Initial GeneTrack commit. Most parts are in, but dependencies will still be a problem.

tools/visualization/genetrack.py

@@ -0,0 +1,139 @@
+#!/usr/bin/env python
+"""
+Run GeneTrack(atlas) with a faked conf file to generate GeneTrack data files.
+
+usage: %prog
+    -l, --label=N: Data label for fit curve/peak plot
+    -1, --fits=N/N/N/N/N,...: Data files (interval format) for fit curve/peak plot
+    -2, --feats=N:M/N/N/N/N/N,...: Data files (interval format) for features.
+    -d, --data=N: Output path for hdf5 and sqlite databases.
+    -o, --output=N: Output path for export file.
+"""
+from galaxy import eggs
+import pkg_resources
+pkg_resources.require("GeneTrack")
+pkg_resources.require("bx-python")
+
+from atlas import commands
+from bx.cookbook import doc_optparse
+import os
+import commands as oscommands
+import tempfile
+
+SIGMA = 20
+WIDTH = 5 * SIGMA
+EXCLUSION_ZONE = 147
+
+def main(label, fit, feats, data_dir, output):
+    os.mkdir(data_dir)
+    conf = DummyConf(
+        __name__=label,
+        CLOBBER = True,
+        DATA_SIZE = 3*10**6,
+        MINIMUM_PEAK_SIZE = 0.1,
+        LOADER_ENABLED = True,
+        FITTER_ENABLED = True,
+        PREDICTOR_ENABLED = True,
+        EXPORTER_ENABLED = True,
+        LOADER = loader,
+        FITTER = fitter,
+        PREDICTOR = predictor,
+        EXPORTER = exporter,
+        HDF_DATABASE = os.path.join( data_dir, "data.hdf" ),
+        SQL_URI = "sqlite:///%s" % os.path.join( data_dir, "features.sqlite" ),
+        SIGMA = SIGMA,
+        WIDTH = WIDTH,
+        DATA_LABEL = label,
+        FIT_LABEL = "%s-SIGMA-%d" % ( label,SIGMA ),
+        PEAK_LABEL = "PRED-%s-SIGMA-%d" % ( label,SIGMA ),
+        EXCLUSION_ZONE = EXCLUSION_ZONE,
+        LEFT_SHIFT = EXCLUSION_ZONE / 2,
+        RIGHT_SHIFT = EXCLUSION_ZONE / 2,
+        EXPORT_LABELS = [ "PRED-%s-SIGMA-%d" % ( label,SIGMA ) ],
+        EXPORT_DIR = os.path.join( data_dir ),
+        DATA_FILE=fit[1],
+        fit=fit,
+        feats=feats,
+        )
+    commands.execute(conf)
+
+# mod454 seems to be a module without a package.  The necessary funcitons are
+# stubbed out here until I'm sure of their final home. INS
+
+def loader( conf ):
+    from atlas import hdf
+    from mod454.schema import Mod454Schema as Schema
+    last_chrom = table = None
+    db = hdf.hdf_open( conf.HDF_DATABASE, mode='a', title='HDF database')
+    gp = hdf.create_group( db=db, name=conf.DATA_LABEL, desc='data group', clobber=conf.CLOBBER ) 
+    fit_meta = conf.fit[2]
+    # iterate over the file and insert into table
+    for line in open( conf.fit[1], "r" ):
+        if line.startswith("chrom"): continue  #Skip possible header
+        if line.startswith("#"): continue
+        fields = line.rstrip('\r\n').split('\t')
+        chrom = fields[fit_meta.chromCol]
+        if chrom != last_chrom:
+            if table: table.flush()
+            table = hdf.create_table( db=db, name=chrom, where=gp, schema=Schema, clobber=False )
+            last_chrom = chrom
+        try:
+            position = int(fields[fit_meta.positionCol])
+            forward = float(fields[fit_meta.forwardCol])
+            reverse = fit_meta.reverseCol > -1 and float(fields[fit_meta.reverseCol]) or 0.0
+            row = ( position, forward, reverse, forward+reverse, )
+            table.append( [ row ] )
+        except ValueError:
+            # Ignore bad lines
+            pass
+    table.flush()
+    db.close()
+    
+def fitter( conf ):
+    from mod454.fitter import fitter as mod454_fitter 
+    return mod454_fitter( conf )
+
+def predictor( conf ):
+    from mod454.predictor import predictor as mod454_predictor
+    return mod454_predictor( conf )
+
+def exporter( conf ):
+    return commands.bed_exporter(conf)
+
+class Bunch( object ):
+    def __init__(self, **kwargs):
+        for key,value in kwargs.items():
+            setattr( self, key, value ) 
+
+class DummyConf( Bunch ):
+    """
+    Fake conf module for genetrack/atlas.
+    """
+    pass
+
+if __name__ == "__main__":
+    options, args = doc_optparse.parse( __doc__ )
+    try:
+        label = options.label
+        fit_name, fit_meta = options.fits.split(':')[0], [int(x)-1 for x in options.fits.split(':')[1:]]
+        fit_meta = Bunch(chromCol=fit_meta[0], positionCol=fit_meta[1], forwardCol=fit_meta[2], reverseCol=fit_meta[3])
+        fit = ( label, fit_name, fit_meta, )
+        # split apart the string into nested lists, preserves order
+        if options.feats:
+            feats = [ ( 
+                    feat_label,
+                    fname,
+                    Bunch(chromCol=int(chromCol)-1, startCol=int(startCol)-1, endCol=int(endCol)-1, 
+                         strandCol=int(strandCol)-1, nameCol=int(nameCol)-1),
+                    ) 
+                 for feat_label, fname, chromCol, startCol, endCol, strandCol, nameCol
+                 in ( feat.split(':') for feat in options.feats.split(',') )]
+        else:
+            feats = []
+        data_dir = options.data
+        output = options.output
+    except:
+        doc_optparse.exception()
+    
+    main(label, fit, feats, data_dir, output)
+    

tools/visualization/genetrack.xml

@@ -0,0 +1,53 @@
+<tool id="genetrack1" name="GeneTrack">
+  
+    <description>Track creator/viewer</description>
+	
+	<code file="genetrack_code.py">
+      <hook exec_after_process="exec_after_process" />
+	</code>
+  	
+    <command interpreter="python">
+        genetrack.py -l $data_label
+        -1 ${fit_data}:${fit_data.metadata.chromCol}:${fit_data.metadata.positionCol}:${fit_data.metadata.forwardCol}:${fit_data.metadata.reverseCol}
+        #if $feature_data
+          -2
+        #end if 
+        #for $data in $feature_data
+          ${data.name}:${data.input}:${data.input.metadata.chromCol}:${data.input.metadata.startCol}:${data.input.metadata.endCol}:${data.input.metadata.strandCol}:${data.input.metadata.nameCol},
+        #end for
+        -d ${genetrack.files_path}
+        -o ${bed_out}
+    </command>
+    
+    <inputs>
+    	  <param name="data_label" type="text" label="Track Label" size="50">
+	        <validator type="regex" message="Please name the track with only alphanumeric characters.">[a-zA-Z0-9]{0,25}</validator>
+    	  </param>
+		  <param name="fit_data" type="data" format="coverage" label="Coverage Dataset" />
+    	  <repeat name="feature_data" title="Features">
+    	  	<param name="input" type="data" format="interval" label="Dataset" />
+    	  	<param name="name" type="text" label="Feature Type (mRNA, ESTs, ORFs, etc.)" size="25">
+	        	<validator type="regex" message="Please name the feature with only alphanumeric characters.">[a-zA-Z0-9]{0,25}</validator>
+			</param>
+    	  </repeat>
+   </inputs>
+
+   <outputs>  
+       <data format="genetrack" name="genetrack" />
+       <data format="bed" name="bed_out" />
+   </outputs>
+
+ <help>
+This tool takes the input Fit Data and creates a peak and curve plot showing 
+the reads and fitness on each basepair.  Features can be plotted below as tracks.
+
+-----
+
+**Syntax**
+
+- **Track Label** is the name of the generated track.
+- **Fit Data** are the datasets to calculate coverage/reads across basepairs and generate a curve.
+- **Features** are additional datasets (interval format) to be plotted below as tracks.
+
+</help>
+</tool>

tools/visualization/genetrack_code.py

@@ -0,0 +1,13 @@
+import sets, os
+from galaxy import eggs
+from galaxy import jobs
+from galaxy.tools.parameters import DataToolParameter
+
+def exec_after_process(app, inp_data, out_data, param_dict, tool=None, stdout=None, stderr=None):
+    """
+    Copy data_label to genetrack.metadata.label
+    """
+    out_data['genetrack'].metadata.label = param_dict['data_label']
+    out_data['genetrack'].info = "Use the link below to view the custom track."
+    out_data['bed_out'].info = ""
+