Mercurial > repos > rliterman > csp2

comparison CSP2/CSP2_env/env-d9b9114564458d9d-741b3de822f2aaca6c6caa4325c4afce/opt/bbmap-39.01-1/kcompress.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 July 16, 2018
7
8 Description: Compresses sequence data into a fasta file containing each kmer
9 exactly once. Allows arbitrary kmer set operations via multiple passes.
10
11 Usage: kcompress.sh in=<reads> out=<contigs> min=<1> max=<2147483647>
12
13 Input parameters:
14 in=<file> Primary input file for reads to use as kmer data.
15 in2=<file> Second input file for paired data.
16 reads=-1 Only process this number of reads, then quit (-1 means all).
17
18 Output parameters:
19 out=<file> Write contigs (in contig mode).
20 showstats=t Print assembly statistics after writing contigs.
21 fuse=0 Fuse output sequences into chunks at least this long,
22 padded with 1 N between sequences.
23
24 Prefiltering parameters:
25 prefilter=0 If set to a positive integer, use a countmin sketch
26 to ignore kmers with depth of that value or lower.
27 prehashes=2 Number of hashes for prefilter.
28 prefiltersize=0.2 (pff) Fraction of memory to use for prefilter.
29 minprobprefilter=t (mpp) Use minprob for the prefilter.
30 prepasses=1 Use this many prefiltering passes; higher be more thorough
31 if the filter is very full. Set to 'auto' to iteratively
32 prefilter until the remaining kmers will fit in memory.
33
34 Hashing parameters:
35 k=31 Kmer length (1 to 31).
36 prealloc=t Pre-allocate memory rather than dynamically growing;
37 faster and more memory-efficient. A float fraction (0-1)
38 may be specified; default is 1.
39 minprob=0.5 Ignore kmers with overall probability of correctness below this.
40 minprobmain=t (mpm) Use minprob for the primary kmer counts.
41 threads=X Spawn X threads (default is number of logical processors).
42
43 Assembly parameters:
44 mincount=1 (min) Only retain kmers that occur at least this many times.
45 maxcount=BIG (max) Only retain kmers that occur at most this many times.
46 requiresamecount (rsc) Only build contigs from kmers with exactly the same count.
47 rcomp=t Store forward and reverse kmers together. Setting this to
48 false will only use forward kmers.
49
50
51 Java Parameters:
52 -Xmx This will set Java's memory usage, overriding autodetection.
53 -Xmx20g will specify 20 gigs of RAM, and -Xmx200m will specify 200 megs.
54 The max is typically 85% of physical memory.
55 -eoom This flag will cause the process to exit if an
56 out-of-memory exception occurs. Requires Java 8u92+.
57 -da Disable assertions.
58 "
59 }
60
61 #This block allows symlinked shellscripts to correctly set classpath.
62 pushd . > /dev/null
63 DIR="${BASH_SOURCE[0]}"
64 while [ -h "$DIR" ]; do
65 cd "$(dirname "$DIR")"
66 DIR="$(readlink "$(basename "$DIR")")"
67 done
68 cd "$(dirname "$DIR")"
69 DIR="$(pwd)/"
70 popd > /dev/null
71
72 #DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )/"
73 CP="$DIR""current/"
74
75 z="-Xmx14g"
76 z2="-Xms14g"
77 set=0
78
79 if [ -z "$1" ] || [[ $1 == -h ]] || [[ $1 == --help ]]; then
80 usage
81 exit
82 fi
83
84 calcXmx () {
85 source "$DIR""/calcmem.sh"
86 setEnvironment
87 parseXmx "$@"
88 if [[ $set == 1 ]]; then
89 return
90 fi
91 freeRam 15000m 84
92 z="-Xmx${RAM}m"
93 z2="-Xms${RAM}m"
94 }
95 calcXmx "$@"
96
97 kcompress() {
98 local CMD="java $EA $EOOM $z $z2 -cp $CP assemble.KmerCompressor $@"
99 echo $CMD >&2
100 eval $CMD
101 }
102
103 kcompress "$@"