laydi/workflows/affy_workflow.py

import gtk
from system import dataset, logger, plots, workflow
from scipy import randn
import cPickle

class AffyWorkflow (workflow.Workflow):

    name = 'Affy Workflow'
    ident = 'affy'
    description = 'Affymetrics Workflow. Analysis of Affy-data.'
    def __init__(self, app):
        workflow.Workflow.__init__(self, app)

        load = workflow.Stage('load', 'Load Data')
        load.add_function(CelFileImportFunction())
        load.add_function(TestDataFunction())
        load.add_function(DatasetLoadFunction())
        self.add_stage(load)

        explore = workflow.Stage('explore', 'Explorative analysis')
        explore.add_function(PCAFunction(self))
        self.add_stage(explore)

        save = workflow.Stage('save', 'Save Data')
        save.add_function(DatasetSaveFunction())
        self.add_stage(save)
        

class TestDataFunction(workflow.Function):  
    def __init__(self):
        workflow.Function.__init__(self, 'test_data', 'Generate Test Data')

    def run(self, data):
        logger.log('notice', 'Injecting foo test data')
        x = randn(20,30)
        X = dataset.Dataset(x)
        return [X, plots.SinePlot()]


class DatasetLoadFunction(workflow.Function):
    """Loader for previously pickled Datasets."""
    def __init__(self):
        workflow.Function.__init__(self, 'load_data', 'Load Pickled Dataset')

    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))
        pkl_filter = gtk.FileFilter()
        pkl_filter.set_name("Python pickled data files (*.pkl)")
        pkl_filter.add_pattern("*.[pP][kK][lL]")
        all_filter = gtk.FileFilter()
        all_filter.set_name("All Files (*.*)")
        all_filter.add_pattern("*")
        chooser.add_filter(pkl_filter)
        chooser.add_filter(all_filter)

        try:
            if chooser.run() == gtk.RESPONSE_OK:
                return [cPickle.load(open(chooser.get_filename()))]
        finally:
            chooser.destroy()


class DatasetSaveFunction(workflow.Function):
    """QND way to save data to file for later import to this program."""
    def __init__(self):
        workflow.Function.__init__(self, 'save_data', 'Save Pickled Dataset')

    def run(self, data):
        if not data:
            logger.log("notice", "No data to save.")
            return

        chooser = gtk.FileChooserDialog(title="Save pickled data...", parent=None,
                                        action=gtk.FILE_CHOOSER_ACTION_SAVE,
                                        buttons=(gtk.STOCK_CANCEL, gtk.RESPONSE_CANCEL,
                                                 gtk.STOCK_SAVE, gtk.RESPONSE_OK))
        pkl_filter = gtk.FileFilter()
        pkl_filter.set_name("Python pickled data files (*.pkl)")
        pkl_filter.add_pattern("*.[pP][kK][lL]")
        all_filter = gtk.FileFilter()
        all_filter.set_name("All Files (*.*)")
        all_filter.add_pattern("*")
        chooser.add_filter(pkl_filter)
        chooser.add_filter(all_filter)
        chooser.set_current_name(data.get_name() + ".pkl")

        try:
            if chooser.run() == gtk.RESPONSE_OK:
                cPickle.dump(data, open(chooser.get_filename(), "w"), protocol=2)
                logger.log("notice", "Saved data to %r." % chooser.get_filename())
        finally:
            chooser.destroy()
                

class CelFileImportFunction(workflow.Function):
    """Loads Affymetrics .CEL-files into matrix."""
    def __init__(self):
        workflow.Function.__init__(self, 'cel_import', 'Import Affy')

    def run(self, data):
        import rpy
        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)

        try:
            if chooser.run() == gtk.RESPONSE_OK:
                rpy.r.library("affy")
    
                silent_eval = rpy.with_mode(rpy.NO_CONVERSION, rpy.r)
                silent_eval('E <- ReadAffy(filenames=c("%s"))' % '", "'.join(chooser.get_filenames()))
                silent_eval('E <- rma(E)')

                m = rpy.r('m <- E@exprs')
    
                vector_eval = rpy.with_mode(rpy.VECTOR_CONVERSION, rpy.r)
                rownames = vector_eval('rownames(m)') 
                colnames = vector_eval('colnames(m)') 

                # We should be nice and clean up after ourselves
                rpy.r.rm(["E", "m"])
                
                if m:
                    data = dataset.Dataset(m, (('ids', rownames), ('filename', colnames)), name="Affymetrics Data")
                    plot = plots.LinePlot(data, "Gene profiles")
                    return [data, plot]
                else:
                    logger.log("notice", "No data loaded from importer.")
        finally:
            chooser.destroy()


class PCAFunction(workflow.Function):
    """Generic PCA function."""
    def __init__(self, wf):
        workflow.Function.__init__(self, 'pca', 'PCA')
        self._workflow = wf

    def run(self, data):
        import rpy
        
        dim_2, dim_1 = data.get_dim_names()
        
    
        silent_eval = rpy.with_mode(rpy.NO_CONVERSION, rpy.r)
        rpy.with_mode(rpy.NO_CONVERSION, rpy.r.assign)("m", data.get_matrix())
        silent_eval("t = prcomp(t(m))")

        T_ids = map(str, range(1, rpy.r("dim(t$x)")[1]+1))
        T = dataset.Dataset(rpy.r("t$x"), [(dim_1, data.get_identifiers(dim_1)),
                                   ("component", T_ids)], name="T")
        P = dataset.Dataset(rpy.r("t$rotation"), [(dim_2, data.get_identifiers(dim_2)),
                                          ("component", T_ids)], name="P")

        # cleanup
        rpy.r.rm(["t", "m"])

        loading_plot = plots.ScatterPlot(P,'ids','component','1','2', "Loadings")
        score_plot = plots.ScatterPlot(T,'filename','component','1','2', "Scores")
        
        return [T, P, loading_plot, score_plot]