changeset 0:642c0da30ca6 draft

Initial upload.
author stemcellcommons
date Thu, 17 Oct 2013 12:47:49 -0400 (2013-10-17)
parents
children c05f607d116c
files macs2_wrapper.py macs2_wrapper.xml
diffstat 2 files changed, 405 insertions(+), 0 deletions(-) [+]
line wrap: on
line diff
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/macs2_wrapper.py	Thu Oct 17 12:47:49 2013 -0400
@@ -0,0 +1,177 @@
+#purpose: macs2 python wrapper
+#author: Ziru Zhou
+#date: November, 2012
+
+import sys, subprocess, tempfile, shutil, glob, os, os.path, gzip
+from galaxy import eggs
+import pkg_resources
+pkg_resources.require( "simplejson" )
+import simplejson
+
+CHUNK_SIZE = 1024
+
+#==========================================================================================
+#functions
+#==========================================================================================
+def gunzip_cat_glob_path( glob_path, target_filename, delete = False ):
+    out = open( target_filename, 'wb' )
+    for filename in glob.glob( glob_path ):
+        fh = gzip.open( filename, 'rb' )
+        while True:
+            data = fh.read( CHUNK_SIZE )
+            if data:
+                out.write( data )
+            else:
+                break
+        fh.close()
+        if delete:
+            os.unlink( filename )
+    out.close()
+
+def xls_to_interval( xls_file, interval_file, header = None ):
+    out = open( interval_file, 'wb' )
+    if header:
+        out.write( '#%s\n' % header )
+    wrote_header = False
+    #From macs readme: Coordinates in XLS is 1-based which is different with BED format.
+    for line in open( xls_file ):
+        #keep all existing comment lines
+        if line.startswith( '#' ):
+            out.write( line )
+	#added for macs2 since there is an extra newline 
+        elif line.startswith( '\n' ):
+            out.write( line )
+        elif not wrote_header:
+            out.write( '#%s' % line )
+	    print line
+            wrote_header = True
+        else:
+            fields = line.split( '\t' )
+            if len( fields ) > 1:
+                fields[1] = str( int( fields[1] ) - 1 )
+            out.write( '\t'.join( fields ) )
+    out.close()
+
+#==========================================================================================
+#main
+#==========================================================================================
+def main():
+    #take in options file and output file names
+    options = simplejson.load( open( sys.argv[1] ) )
+    outputs = simplejson.load( open( sys.argv[2] ) )
+
+    #=================================================================================
+    #parse options and execute macs2
+    #=================================================================================
+    #default inputs that are in every major command
+    experiment_name = '_'.join( options['experiment_name'].split() ) #save experiment name here, it will be used by macs for some file names
+    cmdline = "macs2 %s -t %s" % ( options['command'], ",".join( options['input_chipseq'] ) )
+    if options['input_control']:
+        cmdline = "%s -c %s" % ( cmdline, ",".join( options['input_control'] ) )
+
+    #=================================================================================
+    if (options['command'] == "callpeak"):
+    	output_bed = outputs['output_bed_file']
+    	output_extra_html = outputs['output_extra_file']
+    	output_extra_path = outputs['output_extra_file_path']
+    	output_peaks =  outputs['output_peaks_file']
+   	output_narrowpeaks = outputs['output_narrowpeaks_file']    
+	output_xls_to_interval_peaks_file = outputs['output_xls_to_interval_peaks_file']
+	output_xls_to_interval_negative_peaks_file = outputs['output_xls_to_interval_negative_peaks_file']
+
+	if 'pvalue' in options:
+    		cmdline = "%s --format='%s' --name='%s' --gsize='%s' --bw='%s' --pvalue='%s' --mfold %s %s %s %s" % ( cmdline, options['format'], experiment_name, options['gsize'], options['bw'], options['pvalue'], options['mfoldlo'], options['mfoldhi'], options['nolambda'], options['bdg'] )
+	elif 'qvalue' in options:
+    		cmdline = "%s --format='%s' --name='%s' --gsize='%s' --bw='%s' --qvalue='%s' --mfold %s %s %s %s" % ( cmdline, options['format'], experiment_name, options['gsize'], options['bw'], options['qvalue'], options['mfoldlo'], options['mfoldhi'], options['nolambda'], options['bdg'] )
+		
+	if 'nomodel' in options:
+        	cmdline = "%s --nomodel --shiftsize='%s'" % ( cmdline, options['nomodel'] )
+    #=================================================================================
+    if (options['command'] == "bdgcmp"):
+	output_bdgcmp = outputs['output_bdgcmp_file']
+
+	cmdline = "%s -m %s -p %s -o bdgcmp_out.bdg" % ( cmdline, options['m'], options['pseudocount'] )
+    #=================================================================================
+
+    tmp_dir = tempfile.mkdtemp() #macs makes very messy output, need to contain it into a temp dir, then provide to user
+    stderr_name = tempfile.NamedTemporaryFile().name # redirect stderr here, macs provides lots of info via stderr, make it into a report
+    proc = subprocess.Popen( args=cmdline, shell=True, cwd=tmp_dir, stderr=open( stderr_name, 'wb' ) )
+    proc.wait()
+    #We don't want to set tool run to error state if only warnings or info, e.g. mfold could be decreased to improve model, but let user view macs log
+    #Do not terminate if error code, allow dataset (e.g. log) creation and cleanup
+    if proc.returncode:
+        stderr_f = open( stderr_name )
+        while True:
+            chunk = stderr_f.read( CHUNK_SIZE )
+            if not chunk:
+                stderr_f.close()
+                break
+            sys.stderr.write( chunk )
+    
+    #=================================================================================
+    #copy files created by macs2 to appripriate directory with the provided names
+    #=================================================================================
+
+    #=================================================================================
+    #move files generated by callpeak command 
+    if (options['command'] == "callpeak"):
+    	#run R to create pdf from model script
+    	if os.path.exists( os.path.join( tmp_dir, "%s_model.r" % experiment_name ) ):
+     	   	cmdline = 'R --vanilla --slave < "%s_model.r" > "%s_model.r.log"' % ( experiment_name, experiment_name )
+     	   	proc = subprocess.Popen( args=cmdline, shell=True, cwd=tmp_dir )
+		proc.wait()
+        
+    	#move bed out to proper output file
+    	created_bed_name =  os.path.join( tmp_dir, "%s_peaks.bed" % experiment_name )
+    	if os.path.exists( created_bed_name ):
+        	shutil.move( created_bed_name, output_bed )
+
+    	#OICR peak_xls file
+    	created_peak_xls_file =  os.path.join( tmp_dir, "%s_peaks.xls" % experiment_name )
+    	if os.path.exists( created_peak_xls_file ):
+        	# shutil.copy( created_peak_xls_file, os.path.join ( "/mnt/galaxyData/tmp/", "%s_peaks.xls" % ( os.path.basename(output_extra_path) )))
+		shutil.copyfile( created_peak_xls_file, output_peaks )   
+
+    	#peaks.encodepeaks (narrowpeaks) file
+    	created_narrowpeak_file = os.path.join (tmp_dir, "%s_peaks.encodePeak" % experiment_name )
+    	if os.path.exists( created_narrowpeak_file ):
+		shutil.move (created_narrowpeak_file, output_narrowpeaks )
+
+ 	#parse xls files to interval files as needed
+   	#if 'xls_to_interval' in options:
+    	if (options['xls_to_interval'] == "True"):
+        	create_peak_xls_file = os.path.join( tmp_dir, '%s_peaks.xls' % experiment_name )
+        	if os.path.exists( create_peak_xls_file ):
+            		xls_to_interval( create_peak_xls_file, output_xls_to_interval_peaks_file, header = 'peaks file' )
+        	create_peak_xls_file = os.path.join( tmp_dir, '%s_negative_peaks.xls' % experiment_name )
+        	if os.path.exists( create_peak_xls_file ):
+			print "negative file exists"
+            		xls_to_interval( create_peak_xls_file, output_xls_to_interval_negative_peaks_file, header = 'negative peaks file' )
+
+    	#move all remaining files to extra files path of html file output to allow user download
+    	out_html = open( output_extra_html, 'wb' )
+    	out_html.write( '<html><head><title>Additional output created by MACS (%s)</title></head><body><h3>Additional Files:</h3><p><ul>\n' % experiment_name )
+    	os.mkdir( output_extra_path )
+    	for filename in sorted( os.listdir( tmp_dir ) ):
+    		shutil.move( os.path.join( tmp_dir, filename ), os.path.join( output_extra_path, filename ) )
+        	out_html.write( '<li><a href="%s">%s</a></li>\n' % ( filename, filename ) )
+		#out_html.write( '<li><a href="%s">%s</a>peakxls %s SomethingDifferent tmp_dir %s path %s exp_name %s</li>\n' % ( created_peak_xls_file, filename, filename, tmp_dir, output_extra_path, experiment_name ) )
+    	out_html.write( '</ul></p>\n' )
+    	out_html.write( '<h3>Messages from MACS:</h3>\n<p><pre>%s</pre></p>\n' % open( stderr_name, 'rb' ).read() )
+    	out_html.write( '</body></html>\n' )
+    	out_html.close()
+
+    #=================================================================================
+    #move files generated by bdgcmp command
+    if (options['command'] == "bdgcmp"):
+    	created_bdgcmp_file = os.path.join (tmp_dir, "bdgcmp_out.bdg" )
+    	if os.path.exists( created_bdgcmp_file ):
+		shutil.move (created_bdgcmp_file, output_bdgcmp )
+     
+    #=================================================================================    
+    #cleanup
+    #=================================================================================    
+    os.unlink( stderr_name )
+    os.rmdir( tmp_dir )
+
+if __name__ == "__main__": main()
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/macs2_wrapper.xml	Thu Oct 17 12:47:49 2013 -0400
@@ -0,0 +1,228 @@
+<tool id="macs2_peakcalling" name="MACS2" version="2.0.10">
+  <description>Model-based Analysis of ChIP-Seq</description>
+  <command interpreter="python">macs2_wrapper.py $options_file $outputs_file</command>
+  <inputs>
+    <!--experiment name and option of selecting paired or single end will always be present-->
+    <param name="experiment_name" type="text" value="MACS2 in Galaxy" size="50" label="Experiment Name"/>
+    
+    <!--select one of the 7 major commands offered by macs2-->
+    <conditional name="major_command">
+      <param name="major_command_selector" type="select" label="Select action to be performed">
+	<option value="callpeak">Peak Calling</option>
+	<!--<option value="filterdup">filterdup</option>
+	<option value="randsample">randsample</option>-->
+	<option value="bdgcmp">Compare .bdg Files</option>
+	<!--<option value="bdgdiff">bdgdiff</option>
+	<option value="bdgpeakcall">bdgpeakcall</option>
+	<option value="bdgbroadcall">bdgbroadcall</option>-->
+      </param>
+      <!--callpeak option of macs2-->
+      <when value="callpeak">
+	<!--may need to add a few more formats at later time-->
+        <param name="input_chipseq_file1" type="data" format="bam" label="ChIP-Seq Tag File" />
+        <param name="input_control_file1" type="data" format="bam" optional="True" label="ChIP-Seq Control File" />
+	<param name="gsize" type="float" label="Effective genome size" value="2.7e+9" help="Human: 3.3e+9, Mouse: 3.0e+9, Fly: 1.9e+8, Worm: 1.3e+8 (--gsize)"/>
+	<param name="bw" type="integer" label="Band width" value="300" help="(--bw)"/>
+	<param name="xls_to_interval" label="Parse xls files into into distinct interval files" type="boolean" truevalue="True" falsevalue="False" checked="False"/>
+	<param name="bdg" label="Save fragment pileup, control lambda, -log10pvalue/qvalue in bedGraph" type="boolean" truevalue="-B" falsevalue="" checked="False" help="files located in html report"/>
+
+	<conditional name="pq_options">
+	  <param name="pq_options_selector" type="select" label="Select p-value or q-value" help="default uses q-value">
+	    <option value="qvalue">q-value</option>
+	    <option value="pvalue">p-value</option>
+	  </param>
+	  <when value="pvalue">
+	    <param name="pvalue" type="float" label="p-value cutoff for peak detection" value="1e-2" help="default: 1e-2 (--pvalue)"/>
+	  </when>
+	  <when value="qvalue">
+	    <param name="qvalue" type="float" label="q-value cutoff for peak detection" value="5e-2" help="default: 5e-2 (--qvalue)"/>
+	  </when> 
+	</conditional>
+	<conditional name="advanced_options">
+	  <param name="advanced_options_selector" type="select" label="Display advanced options">
+	    <option value="off">Hide</option>
+	    <option value="on">Display</option>
+	  </param>
+	  <when value="on">
+            <param name="mfoldlo" type="integer" label="Select the regions with MFOLD high-confidence enrichment ratio against background to build model (lower-limit)" value="10" help="(--mfold)"/>
+	    <param name="mfoldhi" type="integer" label="Select the regions with MFOLD high-confidence enrichment ratio against background to build model (upper-limit)" value="30" help="(--mfold)"/>
+	    <param name="nolambda" label="Use fixed background lambda as local lambda for every peak region" type="boolean" truevalue="--nolambda" falsevalue="" checked="False" help="up to 9X more time consuming (--nolambda)"/>
+	  </when>
+	  <when value="off">
+	    <!--display nothing-->
+	  </when>
+	</conditional>
+    	<conditional name="nomodel_type">
+          <param name="nomodel_type_selector" type="select" label="Build Model">
+	   <option value="nomodel">Do not build the shifting model (--nomodel enabled)</option>
+           <option value="create_model" selected="true">Build the shifting model (--nomodel disabled)</option>
+          </param>
+          <when value="nomodel">
+            <param name="shiftsize" type="integer" label="Arbitrary shift size in bp" value="100" help="(--shiftsize)"/>
+          </when>
+        </conditional>
+      </when>
+
+      <!--callpeak option of macs2-->
+      <when value="bdgcmp">
+        <param name="input_chipseq_file1" type="data" format="bam" label="ChIP-Seq Tag File" />
+        <param name="input_control_file1" type="data" format="bam" optional="True" label="ChIP-Seq Control File" />
+	<param name="pseudocount" type="float" label="Set pseudocount" value="0.00001" help="default: 0.00001 (-p)"/>
+
+        <conditional name="bdgcmp_options">
+          <param name="bdgcmp_options_selector" type="select" label="Select action to be performed">
+	    <option value="ppois">ppois</option>
+	    <option value="qpois">qpois</option>
+	    <option value="subtract">subtract</option>
+	    <option value="logFE">logFE</option>
+	    <option value="FE">FE</option>
+	    <option value="logLR">logLR</option>
+          </param>
+	</conditional>
+      </when>
+    </conditional>
+  </inputs>
+
+  <outputs>
+    <!--callpeaks output-->
+    <data name="output_bed_file" format="bed" label="${tool.name}: callpeak on ${on_string} (peaks: bed)">
+      <filter>major_command['major_command_selector'] == 'callpeak'</filter>
+    </data>
+    <data name="output_extra_files" format="html" label="${tool.name}: callpeak on ${on_string} (html report)">
+      <filter>major_command['major_command_selector'] == 'callpeak'</filter>
+    </data>
+    <data name="output_peaks_file" format="xls" label="${tool.name}: callpeak on ${on_string} (peaks: xls)">
+      <filter>major_command['major_command_selector'] == 'callpeak'</filter>
+    </data>
+    <data name="output_narrowpeaks_file" format="txt" label="${tool.name}: callpeak on ${on_string} (peaks: encodePeak)">
+      <filter>major_command['major_command_selector'] == 'callpeak'</filter>
+    </data>
+    <data name="output_xls_to_interval_peaks_file" format="interval" label="${tool.name}: callpeak on ${on_string} (peaks: interval)">
+      <filter>major_command['xls_to_interval'] is True</filter>
+      <filter>major_command['major_command_selector'] == 'callpeak'</filter>
+    </data>
+    <data name="output_xls_to_interval_negative_peaks_file" format="interval" label="${tool.name}: callpeak on ${on_string} (negative peaks: interval)">
+      <filter>major_command['xls_to_interval'] is True</filter>
+      <filter>major_command['input_control_file1'] is not None</filter>
+      <filter>major_command['major_command_selector'] == 'callpeak'</filter>
+    </data>
+    <!--bdgcmp output-->
+    <data name="output_bdgcmp_file" format="bdg" label="${tool.name}: bdgcmp on ${on_string} (bdg)">
+      <filter>major_command['major_command_selector'] == 'bdgcmp'</filter>
+    </data>
+  </outputs>
+  <configfiles>
+
+    <configfile name="outputs_file">&lt;%
+import simplejson
+%&gt;
+##=======================================================================================
+#set $__outputs = { 'command':str( $major_command.major_command_selector ) }
+#if str( $major_command.major_command_selector ) == 'callpeak':
+	#set $__outputs['output_bed_file'] = str( $output_bed_file )
+	#set $__outputs['output_extra_file'] = str( $output_extra_files )
+	#set $__outputs['output_extra_file_path'] = str( $output_extra_files.files_path )
+	#set $__outputs['output_peaks_file'] = str( $output_peaks_file )
+	#set $__outputs['output_narrowpeaks_file'] = str( $output_narrowpeaks_file )
+	#set $__outputs['output_xls_to_interval_peaks_file'] = str( $output_xls_to_interval_peaks_file )
+	#set $__outputs['output_xls_to_interval_negative_peaks_file'] = str( $output_xls_to_interval_negative_peaks_file )
+#end if
+##=======================================================================================
+#if str( $major_command.major_command_selector ) == 'bdgcmp':
+	#set $__outputs['output_bdgcmp_file'] = str( $output_bdgcmp_file )
+#end if
+
+${ simplejson.dumps( __outputs ) }
+    </configfile>
+    <configfile name="options_file">&lt;%
+import simplejson
+%&gt;
+##=======================================================================================
+#set $__options = { 'experiment_name':str( $experiment_name ) }
+##treatment/tag input files and format
+#set $__options['input_chipseq'] = [ str( $major_command.input_chipseq_file1 ) ]
+#set $__options['format'] = $major_command.input_chipseq_file1.extension.upper()
+
+##control/input files
+#set $__options['input_control'] = []
+#if str( $major_command.input_control_file1 ) != 'None':
+	#set $_hole = __options['input_control'].append( str( $major_command.input_control_file1 ) )
+#end if
+
+#if str( $major_command.major_command_selector ) == 'callpeak':
+	#set $__options['command'] = str( "callpeak" )
+	#set $__options['gsize'] = int( $major_command.gsize )
+	#set $__options['bw'] = str( $major_command.bw )
+	#set $__options['bdg'] = str( $major_command.bdg )
+	#set $__options['xls_to_interval'] = str( $major_command.xls_to_interval )
+
+	##advanced options
+	#if str( $major_command.advanced_options.advanced_options_selector ) == 'on':
+		#set $__options['mfoldlo'] = int( $major_command.advanced_options.mfoldlo )
+		#set $__options['mfoldhi'] = int( $major_command.advanced_options.mfoldhi )
+		#set $__options['nolambda'] = str( $major_command.advanced_options.nolambda )
+	#else:
+		#set $__options['mfoldlo'] = int( "10" )
+		#set $__options['mfoldhi'] = int( "30" )
+		#set $__options['nolambda'] = str( "" )		
+	#end if
+
+	##enable xls file options
+	##if str( $major_command.xls_to_interval ) == 'create':
+		##set $__options['xls_to_interval'] = { 'peaks_file': str( $output_xls_to_interval_peaks_file ), 'negative_peaks_file': str( $output_xls_to_interval_negative_peaks_file ) }
+	##end if
+	
+	##pq value select options
+	#if str( $major_command.pq_options.pq_options_selector ) == 'qvalue':
+		#set $__options['qvalue'] = str( $major_command.pq_options.qvalue )
+	#else:
+		#set $__options['pvalue'] = str( $major_command.pq_options.pvalue )
+	#end if
+	
+	##model options
+	#if str( $major_command.nomodel_type.nomodel_type_selector ) == 'nomodel':
+		#set $__options['nomodel'] = str( $major_command.nomodel_type.shiftsize )
+	#end if
+#end if
+##=======================================================================================
+#if str( $major_command.major_command_selector ) == 'bdgcmp':
+	#set $__options['command'] = str( "bdgcmp" )
+	#set $__options['pseudocount'] = float( str( $major_command.pseudocount ) )
+	#set $__options['m'] = str( $major_command.bdgcmp_options.bdgcmp_options_selector )
+#end if
+##=======================================================================================
+
+${ simplejson.dumps( __options ) }
+    </configfile>
+  </configfiles>
+  <tests>
+	<!--none yet for macs2-->
+  </tests>
+  <help>
+**What it does**
+
+With the improvement of sequencing techniques, chromatin immunoprecipitation followed by high throughput sequencing (ChIP-Seq)
+is getting popular to study genome-wide protein-DNA interactions. To address the lack of powerful ChIP-Seq analysis method, we present a novel algorithm, named Model-based Analysis of ChIP-Seq (MACS), for
+identifying transcript factor binding sites. MACS captures the influence of genome complexity to evaluate the significance of enriched ChIP regions, and MACS improves the spatial resolution of
+binding sites through combining the information of both sequencing tag position and orientation. MACS can be easily used for ChIP-Seq data alone, or with control sample with the increase of specificity.
+
+View the original MACS2 documentation: https://github.com/taoliu/MACS/blob/master/README
+
+------
+
+**Usage**
+
+**Peak Calling**: Main MACS2 Function to Call peaks from alignment results.
+
+**Compare .bdg files**: Deduct noise by comparing two signal tracks in bedGraph.
+
+
+------
+
+**Citation**
+
+For the underlying tool, please cite Zhang Y, Liu T, Meyer CA, Eeckhoute J, Johnson DS, Bernstein BE, Nusbaum C, Myers RM, Brown M, Li W, Liu XS. Model-based analysis of ChIP-Seq (MACS). Genome Biol. 2008;9(9):R137.
+
+Integration of MACS2 with Galaxy performed by Ziru Zhou ( ziruzhou@gmail.com ). Please send your comments/questions to modENCODE DCC at help@modencode.org.
+  </help>
+</tool>