Added support for affy cel files.

workflows/go_workflow.py

@@ -6,6 +6,7 @@ import plots
 import dataset
 #import geneontology
 #import gostat
+import rpy
 from scipy import array,randn
 
 class EinarsWorkflow (Workflow):
@@ -16,6 +17,7 @@ class EinarsWorkflow (Workflow):
 
         load = Stage('load', 'Load Data')
         load.add_function(Function('load', 'Load Microarrays'))
+        load.add_function(CelFileImportFunction())
         load.add_function(TestDataFunction())
         self.add_stage(load)
 
@@ -98,3 +100,40 @@ class TestDataFunction(Function):
         axis_1 = ['cols',[]]
         X = dataset.Dataset(x,[axis_0,axis_1])
         return [X, plots.SinePlot(None)]
+
+
+class CelFileImportFunction(Function):
+    def __init__(self):
+        Function.__init__(self, 'cel_import', 'Import Affy')
+
+    def run(self, data):
+        chooser = gtk.FileChooserDialog(title="Select cel files...", parent=None,
+                                        action=gtk.FILE_CHOOSER_ACTION_OPEN,
+                                        buttons=(gtk.STOCK_CANCEL, gtk.RESPONSE_CANCEL,
+                                                 gtk.STOCK_OPEN, gtk.RESPONSE_OK))
+        chooser.set_select_multiple(True)
+        cel_filter = gtk.FileFilter()
+        cel_filter.set_name("Cel Files (*.cel)")
+        cel_filter.add_pattern("*.[cC][eE][lL]")
+        all_filter = gtk.FileFilter()
+        all_filter.set_name("All Files (*.*)")
+        all_filter.add_pattern("*")
+        chooser.add_filter(cel_filter)
+        chooser.add_filter(all_filter)
+
+        if chooser.run() == gtk.RESPONSE_OK:
+            logger.log('debug', "Selected files: %s" % ", ".join(chooser.get_filenames()))
+            rpy.r.library("affy")
+            # hack: we append ";1" to make sure no r-object is returned to python (faster)
+            rpy.r('At.aBatch <- ReadAffy(filenames=c("%s"));1' % '", "'.join(chooser.get_filenames()))
+            # we destroy it immediately to keep it from being on
+            # screen while we do something with the files
+            chooser.destroy()
+            # also here we append ";1"
+            rpy.r('At.eSet <- expresso(At.aBatch, bg.correct=F, summary.method="liwong", pmcorrect.method="pmonly", normalize.method="qspline");1')
+            m = rpy.r('At.m <- exprs(At.eSet)')
+            rownames = rpy.r('rownames(At.m)') 
+            colnames = rpy.r('colnames(At.m)') 
+            return [dataset.Dataset(m, (('ids', rownames), ('filename', colnames)))] 
+        else:
+            chooser.destroy()