|
0
|
1 <tool id="multigps" name="MultiGPS" version="0.5.0.0">
|
|
|
2 <description>analyzes collections of multi-condition ChIP-seq data</description>
|
|
|
3 <requirements>
|
|
|
4 <requirement type="package" version="4.11.0">meme</requirement>
|
|
|
5 <requirement type="package" version="3.11.0">edger</requirement>
|
|
|
6 <requirement type="package" version="2.14">biocbasics</requirement>
|
|
|
7 </requirements>
|
|
|
8 <command>
|
|
|
9 <![CDATA[
|
|
|
10 mkdir multigps_out &&
|
|
|
11 java -jar $__tool_directory__/multigps_v0.5.jar
|
|
|
12 --threads="\${GALAXY_SLOTS:-4}"
|
|
|
13 --verbose
|
|
|
14 --geninfo ${chromInfo}
|
|
|
15 #if str($use_motif_cond.use_motif) == "yes":
|
|
|
16 #set seq_dir = python "$get_seq_dir"
|
|
|
17 --seq "${seq_dir}"
|
|
|
18 --mememinw $use_motif_cond.min_motif_width
|
|
|
19 --mememaxw $use_motif_cond.max_motif_width
|
|
|
20 --design $design_file
|
|
|
21 #if str($gauss_model_smoothing_cond.gauss_model_smoothing) == "yes":
|
|
|
22 --gaussmodelsmoothing
|
|
|
23 --gausssmoothparam $gauss_model_smoothing_cond.gauss_model_smoothing.gauss_smooth
|
|
|
24 #end if
|
|
|
25 --out multigps_out
|
|
|
26 ]]>
|
|
|
27 </command>
|
|
|
28 <configfiles>
|
|
|
29 <configfile name="design_file">
|
|
|
30 <![CDATA[
|
|
1
|
31 #for input in $input_files:
|
|
|
32 #set filename = "{$input['input'].filename}"
|
|
|
33 #set signal_control = $input['signal_control']
|
|
|
34 #set format = $input['input'].ext
|
|
|
35 #set condition_name = $input['condition_name']
|
|
|
36 #set replicate_name = $input['replicate_name']
|
|
|
37 #set line = '\t'.join($filename, $signal_control, $format, $condition_name, $replicate_name])
|
|
|
38 ${line}
|
|
|
39 #end for
|
|
0
|
40 ]]>
|
|
|
41 </configfile>
|
|
|
42 <configfile name="get_seq_dir">
|
|
|
43 <![CDATA[
|
|
|
44 #!/usr/bin/env python
|
|
|
45 import os
|
|
|
46 import tempfile
|
|
|
47
|
|
|
48 # All inputs must have the same dbkey, and MultiGPS requires a directory, not the reference file.
|
|
|
49 if $use_motif_cond.reference_genome_cond.reference_genome_source == "cached":
|
|
|
50 return os.path.split($use_motif_cond.reference_genome_cond.reference_genome.fields.path)[0]
|
|
|
51 else:
|
|
|
52 seq_dir = tempfile.mkdtemp(prefix="tmp-multigps-")
|
|
|
53 # Populate the directory with the history dataset
|
|
|
54 tmp_filename = "%s.fa" % $use_motif_cond.reference_genome_cond.reference_genome.dbkey
|
|
|
55 ln -f -s $use_motif_cond.reference_genome_cond.reference_genome.filename os.path.join(seq_dir, tmp_filename)
|
|
|
56 return seq_dir
|
|
|
57 ]]>
|
|
|
58 </configfile>
|
|
|
59 </configfiles>
|
|
|
60 <inputs>
|
|
|
61 <repeat name="input_files" title="Input Files" min="1">
|
|
|
62 <param name="input" type="data" format="bam,bed,scidx" label="Add file">
|
|
|
63 <validator type="unspecified_build" />
|
|
|
64 </param>
|
|
|
65 <param name="signal_control" type="select" label="Signal or control" display="checkboxes" >
|
|
|
66 <option value="signal" selected="True">Signal</option>
|
|
|
67 <option value="control">Control</option>
|
|
|
68 </param>
|
|
|
69 <param label="Condition name" name="condition_name" type="text" />
|
|
|
70 <param label="Replicate name" name="replicate_name" type="text" />
|
|
|
71
|
|
|
72 <conditional name="use_motif_cond">
|
|
|
73 <param name="use_motif" type="select" label="Perform motif-finding or use a motif-prior?">
|
|
|
74 <option value="yes" selected="True">Yes</option>
|
|
|
75 <option value="no">No</option>
|
|
|
76 </param>
|
|
|
77 <when value="yes">
|
|
|
78 <conditional name="reference_genome_cond">
|
|
|
79 <param name="reference_genome_source" type="select" label="Choose the source for the reference genome">
|
|
|
80 <option value="cached">Locally Cached</option>
|
|
|
81 <option value="history">From History</option>
|
|
|
82 </param>
|
|
|
83 <when value="cached">
|
|
|
84 <param name="reference_genome" type="select" label="Using reference genome">
|
|
|
85 <options from_data_table="fasta_indexes"/>
|
|
|
86 <validator type="no_options" message="A built-in reference genome is not available for the build associated with the selected input file"/>
|
|
|
87 </param>
|
|
|
88 </when>
|
|
|
89 <when value="history">
|
|
|
90 <param name="reference_genome" type="data" format="fasta" label="Using reference genome"/>
|
|
|
91 </when>
|
|
|
92 </conditional>
|
|
|
93 <param name="min_motif_width" type="integer" min="0" value="6" label="Minimum motif width for MEME">
|
|
|
94 <param name="max_motif_width" type="integer" min="0" value="16" label="Maximum motif width for MEME">
|
|
|
95 </when>
|
|
|
96 <when value="no">
|
|
|
97 </conditional>
|
|
|
98
|
|
|
99
|
|
|
100 </repeat>
|
|
|
101
|
|
|
102 <param name="genome_info_file" type="data" format="tabular" label="Genome information dataset" help="What's this? See the help section below."/>
|
|
|
103 <conditional name="gauss_model_smoothing_cond">
|
|
|
104 <param name="gauss_model_smoothing" type="select" label="Turn on Gaussian model smoothing?" help="Smoothing is done with a cubic spline.">
|
|
|
105 <option value="yes" selected="True">Yes</option>
|
|
|
106 <option value="no">No</option>
|
|
|
107 </param>
|
|
|
108 <when value="yes">
|
|
|
109 <param name="gauss_smooth" type="integer" value="3" min="0" label="Smoothing factor" help="Gaussian smoothing standard deviation." />
|
|
|
110 </when>
|
|
|
111 <when value="no" />
|
|
|
112 </conditional>
|
|
|
113 </inputs>
|
|
|
114 <outputs>
|
|
|
115 <data name="output_report" format="txt" label="MultiGPS-Report"/>
|
|
|
116 <data name="output_gff" format="gff" label="MultiGPS-GFF" />
|
|
|
117 <data name="output_gz" format="gz" label="MultiGPS-GZ" />
|
|
|
118 </outputs>
|
|
|
119 <help>
|
|
|
120
|
|
|
121 **What it does**
|
|
|
122
|
|
|
123 MultiGPS is a framework for analyzing collections of multi-condition ChIP-seq datasets and characterizing
|
|
|
124 differential binding events between conditions. MultiGPS encourages consistency in the reported binding
|
|
|
125 event locations across conditions and provides accurate estimation of ChIP enrichment levels at each event.
|
|
|
126 MultiGPS loads all data to memory, so you will need a lot of available memory if you are running analysis
|
|
|
127 over many conditions or large datasets.
|
|
|
128
|
|
|
129 -----
|
|
|
130
|
|
|
131 **Options**
|
|
|
132
|
|
|
133 * **Input Files**
|
|
|
134
|
|
|
135 </help>
|
|
|
136 <citations>
|
|
|
137 <citation type="doi">10.1371/journal.pcbi.1003501</citation>
|
|
|
138 </citations>
|
|
|
139 </tool>
|
