=file Specify the input fasta file, or stdin.
17 out=stdout Destination of primary output; may be directed to a file.
18 gc=file Writes ACGTN content per scaffold to a file.
19 gchist=file Filename to output scaffold gc content histogram.
20 shist=file Filename to output cumulative scaffold length histogram.
21 gcbins=200 Number of bins for gc histogram.
22 n=10 Number of contiguous Ns to signify a break between contigs.
23 k=13 Estimate memory usage of BBMap with this kmer length.
24 minscaf=0 Ignore scaffolds shorter than this.
25 phs=f (printheaderstats) Set to true to print total size of headers.
26 n90=t (printn90) Print the N/L90 metrics.
27 extended=f Print additional metrics such as L90, logsum, and score.
28 pdl=f (printduplicatelines) Set to true to print lines in the
29 scaffold size table where the counts did not change.
30 n_=t This flag will prefix the terms 'contigs' and 'scaffolds'
31 with 'n_' in formats 3-6.
32 addname=f Adds a column for input file name, for formats 3-6.
33
34 Logsum and Powsum:
35 logoffset=1000 Minimum length for calculating log sum.
36 logbase=2 Log base for calculating log sum.
37 logpower=1 Raise the log to a power to increase the weight
38 of longer scaffolds for log sum.
39 powsum=0.25 Use this power of the length to increase weight
40 of longer scaffolds for power sum.
41
42 Assembly Score Metric:
43 score=f Print assembly score.
44 aligned=0.0 Set the fraction of aligned reads (0-1).
45 assemblyscoreminlen=2000 Minimum length of scaffolds to include in
46 assembly score calculation.
47 assemblyscoremaxlen=50000 Maximum length of scaffolds to get bonus points
48 for being long.
49
50
51 format=<0-7> Format of the stats information; default 1.
52 format=0 prints no assembly stats.
53 format=1 uses variable units like MB and KB, and is designed for compatibility with existing tools.
54 format=2 uses only whole numbers of bases, with no commas in numbers, and is designed for machine parsing.
55 format=3 outputs stats in 2 rows of tab-delimited columns: a header row and a data row.
56 format=4 is like 3 but with scaffold data only.
57 format=5 is like 3 but with contig data only.
58 format=6 is like 3 but the header starts with a #.
59 format=7 is like 1 but only prints contig info.
60 format=8 is like 3 but in JSON. You can also just use the 'json' flag.
61
62 gcformat=<0-5> Select GC output format; default 1.
63 gcformat=0: (no base content info printed)
64 gcformat=1: name length A C G T N GC
65 gcformat=2: name GC
66 gcformat=4: name length GC
67 gcformat=5: name length GC logsum powsum
68 Note that in gcformat 1, A+C+G+T=1 even when N is nonzero.
69
70 Please contact Brian Bushnell at bbushnell@lbl.gov if you encounter any problems.
71 "
72 }
73
74 #This block allows symlinked shellscripts to correctly set classpath.
75 pushd . > /dev/null
76 DIR="${BASH_SOURCE[0]}"
77 while [ -h "$DIR" ]; do
78 cd "$(dirname "$DIR")"
79 DIR="$(readlink "$(basename "$DIR")")"
80 done
81 cd "$(dirname "$DIR")"
82 DIR="$(pwd)/"
83 popd > /dev/null
84
85 #DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )/"
86 CP="$DIR""current/"
87
88 z="-Xmx120m"
89 set=0
90
91 if [ -z "$1" ] || [[ $1 == -h ]] || [[ $1 == --help ]]; then
92 usage
93 exit
94 fi
95
96 calcXmx () {
97 source "$DIR""/calcmem.sh"
98 setEnvironment
99 parseXmx "$@"
100 }
101 calcXmx "$@"
102
103 stats() {
104 local CMD="java $EA $EOOM $z -cp $CP jgi.AssemblyStats2 $@"
105 # echo $CMD >&2
106 eval $CMD
107 }
108
109 stats "$@"