Commit 3c7dd73d authored by Andreas Tille's avatar Andreas Tille
Browse files

New upstream version 38.26+dfsg

parent e608f936

README.md

+4 −3
Original line number Diff line number Diff line 
# BBTools bioinformatics tools, including BBMap.

# Author: Brian Bushnell, Jon Rood

# Language: Java

# Author: Brian Bushnell, Jon Rood, Shijie Yao

# Language: Java, Bash

# Information about documentation is in /docs/readme.txt.

# Version 35.85
 


# Version 38.26

a_sample_mt.sh

+21 −6
Original line number Diff line number Diff line 
#!/bin/bash

#a_sample_mt in=<infile> out=<outfile>



usage(){

echo "

Written by Brian Bushnell

Last modified November 19, 2015

Last modified August 22, 2018



Description:  Does nothing.  Should be fast.
@@ -12,7 +11,6 @@ Usage: a_sample_mt.sh in=<input file> out=<output file> 


Input may be fasta or fastq, compressed or uncompressed.





Standard parameters:

in=<file>       Primary input, or read 1 input.

in2=<file>      Read 2 input if reads are in two files.
@@ -30,11 +28,14 @@ None yet! 
Java Parameters:

-Xmx            This will be passed to Java to set memory usage, overriding the program's automatic memory detection.

                -Xmx20g will specify 20 gigs of RAM, and -Xmx200m will specify 200 megs.  The max is typically 85% of physical memory.

-eoom           This flag will cause the process to exit if an out-of-memory exception occurs.  Requires Java 8u92+.

-da             Disable assertions.



Please contact Brian Bushnell at bbushnell@lbl.gov if you encounter any problems.

"

}



#This block allows symlinked shellscripts to correctly set classpath.

pushd . > /dev/null

DIR="${BASH_SOURCE[0]}"

while [ -h "$DIR" ]; do
@@ -51,6 +52,7 @@ CP="$DIR""current/" 
z="-Xmx4g"

z2="-Xms4g"

EA="-ea"

EOOM=""

set=0



if [ -z "$1" ] || [[ $1 == -h ]] || [[ $1 == --help ]]; then
@@ -71,12 +73,25 @@ calcXmx () { 
calcXmx "$@"



a_sample_mt() {

	if [[ $NERSC_HOST == genepool ]]; then

	if [[ $SHIFTER_RUNTIME == 1 ]]; then

		#Ignore NERSC_HOST

		shifter=1

	elif [[ $NERSC_HOST == genepool ]]; then

		module unload oracle-jdk

		module load oracle-jdk/1.7_64bit

		module load oracle-jdk/1.8_144_64bit

		module load pigz

	elif [[ $NERSC_HOST == denovo ]]; then

		module unload java

		module load java/1.8.0_144

		module load pigz

	elif [[ $NERSC_HOST == cori ]]; then

		module use /global/common/software/m342/nersc-builds/denovo/Modules/jgi

		module use /global/common/software/m342/nersc-builds/denovo/Modules/usg

		module unload java

		module load java/1.8.0_144

		module load pigz

	fi

	local CMD="java $EA $z -cp $CP jgi.A_SampleMT $@"

	local CMD="java $EA $EOOM $z -cp $CP jgi.A_SampleMT $@"

	echo $CMD >&2

	eval $CMD

}

addadapters.sh

+22 −6
Original line number Diff line number Diff line 
#!/bin/bash

#addadapters in=<infile> out=<outfile>



function usage(){

usage(){

echo "

Written by Brian Bushnell

Last modified February 17, 2015
@@ -11,13 +10,14 @@ The input is a set of reads, paired or unpaired. 
The output is those same reads with adapter sequence replacing some of the bases in some reads.

For paired reads, adapters are located in the same position in read1 and read2.

This is designed for benchmarking adapter-trimming software, and evaluating methodology.

randomreads.sh is better for paired reads, though, as it actually adds adapters at the correct location,

so that overlap may be used for adapter detection.



Usage:  addadapters.sh in=<file> in2=<file2> out=<outfile> out2=<outfile2> adapters=<file>



in2 and out2 are for paired reads and are optional.

If input is paired and there is only one output file, it will be written interleaved.





Parameters:

ow=f                (overwrite) Overwrites files that already exist.

int=f               (interleaved) Determines whether INPUT file is considered interleaved.
@@ -38,6 +38,7 @@ Please contact Brian Bushnell at bbushnell@lbl.gov if you encounter any problems 
"

}



#This block allows symlinked shellscripts to correctly set classpath.

pushd . > /dev/null

DIR="${BASH_SOURCE[0]}"

while [ -h "$DIR" ]; do
@@ -53,6 +54,7 @@ CP="$DIR""current/" 


z="-Xmx200m"

EA="-ea"

EOOM=""

set=0



if [ -z "$1" ] || [[ $1 == -h ]] || [[ $1 == --help ]]; then
@@ -67,11 +69,25 @@ calcXmx () { 
calcXmx "$@"



function addadapters() {

	if [[ $NERSC_HOST == genepool ]]; then

		module load oracle-jdk/1.7_64bit

	if [[ $SHIFTER_RUNTIME == 1 ]]; then

		#Ignore NERSC_HOST

		shifter=1

	elif [[ $NERSC_HOST == genepool ]]; then

		module unload oracle-jdk

		module load oracle-jdk/1.8_144_64bit

		module load pigz

	elif [[ $NERSC_HOST == denovo ]]; then

		module unload java

		module load java/1.8.0_144

		module load pigz

	elif [[ $NERSC_HOST == cori ]]; then

		module use /global/common/software/m342/nersc-builds/denovo/Modules/jgi

		module use /global/common/software/m342/nersc-builds/denovo/Modules/usg

		module unload java

		module load java/1.8.0_144

		module load pigz

	fi

	local CMD="java $EA $z -cp $CP jgi.AddAdapters $@"

	local CMD="java $EA $EOOM $z -cp $CP jgi.AddAdapters $@"

	echo $CMD >&2

	eval $CMD

}

analyzeaccession.sh

0 → 100755
+80 −0
Original line number Diff line number Diff line 
#!/bin/bash



usage(){

echo "

Written by Brian Bushnell

Last modified August 9, 2018



Description:  Looks at accessions to see how to compress them.



Usage:  analyzeaccession.sh *accession2taxid.gz out=<output file>



Parameters:

lines=-1        If positive, stop after this many lines.



Java Parameters:

-Xmx            This will be passed to Java to set memory usage, overriding the program's automatic memory detection.

                -Xmx20g will specify 20 gigs of RAM, and -Xmx200m will specify 200 megs.  The max is typically 85% of physical memory.

-eoom           This flag will cause the process to exit if an out-of-memory exception occurs.  Requires Java 8u92+.

-da             Disable assertions.



Please contact Brian Bushnell at bbushnell@lbl.gov if you encounter any problems.

"

}



#This block allows symlinked shellscripts to correctly set classpath.

pushd . > /dev/null

DIR="${BASH_SOURCE[0]}"

while [ -h "$DIR" ]; do

  cd "$(dirname "$DIR")"

  DIR="$(readlink "$(basename "$DIR")")"

done

cd "$(dirname "$DIR")"

DIR="$(pwd)/"

popd > /dev/null



#DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )/"

CP="$DIR""current/"



z="-Xmx400m"

z2="-Xms400m"

EA="-ea"

EOOM=""

set=0



if [ -z "$1" ] || [[ $1 == -h ]] || [[ $1 == --help ]]; then

	usage

	exit

fi



calcXmx () {

	source "$DIR""/calcmem.sh"

	parseXmx "$@"

}

calcXmx "$@"



a_sample_mt() {

	if [[ $SHIFTER_RUNTIME == 1 ]]; then

		#Ignore NERSC_HOST

		shifter=1

	elif [[ $NERSC_HOST == genepool ]]; then

		module unload oracle-jdk

		module load oracle-jdk/1.8_144_64bit

		module load pigz

	elif [[ $NERSC_HOST == denovo ]]; then

		module unload java

		module load java/1.8.0_144

		module load pigz

	elif [[ $NERSC_HOST == cori ]]; then

		module use /global/common/software/m342/nersc-builds/denovo/Modules/jgi

		module use /global/common/software/m342/nersc-builds/denovo/Modules/usg

		module unload java

		module load java/1.8.0_144

		module load pigz

	fi

	local CMD="java $EA $EOOM $z -cp $CP tax.AnalyzeAccession $@"

	echo $CMD >&2

	eval $CMD

}



a_sample_mt "$@"

analyzegenes.sh

0 → 100755
+81 −0
Original line number Diff line number Diff line 
#!/bin/bash



usage(){

echo "

Written by Brian Bushnell

Last modified September 27, 2018



Description:  Generates a prokaryotic gene model (.pkm) for gene calling.

Input is fasta and gff files.

The .pkm file may be used by CallGenes.



Usage:  analyzegenes.sh in=x.fa gff=x.gff out=x.pgm



File parameters:

in=<file>       A fasta file or comma-delimited list of fasta files.

gff=<file>      A gff file or comma-delimited list.  This is optional;

                if present, it must match the number of fasta files.

                If absent, a fasta file 'foo.fasta' will imply the

                presence of 'foo.gff'.

out=<file>      Output pgm file.



Please contact Brian Bushnell at bbushnell@lbl.gov if you encounter any problems.

"

}



#This block allows symlinked shellscripts to correctly set classpath.

pushd . > /dev/null

DIR="${BASH_SOURCE[0]}"

while [ -h "$DIR" ]; do

  cd "$(dirname "$DIR")"

  DIR="$(readlink "$(basename "$DIR")")"

done

cd "$(dirname "$DIR")"

DIR="$(pwd)/"

popd > /dev/null



#DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )/"

CP="$DIR""current/"



z="-Xmx1g"

z2="-Xms1g"

EA="-ea"

EOOM=""

set=0



if [ -z "$1" ] || [[ $1 == -h ]] || [[ $1 == --help ]]; then

	usage

	exit

fi



calcXmx () {

	source "$DIR""/calcmem.sh"

	parseXmx "$@"

}

calcXmx "$@"



function analyze() {

	if [[ $SHIFTER_RUNTIME == 1 ]]; then

		#Ignore NERSC_HOST

		shifter=1

	elif [[ $NERSC_HOST == genepool ]]; then

		module unload oracle-jdk

		module load oracle-jdk/1.8_144_64bit

		module load pigz

	elif [[ $NERSC_HOST == denovo ]]; then

		module unload java

		module load java/1.8.0_144

		module load pigz

	elif [[ $NERSC_HOST == cori ]]; then

		module use /global/common/software/m342/nersc-builds/denovo/Modules/jgi

		module use /global/common/software/m342/nersc-builds/denovo/Modules/usg

		module unload java

		module load java/1.8.0_144

		module load pigz

	fi

	local CMD="java $EA $EOOM $z $z2 -cp $CP prok.AnalyzeGenes $@"

	echo $CMD >&2

	eval $CMD

}



analyze "$@"