Mercurial > repos > rliterman > csp2
comparison CSP2/CSP2_env/env-d9b9114564458d9d-741b3de822f2aaca6c6caa4325c4afce/opt/bbmap-39.01-1/loglog.sh @ 69:33d812a61356
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 24, 2020 | |
| 7 | |
| 8 Description: Estimates cardinality of unique kmers in sequence data. | |
| 9 See also kmercountmulti.sh. | |
| 10 | |
| 11 Usage: loglog.sh in=<file> k=<31> | |
| 12 | |
| 13 Parameters: | |
| 14 in=<file> (in1) Input file, or comma-delimited list of files. | |
| 15 in2=<file> ptional second file for paired reads. | |
| 16 k=31 Use this kmer length for counting. | |
| 17 buckets=2048 Use this many buckets for counting; higher decreases | |
| 18 variance, for large datasets. Must be a power of 2. | |
| 19 seed=-1 Use this seed for hash functions. A negative number forces | |
| 20 a random seed. | |
| 21 minprob=0 Set to a value between 0 and 1 to exclude kmers with a lower | |
| 22 probability of being correct. | |
| 23 | |
| 24 | |
| 25 Shortcuts: | |
| 26 The # symbol will be substituted for 1 and 2. | |
| 27 For example: | |
| 28 loglog.sh in=read#.fq | |
| 29 ...is equivalent to: | |
| 30 loglog.sh in1=read1.fq in2=read2.fq | |
| 31 | |
| 32 Java Parameters: | |
| 33 -Xmx This will set Java's memory usage, overriding autodetection. | |
| 34 -Xmx20g will specify 20 gigs of RAM, and -Xmx200m will specify 200 megs. | |
| 35 The max is typically 85% of physical memory. | |
| 36 -eoom This flag will cause the process to exit if an out-of-memory | |
| 37 exception occurs. Requires Java 8u92+. | |
| 38 -da Disable assertions. | |
| 39 | |
| 40 Supported input formats are fastq, fasta, scarf, sam, and bam. | |
| 41 Supported compression formats are gzip and bz2. | |
| 42 To read from stdin, set 'in=stdin'. The format should be specified with an extension, like 'in=stdin.fq.gz' | |
| 43 | |
| 44 Please contact Brian Bushnell at bbushnell@lbl.gov if you encounter any problems. | |
| 45 " | |
| 46 } | |
| 47 | |
| 48 #This block allows symlinked shellscripts to correctly set classpath. | |
| 49 pushd . > /dev/null | |
| 50 DIR="${BASH_SOURCE[0]}" | |
| 51 while [ -h "$DIR" ]; do | |
| 52 cd "$(dirname "$DIR")" | |
| 53 DIR="$(readlink "$(basename "$DIR")")" | |
| 54 done | |
| 55 cd "$(dirname "$DIR")" | |
| 56 DIR="$(pwd)/" | |
| 57 popd > /dev/null | |
| 58 | |
| 59 #DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )/" | |
| 60 CP="$DIR""current/" | |
| 61 | |
| 62 z="-Xmx200m" | |
| 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 loglog() { | |
| 78 local CMD="java $EA $EOOM $z -cp $CP cardinality.LogLogWrapper $@" | |
| 79 echo $CMD >&2 | |
| 80 eval $CMD | |
| 81 } | |
| 82 | |
| 83 loglog "$@" |