|
0
|
1 <tool id="bamSplit" name="Split" version="2.4.1">
|
|
|
2 <description>BAM datasets on variety of attributes</description>
|
|
|
3 <macros>
|
|
|
4 <xml name="macro_output_type">
|
|
|
5 <param name="output_type" type="select" label="output as">
|
|
|
6 <option value="history_items">history items</option>
|
|
|
7 <option value="dataset_collection">dataset_collection</option>
|
|
|
8 </param>
|
|
|
9 </xml>
|
|
|
10 </macros>
|
|
|
11 <requirements>
|
|
|
12 <requirement type="package" version="2.4.0">bamtools</requirement>
|
|
|
13 </requirements>
|
|
|
14 <command>
|
|
|
15 <![CDATA[
|
|
|
16 echo "BAM" > $report &&
|
|
|
17 #for $bam_count, $input_bam in enumerate( $input_bams ):
|
|
|
18 ln -s "${input_bam}" "localbam_${bam_count}.bam" &&
|
|
|
19 ln -s "${input_bam.metadata.bam_index}" "localbam_${bam_count}.bam.bai" &&
|
|
|
20 #end for
|
|
|
21 bamtools
|
|
|
22 split
|
|
|
23 #if str ( $analysis_type.analysis_type_selector ) == "-tag" :
|
|
|
24 ${analysis_type.analysis_type_selector} "${analysis_type.tag_name}"
|
|
|
25 #else
|
|
|
26 ${analysis_type.analysis_type_selector}
|
|
|
27 #end if
|
|
|
28 -stub split_bam
|
|
|
29 #for $bam_count, $input_bam in enumerate( $input_bams ):
|
|
|
30 -in "localbam_${bam_count}.bam"
|
|
|
31 #end for
|
|
|
32 #if str ( $analysis_type.analysis_type_selector ) == "-reference" and $analysis_type.output_type == "dataset_collection":
|
|
|
33 #import re
|
|
|
34 #set $name = $re.sub('\W','_',$re.sub('\.bam$','',$input_bams[0].name))
|
|
|
35 #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(',')])
|
|
|
36 && (export I=0;
|
|
|
37 for i in $ref_list;
|
|
|
38 do I=\$((++I)); SN=`printf "split_bam.REF_%s.bam" "\$i"`;
|
|
|
39 if [ -e \$SN ];
|
|
|
40 then FN=`printf "split_bam%05d%s.%s.bam" \$((I)) "$name" "\$i"`;
|
|
|
41 mv \$SN \$FN;
|
|
|
42 fi;
|
|
|
43 done)
|
|
|
44 #end if
|
|
|
45 ]]>
|
|
|
46 </command>
|
|
|
47 <inputs>
|
|
|
48 <param name="input_bams" type="data" format="bam" label="BAM dataset(s) to filter" min="1" multiple="True"/>
|
|
|
49 <conditional name="analysis_type">
|
|
|
50 <param name="analysis_type_selector" type="select" label="Split BAM dataset(s) by" help="See help below for explanation of each option">
|
|
|
51 <option value="-mapped">Mapping status (-mapped)</option>
|
|
|
52 <option value="-paired">Pairing status (-paired)</option>
|
|
|
53 <option value="-reference">Reference name (-reference)</option>
|
|
|
54 <option value="-tag">Specific tag (-tag)</option>
|
|
|
55 </param>
|
|
|
56 <when value="-mapped" />
|
|
|
57 <when value="-paired" />
|
|
|
58 <when value="-reference" >
|
|
|
59 <expand macro="macro_output_type" />
|
|
|
60 </when>
|
|
|
61 <when value="-tag">
|
|
|
62 <param name="tag_name" type="text" value="NM" label="Enter tag name here" help="For example, to split on NM tag enter "NM""/>
|
|
|
63 <expand macro="macro_output_type" />
|
|
|
64 </when>
|
|
|
65 </conditional>
|
|
|
66 </inputs>
|
|
|
67 <outputs>
|
|
|
68 <data format="txt" name="report" label="BAMSplitter Run" hidden="true">
|
|
|
69 <discover_datasets pattern="split_bam\.(?P<designation>.+)\.bam" ext="bam" visible="true"/>
|
|
|
70 <filter>analysis_type['analysis_type_selector'] in ('-mapped','-paired') or analysis_type['output_type'] != 'dataset_collection'</filter>
|
|
|
71 </data>
|
|
|
72 <collection name="output_bams" type="list" label="${input_bams[0].name} by ${analysis_type.analysis_type_selector.replace('-','')}">
|
|
|
73 <discover_datasets pattern="split_bam\d*(?P<designation>.+)\.bam" ext="bam" visible="false"/>
|
|
|
74 <filter>analysis_type['analysis_type_selector'] in ('-reference','-tag') and analysis_type['output_type'] == 'dataset_collection'</filter>
|
|
|
75 </collection>
|
|
|
76 </outputs>
|
|
|
77 <tests>
|
|
|
78 <test>
|
|
|
79 <param name="input_bams" ftype="bam" value="bamtools-input1.bam"/>
|
|
|
80 <param name="analysis_type_selector" value="-mapped"/>
|
|
|
81 <output name="report">
|
|
|
82 <assert_contents>
|
|
|
83 <has_line line="BAM" />
|
|
|
84 </assert_contents>
|
|
|
85 <discovered_dataset designation="MAPPED" file="bamtools-split-test1.bam" ftype="bam"/>
|
|
|
86 </output>
|
|
|
87 </test>
|
|
|
88 <test>
|
|
|
89 <param name="input_bams" ftype="bam" value="bamtools-input2.bam"/>
|
|
|
90 <param name="analysis_type_selector" value="-reference"/>
|
|
|
91 <param name="output_type" value="dataset_collection"/>
|
|
|
92 <output_collection name="output_bams" type="list">
|
|
|
93 <element name="bamtools_input2.chr1" file="bamtools_input2.chr1" compare="sim_size" delta="500" />
|
|
|
94 </output_collection>
|
|
|
95 </test>
|
|
|
96
|
|
|
97 </tests>
|
|
|
98 <help>
|
|
|
99 **What is does**
|
|
|
100
|
|
|
101 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).
|
|
|
102
|
|
|
103 -----
|
|
|
104
|
|
|
105 .. class:: warningmark
|
|
|
106
|
|
|
107 **DANGER: Multiple Outputs**
|
|
|
108
|
|
|
109 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.
|
|
|
110
|
|
|
111 -----
|
|
|
112
|
|
|
113 **How it works**
|
|
|
114
|
|
|
115 The following options can be specified via "**Split BAM dataset(s) by**" dropdown::
|
|
|
116
|
|
|
117 Mapping status (-mapped) split mapped/unmapped and generate two output files
|
|
|
118 named (MAPPED) and (UNMAPPED) containing mapped and unmapped
|
|
|
119 reads, respectively.
|
|
|
120
|
|
|
121 Pairing status (-paired) split single-end/paired-end alignments and generate two output files
|
|
|
122 named (SINGLE_END) and (PAIRED_END) containing paired and unpaired
|
|
|
123 reads, respectively.
|
|
|
124
|
|
|
125 Reference name (-reference) split alignments by reference name. In cases of unfinished genomes with
|
|
|
126 very large number of reference sequences (scaffolds) it can generate
|
|
|
127 thousands (if not millions) of output datasets.
|
|
|
128
|
|
|
129 Specific tag (-tag) split alignments based on all values of TAG encountered. Choosing this
|
|
|
130 option from the menu will allow you to enter the tag name. As was the
|
|
|
131 case with the reference splitting above, this option can produce very
|
|
|
132 large number of outputs if a tag has a large number of unique values.
|
|
|
133
|
|
|
134 -----
|
|
|
135
|
|
|
136 .. class:: infomark
|
|
|
137
|
|
|
138 **More information**
|
|
|
139
|
|
|
140 Additional information about BAMtools can be found at https://github.com/pezmaster31/bamtools/wiki
|
|
|
141
|
|
|
142 </help>
|
|
|
143 <citations>
|
|
|
144 <citation type="doi">10.1093/bioinformatics/btr174</citation>
|
|
|
145 </citations>
|
|
|
146 </tool>
|
