Mercurial > repos > iuc > bamtools_split_ref
changeset 0:09470ab960f1 draft
"planemo upload for repository https://github.com/galaxyproject/tools-iuc/tree/master/tool_collections/bamtools/bamtools_split commit 1cacd14cfb2429fefa91fdd05c82d2e36d23283a"
| author | iuc |
|---|---|
| date | Sat, 27 Nov 2021 10:03:33 +0000 |
| parents | |
| children | 9dbf707bebb0 |
| files | bamtools_split_ref.xml macros.xml test-data/bamtools-input-paired.bam test-data/bamtools-input-tags.bam test-data/bamtools-input1.bam test-data/bamtools-input2.bam test-data/bamtools-split-test1.bam test-data/bamtools_input2.chr1 test-data/bamtools_input_tags.TAG_XG_N.bam test-data/bamtools_input_tags.TAG_XG_V.bam test-data/split_bam.MAPPED.bam test-data/split_bam.PAIRED_END.bam test-data/split_bam.SINGLE_END.bam test-data/split_bam.UNMAPPED.bam |
| diffstat | 14 files changed, 105 insertions(+), 0 deletions(-) [+] |
line wrap: on
line diff
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/bamtools_split_ref.xml Sat Nov 27 10:03:33 2021 +0000 @@ -0,0 +1,89 @@ +<tool id="bamtools_split_ref" name="Split BAM by Reference" version="@TOOL_VERSION@+galaxy@VERSION_SUFFIX@"> + <description>into dataset list collection</description> + <macros> + <import>macros.xml</import> + </macros> + <expand macro="requirements" /> + <command> + <![CDATA[ + ln -s '${input_bam}' 'localbam.bam' && + ln -s '${input_bam.metadata.bam_index}' 'localbam.bam.bai' && + bamtools split -reference + -in localbam.bam + -stub split_bam + ## Preserve order from metadata in the output collection + #import re + #set $name = $re.sub('\W','_',$re.sub('\.bam$','',$input_bam.name)) + #if str($refs) != 'None': + #set $ref_list = ' '.join(str($refs).split(",")) + #else + #set $ref_list = ' '.join([$re.sub('^.*__sq__(.+)__sq__.*$','\\1',n) if n.find('__sq__') >= 0 else n for n in str($input_bam.metadata.reference_names).split(',')]) + #end if + && mkdir -p outputs + && (export I=0; + for i in $ref_list; + do I=\$((++I)); SN=`printf "split_bam.REF_%s.bam" "\$i"`; + if [ -e \$SN ]; then FN=`printf "outputs/split_bam%05d%s.%s.bam" \$((I)) '$name' "\$i"`; mv \$SN \$FN; fi; + done) + ]]> + </command> + <inputs> + <param name="input_bam" type="data" format="bam" label="BAM dataset to split by reference"/> + <param name="refs" type="select" optional="True" multiple="True" label="Select references (chromosomes and contigs) you would like to restrict bam to" > + <help><![CDATA[Click and type in the box above to see options. You can select multiple entries. + If "No options available" is displayed, you need to re-detect metadata on the input dataset. + ]]></help> + <options> + <filter type="data_meta" ref="input_bam" key="reference_names" /> + </options> + </param> + </inputs> + <outputs> + <collection name="output_bams" type="list" label="${input_bam.name} Split List"> + <discover_datasets pattern="split_bam\d*(?P<designation>.+)\.bam" ext="bam" directory="outputs" visible="false"/> + </collection> + </outputs> + <tests> + <test> + <param name="input_bam" ftype="bam" value="bamtools-input2.bam"/> + <output_collection name="output_bams" type="list"> + <element name="bamtools_input2.chr1" file="bamtools_input2.chr1" compare="sim_size" delta="500" /> + </output_collection> + </test> + </tests> + <help> +**What is does** + +BAMTools split is a utility for splitting BAM files. It is based on BAMtools suite of tools by Derek Barnett (https://github.com/pezmaster31/bamtools). + +----- + +.. class:: warningmark + +**DANGER: Multiple Outputs** + +As described below, splitting a BAM dataset(s) on reference name or a tag value can produce very large numbers of outputs. Read below and know what you are doing. + +----- + +**How it works** + +Split alignments by reference name into a dataset list collection. The collection will be in the same order as the input BAM references. + +In cases of unfinished genomes with very large number of reference sequences (scaffolds) +it can generate thousands (if not millions) of output datasets. + + +----- + +.. class:: infomark + +**More information** + +Additional information about BAMtools can be found at https://github.com/pezmaster31/bamtools/wiki + + </help> + <citations> + <citation type="doi">10.1093/bioinformatics/btr174</citation> + </citations> +</tool>
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/macros.xml Sat Nov 27 10:03:33 2021 +0000 @@ -0,0 +1,16 @@ +<?xml version="1.0"?> +<macros> + <token name="@TOOL_VERSION@">2.5.1</token> + <token name="@VERSION_SUFFIX@">0</token> + <xml name="requirements"> + <requirements> + <requirement type="package" version="@TOOL_VERSION@">bamtools</requirement> + <requirement type="package" version="1.4.1">samtools</requirement> + </requirements> + </xml> + <xml name="citations"> + <citations> + <citation type="doi">10.1093/bioinformatics/btr174</citation> + </citations> + </xml> +</macros> \ No newline at end of file