|
jpayne@69
|
1 #!/bin/bash
|
|
jpayne@69
|
2
|
|
jpayne@69
|
3 usage(){
|
|
jpayne@69
|
4 echo "
|
|
jpayne@69
|
5 Written by Brian Bushnell
|
|
jpayne@69
|
6 Last modified October 4, 2019
|
|
jpayne@69
|
7
|
|
jpayne@69
|
8 Description: Discards all but the best copy of a ribosomal gene per TaxID.
|
|
jpayne@69
|
9 Gene sequences should be named like this: >tid|123|whatever
|
|
jpayne@69
|
10 Sequences are selected based on the number of fully defined bases.
|
|
jpayne@69
|
11
|
|
jpayne@69
|
12 Usage: keepbest.sh in=<input file> out=<output file> rate=<float>
|
|
jpayne@69
|
13
|
|
jpayne@69
|
14 Input may be fasta or fastq, compressed or uncompressed.
|
|
jpayne@69
|
15
|
|
jpayne@69
|
16 Standard parameters:
|
|
jpayne@69
|
17 in=<file> Input sequences.
|
|
jpayne@69
|
18 out=<file> Output sequences.
|
|
jpayne@69
|
19 overwrite=f (ow) Set to false to force the program to abort rather than
|
|
jpayne@69
|
20 overwrite an existing file.
|
|
jpayne@69
|
21 ziplevel=2 (zl) Set to 1 (lowest) through 9 (max) to change compression
|
|
jpayne@69
|
22 level; lower compression is faster.
|
|
jpayne@69
|
23
|
|
jpayne@69
|
24 Processing parameters:
|
|
jpayne@69
|
25 maxlen=1600 Prefer sequences shorter than this.
|
|
jpayne@69
|
26
|
|
jpayne@69
|
27 Java Parameters:
|
|
jpayne@69
|
28 -Xmx This will set Java's memory usage, overriding autodetection.
|
|
jpayne@69
|
29 -Xmx20g will specify 20 gigs of RAM, and -Xmx200m will
|
|
jpayne@69
|
30 specify 200 megs. The max is typically 85% of physical memory.
|
|
jpayne@69
|
31 -eoom This flag will cause the process to exit if an out-of-memory
|
|
jpayne@69
|
32 exception occurs. Requires Java 8u92+.
|
|
jpayne@69
|
33 -da Disable assertions.
|
|
jpayne@69
|
34
|
|
jpayne@69
|
35 Please contact Brian Bushnell at bbushnell@lbl.gov if you encounter any problems.
|
|
jpayne@69
|
36 "
|
|
jpayne@69
|
37 }
|
|
jpayne@69
|
38
|
|
jpayne@69
|
39 #This block allows symlinked shellscripts to correctly set classpath.
|
|
jpayne@69
|
40 pushd . > /dev/null
|
|
jpayne@69
|
41 DIR="${BASH_SOURCE[0]}"
|
|
jpayne@69
|
42 while [ -h "$DIR" ]; do
|
|
jpayne@69
|
43 cd "$(dirname "$DIR")"
|
|
jpayne@69
|
44 DIR="$(readlink "$(basename "$DIR")")"
|
|
jpayne@69
|
45 done
|
|
jpayne@69
|
46 cd "$(dirname "$DIR")"
|
|
jpayne@69
|
47 DIR="$(pwd)/"
|
|
jpayne@69
|
48 popd > /dev/null
|
|
jpayne@69
|
49
|
|
jpayne@69
|
50 #DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )/"
|
|
jpayne@69
|
51 CP="$DIR""current/"
|
|
jpayne@69
|
52
|
|
jpayne@69
|
53 z="-Xmx8g"
|
|
jpayne@69
|
54 z2="-Xms8g"
|
|
jpayne@69
|
55 set=0
|
|
jpayne@69
|
56
|
|
jpayne@69
|
57 if [ -z "$1" ] || [[ $1 == -h ]] || [[ $1 == --help ]]; then
|
|
jpayne@69
|
58 usage
|
|
jpayne@69
|
59 exit
|
|
jpayne@69
|
60 fi
|
|
jpayne@69
|
61
|
|
jpayne@69
|
62 calcXmx () {
|
|
jpayne@69
|
63 source "$DIR""/calcmem.sh"
|
|
jpayne@69
|
64 setEnvironment
|
|
jpayne@69
|
65 parseXmx "$@"
|
|
jpayne@69
|
66 if [[ $set == 1 ]]; then
|
|
jpayne@69
|
67 return
|
|
jpayne@69
|
68 fi
|
|
jpayne@69
|
69 freeRam 8g 84
|
|
jpayne@69
|
70 z="-Xmx${RAM}m"
|
|
jpayne@69
|
71 z2="-Xms${RAM}m"
|
|
jpayne@69
|
72 }
|
|
jpayne@69
|
73 calcXmx "$@"
|
|
jpayne@69
|
74
|
|
jpayne@69
|
75 keepbest() {
|
|
jpayne@69
|
76 local CMD="java $EA $EOOM $z -cp $CP jgi.KeepBestCopy $@"
|
|
jpayne@69
|
77 echo $CMD >&2
|
|
jpayne@69
|
78 eval $CMD
|
|
jpayne@69
|
79 }
|
|
jpayne@69
|
80
|
|
jpayne@69
|
81 keepbest "$@"
|
