Example #1
0
class OWGEODatasets(OWWidget):
    name = "GEO Data Sets"
    description = DESCRIPTION
    icon = "../widgets/icons/GEODataSets.svg"
    priority = PRIORITY

    inputs = []
    outputs = [("Expression Data", Orange.data.Table)]

    settingsList = [
        "outputRows", "mergeSpots", "gdsSelectionStates", "splitterSettings",
        "currentGds", "autoCommit", "datasetNames"
    ]

    outputRows = Setting(True)
    mergeSpots = Setting(True)
    gdsSelectionStates = Setting({})
    currentGds = Setting(None)
    datasetNames = Setting({})
    splitterSettings = Setting((
        b'\x00\x00\x00\xff\x00\x00\x00\x00\x00\x00\x00\x02\x00\x00\x01\xea\x00\x00\x00\xd7\x01\x00\x00\x00\x07\x01\x00\x00\x00\x02',
        b'\x00\x00\x00\xff\x00\x00\x00\x00\x00\x00\x00\x02\x00\x00\x01\xb5\x00\x00\x02\x10\x01\x00\x00\x00\x07\x01\x00\x00\x00\x01'
    ))

    autoCommit = Setting(False)

    def __init__(self, parent=None, signalManager=None, name=" GEO Data Sets"):
        OWWidget.__init__(self, parent, signalManager, name)

        self.selectionChanged = False
        self.filterString = ""
        self.datasetName = ""

        ## GUI
        box = gui.widgetBox(self.controlArea, "Info", addSpace=True)
        self.infoBox = gui.widgetLabel(box, "Initializing\n\n")

        box = gui.widgetBox(self.controlArea, "Output", addSpace=True)
        gui.radioButtonsInBox(box,
                              self,
                              "outputRows",
                              ["Genes in rows", "Samples in rows"],
                              "Rows",
                              callback=self.commitIf)

        gui.checkBox(box,
                     self,
                     "mergeSpots",
                     "Merge spots of same gene",
                     callback=self.commitIf)

        gui.separator(box)
        self.nameEdit = gui.lineEdit(
            box,
            self,
            "datasetName",
            "Data set name",
            tooltip="Override the default output data set name",
            callback=self.onNameEdited)
        self.nameEdit.setPlaceholderText("")

        if sys.version_info < (3, ):
            box = gui.widgetBox(self.controlArea, "Commit", addSpace=True)
            self.commitButton = gui.button(box,
                                           self,
                                           "Commit",
                                           callback=self.commit)
            cb = gui.checkBox(box, self, "autoCommit", "Commit on any change")
            gui.setStopper(self, self.commitButton, cb, "selectionChanged",
                           self.commit)
        else:
            gui.auto_commit(self.controlArea,
                            self,
                            "autoCommit",
                            "Commit",
                            box="Commit")
            self.commitIf = self.commit

        gui.rubber(self.controlArea)

        gui.widgetLabel(self.mainArea, "Filter")
        self.filterLineEdit = QLineEdit(textChanged=self.filter)
        self.completer = TokenListCompleter(self,
                                            caseSensitivity=Qt.CaseInsensitive)
        self.filterLineEdit.setCompleter(self.completer)

        self.mainArea.layout().addWidget(self.filterLineEdit)

        splitter = QSplitter(Qt.Vertical, self.mainArea)
        self.mainArea.layout().addWidget(splitter)
        self.treeWidget = QTreeView(splitter)

        self.treeWidget.setSelectionMode(QTreeView.SingleSelection)
        self.treeWidget.setRootIsDecorated(False)
        self.treeWidget.setSortingEnabled(True)
        self.treeWidget.setAlternatingRowColors(True)
        self.treeWidget.setUniformRowHeights(True)
        self.treeWidget.setEditTriggers(QTreeView.NoEditTriggers)

        linkdelegate = LinkStyledItemDelegate(self.treeWidget)
        self.treeWidget.setItemDelegateForColumn(1, linkdelegate)
        self.treeWidget.setItemDelegateForColumn(8, linkdelegate)
        self.treeWidget.setItemDelegateForColumn(
            0, gui.IndicatorItemDelegate(self.treeWidget, role=Qt.DisplayRole))

        proxyModel = MySortFilterProxyModel(self.treeWidget)
        self.treeWidget.setModel(proxyModel)
        self.treeWidget.selectionModel().selectionChanged.connect(
            self.updateSelection)
        self.treeWidget.viewport().setMouseTracking(True)

        splitterH = QSplitter(Qt.Horizontal, splitter)

        box = gui.widgetBox(splitterH, "Description")
        self.infoGDS = gui.widgetLabel(box, "")
        self.infoGDS.setWordWrap(True)
        gui.rubber(box)

        box = gui.widgetBox(splitterH, "Sample Annotations")
        self.annotationsTree = QTreeWidget(box)
        self.annotationsTree.setHeaderLabels(
            ["Type (Sample annotations)", "Sample count"])
        self.annotationsTree.setRootIsDecorated(True)
        box.layout().addWidget(self.annotationsTree)
        self.annotationsTree.itemChanged.connect(
            self.annotationSelectionChanged)
        self._annotationsUpdating = False
        self.splitters = splitter, splitterH

        for sp, setting in zip(self.splitters, self.splitterSettings):
            sp.splitterMoved.connect(self.splitterMoved)
            sp.restoreState(setting)

        self.searchKeys = [
            "dataset_id", "title", "platform_organism", "description"
        ]

        self.gds = []
        self.gds_info = None

        self.resize(1000, 600)

        self.setBlocking(True)
        self.setEnabled(False)
        self.progressBarInit()

        self._executor = ThreadExecutor()

        func = partial(get_gds_model,
                       methodinvoke(self, "_setProgress", (float, )))
        self._inittask = Task(function=func)
        self._inittask.finished.connect(self._initializemodel)
        self._executor.submit(self._inittask)

        self._datatask = None

    @Slot(float)
    def _setProgress(self, value):
        self.progressBarValue = value

    def _initializemodel(self):
        assert self.thread() is QThread.currentThread()
        model, self.gds_info, self.gds = self._inittask.result()
        model.setParent(self)

        proxy = self.treeWidget.model()
        proxy.setFilterKeyColumn(0)
        proxy.setFilterRole(TextFilterRole)
        proxy.setFilterCaseSensitivity(False)
        proxy.setFilterFixedString(self.filterString)

        proxy.setSourceModel(model)
        proxy.sort(0, Qt.DescendingOrder)

        self.progressBarFinished()
        self.setBlocking(False)
        self.setEnabled(True)

        filter_items = " ".join(gds[key] for gds in self.gds
                                for key in self.searchKeys)
        tr_chars = ",.:;!?(){}[]_-+\\|/%#@$^&*<>~`"
        tr_table = str.maketrans(tr_chars, " " * len(tr_chars))
        filter_items = filter_items.translate(tr_table)

        filter_items = sorted(set(filter_items.split(" ")))
        filter_items = [item for item in filter_items if len(item) > 3]

        self.completer.setTokenList(filter_items)

        if self.currentGds:
            current_id = self.currentGds["dataset_id"]
            gdss = [(i, qunpack(proxy.data(proxy.index(i, 1), Qt.DisplayRole)))
                    for i in range(proxy.rowCount())]
            current = [i for i, data in gdss if data and data == current_id]
            if current:
                current_index = proxy.index(current[0], 0)
                self.treeWidget.selectionModel().select(
                    current_index,
                    QItemSelectionModel.Select | QItemSelectionModel.Rows)
                self.treeWidget.scrollTo(current_index,
                                         QTreeView.PositionAtCenter)

        for i in range(8):
            self.treeWidget.resizeColumnToContents(i)

        self.treeWidget.setColumnWidth(
            1, min(self.treeWidget.columnWidth(1), 300))
        self.treeWidget.setColumnWidth(
            2, min(self.treeWidget.columnWidth(2), 200))

        self.updateInfo()

    def updateInfo(self):
        gds_info = self.gds_info
        text = ("%i datasets\n%i datasets cached\n" %
                (len(gds_info),
                 len(glob.glob(serverfiles.localpath("GEO") + "/GDS*"))))
        filtered = self.treeWidget.model().rowCount()
        if len(self.gds) != filtered:
            text += ("%i after filtering") % filtered
        self.infoBox.setText(text)

    def updateSelection(self, *args):
        current = self.treeWidget.selectedIndexes()
        mapToSource = self.treeWidget.model().mapToSource
        current = [mapToSource(index).row() for index in current]
        if current:
            self.currentGds = self.gds[current[0]]
            self.setAnnotations(self.currentGds)
            self.infoGDS.setText(self.currentGds.get("description", ""))
            self.nameEdit.setPlaceholderText(self.currentGds["title"])
            self.datasetName = \
                self.datasetNames.get(self.currentGds["dataset_id"], "")
        else:
            self.currentGds = None
            self.nameEdit.setPlaceholderText("")
            self.datasetName = ""

        self.commitIf()

    def setAnnotations(self, gds):
        self._annotationsUpdating = True
        self.annotationsTree.clear()

        annotations = defaultdict(set)
        subsetscount = {}
        for desc in gds["subsets"]:
            annotations[desc["type"]].add(desc["description"])
            subsetscount[desc["description"]] = str(len(desc["sample_id"]))

        for type, subsets in annotations.items():
            key = (gds["dataset_id"], type)
            subsetItem = QTreeWidgetItem(self.annotationsTree, [type])
            subsetItem.setFlags(subsetItem.flags() | Qt.ItemIsUserCheckable
                                | Qt.ItemIsTristate)
            subsetItem.setCheckState(
                0, self.gdsSelectionStates.get(key, Qt.Checked))
            subsetItem.key = key
            for subset in subsets:
                key = (gds["dataset_id"], type, subset)
                item = QTreeWidgetItem(
                    subsetItem, [subset, subsetscount.get(subset, "")])
                item.setFlags(item.flags() | Qt.ItemIsUserCheckable)
                item.setCheckState(
                    0, self.gdsSelectionStates.get(key, Qt.Checked))
                item.key = key
        self._annotationsUpdating = False
        self.annotationsTree.expandAll()
        for i in range(self.annotationsTree.columnCount()):
            self.annotationsTree.resizeColumnToContents(i)

    def annotationSelectionChanged(self, item, column):
        if self._annotationsUpdating:
            return
        for i in range(self.annotationsTree.topLevelItemCount()):
            item = self.annotationsTree.topLevelItem(i)
            self.gdsSelectionStates[item.key] = item.checkState(0)
            for j in range(item.childCount()):
                child = item.child(j)
                self.gdsSelectionStates[child.key] = child.checkState(0)

    def filter(self):
        filter_string = unicode(self.filterLineEdit.text())
        proxyModel = self.treeWidget.model()
        if proxyModel:
            strings = filter_string.lower().strip().split()
            proxyModel.setFilterFixedStrings(strings)
            self.updateInfo()

    def selectedSamples(self):
        """
        Return the currently selected sample annotations.

        The return value is a list of selected (sample type, sample value)
        tuples.

        .. note:: if some Sample annotation type has no selected values.
                  this method will return all values for it.

        """
        samples = []
        unused_types = []
        used_types = []
        for stype in childiter(self.annotationsTree.invisibleRootItem()):
            selected_values = []
            all_values = []
            for sval in childiter(stype):
                value = (str(stype.text(0)), str(sval.text(0)))
                if self.gdsSelectionStates.get(sval.key, True):
                    selected_values.append(value)
                all_values.append(value)
            if selected_values:
                samples.extend(selected_values)
                used_types.append(str(stype.text(0)))
            else:
                # If no sample of sample type is selected we don't filter
                # on it.
                samples.extend(all_values)
                unused_types.append(str(stype.text(0)))

        return samples, used_types

    def commitIf(self):
        if self.autoCommit:
            self.commit()
        else:
            self.selectionChanged = True

    @Slot(int, int)
    def progressCompleted(self, value, total):
        if total > 0:
            self.progressBarSet(100. * value / total, processEvents=False)
        else:
            pass
            # TODO: report 'indeterminate progress'

    def commit(self):
        if self.currentGds:
            self.error(0)
            sample_type = None
            self.progressBarInit(processEvents=None)

            _, groups = self.selectedSamples()
            if len(groups) == 1 and self.outputRows:
                sample_type = groups[0]

            self.setEnabled(False)
            self.setBlocking(True)

            progress = methodinvoke(self, "progressCompleted", (int, int))

            def get_data(gds_id, report_genes, transpose, sample_type, title):
                gds_ensure_downloaded(gds_id, progress)
                gds = geo.GDS(gds_id)
                data = gds.getdata(report_genes=report_genes,
                                   transpose=transpose,
                                   sample_type=sample_type)
                data.name = title
                return data

            get_data = partial(get_data,
                               self.currentGds["dataset_id"],
                               report_genes=self.mergeSpots,
                               transpose=self.outputRows,
                               sample_type=sample_type,
                               title=self.datasetName
                               or self.currentGds["title"])
            self._datatask = Task(function=get_data)
            self._datatask.finished.connect(self._on_dataready)
            self._executor.submit(self._datatask)

    def _on_dataready(self):
        self.setEnabled(True)
        self.setBlocking(False)
        self.progressBarFinished(processEvents=False)

        try:
            data = self._datatask.result()
        except urlrequest.URLError as error:
            self.error(0, ("Error while connecting to the NCBI ftp server! "
                           "'%s'" % error))
            sys.excepthook(type(error), error, getattr(error, "__traceback__"))
            return
        finally:
            self._datatask = None

        data_name = data.name
        samples, _ = self.selectedSamples()

        self.warning(0)
        message = None
        if self.outputRows:

            def samplesinst(ex):
                out = []
                for meta in data.domain.metas:
                    out.append((meta.name, ex[meta].value))

                if data.domain.class_var.name != 'class':
                    out.append((data.domain.class_var.name,
                                ex[data.domain.class_var].value))

                return out

            samples = set(samples)
            mask = [samples.issuperset(samplesinst(ex)) for ex in data]
            data = data[numpy.array(mask, dtype=bool)]
            if len(data) == 0:
                message = "No samples with selected sample annotations."
        else:
            samples = set(samples)
            domain = Orange.data.Domain([
                attr for attr in data.domain.attributes
                if samples.issuperset(attr.attributes.items())
            ], data.domain.class_var, data.domain.metas)
            #             domain.addmetas(data.domain.getmetas())

            if len(domain.attributes) == 0:
                message = "No samples with selected sample annotations."
            stypes = set(s[0] for s in samples)
            for attr in domain.attributes:
                attr.attributes = dict(
                    (key, value) for key, value in attr.attributes.items()
                    if key in stypes)
            data = Orange.data.Table(domain, data)

        if message is not None:
            self.warning(0, message)

        data_hints.set_hint(data, "taxid", self.currentGds.get("taxid", ""),
                            10.0)
        data_hints.set_hint(data, "genesinrows", self.outputRows, 10.0)

        data.name = data_name
        self.send("Expression Data", data)

        model = self.treeWidget.model().sourceModel()
        row = self.gds.index(self.currentGds)

        model.setData(model.index(row, 0), " ", Qt.DisplayRole)

        self.updateInfo()
        self.selectionChanged = False

    def splitterMoved(self, *args):
        self.splitterSettings = [
            bytes(sp.saveState()) for sp in self.splitters
        ]

    def send_report(self):
        self.report_items("GEO Dataset",
                          [("ID", self.currentGds['dataset_id']),
                           ("Title", self.currentGds['title']),
                           ("Organism", self.currentGds['sample_organism'])])
        self.report_items("Data",
                          [("Samples", self.currentGds['sample_count']),
                           ("Features", self.currentGds['feature_count']),
                           ("Genes", self.currentGds['gene_count'])])
        self.report_name("Sample annotations")
        subsets = defaultdict(list)
        for subset in self.currentGds['subsets']:
            subsets[subset['type']].append(
                (subset['description'], len(subset['sample_id'])))
        self.report_html += "<ul>"
        for type in subsets:
            self.report_html += "<b>" + type + ":</b></br>"
            for desc, count in subsets[type]:
                self.report_html += 9 * "&nbsp" + "<b>{}:</b> {}</br>".format(
                    desc, count)
        self.report_html += "</ul>"

    def onDeleteWidget(self):
        if self._inittask:
            self._inittask.future().cancel()
            self._inittask.finished.disconnect(self._initializemodel)
        if self._datatask:
            self._datatask.future().cancel()
            self._datatask.finished.disconnect(self._on_dataready)
        self._executor.shutdown(wait=False)

        super(OWGEODatasets, self).onDeleteWidget()

    def onNameEdited(self):
        if self.currentGds:
            gds_id = self.currentGds["dataset_id"]
            self.datasetNames[gds_id] = unicode(self.nameEdit.text())
            self.commitIf()
Example #2
0
class OWGEODatasets(OWWidget):
    settingsList = ["outputRows", "mergeSpots", "gdsSelectionStates",
                    "splitterSettings", "currentGds", "autoCommit",
                    "datasetNames"]

    def __init__(self, parent=None, signalManager=None, name=" GEO Data Sets"):
        OWWidget.__init__(self, parent, signalManager, name)

        self.outputs = [("Expression Data", ExampleTable)]

        ## Settings
        self.selectedAnnotation = 0
        self.includeIf = False
        self.minSamples = 3
        self.autoCommit = False
        self.outputRows = 0
        self.mergeSpots = True
        self.filterString = ""
        self.currentGds = None
        self.selectionChanged = False
        self.autoCommit = False
        self.gdsSelectionStates = {}
        self.splitterSettings = [
            '\x00\x00\x00\xff\x00\x00\x00\x00\x00\x00\x00\x02\x00\x00\x01\xea\x00\x00\x00\xd7\x01\x00\x00\x00\x07\x01\x00\x00\x00\x02',
            '\x00\x00\x00\xff\x00\x00\x00\x00\x00\x00\x00\x02\x00\x00\x01\xb5\x00\x00\x02\x10\x01\x00\x00\x00\x07\x01\x00\x00\x00\x01'
        ]
        self.datasetNames = {}
        self.loadSettings()

        self.datasetName = ""

        ## GUI
        self.infoBox = OWGUI.widgetLabel(
            OWGUI.widgetBox(self.controlArea, "Info", addSpace=True),
            "Initializing\n\n"
        )

        box = OWGUI.widgetBox(self.controlArea, "Output", addSpace=True)
        OWGUI.radioButtonsInBox(box, self, "outputRows",
                                ["Genes or spots", "Samples"], "Rows",
                                callback=self.commitIf)
        OWGUI.checkBox(box, self, "mergeSpots", "Merge spots of same gene",
                       callback=self.commitIf)

        OWGUI.separator(box)
        self.nameEdit = OWGUI.lineEdit(
            box, self, "datasetName", "Data set name",
            tooltip="Override the default output data set name",
            callback=self.onNameEdited
        )
        self.nameEdit.setPlaceholderText("")

        box = OWGUI.widgetBox(self.controlArea, "Commit", addSpace=True)
        self.commitButton = OWGUI.button(box, self, "Commit",
                                         callback=self.commit)
        cb = OWGUI.checkBox(box, self, "autoCommit", "Commit on any change")
        OWGUI.setStopper(self, self.commitButton, cb, "selectionChanged",
                         self.commit)
        OWGUI.rubber(self.controlArea)

        self.filterLineEdit = OWGUIEx.lineEditHint(
            self.mainArea, self, "filterString", "Filter",
            caseSensitive=False, matchAnywhere=True,
            callback=self.filter,  delimiters=" ")

        splitter = QSplitter(Qt.Vertical, self.mainArea)
        self.mainArea.layout().addWidget(splitter)
        self.treeWidget = QTreeView(splitter)

        self.treeWidget.setSelectionMode(QAbstractItemView.SingleSelection)
        self.treeWidget.setRootIsDecorated(False)
        self.treeWidget.setSortingEnabled(True)
        self.treeWidget.setAlternatingRowColors(True)
        self.treeWidget.setUniformRowHeights(True)
        self.treeWidget.setEditTriggers(QTreeView.NoEditTriggers)

        linkdelegate = LinkStyledItemDelegate(self.treeWidget)
        self.treeWidget.setItemDelegateForColumn(1, linkdelegate)
        self.treeWidget.setItemDelegateForColumn(8, linkdelegate)
        self.treeWidget.setItemDelegateForColumn(
            0, OWGUI.IndicatorItemDelegate(self.treeWidget,
                                           role=Qt.DisplayRole))

        proxyModel = MySortFilterProxyModel(self.treeWidget)
        self.treeWidget.setModel(proxyModel)
        self.treeWidget.selectionModel().selectionChanged.connect(
            self.updateSelection
        )
        self.treeWidget.viewport().setMouseTracking(True)

        splitterH = QSplitter(Qt.Horizontal, splitter)

        box = OWGUI.widgetBox(splitterH, "Description")
        self.infoGDS = OWGUI.widgetLabel(box, "")
        self.infoGDS.setWordWrap(True)
        OWGUI.rubber(box)

        box = OWGUI.widgetBox(splitterH, "Sample Annotations")
        self.annotationsTree = QTreeWidget(box)
        self.annotationsTree.setHeaderLabels(
            ["Type (Sample annotations)", "Sample count"]
        )
        self.annotationsTree.setRootIsDecorated(True)
        box.layout().addWidget(self.annotationsTree)
        self.annotationsTree.itemChanged.connect(
            self.annotationSelectionChanged
        )
        self._annotationsUpdating = False
        self.splitters = splitter, splitterH

        for sp, setting in zip(self.splitters, self.splitterSettings):
            sp.splitterMoved.connect(self.splitterMoved)
            sp.restoreState(setting)

        self.searchKeys = ["dataset_id", "title", "platform_organism",
                           "description"]

        self.gds = []
        self.gds_info = None

        self.resize(1000, 600)

        self.setBlocking(True)
        self.setEnabled(False)
        self.progressBarInit()

        self._executor = ThreadExecutor()

        func = partial(get_gds_model,
                       methodinvoke(self, "_setProgress", (float,)))
        self._inittask = Task(function=func)
        self._inittask.finished.connect(self._initializemodel)
        self._executor.submit(self._inittask)

        self._datatask = None

    @pyqtSlot(float)
    def _setProgress(self, value):
        self.progressBarValue = value

    def _initializemodel(self):
        assert self.thread() is QThread.currentThread()
        model, self.gds_info, self.gds = self._inittask.result()
        model.setParent(self)

        proxy = self.treeWidget.model()
        proxy.setFilterKeyColumn(0)
        proxy.setFilterRole(TextFilterRole)
        proxy.setFilterCaseSensitivity(False)
        proxy.setFilterFixedString(self.filterString)

        proxy.setSourceModel(model)
        proxy.sort(0, Qt.DescendingOrder)

        self.progressBarFinished()
        self.setBlocking(False)
        self.setEnabled(True)

        filter_items = " ".join(
            gds[key] for gds in self.gds for key in self.searchKeys
        )
        tr_chars = ",.:;!?(){}[]_-+\\|/%#@$^&*<>~`"
        tr_table = string.maketrans(tr_chars, " " * len(tr_chars))
        filter_items = filter_items.translate(tr_table)

        filter_items = sorted(set(filter_items.split(" ")))
        filter_items = [item for item in filter_items if len(item) > 3]
        self.filterLineEdit.setItems(filter_items)

        if self.currentGds:
            gdss = [(i, proxy.data(proxy.index(i, 1), Qt.DisplayRole))
                    for i in range(proxy.rowCount())]
            current = [i for i, variant in gdss
                       if variant.isValid() and
                       str(variant.toString()) == self.currentGds["dataset_id"]]
            if current:
                current_index = proxy.index(current[0], 0)
                self.treeWidget.selectionModel().select(
                    current_index,
                    QItemSelectionModel.Select |
                    QItemSelectionModel.Rows
                )
                self.treeWidget.scrollTo(current_index)

        for i in range(8):
            self.treeWidget.resizeColumnToContents(i)

        self.treeWidget.setColumnWidth(
            1, min(self.treeWidget.columnWidth(1), 300))
        self.treeWidget.setColumnWidth(
            2, min(self.treeWidget.columnWidth(2), 200))

        self.updateInfo()

    def updateInfo(self):
        gds_info = self.gds_info
        text = ("%i datasets\n%i datasets cached\n" %
                (len(gds_info),
                 len(glob.glob(serverfiles.localpath("GEO") + "/GDS*"))))
        filtered = self.treeWidget.model().rowCount()
        if len(self.gds) != filtered:
            text += ("%i after filtering") % filtered
        self.infoBox.setText(text)

    def updateSelection(self, *args):
        current = self.treeWidget.selectedIndexes()
        mapToSource = self.treeWidget.model().mapToSource
        current = [mapToSource(index).row() for index in current]
        if current:
            self.currentGds = self.gds[current[0]]
            self.setAnnotations(self.currentGds)
            self.infoGDS.setText(self.currentGds.get("description", ""))
            self.nameEdit.setPlaceholderText(self.currentGds["title"])
            self.datasetName = \
                self.datasetNames.get(self.currentGds["dataset_id"], "")
        else:
            self.currentGds = None
            self.nameEdit.setPlaceholderText("")
            self.datasetName = ""

        self.commitIf()

    def setAnnotations(self, gds):
        self._annotationsUpdating = True
        self.annotationsTree.clear()

        annotations = defaultdict(set)
        subsetscount = {}
        for desc in gds["subsets"]:
            annotations[desc["type"]].add(desc["description"])
            subsetscount[desc["description"]] = str(len(desc["sample_id"]))

        for type, subsets in annotations.items():
            key = (gds["dataset_id"], type)
            subsetItem = QTreeWidgetItem(self.annotationsTree, [type])
            subsetItem.setFlags(subsetItem.flags() | Qt.ItemIsUserCheckable |
                                Qt.ItemIsTristate)
            subsetItem.setCheckState(
                0, self.gdsSelectionStates.get(key, Qt.Checked)
            )
            subsetItem.key = key
            for subset in subsets:
                key = (gds["dataset_id"], type, subset)
                item = QTreeWidgetItem(
                    subsetItem, [subset, subsetscount.get(subset, "")]
                )
                item.setFlags(item.flags() | Qt.ItemIsUserCheckable)
                item.setCheckState(
                    0, self.gdsSelectionStates.get(key, Qt.Checked)
                )
                item.key = key
        self._annotationsUpdating = False
        self.annotationsTree.expandAll()
        for i in range(self.annotationsTree.columnCount()):
            self.annotationsTree.resizeColumnToContents(i)

    def annotationSelectionChanged(self, item, column):
        if self._annotationsUpdating:
            return
        for i in range(self.annotationsTree.topLevelItemCount()):
            item = self.annotationsTree.topLevelItem(i)
            self.gdsSelectionStates[item.key] = item.checkState(0)
            for j in range(item.childCount()):
                child = item.child(j)
                self.gdsSelectionStates[child.key] = child.checkState(0)

    def filter(self):
        filter_string = unicode(self.filterLineEdit.text(), errors="ignore")
        proxyModel = self.treeWidget.model()
        if proxyModel:
            strings = filter_string.lower().strip().split()
            proxyModel.setFilterFixedStrings(strings)
            self.updateInfo()

    def selectedSamples(self):
        """
        Return the currently selected sample annotations.

        The return value is a list of selected (sample type, sample value)
        tuples.

        .. note:: if some Sample annotation type has no selected values.
                  this method will return all values for it.

        """
        samples = []
        unused_types = []
        used_types = []
        for stype in childiter(self.annotationsTree.invisibleRootItem()):
            selected_values = []
            all_values = []
            for sval in childiter(stype):
                value = (str(stype.text(0)), str(sval.text(0)))
                if self.gdsSelectionStates.get(sval.key, True):
                    selected_values.append(value)
                all_values.append(value)
            if selected_values:
                samples.extend(selected_values)
                used_types.append(str(stype.text(0)))
            else:
                # If no sample of sample type is selected we don't filter
                # on it.
                samples.extend(all_values)
                unused_types.append(str(stype.text(0)))

        return samples, used_types

    def commitIf(self):
        if self.autoCommit:
            self.commit()
        else:
            self.selectionChanged = True

    def commit(self):
        if self.currentGds:
            self.error(0)
            sample_type = None
            self.progressBarInit()
            self.progressBarSet(10)

            _, groups = self.selectedSamples()
            if len(groups) == 1 and self.outputRows:
                sample_type = groups[0]

            self.setEnabled(False)
            self.setBlocking(True)

            def get_data(gds_id, report_genes, transpose, sample_type, title):
                gds = geo.GDS(gds_id)
                data = gds.getdata(
                    report_genes=report_genes, transpose=transpose,
                    sample_type=sample_type
                )
                data.name = title
                return data

            get_data = partial(
                get_data, self.currentGds["dataset_id"],
                report_genes=self.mergeSpots,
                transpose=self.outputRows,
                sample_type=sample_type,
                title=self.datasetName or self.currentGds["title"]
            )
            self._datatask = Task(function=get_data)
            self._datatask.finished.connect(self._on_dataready)
            self._executor.submit(self._datatask)

    def _on_dataready(self):
        self.setEnabled(True)
        self.setBlocking(False)

        self.progressBarSet(50)

        try:
            data = self._datatask.result()
        except urllib2.URLError as error:
            self.error(0, "Error while connecting to the NCBI ftp server! %r" %
                       error)
            self._datatask = None
            self.progressBarFinished()
            return

        self._datatask = None

        data_name = data.name
        samples, _ = self.selectedSamples()

        self.warning(0)
        message = None
        if self.outputRows:
            def samplesinst(ex):
                out = []
                for i, a in data.domain.get_metas().items():
                    out.append((a.name, ex[i].value))
                if data.domain.class_var.name != 'class':
                    out.append((data.domain.class_var.name, ex[-1].value))
                return out
            samples = set(samples)

            select = [1 if samples.issuperset(samplesinst(ex)) else 0
                      for ex in data]
            data = data.select(select)
            if len(data) == 0:
                message = "No samples with selected sample annotations."
        else:
            samples = set(samples)
            domain = orange.Domain(
                [attr for attr in data.domain.attributes
                 if samples.issuperset(attr.attributes.items())],
                data.domain.classVar
            )
            domain.addmetas(data.domain.getmetas())
            if len(domain.attributes) == 0:
                message = "No samples with selected sample annotations."
            stypes = set(s[0] for s in samples)
            for attr in domain.attributes:
                attr.attributes = dict(
                    (key, value) for key, value in attr.attributes.items()
                    if key in stypes
                )
            data = orange.ExampleTable(domain, data)

        if message is not None:
            self.warning(0, message)

        data_hints.set_hint(data, "taxid", self.currentGds.get("taxid", ""),
                            10.0)
        data_hints.set_hint(data, "genesinrows", self.outputRows, 10.0)

        self.progressBarFinished()
        data.name = data_name
        self.send("Expression Data", data)

        model = self.treeWidget.model().sourceModel()
        row = self.gds.index(self.currentGds)

        model.setData(model.index(row, 0),  QVariant(" "), Qt.DisplayRole)

        self.updateInfo()
        self.selectionChanged = False

    def splitterMoved(self, *args):
        self.splitterSettings = [str(sp.saveState()) for sp in self.splitters]

    def onDeleteWidget(self):
        if self._inittask:
            self._inittask.future().cancel()
            self._inittask.finished.disconnect(self._initializemodel)
        if self._datatask:
            self._datatask.future().cancel()
            self._datatask.finished.disconnect(self._on_dataready)
        self._executor.shutdown(wait=False)

        super(OWGEODatasets, self).onDeleteWidget()

    def onNameEdited(self):
        if self.currentGds:
            gds_id = self.currentGds["dataset_id"]
            self.datasetNames[gds_id] = unicode(self.nameEdit.text())
            self.commitIf()
class OWDatabasesUpdate(OWWidget):

    name = "Databases Update"
    description = "Update local systems biology databases."
    icon = "../widgets/icons/Databases.svg"
    priority = 10

    inputs = []
    outputs = []

    want_main_area = False

    def __init__(self, parent=None, signalManager=None,
                 name="Databases update"):
        OWWidget.__init__(self, parent, signalManager, name,
                          wantMainArea=False)

        self.searchString = ""

        fbox = gui.widgetBox(self.controlArea, "Filter")
        self.completer = TokenListCompleter(
            self, caseSensitivity=Qt.CaseInsensitive)
        self.lineEditFilter = QLineEdit(textChanged=self.SearchUpdate)
        self.lineEditFilter.setCompleter(self.completer)

        fbox.layout().addWidget(self.lineEditFilter)

        box = gui.widgetBox(self.controlArea, "Files")

        self.filesView = QTreeWidget(self)
        self.filesView.setHeaderLabels(
            ["", "Data Source", "Update", "Last Updated", "Size"])

        self.filesView.setRootIsDecorated(False)
        self.filesView.setUniformRowHeights(True)
        self.filesView.setSelectionMode(QAbstractItemView.NoSelection)
        self.filesView.setSortingEnabled(True)
        self.filesView.sortItems(1, Qt.AscendingOrder)
        self.filesView.setItemDelegateForColumn(
            0, UpdateOptionsItemDelegate(self.filesView))

        self.filesView.model().layoutChanged.connect(self.SearchUpdate)

        box.layout().addWidget(self.filesView)

        box = gui.widgetBox(self.controlArea, orientation="horizontal")
        self.updateButton = gui.button(
            box, self, "Update all",
            callback=self.UpdateAll,
            tooltip="Update all updatable files",
         )

        self.downloadButton = gui.button(
            box, self, "Download all",
            callback=self.DownloadFiltered,
            tooltip="Download all filtered files shown"
        )

        self.cancelButton = gui.button(
            box, self, "Cancel", callback=self.Cancel,
            tooltip="Cancel scheduled downloads/updates."
        )

        self.retryButton = gui.button(
            box, self, "Reconnect", callback=self.RetrieveFilesList
        )
        self.retryButton.hide()

        gui.rubber(box)
        self.warning(0)

        box = gui.widgetBox(self.controlArea, orientation="horizontal")
        gui.rubber(box)

        self.infoLabel = QLabel()
        self.infoLabel.setAlignment(Qt.AlignCenter)

        self.controlArea.layout().addWidget(self.infoLabel)
        self.infoLabel.setSizePolicy(QSizePolicy.Expanding, QSizePolicy.Fixed)

        self.updateItems = []

        self.resize(800, 600)

        self.progress = ProgressState(self, maximum=3)
        self.progress.valueChanged.connect(self._updateProgress)
        self.progress.rangeChanged.connect(self._updateProgress)
        self.executor = ThreadExecutor(
            threadPool=QThreadPool(maxThreadCount=2)
        )

        task = Task(self, function=self.RetrieveFilesList)
        task.exceptionReady.connect(self.HandleError)
        task.start()

        self._tasks = []
        self._haveProgress = False

    def RetrieveFilesList(self):
        self.retryButton.hide()
        self.warning(0)
        self.progress.setRange(0, 3)

        task = Task(function=partial(retrieveFilesList, methodinvoke(self.progress, "advance")))

        task.resultReady.connect(self.SetFilesList)
        task.exceptionReady.connect(self.HandleError)

        self.executor.submit(task)

        self.setEnabled(False)

    def SetFilesList(self, serverInfo):
        """
        Set the files to show.
        """
        self.setEnabled(True)

        localInfo = serverfiles.allinfo()
        all_tags = set()

        self.filesView.clear()
        self.updateItems = []

        for item in join_info_dict(localInfo, serverInfo):
            tree_item = UpdateTreeWidgetItem(item)
            options_widget = UpdateOptionsWidget(item.state)
            options_widget.item = item

            options_widget.installClicked.connect(
                partial(self.SubmitDownloadTask, item.domain, item.filename)
            )
            options_widget.removeClicked.connect(
                partial(self.SubmitRemoveTask, item.domain, item.filename)
            )

            self.updateItems.append((item, tree_item, options_widget))
            all_tags.update(item.tags)

        self.filesView.addTopLevelItems(
            [tree_item for _, tree_item, _ in self.updateItems]
        )

        for item, tree_item, options_widget in self.updateItems:
            self.filesView.setItemWidget(tree_item, 0, options_widget)

            # Add an update button if the file is updateable
            if item.state == OUTDATED:
                button = QToolButton(
                    None, text="Update",
                    maximumWidth=120,
                    minimumHeight=20,
                    maximumHeight=20
                )

                if sys.platform == "darwin":
                    button.setAttribute(Qt.WA_MacSmallSize)

                button.clicked.connect(
                    partial(self.SubmitDownloadTask, item.domain,
                            item.filename)
                )

                self.filesView.setItemWidget(tree_item, 2, button)

        self.progress.advance()

        self.filesView.setColumnWidth(0, self.filesView.sizeHintForColumn(0))

        for column in range(1, 4):
            contents_hint = self.filesView.sizeHintForColumn(column)
            header_hint = self.filesView.header().sectionSizeHint(column)
            width = max(min(contents_hint, 400), header_hint)
            self.filesView.setColumnWidth(column, width)

        hints = [hint for hint in sorted(all_tags) if not hint.startswith("#")]
        self.completer.setTokenList(hints)
        self.SearchUpdate()
        self.UpdateInfoLabel()
        self.toggleButtons()
        self.cancelButton.setEnabled(False)

        self.progress.setRange(0, 0)

    def buttonCheck(self, selected_items, state, button):
        for item in selected_items:
            if item.state != state:
                button.setEnabled(False)
            else:
                button.setEnabled(True)
                break

    def toggleButtons(self):
        selected_items = [item for item, tree_item, _ in self.updateItems if not tree_item.isHidden()]
        self.buttonCheck(selected_items, OUTDATED, self.updateButton)
        self.buttonCheck(selected_items, AVAILABLE, self.downloadButton)

    def HandleError(self, exception):
        if isinstance(exception, ConnectionError):
            self.warning(0,
                       "Could not connect to server! Check your connection "
                       "and try to reconnect.")
            self.SetFilesList({})
            self.retryButton.show()
        else:
            sys.excepthook(type(exception), exception, None)
            self.progress.setRange(0, 0)
            self.setEnabled(True)

    def UpdateInfoLabel(self):
        local = [item for item, tree_item, _ in self.updateItems
                 if item.state != AVAILABLE and not tree_item.isHidden()]
        size = sum(float(item.size) for item in local)

        onServer = [item for item, tree_item, _ in self.updateItems if not tree_item.isHidden()]
        sizeOnServer = sum(float(item.size) for item in onServer)

        text = ("%i items, %s (on server: %i items, %s)" %
                (len(local),
                 sizeof_fmt(size),
                 len(onServer),
                 sizeof_fmt(sizeOnServer)))

        self.infoLabel.setText(text)

    def UpdateAll(self):
        self.warning(0)
        for item, tree_item, _ in self.updateItems:
            if item.state == OUTDATED and not tree_item.isHidden():
                self.SubmitDownloadTask(item.domain, item.filename)

    def DownloadFiltered(self):
        # TODO: submit items in the order shown.
        for item, tree_item, _ in self.updateItems:
            if not tree_item.isHidden() and item.state in \
                    [AVAILABLE, OUTDATED]:
                self.SubmitDownloadTask(item.domain, item.filename)

    def SearchUpdate(self, searchString=None):
        strings = str(self.lineEditFilter.text()).split()
        for item, tree_item, _ in self.updateItems:
            hide = not all(UpdateItem_match(item, string)
                           for string in strings)
            tree_item.setHidden(hide)
        self.UpdateInfoLabel()
        self.toggleButtons()

    def SubmitDownloadTask(self, domain, filename):
        """
        Submit the (domain, filename) to be downloaded/updated.
        """
        self.cancelButton.setEnabled(True)

        index = self.updateItemIndex(domain, filename)
        _, tree_item, opt_widget = self.updateItems[index]

        sf = LocalFiles(serverfiles.PATH, serverfiles.ServerFiles())
        task = DownloadTask(domain, filename, sf)

        self.progress.adjustRange(0, 100)

        pb = ItemProgressBar(self.filesView)
        pb.setRange(0, 100)
        pb.setTextVisible(False)

        task.advanced.connect(pb.advance)
        task.advanced.connect(self.progress.advance)
        task.finished.connect(pb.hide)
        task.finished.connect(self.onDownloadFinished, Qt.QueuedConnection)
        task.exception.connect(self.onDownloadError, Qt.QueuedConnection)

        self.filesView.setItemWidget(tree_item, 2, pb)

        # Clear the text so it does not show behind the progress bar.
        tree_item.setData(2, Qt.DisplayRole, "")
        pb.show()

        # Disable the options widget
        opt_widget.setEnabled(False)
        self._tasks.append(task)

        self.executor.submit(task)

    def EndDownloadTask(self, task):
        future = task.future()
        index = self.updateItemIndex(task.domain, task.filename)
        item, tree_item, opt_widget = self.updateItems[index]

        self.filesView.removeItemWidget(tree_item, 2)
        opt_widget.setEnabled(True)

        if future.cancelled():
            # Restore the previous state
            tree_item.setUpdateItem(item)
            opt_widget.setState(item.state)

        elif future.exception():
            tree_item.setUpdateItem(item)
            opt_widget.setState(item.state)

            # Show the exception string in the size column.
            self.warning(0, "Error while downloading. Check your connection "
                            "and retry.")

            # recreate button for download
            button = QToolButton(
                None, text="Retry",
                maximumWidth=120,
                minimumHeight=20,
                maximumHeight=20
            )

            if sys.platform == "darwin":
                button.setAttribute(Qt.WA_MacSmallSize)

            button.clicked.connect(
                partial(self.SubmitDownloadTask, item.domain,
                        item.filename)
            )

            self.filesView.setItemWidget(tree_item, 2, button)

        else:
            # get the new updated info dict and replace the the old item
            self.warning(0)
            info = serverfiles.info(item.domain, item.filename)
            new_item = update_item_from_info(item.domain, item.filename,
                                             info, info)

            self.updateItems[index] = (new_item, tree_item, opt_widget)

            tree_item.setUpdateItem(new_item)
            opt_widget.setState(new_item.state)

            self.UpdateInfoLabel()

    def SubmitRemoveTask(self, domain, filename):
        serverfiles.LOCALFILES.remove(domain, filename)
        index = self.updateItemIndex(domain, filename)
        item, tree_item, opt_widget = self.updateItems[index]

        if item.info_server:
            new_item = item._replace(state=AVAILABLE, local=None,
                                      info_local=None)
        else:
            new_item = item._replace(local=None, info_local=None)
            # Disable the options widget. No more actions can be performed
            # for the item.
            opt_widget.setEnabled(False)

        tree_item.setUpdateItem(new_item)
        opt_widget.setState(new_item.state)
        self.updateItems[index] = (new_item, tree_item, opt_widget)

        self.UpdateInfoLabel()

    def Cancel(self):
        """
        Cancel all pending update/download tasks (that have not yet started).
        """
        for task in self._tasks:
            task.future().cancel()

    def onDeleteWidget(self):
        self.Cancel()
        self.executor.shutdown(wait=False)
        OWWidget.onDeleteWidget(self)

    def onDownloadFinished(self):
        # on download completed/canceled/error
        assert QThread.currentThread() is self.thread()
        for task in list(self._tasks):
            future = task.future()
            if future.done():
                self.EndDownloadTask(task)
                self._tasks.remove(task)

        if not self._tasks:
            # Clear/reset the overall progress
            self.progress.setRange(0, 0)
            self.cancelButton.setEnabled(False)

    def onDownloadError(self, exc_info):
        sys.excepthook(*exc_info)
        self.warning(0, "Error while downloading. Check your connection and "
                        "retry.")

    def updateItemIndex(self, domain, filename):
        for i, (item, _, _) in enumerate(self.updateItems):
            if item.domain == domain and item.filename == filename:
                return i
        raise ValueError("%r, %r not in update list" % (domain, filename))

    def _updateProgress(self, *args):
        rmin, rmax = self.progress.range()
        if rmin != rmax:
            if not self._haveProgress:
                self._haveProgress = True
                self.progressBarInit()

            self.progressBarSet(self.progress.ratioCompleted() * 100,
                                processEvents=None)
        if rmin == rmax:
            self._haveProgress = False
            self.progressBarFinished()
Example #4
0
class OWGEODatasets(OWWidget):
    name = "GEO Data Sets"
    description = DESCRIPTION
    icon = "../widgets/icons/GEODataSets.svg"
    priority = PRIORITY

    inputs = []
    outputs = [("Expression Data", Orange.data.Table)]

    settingsList = ["outputRows", "mergeSpots", "gdsSelectionStates",
                    "splitterSettings", "currentGds", "autoCommit",
                    "datasetNames"]

    outputRows = Setting(True)
    mergeSpots = Setting(True)
    gdsSelectionStates = Setting({})
    currentGds = Setting(None)
    datasetNames = Setting({})
    splitterSettings = Setting(
        (b'\x00\x00\x00\xff\x00\x00\x00\x00\x00\x00\x00\x02\x00\x00\x01\xea\x00\x00\x00\xd7\x01\x00\x00\x00\x07\x01\x00\x00\x00\x02',
         b'\x00\x00\x00\xff\x00\x00\x00\x00\x00\x00\x00\x02\x00\x00\x01\xb5\x00\x00\x02\x10\x01\x00\x00\x00\x07\x01\x00\x00\x00\x01')
    )

    autoCommit = Setting(False)

    def __init__(self, parent=None, signalManager=None, name=" GEO Data Sets"):
        OWWidget.__init__(self, parent, signalManager, name)

        self.selectionChanged = False
        self.filterString = ""
        self.datasetName = ""

        ## GUI
        box = gui.widgetBox(self.controlArea, "Info", addSpace=True)
        self.infoBox = gui.widgetLabel(box, "Initializing\n\n")

        box = gui.widgetBox(self.controlArea, "Output", addSpace=True)
        gui.radioButtonsInBox(box, self, "outputRows",
                              ["Genes in rows", "Samples in rows"], "Rows",
                              callback=self.commitIf)

        gui.checkBox(box, self, "mergeSpots", "Merge spots of same gene",
                     callback=self.commitIf)

        gui.separator(box)
        self.nameEdit = gui.lineEdit(
            box, self, "datasetName", "Data set name",
            tooltip="Override the default output data set name",
            callback=self.onNameEdited
        )
        self.nameEdit.setPlaceholderText("")

        if sys.version_info < (3, ):
            box = gui.widgetBox(self.controlArea, "Commit", addSpace=True)
            self.commitButton = gui.button(
                box, self, "Commit", callback=self.commit)
            cb = gui.checkBox(box, self, "autoCommit", "Commit on any change")
            gui.setStopper(self, self.commitButton, cb, "selectionChanged",
                           self.commit)
        else:
            gui.auto_commit(self.controlArea, self, "autoCommit", "Commit",
                            box="Commit")
            self.commitIf = self.commit

        gui.rubber(self.controlArea)

        gui.widgetLabel(self.mainArea, "Filter")
        self.filterLineEdit = QLineEdit(
            textChanged=self.filter
        )
        self.completer = TokenListCompleter(
            self, caseSensitivity=Qt.CaseInsensitive
        )
        self.filterLineEdit.setCompleter(self.completer)

        self.mainArea.layout().addWidget(self.filterLineEdit)

        splitter = QSplitter(Qt.Vertical, self.mainArea)
        self.mainArea.layout().addWidget(splitter)
        self.treeWidget = QTreeView(splitter)

        self.treeWidget.setSelectionMode(QTreeView.SingleSelection)
        self.treeWidget.setRootIsDecorated(False)
        self.treeWidget.setSortingEnabled(True)
        self.treeWidget.setAlternatingRowColors(True)
        self.treeWidget.setUniformRowHeights(True)
        self.treeWidget.setEditTriggers(QTreeView.NoEditTriggers)

        linkdelegate = LinkStyledItemDelegate(self.treeWidget)
        self.treeWidget.setItemDelegateForColumn(1, linkdelegate)
        self.treeWidget.setItemDelegateForColumn(8, linkdelegate)
        self.treeWidget.setItemDelegateForColumn(
            0, gui.IndicatorItemDelegate(self.treeWidget,
                                         role=Qt.DisplayRole))

        proxyModel = MySortFilterProxyModel(self.treeWidget)
        self.treeWidget.setModel(proxyModel)
        self.treeWidget.selectionModel().selectionChanged.connect(
            self.updateSelection
        )
        self.treeWidget.viewport().setMouseTracking(True)

        splitterH = QSplitter(Qt.Horizontal, splitter)

        box = gui.widgetBox(splitterH, "Description")
        self.infoGDS = gui.widgetLabel(box, "")
        self.infoGDS.setWordWrap(True)
        gui.rubber(box)

        box = gui.widgetBox(splitterH, "Sample Annotations")
        self.annotationsTree = QTreeWidget(box)
        self.annotationsTree.setHeaderLabels(
            ["Type (Sample annotations)", "Sample count"]
        )
        self.annotationsTree.setRootIsDecorated(True)
        box.layout().addWidget(self.annotationsTree)
        self.annotationsTree.itemChanged.connect(
            self.annotationSelectionChanged
        )
        self._annotationsUpdating = False
        self.splitters = splitter, splitterH

        for sp, setting in zip(self.splitters, self.splitterSettings):
            sp.splitterMoved.connect(self.splitterMoved)
            sp.restoreState(setting)

        self.searchKeys = ["dataset_id", "title", "platform_organism",
                           "description"]

        self.gds = []
        self.gds_info = None

        self.resize(1000, 600)

        self.setBlocking(True)
        self.setEnabled(False)
        self.progressBarInit()

        self._executor = ThreadExecutor()

        func = partial(get_gds_model,
                       methodinvoke(self, "_setProgress", (float,)))
        self._inittask = Task(function=func)
        self._inittask.finished.connect(self._initializemodel)
        self._executor.submit(self._inittask)

        self._datatask = None

    @Slot(float)
    def _setProgress(self, value):
        self.progressBarValue = value

    def _initializemodel(self):
        assert self.thread() is QThread.currentThread()
        model, self.gds_info, self.gds = self._inittask.result()
        model.setParent(self)

        proxy = self.treeWidget.model()
        proxy.setFilterKeyColumn(0)
        proxy.setFilterRole(TextFilterRole)
        proxy.setFilterCaseSensitivity(False)
        proxy.setFilterFixedString(self.filterString)

        proxy.setSourceModel(model)
        proxy.sort(0, Qt.DescendingOrder)

        self.progressBarFinished()
        self.setBlocking(False)
        self.setEnabled(True)

        filter_items = " ".join(
            gds[key] for gds in self.gds for key in self.searchKeys
        )
        tr_chars = ",.:;!?(){}[]_-+\\|/%#@$^&*<>~`"
        tr_table = str.maketrans(tr_chars, " " * len(tr_chars))
        filter_items = filter_items.translate(tr_table)

        filter_items = sorted(set(filter_items.split(" ")))
        filter_items = [item for item in filter_items if len(item) > 3]

        self.completer.setTokenList(filter_items)

        if self.currentGds:
            current_id = self.currentGds["dataset_id"]
            gdss = [(i, qunpack(proxy.data(proxy.index(i, 1), Qt.DisplayRole)))
                    for i in range(proxy.rowCount())]
            current = [i for i, data in gdss if data and data == current_id]
            if current:
                current_index = proxy.index(current[0], 0)
                self.treeWidget.selectionModel().select(
                    current_index,
                    QItemSelectionModel.Select | QItemSelectionModel.Rows
                )
                self.treeWidget.scrollTo(
                    current_index, QTreeView.PositionAtCenter)

        for i in range(8):
            self.treeWidget.resizeColumnToContents(i)

        self.treeWidget.setColumnWidth(
            1, min(self.treeWidget.columnWidth(1), 300))
        self.treeWidget.setColumnWidth(
            2, min(self.treeWidget.columnWidth(2), 200))

        self.updateInfo()

    def updateInfo(self):
        gds_info = self.gds_info
        text = ("%i datasets\n%i datasets cached\n" %
                (len(gds_info),
                 len(glob.glob(serverfiles.localpath("GEO") + "/GDS*"))))
        filtered = self.treeWidget.model().rowCount()
        if len(self.gds) != filtered:
            text += ("%i after filtering") % filtered
        self.infoBox.setText(text)

    def updateSelection(self, *args):
        current = self.treeWidget.selectedIndexes()
        mapToSource = self.treeWidget.model().mapToSource
        current = [mapToSource(index).row() for index in current]
        if current:
            self.currentGds = self.gds[current[0]]
            self.setAnnotations(self.currentGds)
            self.infoGDS.setText(self.currentGds.get("description", ""))
            self.nameEdit.setPlaceholderText(self.currentGds["title"])
            self.datasetName = \
                self.datasetNames.get(self.currentGds["dataset_id"], "")
        else:
            self.currentGds = None
            self.nameEdit.setPlaceholderText("")
            self.datasetName = ""

        self.commitIf()

    def setAnnotations(self, gds):
        self._annotationsUpdating = True
        self.annotationsTree.clear()

        annotations = defaultdict(set)
        subsetscount = {}
        for desc in gds["subsets"]:
            annotations[desc["type"]].add(desc["description"])
            subsetscount[desc["description"]] = str(len(desc["sample_id"]))

        for type, subsets in annotations.items():
            key = (gds["dataset_id"], type)
            subsetItem = QTreeWidgetItem(self.annotationsTree, [type])
            subsetItem.setFlags(subsetItem.flags() | Qt.ItemIsUserCheckable |
                                Qt.ItemIsTristate)
            subsetItem.setCheckState(
                0, self.gdsSelectionStates.get(key, Qt.Checked)
            )
            subsetItem.key = key
            for subset in subsets:
                key = (gds["dataset_id"], type, subset)
                item = QTreeWidgetItem(
                    subsetItem, [subset, subsetscount.get(subset, "")]
                )
                item.setFlags(item.flags() | Qt.ItemIsUserCheckable)
                item.setCheckState(
                    0, self.gdsSelectionStates.get(key, Qt.Checked)
                )
                item.key = key
        self._annotationsUpdating = False
        self.annotationsTree.expandAll()
        for i in range(self.annotationsTree.columnCount()):
            self.annotationsTree.resizeColumnToContents(i)

    def annotationSelectionChanged(self, item, column):
        if self._annotationsUpdating:
            return
        for i in range(self.annotationsTree.topLevelItemCount()):
            item = self.annotationsTree.topLevelItem(i)
            self.gdsSelectionStates[item.key] = item.checkState(0)
            for j in range(item.childCount()):
                child = item.child(j)
                self.gdsSelectionStates[child.key] = child.checkState(0)

    def filter(self):
        filter_string = unicode(self.filterLineEdit.text())
        proxyModel = self.treeWidget.model()
        if proxyModel:
            strings = filter_string.lower().strip().split()
            proxyModel.setFilterFixedStrings(strings)
            self.updateInfo()

    def selectedSamples(self):
        """
        Return the currently selected sample annotations.

        The return value is a list of selected (sample type, sample value)
        tuples.

        .. note:: if some Sample annotation type has no selected values.
                  this method will return all values for it.

        """
        samples = []
        unused_types = []
        used_types = []
        for stype in childiter(self.annotationsTree.invisibleRootItem()):
            selected_values = []
            all_values = []
            for sval in childiter(stype):
                value = (str(stype.text(0)), str(sval.text(0)))
                if self.gdsSelectionStates.get(sval.key, True):
                    selected_values.append(value)
                all_values.append(value)
            if selected_values:
                samples.extend(selected_values)
                used_types.append(str(stype.text(0)))
            else:
                # If no sample of sample type is selected we don't filter
                # on it.
                samples.extend(all_values)
                unused_types.append(str(stype.text(0)))

        return samples, used_types

    def commitIf(self):
        if self.autoCommit:
            self.commit()
        else:
            self.selectionChanged = True

    @Slot(int, int)
    def progressCompleted(self, value, total):
        if total > 0:
            self.progressBarSet(100. * value / total, processEvents=False)
        else:
            pass
            # TODO: report 'indeterminate progress'

    def commit(self):
        if self.currentGds:
            self.error(0)
            sample_type = None
            self.progressBarInit(processEvents=None)

            _, groups = self.selectedSamples()
            if len(groups) == 1 and self.outputRows:
                sample_type = groups[0]

            self.setEnabled(False)
            self.setBlocking(True)

            progress = methodinvoke(self, "progressCompleted", (int, int))

            def get_data(gds_id, report_genes, transpose, sample_type, title):
                gds_ensure_downloaded(gds_id, progress)
                gds = geo.GDS(gds_id)
                data = gds.getdata(
                    report_genes=report_genes, transpose=transpose,
                    sample_type=sample_type
                )
                data.name = title
                return data

            get_data = partial(
                get_data, self.currentGds["dataset_id"],
                report_genes=self.mergeSpots,
                transpose=self.outputRows,
                sample_type=sample_type,
                title=self.datasetName or self.currentGds["title"]
            )
            self._datatask = Task(function=get_data)
            self._datatask.finished.connect(self._on_dataready)
            self._executor.submit(self._datatask)

    def _on_dataready(self):
        self.setEnabled(True)
        self.setBlocking(False)
        self.progressBarFinished(processEvents=False)

        try:
            data = self._datatask.result()
        except urlrequest.URLError as error:
            self.error(0, ("Error while connecting to the NCBI ftp server! "
                           "'%s'" % error))
            sys.excepthook(type(error), error, getattr(error, "__traceback__"))
            return
        finally:
            self._datatask = None

        data_name = data.name
        samples, _ = self.selectedSamples()

        self.warning(0)
        message = None
        if self.outputRows:
            def samplesinst(ex):
                out = []
                for meta in data.domain.metas:
                    out.append((meta.name, ex[meta].value))

                if data.domain.class_var.name != 'class':
                    out.append((data.domain.class_var.name,
                                ex[data.domain.class_var].value))

                return out
            samples = set(samples)
            mask = [samples.issuperset(samplesinst(ex)) for ex in data]
            data = data[numpy.array(mask, dtype=bool)]
            if len(data) == 0:
                message = "No samples with selected sample annotations."
        else:
            samples = set(samples)
            domain = Orange.data.Domain(
                [attr for attr in data.domain.attributes
                 if samples.issuperset(attr.attributes.items())],
                data.domain.class_var,
                data.domain.metas
            )
#             domain.addmetas(data.domain.getmetas())

            if len(domain.attributes) == 0:
                message = "No samples with selected sample annotations."
            stypes = set(s[0] for s in samples)
            for attr in domain.attributes:
                attr.attributes = dict(
                    (key, value) for key, value in attr.attributes.items()
                    if key in stypes
                )
            data = Orange.data.Table(domain, data)

        if message is not None:
            self.warning(0, message)

        data_hints.set_hint(data, "taxid", self.currentGds.get("taxid", ""),
                            10.0)
        data_hints.set_hint(data, "genesinrows", self.outputRows, 10.0)

        data.name = data_name
        self.send("Expression Data", data)

        model = self.treeWidget.model().sourceModel()
        row = self.gds.index(self.currentGds)

        model.setData(model.index(row, 0),  " ", Qt.DisplayRole)

        self.updateInfo()
        self.selectionChanged = False

    def splitterMoved(self, *args):
        self.splitterSettings = [bytes(sp.saveState()) for sp in self.splitters]

    def send_report(self):
        self.report_items("GEO Dataset",
                          [("ID", self.currentGds['dataset_id']), ("Title", self.currentGds['title']),
                           ("Organism", self.currentGds['sample_organism'])])
        self.report_items("Data", [("Samples", self.currentGds['sample_count']),
                                   ("Features", self.currentGds['feature_count']),
                                   ("Genes", self.currentGds['gene_count'])])
        self.report_name("Sample annotations")
        subsets = defaultdict(list)
        for subset in self.currentGds['subsets']:
            subsets[subset['type']].append((subset['description'], len(subset['sample_id'])))
        self.report_html += "<ul>"
        for type in subsets:
            self.report_html += "<b>" + type + ":</b></br>"
            for desc, count in subsets[type]:
                self.report_html += 9 * "&nbsp" + "<b>{}:</b> {}</br>".format(desc, count)
        self.report_html += "</ul>"

    def onDeleteWidget(self):
        if self._inittask:
            self._inittask.future().cancel()
            self._inittask.finished.disconnect(self._initializemodel)
        if self._datatask:
            self._datatask.future().cancel()
            self._datatask.finished.disconnect(self._on_dataready)
        self._executor.shutdown(wait=False)

        super(OWGEODatasets, self).onDeleteWidget()

    def onNameEdited(self):
        if self.currentGds:
            gds_id = self.currentGds["dataset_id"]
            self.datasetNames[gds_id] = unicode(self.nameEdit.text())
            self.commitIf()
Example #5
0
class OWDatabasesUpdate(OWWidget):

    name = "Databases Update"
    description = "Update local systems biology databases."
    icon = "../widgets/icons/Databases.svg"
    priority = 10

    inputs = []
    outputs = []

    want_main_area = False

    def __init__(self, parent=None, signalManager=None,
                 name="Databases update", domains=None):
        OWWidget.__init__(self, parent, signalManager, name,
                          wantMainArea=False)

        self.searchString = ""
        self.accessCode = ""
        self.domains = domains or DOMAINS
        self.serverFiles = serverfiles.ServerFiles()

        fbox = gui.widgetBox(self.controlArea, "Filter")
        self.completer = TokenListCompleter(
            self, caseSensitivity=Qt.CaseInsensitive)
        self.lineEditFilter = QLineEdit(textChanged=self.SearchUpdate)
        self.lineEditFilter.setCompleter(self.completer)

        fbox.layout().addWidget(self.lineEditFilter)

        box = gui.widgetBox(self.controlArea, "Files")

        self.filesView = QTreeWidget(self)
        self.filesView.setHeaderLabels(
            ["", "Data Source", "Update", "Last Updated", "Size"])

        self.filesView.setRootIsDecorated(False)
        self.filesView.setUniformRowHeights(True)
        self.filesView.setSelectionMode(QAbstractItemView.NoSelection)
        self.filesView.setSortingEnabled(True)
        self.filesView.sortItems(1, Qt.AscendingOrder)
        self.filesView.setItemDelegateForColumn(
            0, UpdateOptionsItemDelegate(self.filesView))

        self.filesView.model().layoutChanged.connect(self.SearchUpdate)

        box.layout().addWidget(self.filesView)

        box = gui.widgetBox(self.controlArea, orientation="horizontal")
        self.updateButton = gui.button(
            box, self, "Update all",
            callback=self.UpdateAll,
            tooltip="Update all updatable files",
         )

        self.downloadButton = gui.button(
            box, self, "Download all",
            callback=self.DownloadFiltered,
            tooltip="Download all filtered files shown"
        )

        self.cancelButton = gui.button(
            box, self, "Cancel", callback=self.Cancel,
            tooltip="Cancel scheduled downloads/updates."
        )

        self.retryButton = gui.button(
            box, self, "Reconnect", callback=self.RetrieveFilesList
        )
        self.retryButton.hide()

        gui.rubber(box)

        gui.lineEdit(box, self, "accessCode", "Access Code",
                     orientation="horizontal",
                     callback=self.RetrieveFilesList)

        self.warning(0)

        box = gui.widgetBox(self.controlArea, orientation="horizontal")
        gui.rubber(box)

        self.infoLabel = QLabel()
        self.infoLabel.setAlignment(Qt.AlignCenter)

        self.controlArea.layout().addWidget(self.infoLabel)
        self.infoLabel.setSizePolicy(QSizePolicy.Expanding, QSizePolicy.Fixed)

        self.updateItems = []

        self.resize(800, 600)

        self.progress = ProgressState(self, maximum=3)
        self.progress.valueChanged.connect(self._updateProgress)
        self.progress.rangeChanged.connect(self._updateProgress)
        self.executor = ThreadExecutor(
            threadPool=QThreadPool(maxThreadCount=2)
        )

        task = Task(self, function=self.RetrieveFilesList)
        task.exceptionReady.connect(self.HandleError)
        task.start()

        self._tasks = []
        self._haveProgress = False

    def RetrieveFilesList(self):
        self.retryButton.hide()
        self.warning(0)
        self.progress.setRange(0, 3)
        self.serverFiles = serverfiles.ServerFiles(access_code=self.accessCode)

        task = Task(function=partial(retrieveFilesList, self.serverFiles,
                                     self.domains,
                                     methodinvoke(self.progress, "advance")))

        task.resultReady.connect(self.SetFilesList)
        task.exceptionReady.connect(self.HandleError)

        self.executor.submit(task)

        self.setEnabled(False)

    def SetFilesList(self, serverInfo):
        """
        Set the files to show.
        """
        self.setEnabled(True)

        domains = serverInfo.keys()
        if not domains:
            if self.domains:
                domains = self.domains
            else:
                domains = serverfiles.listdomains()

        localInfo = dict([(dom, serverfiles.allinfo(dom)) for dom in domains])

        all_tags = set()

        self.filesView.clear()
        self.updateItems = []

        for item in join_info_dict(localInfo, serverInfo):
            tree_item = UpdateTreeWidgetItem(item)
            options_widget = UpdateOptionsWidget(item.state)
            options_widget.item = item

            options_widget.installClicked.connect(
                partial(self.SubmitDownloadTask, item.domain, item.filename)
            )
            options_widget.removeClicked.connect(
                partial(self.SubmitRemoveTask, item.domain, item.filename)
            )

            self.updateItems.append((item, tree_item, options_widget))
            all_tags.update(item.tags)

        self.filesView.addTopLevelItems(
            [tree_item for _, tree_item, _ in self.updateItems]
        )

        for item, tree_item, options_widget in self.updateItems:
            self.filesView.setItemWidget(tree_item, 0, options_widget)

            # Add an update button if the file is updateable
            if item.state == OUTDATED:
                button = QToolButton(
                    None, text="Update",
                    maximumWidth=120,
                    minimumHeight=20,
                    maximumHeight=20
                )

                if sys.platform == "darwin":
                    button.setAttribute(Qt.WA_MacSmallSize)

                button.clicked.connect(
                    partial(self.SubmitDownloadTask, item.domain,
                            item.filename)
                )

                self.filesView.setItemWidget(tree_item, 2, button)

        self.progress.advance()

        self.filesView.setColumnWidth(0, self.filesView.sizeHintForColumn(0))

        for column in range(1, 4):
            contents_hint = self.filesView.sizeHintForColumn(column)
            header_hint = self.filesView.header().sectionSizeHint(column)
            width = max(min(contents_hint, 400), header_hint)
            self.filesView.setColumnWidth(column, width)

        hints = [hint for hint in sorted(all_tags) if not hint.startswith("#")]
        self.completer.setTokenList(hints)
        self.SearchUpdate()
        self.UpdateInfoLabel()
        self.toggleButtons()
        self.cancelButton.setEnabled(False)

        self.progress.setRange(0, 0)

    def buttonCheck(self, selected_items, state, button):
        for item in selected_items:
            if item.state != state:
                button.setEnabled(False)
            else:
                button.setEnabled(True)
                break

    def toggleButtons(self):
        selected_items = [item for item, tree_item, _ in self.updateItems if not tree_item.isHidden()]
        self.buttonCheck(selected_items, OUTDATED, self.updateButton)
        self.buttonCheck(selected_items, AVAILABLE, self.downloadButton)

    def HandleError(self, exception):
        if isinstance(exception, ConnectionError):
            self.warning(0,
                       "Could not connect to server! Check your connection "
                       "and try to reconnect.")
            self.SetFilesList({})
            self.retryButton.show()
        else:
            sys.excepthook(type(exception), exception.args, None)
            self.progress.setRange(0, 0)
            self.setEnabled(True)

    def UpdateInfoLabel(self):
        local = [item for item, tree_item, _ in self.updateItems
                 if item.state != AVAILABLE and not tree_item.isHidden()]
        size = sum(float(item.size) for item in local)

        onServer = [item for item, tree_item, _ in self.updateItems if not tree_item.isHidden()]
        sizeOnServer = sum(float(item.size) for item in onServer)

        text = ("%i items, %s (on server: %i items, %s)" %
                (len(local),
                 sizeof_fmt(size),
                 len(onServer),
                 sizeof_fmt(sizeOnServer)))

        self.infoLabel.setText(text)

    def UpdateAll(self):
        self.warning(0)
        for item, tree_item, _ in self.updateItems:
            if item.state == OUTDATED and not tree_item.isHidden():
                self.SubmitDownloadTask(item.domain, item.filename)

    def DownloadFiltered(self):
        # TODO: submit items in the order shown.
        for item, tree_item, _ in self.updateItems:
            if not tree_item.isHidden() and item.state in \
                    [AVAILABLE, OUTDATED]:
                self.SubmitDownloadTask(item.domain, item.filename)

    def SearchUpdate(self, searchString=None):
        strings = str(self.lineEditFilter.text()).split()
        for item, tree_item, _ in self.updateItems:
            hide = not all(UpdateItem_match(item, string)
                           for string in strings)
            tree_item.setHidden(hide)
        self.UpdateInfoLabel()
        self.toggleButtons()

    def SubmitDownloadTask(self, domain, filename):
        """
        Submit the (domain, filename) to be downloaded/updated.
        """
        self.cancelButton.setEnabled(True)

        index = self.updateItemIndex(domain, filename)
        _, tree_item, opt_widget = self.updateItems[index]

        if self.accessCode:
            sf = serverfiles.ServerFiles(access_code=self.accessCode)
        else:
            sf = serverfiles.ServerFiles()

        task = DownloadTask(domain, filename, sf)

        self.progress.adjustRange(0, 100)

        pb = ItemProgressBar(self.filesView)
        pb.setRange(0, 100)
        pb.setTextVisible(False)

        task.advanced.connect(pb.advance)
        task.advanced.connect(self.progress.advance)
        task.finished.connect(pb.hide)
        task.finished.connect(self.onDownloadFinished, Qt.QueuedConnection)
        task.exception.connect(self.onDownloadError, Qt.QueuedConnection)

        self.filesView.setItemWidget(tree_item, 2, pb)

        # Clear the text so it does not show behind the progress bar.
        tree_item.setData(2, Qt.DisplayRole, "")
        pb.show()

        # Disable the options widget
        opt_widget.setEnabled(False)
        self._tasks.append(task)

        self.executor.submit(task)

    def EndDownloadTask(self, task):
        future = task.future()
        index = self.updateItemIndex(task.domain, task.filename)
        item, tree_item, opt_widget = self.updateItems[index]

        self.filesView.removeItemWidget(tree_item, 2)
        opt_widget.setEnabled(True)

        if future.cancelled():
            # Restore the previous state
            tree_item.setUpdateItem(item)
            opt_widget.setState(item.state)

        elif future.exception():
            tree_item.setUpdateItem(item)
            opt_widget.setState(item.state)

            # Show the exception string in the size column.
            self.warning(0, "Error while downloading. Check your connection "
                            "and retry.")

            # recreate button for download
            button = QToolButton(
                None, text="Retry",
                maximumWidth=120,
                minimumHeight=20,
                maximumHeight=20
            )

            if sys.platform == "darwin":
                button.setAttribute(Qt.WA_MacSmallSize)

            button.clicked.connect(
                partial(self.SubmitDownloadTask, item.domain,
                        item.filename)
            )

            self.filesView.setItemWidget(tree_item, 2, button)

        else:
            # get the new updated info dict and replace the the old item
            self.warning(0)
            info = serverfiles.info(item.domain, item.filename)
            new_item = update_item_from_info(item.domain, item.filename,
                                             info, info)

            self.updateItems[index] = (new_item, tree_item, opt_widget)

            tree_item.setUpdateItem(new_item)
            opt_widget.setState(new_item.state)

            self.UpdateInfoLabel()

    def SubmitRemoveTask(self, domain, filename):
        serverfiles.remove(domain, filename)
        index = self.updateItemIndex(domain, filename)
        item, tree_item, opt_widget = self.updateItems[index]

        if item.info_server:
            new_item = item._replace(state=AVAILABLE, local=None,
                                      info_local=None)
        else:
            new_item = item._replace(local=None, info_local=None)
            # Disable the options widget. No more actions can be performed
            # for the item.
            opt_widget.setEnabled(False)

        tree_item.setUpdateItem(new_item)
        opt_widget.setState(new_item.state)
        self.updateItems[index] = (new_item, tree_item, opt_widget)

        self.UpdateInfoLabel()

    def Cancel(self):
        """
        Cancel all pending update/download tasks (that have not yet started).
        """
        for task in self._tasks:
            task.future().cancel()

    def onDeleteWidget(self):
        self.Cancel()
        self.executor.shutdown(wait=False)
        OWWidget.onDeleteWidget(self)

    def onDownloadFinished(self):
        # on download completed/canceled/error
        assert QThread.currentThread() is self.thread()
        for task in list(self._tasks):
            future = task.future()
            if future.done():
                self.EndDownloadTask(task)
                self._tasks.remove(task)

        if not self._tasks:
            # Clear/reset the overall progress
            self.progress.setRange(0, 0)
            self.cancelButton.setEnabled(False)

    def onDownloadError(self, exc_info):
        sys.excepthook(*exc_info)
        self.warning(0, "Error while downloading. Check your connection and "
                        "retry.")

    def updateItemIndex(self, domain, filename):
        for i, (item, _, _) in enumerate(self.updateItems):
            if item.domain == domain and item.filename == filename:
                return i
        raise ValueError("%r, %r not in update list" % (domain, filename))

    def _updateProgress(self, *args):
        rmin, rmax = self.progress.range()
        if rmin != rmax:
            if not self._haveProgress:
                self._haveProgress = True
                self.progressBarInit()

            self.progressBarSet(self.progress.ratioCompleted() * 100,
                                processEvents=None)
        if rmin == rmax:
            self._haveProgress = False
            self.progressBarFinished()
Example #6
0
class OWGEODatasets(OWWidget):
    settingsList = ["outputRows", "mergeSpots", "gdsSelectionStates",
                    "splitterSettings", "currentGds", "autoCommit",
                    "datasetNames"]

    def __init__(self, parent=None, signalManager=None, name=" GEO Data Sets"):
        OWWidget.__init__(self, parent, signalManager, name)

        self.outputs = [("Expression Data", ExampleTable)]

        ## Settings
        self.selectedAnnotation = 0
        self.includeIf = False
        self.minSamples = 3
        self.autoCommit = False
        self.outputRows = 1
        self.mergeSpots = True
        self.filterString = ""
        self.currentGds = None
        self.selectionChanged = False
        self.autoCommit = False
        self.gdsSelectionStates = {}
        self.splitterSettings = [
            '\x00\x00\x00\xff\x00\x00\x00\x00\x00\x00\x00\x02\x00\x00\x01\xea\x00\x00\x00\xd7\x01\x00\x00\x00\x07\x01\x00\x00\x00\x02',
            '\x00\x00\x00\xff\x00\x00\x00\x00\x00\x00\x00\x02\x00\x00\x01\xb5\x00\x00\x02\x10\x01\x00\x00\x00\x07\x01\x00\x00\x00\x01'
        ]
        self.datasetNames = {}
        self.loadSettings()

        self.datasetName = ""

        ## GUI
        self.infoBox = OWGUI.widgetLabel(
            OWGUI.widgetBox(self.controlArea, "Info", addSpace=True),
            "Initializing\n\n"
        )

        box = OWGUI.widgetBox(self.controlArea, "Output", addSpace=True)
        OWGUI.radioButtonsInBox(box, self, "outputRows",
                                ["Genes in rows", "Samples in rows"], "Rows",
                                callback=self.commitIf)
        OWGUI.checkBox(box, self, "mergeSpots", "Merge spots of same gene",
                       callback=self.commitIf)

        OWGUI.separator(box)
        self.nameEdit = OWGUI.lineEdit(
            box, self, "datasetName", "Data set name",
            tooltip="Override the default output data set name",
            callback=self.onNameEdited
        )
        self.nameEdit.setPlaceholderText("")

        box = OWGUI.widgetBox(self.controlArea, "Commit", addSpace=True)
        self.commitButton = OWGUI.button(box, self, "Commit",
                                         callback=self.commit)
        cb = OWGUI.checkBox(box, self, "autoCommit", "Commit on any change")
        OWGUI.setStopper(self, self.commitButton, cb, "selectionChanged",
                         self.commit)
        OWGUI.rubber(self.controlArea)

        self.filterLineEdit = OWGUIEx.lineEditHint(
            self.mainArea, self, "filterString", "Filter",
            caseSensitive=False, matchAnywhere=True,
            callback=self.filter,  delimiters=" ")

        splitter = QSplitter(Qt.Vertical, self.mainArea)
        self.mainArea.layout().addWidget(splitter)
        self.treeWidget = QTreeView(splitter)

        self.treeWidget.setSelectionMode(QAbstractItemView.SingleSelection)
        self.treeWidget.setRootIsDecorated(False)
        self.treeWidget.setSortingEnabled(True)
        self.treeWidget.setAlternatingRowColors(True)
        self.treeWidget.setUniformRowHeights(True)
        self.treeWidget.setEditTriggers(QTreeView.NoEditTriggers)

        linkdelegate = LinkStyledItemDelegate(self.treeWidget)
        self.treeWidget.setItemDelegateForColumn(1, linkdelegate)
        self.treeWidget.setItemDelegateForColumn(8, linkdelegate)
        self.treeWidget.setItemDelegateForColumn(
            0, OWGUI.IndicatorItemDelegate(self.treeWidget,
                                           role=Qt.DisplayRole))

        proxyModel = MySortFilterProxyModel(self.treeWidget)
        self.treeWidget.setModel(proxyModel)
        self.treeWidget.selectionModel().selectionChanged.connect(
            self.updateSelection
        )
        self.treeWidget.viewport().setMouseTracking(True)

        splitterH = QSplitter(Qt.Horizontal, splitter)

        box = OWGUI.widgetBox(splitterH, "Description")
        self.infoGDS = OWGUI.widgetLabel(box, "")
        self.infoGDS.setWordWrap(True)
        OWGUI.rubber(box)

        box = OWGUI.widgetBox(splitterH, "Sample Annotations")
        self.annotationsTree = QTreeWidget(box)
        self.annotationsTree.setHeaderLabels(
            ["Type (Sample annotations)", "Sample count"]
        )
        self.annotationsTree.setRootIsDecorated(True)
        box.layout().addWidget(self.annotationsTree)
        self.annotationsTree.itemChanged.connect(
            self.annotationSelectionChanged
        )
        self._annotationsUpdating = False
        self.splitters = splitter, splitterH

        for sp, setting in zip(self.splitters, self.splitterSettings):
            sp.splitterMoved.connect(self.splitterMoved)
            sp.restoreState(setting)

        self.searchKeys = ["dataset_id", "title", "platform_organism",
                           "description"]

        self.gds = []
        self.gds_info = None

        self.resize(1000, 600)

        self.setBlocking(True)
        self.setEnabled(False)
        self.progressBarInit()

        self._executor = ThreadExecutor()

        func = partial(get_gds_model,
                       methodinvoke(self, "_setProgress", (float,)))
        self._inittask = Task(function=func)
        self._inittask.finished.connect(self._initializemodel)
        self._executor.submit(self._inittask)

        self._datatask = None

    @pyqtSlot(float)
    def _setProgress(self, value):
        self.progressBarValue = value

    def _initializemodel(self):
        assert self.thread() is QThread.currentThread()
        model, self.gds_info, self.gds = self._inittask.result()
        model.setParent(self)

        proxy = self.treeWidget.model()
        proxy.setFilterKeyColumn(0)
        proxy.setFilterRole(TextFilterRole)
        proxy.setFilterCaseSensitivity(False)
        proxy.setFilterFixedString(self.filterString)

        proxy.setSourceModel(model)
        proxy.sort(0, Qt.DescendingOrder)

        self.progressBarFinished()
        self.setBlocking(False)
        self.setEnabled(True)

        filter_items = " ".join(
            gds[key] for gds in self.gds for key in self.searchKeys
        )
        tr_chars = ",.:;!?(){}[]_-+\\|/%#@$^&*<>~`"
        tr_table = string.maketrans(tr_chars, " " * len(tr_chars))
        filter_items = filter_items.translate(tr_table)

        filter_items = sorted(set(filter_items.split(" ")))
        filter_items = [item for item in filter_items if len(item) > 3]
        self.filterLineEdit.setItems(filter_items)

        if self.currentGds:
            gdss = [(i, proxy.data(proxy.index(i, 1), Qt.DisplayRole))
                    for i in range(proxy.rowCount())]
            current = [i for i, variant in gdss
                       if variant.isValid() and
                       str(variant.toString()) == self.currentGds["dataset_id"]]
            if current:
                current_index = proxy.index(current[0], 0)
                self.treeWidget.selectionModel().select(
                    current_index,
                    QItemSelectionModel.Select |
                    QItemSelectionModel.Rows
                )
                self.treeWidget.scrollTo(current_index)

        for i in range(8):
            self.treeWidget.resizeColumnToContents(i)

        self.treeWidget.setColumnWidth(
            1, min(self.treeWidget.columnWidth(1), 300))
        self.treeWidget.setColumnWidth(
            2, min(self.treeWidget.columnWidth(2), 200))

        self.updateInfo()

    def updateInfo(self):
        gds_info = self.gds_info
        text = ("%i datasets\n%i datasets cached\n" %
                (len(gds_info),
                 len(glob.glob(serverfiles.localpath("GEO") + "/GDS*"))))
        filtered = self.treeWidget.model().rowCount()
        if len(self.gds) != filtered:
            text += ("%i after filtering") % filtered
        self.infoBox.setText(text)

    def updateSelection(self, *args):
        current = self.treeWidget.selectedIndexes()
        mapToSource = self.treeWidget.model().mapToSource
        current = [mapToSource(index).row() for index in current]
        if current:
            self.currentGds = self.gds[current[0]]
            self.setAnnotations(self.currentGds)
            self.infoGDS.setText(self.currentGds.get("description", ""))
            self.nameEdit.setPlaceholderText(self.currentGds["title"])
            self.datasetName = \
                self.datasetNames.get(self.currentGds["dataset_id"], "")
        else:
            self.currentGds = None
            self.nameEdit.setPlaceholderText("")
            self.datasetName = ""

        self.commitIf()

    def setAnnotations(self, gds):
        self._annotationsUpdating = True
        self.annotationsTree.clear()

        annotations = defaultdict(set)
        subsetscount = {}
        for desc in gds["subsets"]:
            annotations[desc["type"]].add(desc["description"])
            subsetscount[desc["description"]] = str(len(desc["sample_id"]))

        for type, subsets in annotations.items():
            key = (gds["dataset_id"], type)
            subsetItem = QTreeWidgetItem(self.annotationsTree, [type])
            subsetItem.setFlags(subsetItem.flags() | Qt.ItemIsUserCheckable |
                                Qt.ItemIsTristate)
            subsetItem.setCheckState(
                0, self.gdsSelectionStates.get(key, Qt.Checked)
            )
            subsetItem.key = key
            for subset in subsets:
                key = (gds["dataset_id"], type, subset)
                item = QTreeWidgetItem(
                    subsetItem, [subset, subsetscount.get(subset, "")]
                )
                item.setFlags(item.flags() | Qt.ItemIsUserCheckable)
                item.setCheckState(
                    0, self.gdsSelectionStates.get(key, Qt.Checked)
                )
                item.key = key
        self._annotationsUpdating = False
        self.annotationsTree.expandAll()
        for i in range(self.annotationsTree.columnCount()):
            self.annotationsTree.resizeColumnToContents(i)

    def annotationSelectionChanged(self, item, column):
        if self._annotationsUpdating:
            return
        for i in range(self.annotationsTree.topLevelItemCount()):
            item = self.annotationsTree.topLevelItem(i)
            self.gdsSelectionStates[item.key] = item.checkState(0)
            for j in range(item.childCount()):
                child = item.child(j)
                self.gdsSelectionStates[child.key] = child.checkState(0)

    def filter(self):
        filter_string = unicode(self.filterLineEdit.text(), errors="ignore")
        proxyModel = self.treeWidget.model()
        if proxyModel:
            strings = filter_string.lower().strip().split()
            proxyModel.setFilterFixedStrings(strings)
            self.updateInfo()

    def selectedSamples(self):
        """
        Return the currently selected sample annotations.

        The return value is a list of selected (sample type, sample value)
        tuples.

        .. note:: if some Sample annotation type has no selected values.
                  this method will return all values for it.

        """
        samples = []
        unused_types = []
        used_types = []
        for stype in childiter(self.annotationsTree.invisibleRootItem()):
            selected_values = []
            all_values = []
            for sval in childiter(stype):
                value = (str(stype.text(0)), str(sval.text(0)))
                if self.gdsSelectionStates.get(sval.key, True):
                    selected_values.append(value)
                all_values.append(value)
            if selected_values:
                samples.extend(selected_values)
                used_types.append(str(stype.text(0)))
            else:
                # If no sample of sample type is selected we don't filter
                # on it.
                samples.extend(all_values)
                unused_types.append(str(stype.text(0)))

        return samples, used_types

    def commitIf(self):
        if self.autoCommit:
            self.commit()
        else:
            self.selectionChanged = True

    def commit(self):
        if self.currentGds:
            self.error(0)
            sample_type = None
            self.progressBarInit()
            self.progressBarSet(10)

            _, groups = self.selectedSamples()
            if len(groups) == 1 and self.outputRows:
                sample_type = groups[0]

            self.setEnabled(False)
            self.setBlocking(True)

            def get_data(gds_id, report_genes, transpose, sample_type, title):
                gds = geo.GDS(gds_id)
                data = gds.getdata(
                    report_genes=report_genes, transpose=transpose,
                    sample_type=sample_type
                )
                data.name = title
                return data

            get_data = partial(
                get_data, self.currentGds["dataset_id"],
                report_genes=self.mergeSpots,
                transpose=self.outputRows,
                sample_type=sample_type,
                title=self.datasetName or self.currentGds["title"]
            )
            self._datatask = Task(function=get_data)
            self._datatask.finished.connect(self._on_dataready)
            self._executor.submit(self._datatask)

    def _on_dataready(self):
        self.setEnabled(True)
        self.setBlocking(False)

        self.progressBarSet(50)

        try:
            data = self._datatask.result()
        except urllib2.URLError as error:
            self.error(0, "Error while connecting to the NCBI ftp server! %r" %
                       error)
            self._datatask = None
            self.progressBarFinished()
            return

        self._datatask = None

        data_name = data.name
        samples, _ = self.selectedSamples()

        self.warning(0)
        message = None
        if self.outputRows:
            def samplesinst(ex):
                out = []
                for i, a in data.domain.get_metas().items():
                    out.append((a.name, ex[i].value))
                if data.domain.class_var.name != 'class':
                    out.append((data.domain.class_var.name, ex[-1].value))
                return out
            samples = set(samples)

            select = [1 if samples.issuperset(samplesinst(ex)) else 0
                      for ex in data]
            data = data.select(select)
            if len(data) == 0:
                message = "No samples with selected sample annotations."
        else:
            samples = set(samples)
            domain = orange.Domain(
                [attr for attr in data.domain.attributes
                 if samples.issuperset(attr.attributes.items())],
                data.domain.classVar
            )
            domain.addmetas(data.domain.getmetas())
            if len(domain.attributes) == 0:
                message = "No samples with selected sample annotations."
            stypes = set(s[0] for s in samples)
            for attr in domain.attributes:
                attr.attributes = dict(
                    (key, value) for key, value in attr.attributes.items()
                    if key in stypes
                )
            data = orange.ExampleTable(domain, data)

        if message is not None:
            self.warning(0, message)

        data_hints.set_hint(data, "taxid", self.currentGds.get("taxid", ""),
                            10.0)
        data_hints.set_hint(data, "genesinrows", self.outputRows, 10.0)

        self.progressBarFinished()
        data.name = data_name
        self.send("Expression Data", data)

        model = self.treeWidget.model().sourceModel()
        row = self.gds.index(self.currentGds)

        model.setData(model.index(row, 0),  QVariant(" "), Qt.DisplayRole)

        self.updateInfo()
        self.selectionChanged = False

    def splitterMoved(self, *args):
        self.splitterSettings = [str(sp.saveState()) for sp in self.splitters]

    def onDeleteWidget(self):
        if self._inittask:
            self._inittask.future().cancel()
            self._inittask.finished.disconnect(self._initializemodel)
        if self._datatask:
            self._datatask.future().cancel()
            self._datatask.finished.disconnect(self._on_dataready)
        self._executor.shutdown(wait=False)

        super(OWGEODatasets, self).onDeleteWidget()

    def onNameEdited(self):
        if self.currentGds:
            gds_id = self.currentGds["dataset_id"]
            self.datasetNames[gds_id] = unicode(self.nameEdit.text())
            self.commitIf()
Example #7
0
class OWGeneNetwork(OWWidget.OWWidget):
    settingsList = [
        "taxid", "use_attr_names", "network_source", "include_neighborhood",
        "min_score"
    ]
    contextHandlers = {
        "":
        OWWidget.DomainContextHandler(
            "", ["taxid", "gene_var_index", "use_attr_names"])
    }

    def __init__(self, parent=None, signalManager=None, title="Gene Network"):
        super(OWGeneNetwork, self).__init__(parent,
                                            signalManager,
                                            title,
                                            wantMainArea=False,
                                            resizingEnabled=False)

        self.taxid = "9606"
        self.gene_var_index = -1
        self.use_attr_names = False
        self.network_source = 1
        self.include_neighborhood = True
        self.autocommit = False
        self.min_score = 0.9
        self.loadSettings()

        self.taxids = taxonomy.common_taxids()
        self.current_taxid_index = self.taxids.index(self.taxid)

        self.data = None
        self.geneinfo = None
        self.nettask = None
        self._invalidated = False

        box = OWGUI.widgetBox(self.controlArea, "Info")
        self.info = OWGUI.widgetLabel(box, "No data on input\n")

        box = OWGUI.widgetBox(self.controlArea, "Organism")
        self.organism_cb = OWGUI.comboBox(box,
                                          self,
                                          "current_taxid_index",
                                          items=map(taxonomy.name,
                                                    self.taxids),
                                          callback=self._update_organism)
        box = OWGUI.widgetBox(self.controlArea, "Genes")
        self.genes_cb = OWGUI.comboBox(box,
                                       self,
                                       "gene_var_index",
                                       callback=self._update_query_genes)
        self.varmodel = OWItemModels.VariableListModel()
        self.genes_cb.setModel(self.varmodel)

        OWGUI.checkBox(box,
                       self,
                       "use_attr_names",
                       "Use attribute names",
                       callback=self._update_query_genes)

        box = OWGUI.widgetBox(self.controlArea, "Network")
        OWGUI.comboBox(box,
                       self,
                       "network_source",
                       items=[s.name for s in SOURCES],
                       callback=self._on_source_db_changed)
        OWGUI.checkBox(box,
                       self,
                       "include_neighborhood",
                       "Include immediate gene neighbors",
                       callback=self.invalidate)
        self.score_spin = OWGUI.doubleSpin(box,
                                           self,
                                           "min_score",
                                           0.0,
                                           1.0,
                                           step=0.001,
                                           label="Minimal edge score",
                                           callback=self.invalidate)
        self.score_spin.setEnabled(SOURCES[self.network_source].score_filter)

        box = OWGUI.widgetBox(self.controlArea, "Commit")
        OWGUI.button(box, self, "Commit", callback=self.commit, default=True)

        self.executor = ThreadExecutor()

    def set_data(self, data):
        self.closeContext()
        self.data = data
        if data is not None:
            self.varmodel[:] = string_variables(data.domain)
            taxid = data_hints.get_hint(data, "taxid", default=self.taxid)

            if taxid in self.taxids:
                self.set_organism(self.taxids.index(taxid))

            self.use_attr_names = data_hints.get_hint(
                data, "genesinrows", default=self.use_attr_names)

            if not (0 <= self.gene_var_index < len(self.varmodel)):
                self.gene_var_index = len(self.varmodel) - 1

            self.openContext("", data)
            self.invalidate()
            self.commit()
        else:
            self.varmodel[:] = []
            self.send("Network", None)

    def set_source_db(self, dbindex):
        self.network_source = dbindex
        self.invalidate()

    def set_organism(self, index):
        self.current_taxid_index = index
        self.taxid = self.taxids[index]
        self.invalidate()

    def set_gene_var(self, index):
        self.gene_var_index = index
        self.invalidate()

    def query_genes(self):
        if self.use_attr_names:
            if self.data is not None:
                return [var.name for var in self.data.domain.attributes]
            else:
                return []
        elif self.gene_var_index >= 0:
            var = self.varmodel[self.gene_var_index]
            genes = [
                str(inst[var]) for inst in self.data
                if not inst[var].isSpecial()
            ]
            return list(unique(genes))
        else:
            return []

    def invalidate(self):
        self._invalidated = True

        if self.nettask is not None:
            self.nettask.finished.disconnect(self._on_result_ready)
            self.nettask.future().cancel()
            self.nettask = None

        if self.autocommit:
            QTimer.singleShot(10, self._maybe_commit)

    @Slot()
    def _maybe_commit(self):
        if self._invalidated:
            self.commit()

    @Slot()
    def advance(self):
        self.progressBarValue = (self.progressBarValue + 1) % 100

    @Slot(float)
    def set_progress(self, value):
        self.progressBarValue = value

    def commit(self):
        include_neighborhood = self.include_neighborhood
        query_genes = self.query_genes()
        source = SOURCES[self.network_source]
        if source.score_filter:
            min_score = self.min_score
            assert source.name == "STRING"
            min_score = min_score * 1000
        else:
            min_score = None

        taxid = self.taxid
        progress = methodinvoke(self, "advance")
        if self.geneinfo is None:
            self.geneinfo = self.executor.submit(fetch_ncbi_geneinfo, taxid,
                                                 progress)

        geneinfo_f = self.geneinfo
        taxmap = source.tax_mapping
        db_taxid = taxmap.get(taxid, taxid)

        if db_taxid is None:
            raise ValueError("invalid taxid for this network")

        def fetch_network():
            geneinfo = geneinfo_f.result()
            ppidb = fetch_ppidb(source, db_taxid, progress)
            return get_gene_network(ppidb,
                                    geneinfo,
                                    db_taxid,
                                    query_genes,
                                    include_neighborhood=include_neighborhood,
                                    min_score=min_score,
                                    progress=methodinvoke(
                                        self, "set_progress", (float, )))

        self.nettask = Task(function=fetch_network)
        self.nettask.finished.connect(self._on_result_ready)

        self.executor.submit(self.nettask)

        self.setBlocking(True)
        self.setEnabled(False)
        self.progressBarInit()
        self._invalidated = False
        self._update_info()

    @Slot(object)
    def _on_result_ready(self, ):
        self.progressBarFinished()
        self.setBlocking(False)
        self.setEnabled(True)
        net = self.nettask.result()
        self._update_info()
        self.send("Network", net)

    def _on_source_db_changed(self):
        source = SOURCES[self.network_source]
        self.score_spin.setEnabled(source.score_filter)
        self.invalidate()

    def _update_organism(self):
        self.taxid = self.taxids[self.current_taxid_index]
        if self.geneinfo is not None:
            self.geneinfo.cancel()
        self.geneinfo = None
        self.invalidate()

    def _update_query_genes(self):
        self.invalidate()

    def _update_info(self):
        if self.data is None:
            self.info.setText("No data on input\n")
        else:
            names = self.query_genes()
            lines = ["%i unique genes on input" % len(set(names))]
            if self.nettask is not None:
                if not self.nettask.future().done():
                    lines.append("Retrieving ...")
                else:
                    net = self.nettask.result()
                    lines.append("%i nodes %i edges" %
                                 (len(net.nodes()), len(net.edges())))
            else:
                lines.append("")

            self.info.setText("\n".join(lines))
Example #8
0
class OWGeneNetwork(OWWidget.OWWidget):
    settingsList = ["taxid", "use_attr_names", "network_source",
                    "include_neighborhood", "min_score"]
    contextHandlers = {
        "": OWWidget.DomainContextHandler(
            "", ["taxid", "gene_var_index", "use_attr_names"]
        )
    }

    def __init__(self, parent=None, signalManager=None, title="Gene Network"):
        super(OWGeneNetwork, self).__init__(
            parent, signalManager, title, wantMainArea=False,
            resizingEnabled=False
        )

        self.taxid = "9606"
        self.gene_var_index = -1
        self.use_attr_names = False
        self.network_source = 1
        self.include_neighborhood = True
        self.autocommit = False
        self.min_score = 0.9
        self.loadSettings()

        self.taxids = taxonomy.common_taxids()
        self.current_taxid_index = self.taxids.index(self.taxid)

        self.data = None
        self.geneinfo = None
        self.nettask = None
        self._invalidated = False

        box = OWGUI.widgetBox(self.controlArea, "Info")
        self.info = OWGUI.widgetLabel(box, "No data on input\n")

        box = OWGUI.widgetBox(self.controlArea, "Organism")
        self.organism_cb = OWGUI.comboBox(
            box, self, "current_taxid_index",
            items=map(taxonomy.name, self.taxids),
            callback=self._update_organism
        )
        box = OWGUI.widgetBox(self.controlArea, "Genes")
        self.genes_cb = OWGUI.comboBox(
            box, self, "gene_var_index", callback=self._update_query_genes
        )
        self.varmodel = OWItemModels.VariableListModel()
        self.genes_cb.setModel(self.varmodel)

        OWGUI.checkBox(
            box, self, "use_attr_names",
            "Use attribute names",
            callback=self._update_query_genes
        )

        box = OWGUI.widgetBox(self.controlArea, "Network")
        OWGUI.comboBox(
            box, self, "network_source",
            items=[s.name for s in SOURCES],
            callback=self._on_source_db_changed
        )
        OWGUI.checkBox(
            box, self, "include_neighborhood",
            "Include immediate gene neighbors",
            callback=self.invalidate
        )
        self.score_spin = OWGUI.doubleSpin(
            box, self, "min_score", 0.0, 1.0, step=0.001,
            label="Minimal edge score",
            callback=self.invalidate
        )
        self.score_spin.setEnabled(SOURCES[self.network_source].score_filter)

        box = OWGUI.widgetBox(self.controlArea, "Commit")
        OWGUI.button(box, self, "Commit", callback=self.commit, default=True)

        self.executor = ThreadExecutor()

    def set_data(self, data):
        self.closeContext()
        self.data = data
        if data is not None:
            self.varmodel[:] = string_variables(data.domain)
            taxid = data_hints.get_hint(data, "taxid", default=self.taxid)

            if taxid in self.taxids:
                self.set_organism(self.taxids.index(taxid))

            self.use_attr_names = data_hints.get_hint(
                data, "genesinrows", default=self.use_attr_names
            )

            if not (0 <= self.gene_var_index < len(self.varmodel)):
                self.gene_var_index = len(self.varmodel) - 1

            self.openContext("", data)
            self.invalidate()
            self.commit()
        else:
            self.varmodel[:] = []
            self.send("Network", None)

    def set_source_db(self, dbindex):
        self.network_source = dbindex
        self.invalidate()

    def set_organism(self, index):
        self.current_taxid_index = index
        self.taxid = self.taxids[index]
        self.invalidate()

    def set_gene_var(self, index):
        self.gene_var_index = index
        self.invalidate()

    def query_genes(self):
        if self.use_attr_names:
            if self.data is not None:
                return [var.name for var in self.data.domain.attributes]
            else:
                return []
        elif self.gene_var_index >= 0:
            var = self.varmodel[self.gene_var_index]
            genes = [str(inst[var]) for inst in self.data
                     if not inst[var].isSpecial()]
            return list(unique(genes))
        else:
            return []

    def invalidate(self):
        self._invalidated = True

        if self.nettask is not None:
            self.nettask.finished.disconnect(self._on_result_ready)
            self.nettask.future().cancel()
            self.nettask = None

        if self.autocommit:
            QTimer.singleShot(10, self._maybe_commit)

    @Slot()
    def _maybe_commit(self):
        if self._invalidated:
            self.commit()

    @Slot()
    def advance(self):
        self.progressBarValue = (self.progressBarValue + 1) % 100

    @Slot(float)
    def set_progress(self, value):
        self.progressBarValue = value

    def commit(self):
        include_neighborhood = self.include_neighborhood
        query_genes = self.query_genes()
        source = SOURCES[self.network_source]
        if source.score_filter:
            min_score = self.min_score
            assert source.name == "STRING"
            min_score = min_score * 1000
        else:
            min_score = None

        taxid = self.taxid
        progress = methodinvoke(self, "advance")
        if self.geneinfo is None:
            self.geneinfo = self.executor.submit(
                fetch_ncbi_geneinfo, taxid, progress
            )

        geneinfo_f = self.geneinfo
        taxmap = source.tax_mapping
        db_taxid = taxmap.get(taxid, taxid)

        if db_taxid is None:
            raise ValueError("invalid taxid for this network")

        def fetch_network():
            geneinfo = geneinfo_f.result()
            ppidb = fetch_ppidb(source, db_taxid, progress)
            return get_gene_network(ppidb, geneinfo, db_taxid, query_genes,
                                    include_neighborhood=include_neighborhood,
                                    min_score=min_score,
                                    progress=methodinvoke(self, "set_progress", (float,)))

        self.nettask = Task(function=fetch_network)
        self.nettask.finished.connect(self._on_result_ready)

        self.executor.submit(self.nettask)

        self.setBlocking(True)
        self.setEnabled(False)
        self.progressBarInit()
        self._invalidated = False
        self._update_info()

    @Slot(object)
    def _on_result_ready(self,):
        self.progressBarFinished()
        self.setBlocking(False)
        self.setEnabled(True)
        net = self.nettask.result()
        self._update_info()
        self.send("Network", net)

    def _on_source_db_changed(self):
        source = SOURCES[self.network_source]
        self.score_spin.setEnabled(source.score_filter)
        self.invalidate()

    def _update_organism(self):
        self.taxid = self.taxids[self.current_taxid_index]
        if self.geneinfo is not None:
            self.geneinfo.cancel()
        self.geneinfo = None
        self.invalidate()

    def _update_query_genes(self):
        self.invalidate()

    def _update_info(self):
        if self.data is None:
            self.info.setText("No data on input\n")
        else:
            names = self.query_genes()
            lines = ["%i unique genes on input" % len(set(names))]
            if self.nettask is not None:
                if not self.nettask.future().done():
                    lines.append("Retrieving ...")
                else:
                    net = self.nettask.result()
                    lines.append("%i nodes %i edges" %
                                 (len(net.nodes()), len(net.edges())))
            else:
                lines.append("")

            self.info.setText("\n".join(lines))
Example #9
0
class OWGeneInfo(OWWidget):
    settingsList = ["organismIndex", "geneAttr", "useAttr", "autoCommit",
                    "taxid"]
    contextHandlers = {
        "": DomainContextHandler(
            "", ["organismIndex", "geneAttr", "useAttr", "useAltSource",
                 "taxid"]
        )
    }

    def __init__(self, parent=None, signalManager=None, name="Gene Info"):
        OWWidget.__init__(self, parent, signalManager, name)

        self.inputs = [("Examples", Orange.data.Table, self.setData)]
        self.outputs = [("Selected Examples", Orange.data.Table)]

        self.organismIndex = 0
        self.taxid = None
        self.geneAttr = 0
        self.useAttr = False
        self.autoCommit = False
        self.searchString = ""
        self.selectionChangedFlag = False
        self.useAltSource = 0
        self.loadSettings()

        self.__initialized = False
        self.initfuture = None
        self.itemsfuture = None

        self.infoLabel = OWGUI.widgetLabel(
            OWGUI.widgetBox(self.controlArea, "Info", addSpace=True),
            "Initializing\n"
        )

        self.organisms = None
        self.organismBox = OWGUI.widgetBox(
            self.controlArea, "Organism", addSpace=True)

        self.organismComboBox = OWGUI.comboBox(
            self.organismBox, self, "organismIndex",
            callback=self._onSelectedOrganismChanged,
            debuggingEnabled=0)

        # For now only support one alt source, with a checkbox
        # In the future this can be extended to multiple selections
        self.altSourceCheck = OWGUI.checkBox(self.organismBox, self,
                            "useAltSource", "Show information from dictyBase",
                            callback=self.onAltSourceChange,
#                            debuggingEnabled=0,
                            )
        self.altSourceCheck.hide()

        box = OWGUI.widgetBox(self.controlArea, "Gene names", addSpace=True)
        self.geneAttrComboBox = OWGUI.comboBox(
            box, self, "geneAttr",
            "Gene atttibute", callback=self.updateInfoItems
        )
        OWGUI.checkBox(box, self, "useAttr", "Use attribute names",
                       callback=self.updateInfoItems,
                       disables=[(-1, self.geneAttrComboBox)])

        self.geneAttrComboBox.setDisabled(bool(self.useAttr))

        box = OWGUI.widgetBox(self.controlArea, "Commit", addSpace=True)
        b = OWGUI.button(box, self, "Commit", callback=self.commit)
        c = OWGUI.checkBox(box, self, "autoCommit", "Commit on change")
        OWGUI.setStopper(self, b, c, "selectionChangedFlag",
                         callback=self.commit)

        # A label for dictyExpress link
        self.dictyExpressBox = OWGUI.widgetBox(
            self.controlArea, "Dicty Express")
        self.linkLabel = OWGUI.widgetLabel(self.dictyExpressBox, "")
        self.linkLabel.setOpenExternalLinks(False)
        self.connect(self.linkLabel, SIGNAL("linkActivated(QString)"),
                     self.onDictyExpressLink)
        self.dictyExpressBox.hide()

        OWGUI.rubber(self.controlArea)

        OWGUI.lineEdit(self.mainArea, self, "searchString", "Filter",
                       callbackOnType=True, callback=self.searchUpdate)

        self.treeWidget = QTreeView(self.mainArea)
        self.treeWidget.setRootIsDecorated(False)
        self.treeWidget.setSelectionMode(
            QAbstractItemView.ExtendedSelection)
        self.treeWidget.setItemDelegate(
            LinkStyledItemDelegate(self.treeWidget))
        self.treeWidget.setUniformRowHeights(True)
        self.treeWidget.viewport().setMouseTracking(True)
        self.treeWidget.setSortingEnabled(True)
        self.mainArea.layout().addWidget(self.treeWidget)

        box = OWGUI.widgetBox(self.mainArea, "",
                              orientation="horizontal")
        OWGUI.button(box, self, "Select Filtered",
                     callback=self.selectFiltered)
        OWGUI.button(box, self, "Clear Selection",
                     callback=self.treeWidget.clearSelection)

        self.resize(1000, 700)

        self.geneinfo = []
        self.cells = []
        self.row2geneinfo = {}
        self.data = None

        # : (# input genes, # matches genes)
        self.matchedInfo = 0, 0
        self.selectionUpdateInProgress = False

        self.setBlocking(True)
        self.executor = ThreadExecutor(self)

        self.progressBarInit()

        task = Task(
            function=partial(
                taxonomy.ensure_downloaded,
                callback=methodinvoke(self, "advance", ())
            )
        )

        task.resultReady.connect(self.initialize)
        task.exceptionReady.connect(self._onInitializeError)

        self.initfuture = self.executor.submit(task)

    @Slot()
    def advance(self):
        assert self.thread() is QThread.currentThread()

        self.progressBarSet(self.progressBarValue + 1,
                            processEventsFlags=None)

    def initialize(self):
        if self.__initialized:
            # Already initialized
            return

        self.progressBarFinished()

        self.organisms = sorted(
            set([name.split(".")[-2] for name in
                 serverfiles.listfiles("NCBI_geneinfo")] +
                gene.NCBIGeneInfo.essential_taxids())
        )

        self.organismComboBox.addItems(
            [taxonomy.name(tax_id) for tax_id in self.organisms]
        )
        if self.taxid in self.organisms:
            self.organismIndex = self.organisms.index(self.taxid)

        self.infoLabel.setText("No data on input\n")
        self.__initialized = True
        self.initfuture = None

        self.setBlocking(False)

    def _onInitializeError(self, exc):
        sys.excepthook(type(exc), exc.args, None)
        self.error(0, "Could not download the necessary files.")

    def _onSelectedOrganismChanged(self):
        self.taxid = self.organisms[self.organismIndex]
        if self.data is not None:
            self.updateInfoItems()

    def setData(self, data=None):
        if not self.__initialized:
            self.initfuture.result()
            self.initialize()

        if self.itemsfuture is not None:
            raise Exception("Already processing")

        self.closeContext()
        self.data = data

        if data:
            self.geneAttrComboBox.clear()
            self.attributes = \
                [attr for attr in (data.domain.variables +
                                   data.domain.getmetas().values())
                 if isinstance(attr, (Orange.feature.String,
                                      Orange.feature.Discrete))]

            self.geneAttrComboBox.addItems(
                [attr.name for attr in self.attributes]
            )

            self.taxid = data_hints.get_hint(self.data, "taxid", self.taxid)
            self.useAttr = data_hints.get_hint(
                self.data, "genesinrows", self.useAttr)

            self.openContext("", data)
            self.geneAttr = min(self.geneAttr, len(self.attributes) - 1)

            if self.taxid in self.organisms:
                self.organismIndex = self.organisms.index(self.taxid)

            self.updateInfoItems()
        else:
            self.clear()

    def infoSource(self):
        """ Return the current selected info source getter function from
        INFO_SOURCES
        """
        org = self.organisms[min(self.organismIndex, len(self.organisms) - 1)]
        if org not in INFO_SOURCES:
            org = "default"
        sources = INFO_SOURCES[org]
        name, func = sources[min(self.useAltSource, len(sources) - 1)]
        return name, func

    def inputGenes(self):
        if self.useAttr:
            genes = [attr.name for attr in self.data.domain.attributes]
        elif self.attributes:
            attr = self.attributes[self.geneAttr]
            genes = [str(ex[attr]) for ex in self.data
                     if not ex[attr].isSpecial()]
        else:
            genes = []
        return genes

    def updateInfoItems(self):
        self.warning(0)
        if not self.data:
            return

        genes = self.inputGenes()
        if self.useAttr:
            genes = [attr.name for attr in self.data.domain.attributes]
        elif self.attributes:
            attr = self.attributes[self.geneAttr]
            genes = [str(ex[attr]) for ex in self.data
                     if not ex[attr].isSpecial()]
        else:
            genes = []
        if not genes:
            self.warning(0, "Could not extract genes from input dataset.")

        self.warning(1)
        org = self.organisms[min(self.organismIndex, len(self.organisms) - 1)]
        source_name, info_getter = self.infoSource()

        self.error(0)

        self.updateDictyExpressLink(genes, show=org == "352472")
        self.altSourceCheck.setVisible(org == "352472")

        self.progressBarInit()
        self.setBlocking(True)
        self.setEnabled(False)
        self.infoLabel.setText("Retrieving info records.\n")

        self.genes = genes

        task = Task(
            function=partial(
                info_getter, org, genes,
                advance=methodinvoke(self, "advance", ()))
        )
        self.itemsfuture = self.executor.submit(task)
        task.finished.connect(self._onItemsCompleted)

    def _onItemsCompleted(self):
        self.setBlocking(False)
        self.progressBarFinished()
        self.setEnabled(True)

        try:
            schema, geneinfo = self.itemsfuture.result()
        finally:
            self.itemsfuture = None

        self.geneinfo = geneinfo = list(zip(self.genes, geneinfo))
        self.cells = cells = []
        self.row2geneinfo = {}
        links = []
        for i, (_, gi) in enumerate(geneinfo):
            if gi:
                row = []
                for _, item in zip(schema, gi):
                    if isinstance(item, Link):
                        # TODO: This should be handled by delegates
                        row.append(item.text)
                        links.append(item.link)
                    else:
                        row.append(item)
                cells.append(row)
                self.row2geneinfo[len(cells) - 1] = i

        model = TreeModel(cells, [str(col) for col in schema], None)

        model.setColumnLinks(0, links)
        proxyModel = QSortFilterProxyModel(self)
        proxyModel.setSourceModel(model)
        self.treeWidget.setModel(proxyModel)
        self.connect(self.treeWidget.selectionModel(),
                     SIGNAL("selectionChanged(QItemSelection , QItemSelection )"),
                     self.commitIf)

        for i in range(7):
            self.treeWidget.resizeColumnToContents(i)
            self.treeWidget.setColumnWidth(
                i, min(self.treeWidget.columnWidth(i), 200)
            )

        self.infoLabel.setText("%i genes\n%i matched NCBI's IDs" %
                               (len(self.genes), len(cells)))
        self.matchedInfo = len(self.genes), len(cells)

    def clear(self):
        self.infoLabel.setText("No data on input\n")
        self.treeWidget.setModel(
            TreeModel([], ["NCBI ID", "Symbol", "Locus Tag",
                           "Chromosome", "Description", "Synonyms",
                           "Nomenclature"], self.treeWidget))

        self.geneAttrComboBox.clear()
        self.send("Selected Examples", None)

    def commitIf(self, *args):
        if self.autoCommit and not self.selectionUpdateInProgress:
            self.commit()
        else:
            self.selectionChangedFlag = True

    def commit(self):
        if not self.data:
            return
        model = self.treeWidget.model()
        mapToSource = model.mapToSource
        selectedRows = self.treeWidget.selectedIndexes()
        selectedRows = [mapToSource(index).row() for index in selectedRows]
        model = model.sourceModel()

        selectedGeneids = [self.row2geneinfo[row] for row in selectedRows]
        selectedIds = [self.geneinfo[i][0] for i in selectedGeneids]
        selectedIds = set(selectedIds)
        gene2row = dict((self.geneinfo[self.row2geneinfo[row]][0], row)
                        for row in selectedRows)

        if self.useAttr:
            def is_selected(attr):
                return attr.name in selectedIds
            attrs = [attr for attr in self.data.domain.attributes
                     if is_selected(attr)]
            domain = Orange.data.Domain(attrs, self.data.domain.classVar)
            domain.addmetas(self.data.domain.getmetas())
            newdata = Orange.data.Table(domain, self.data)
            self.send("Selected Examples", newdata)
        elif self.attributes:
            attr = self.attributes[self.geneAttr]
            examples = [ex for ex in self.data if str(ex[attr]) in selectedIds]
            # Add gene info
            domain = Orange.data.Domain(
                self.data.domain, self.data.domain.classVar)
            domain.addmetas(self.data.domain.getmetas())
            n_columns = model.columnCount()

            headers = [str(model.headerData(i, Qt.Horizontal, Qt.DisplayRole)
                           .toString())
                       for i in range(n_columns)]
            new_meta_attrs = [(Orange.feature.Descriptor.new_meta_id(),
                               Orange.feature.String(name))
                              for name in headers]
            domain.addmetas(dict(new_meta_attrs))
            examples = [Orange.data.Instance(domain, ex) for ex in examples]
            for ex in examples:
                for i, (_, meta) in enumerate(new_meta_attrs):
                    index = model.index(gene2row[str(ex[attr])], i)
                    ex[meta] = str(
                        model.data(index, Qt.DisplayRole).toString()
                    )

            if examples:
                newdata = Orange.data.Table(examples)
            else:
                newdata = None
            self.send("Selected Examples", newdata)
        else:
            self.send("Selected Examples", None)

    def rowFiltered(self, row):
        searchStrings = self.searchString.lower().split()
        row = unicode(" ".join(self.cells[row]).lower(), errors="ignore")
        return not all([s in row for s in searchStrings])

    def searchUpdate(self):
        if not self.data:
            return
        searchStrings = self.searchString.lower().split()
        index = self.treeWidget.model().sourceModel().index
        mapFromSource = self.treeWidget.model().mapFromSource
        for i, row in enumerate(self.cells):
            row = unicode(" ".join(row).lower(), errors="ignore")
            self.treeWidget.setRowHidden(
                mapFromSource(index(i, 0)).row(),
                QModelIndex(),
                not all([s in row for s in searchStrings]))

    def selectFiltered(self):
        if not self.data:
            return
        itemSelection = QItemSelection()

        index = self.treeWidget.model().sourceModel().index
        mapFromSource = self.treeWidget.model().mapFromSource
        for i, row in enumerate(self.cells):
            if not self.rowFiltered(i):
                itemSelection.select(mapFromSource(index(i, 0)),
                                     mapFromSource(index(i, 0)))
        self.treeWidget.selectionModel().select(
            itemSelection,
            QItemSelectionModel.Select | QItemSelectionModel.Rows)

    def sendReport(self):
        from Orange.OrangeWidgets import OWReport
        genes, matched = self.matchedInfo

        if self.organisms:
            org = self.organisms[min(self.organismIndex,
                                     len(self.organisms) - 1)]
            org_name = taxonomy.name(org)
        else:
            org = None
            org_name = None
        if self.data is not None:
            self.reportRaw(
                "<p>Input: %i genes of which %i (%.1f%%) matched NCBI synonyms"
                "<br>"
                "Organism: %s"
                "<br>"
                "Filter: %s"
                "</p>" % (genes, matched, 100.0 * matched / genes, org_name,
                          self.searchString)
            )
            self.reportSubsection("Gene list")
            self.reportRaw(reportItemView(self.treeWidget))
        else:
            self.reportRaw("<p>No input</p>")

    def updateDictyExpressLink(self, genes, show=False):
        def fix(ddb):
            if ddb.startswith("DDB"):
                if not ddb.startswith("DDB_G"):
                    ddb = ddb.replace("DDB", "DDB_G")
                return ddb
            return None
        if show:
            genes = [fix(gene) for gene in genes if fix(gene)]
            link1 = '<a href="http://dictyexpress.biolab.si/run/index.php?gene=%s">Microarray profile</a>'
            link2 = '<a href="http://dictyexpress.biolab.si/run/index.php?gene=%s&db=rnaseq">RNA-Seq profile</a>'
            self.linkLabel.setText(link1 + "<br/>" + link2)

            show = any(genes)

        if show:
            self.dictyExpressBox.show()
        else:
            self.dictyExpressBox.hide()

    def onDictyExpressLink(self, link):
        if not self.data:
            return

        selectedIndexes = self.treeWidget.selectedIndexes()
        if not len(selectedIndexes):
            QMessageBox.information(
                self, "No gene ids selected",
                "Please select some genes and try again."
            )
            return
        model = self.treeWidget.model()
        mapToSource = model.mapToSource
        selectedRows = self.treeWidget.selectedIndexes()
        selectedRows = [mapToSource(index).row() for index in selectedRows]
        model = model.sourceModel()

        selectedGeneids = [self.row2geneinfo[row] for row in selectedRows]
        selectedIds = [self.geneinfo[i][0] for i in selectedGeneids]
        selectedIds = set(selectedIds)

        def fix(ddb):
            if ddb.startswith("DDB"):
                if not ddb.startswith("DDB_G"):
                    ddb = ddb.replace("DDB", "DDB_G")
                return ddb
            return None

        genes = [fix(gene) for gene in selectedIds if fix(gene)]
        url = str(link) % " ".join(genes)
        QDesktopServices.openUrl(QUrl(url))

    def onAltSourceChange(self):
        self.updateInfoItems()

    def onDeleteWidget(self):
        OWWidget.onDeleteWidget(self)

        # try to cancel pending tasks
        if self.initfuture:
            self.initfuture.cancel()
        if self.itemsfuture:
            self.itemsfuture.cancel()

        self.executor.shutdown()
class OWKEGGPathwayBrowser(OWWidget):
    settingsList = ["organismIndex", "geneAttrIndex", "autoCommit",
                    "autoResize", "useReference", "useAttrNames",
                    "showOrthology"]

    contextHandlers = {
        "": DomainContextHandler(
            "",
            [ContextField("organismIndex",
                          DomainContextHandler.Required +
                          DomainContextHandler.IncludeMetaAttributes),
             ContextField("geneAttrIndex",
                          DomainContextHandler.Required +
                          DomainContextHandler.IncludeMetaAttributes),
             ContextField("useAttrNames",
                          DomainContextHandler.Required +
                          DomainContextHandler.IncludeMetaAttributes)]
        )
    }

    def __init__(self, parent=None, signalManager=None, name="KEGG Pathways"):
        OWWidget.__init__(self, parent, signalManager, name, wantGraph=True)
        self.inputs = [("Examples", Orange.data.Table, self.SetData),
                       ("Reference", Orange.data.Table, self.SetRefData)]
        self.outputs = [("Selected Examples", Orange.data.Table),
                        ("Unselected Examples", Orange.data.Table)]
        self.organismIndex = 0
        self.geneAttrIndex = 0
        self.autoCommit = False
        self.autoResize = True
        self.useReference = False
        self.useAttrNames = 0
        self.showOrthology = True

        self.loadSettings()

        self.organismCodes = []
        self._changedFlag = False

        self.controlArea.setMaximumWidth(250)
        box = OWGUI.widgetBox(self.controlArea, "Info")
        self.infoLabel = OWGUI.widgetLabel(box, "No data on input\n")

        # Organism selection.
        box = OWGUI.widgetBox(self.controlArea, "Organism")
        self.organismComboBox = OWGUI.comboBox(
            box, self, "organismIndex",
            items=[],
            callback=self.Update,
            addSpace=True,
            debuggingEnabled=0,
            tooltip="Select the organism of the input genes")

        # Selection of genes attribute
        box = OWGUI.widgetBox(self.controlArea, "Gene attribute")
        self.geneAttrCandidates = VariableListModel(parent=self)
        self.geneAttrCombo = OWGUI.comboBox(
            box, self, "geneAttrIndex", callback=self.Update)
        self.geneAttrCombo.setModel(self.geneAttrCandidates)

        OWGUI.checkBox(box, self, "useAttrNames",
                       "Use variable names",
                       disables=[(-1, self.geneAttrCombo)],
                       callback=self.Update)

        self.geneAttrCombo.setDisabled(bool(self.useAttrNames))

        OWGUI.separator(self.controlArea)

        OWGUI.checkBox(self.controlArea, self, "useReference",
                       "From signal",
                       box="Reference",
                       callback=self.Update)

        OWGUI.separator(self.controlArea)

        OWGUI.checkBox(self.controlArea, self, "showOrthology",
                       "Show pathways in full orthology",
                       box="Orthology",
                       callback=self.UpdateListView)

        OWGUI.checkBox(self.controlArea, self, "autoResize",
                       "Resize to fit",
                       box="Image",
                       callback=self.UpdatePathwayViewTransform)

        box = OWGUI.widgetBox(self.controlArea, "Cache Control")

        OWGUI.button(box, self, "Clear cache",
                     callback=self.ClearCache,
                     tooltip="Clear all locally cached KEGG data.")

        OWGUI.separator(self.controlArea)

        box = OWGUI.widgetBox(self.controlArea, "Selection")
        cb = OWGUI.checkBox(box, self, "autoCommit", "Commit on update")
        button = OWGUI.button(box, self, "Commit", callback=self.Commit,
                              default=True)
        OWGUI.setStopper(self, button, cb, "_changedFlag", self.Commit)

        OWGUI.rubber(self.controlArea)

        spliter = QSplitter(Qt.Vertical, self.mainArea)
        self.pathwayView = PathwayView(self, spliter)
        self.pathwayView.scene().selectionChanged.connect(
            self._onSelectionChanged
        )
        self.mainArea.layout().addWidget(spliter)

        self.listView = QTreeWidget(spliter)
        spliter.addWidget(self.listView)

        self.listView.setAllColumnsShowFocus(1)
        self.listView.setColumnCount(4)
        self.listView.setHeaderLabels(["Pathway", "P value",
                                       "Genes", "Reference"])

        self.listView.setSelectionMode(QTreeWidget.SingleSelection)

        self.listView.setSortingEnabled(True)

        self.listView.setMaximumHeight(200)

        self.connect(self.listView,
                     SIGNAL("itemSelectionChanged()"),
                     self.UpdatePathwayView)

        self.connect(self.graphButton,
                     SIGNAL("clicked()"),
                     self.saveGraph)

        select = QAction(
            "Select All", self,
            shortcut=QKeySequence.SelectAll
        )
        select.triggered.connect(self.selectAll)
        self.addAction(select)

        self.data = None
        self.refData = None

        self.resize(800, 600)

        self.connect(self,
                     SIGNAL("widgetStateChanged(QString, int, QString)"),
                     self.onStateChange)

        self.has_new_data = False
        self.has_new_reference_set = False

        self._executor = ThreadExecutor()
        self.setEnabled(False)
        self.setBlocking(True)
        QTimer.singleShot(0, self._initialize)
        self.infoLabel.setText("Fetching organism definitions\n")

    def _initialize(self):
        # First try to import slumber to see if we can even use the
        # kegg module.
        try:
            import slumber
        except ImportError:
            QMessageBox.warning(self,
                "'slumber' library required.",
                '<p>Please install '
                '<a href="http://pypi.python.org/pypi/slumber">slumber</a> '
                'library to use KEGG Pathways widget.</p>'
            )
            self.infoLabel.setText(
                '<p>Please install '
                '<a href="http://pypi.python.org/pypi/slumber">slumber</a> '
                'library to use KEGG Pathways widget.</p>'
            )
            self.error(0, "Missing slumber/requests library")
            return

        progress = methodinvoke(self, "setProgress", (float,))

        def get_genome():
            """Return a KEGGGenome with the common org entries precached."""
            genome = kegg.KEGGGenome()

            essential = genome.essential_organisms()
            common = genome.common_organisms()
            # Remove duplicates of essential from common.
            # (essential + common list as defined here will be used in the
            # GUI.)
            common = [c for c in common if c not in essential]

            # TODO: Add option to specify additional organisms not
            # in the common list.

            keys = map(genome.org_code_to_entry_key, essential + common)

            genome.pre_cache(keys, progress_callback=progress)
            return (keys, genome)

        self._genomeTask = task = Task(function=get_genome)
        task.finished.connect(self._initializeOrganisms)

        self.progressBarInit()
        self._executor.submit(task)

    def _initializeOrganisms(self):
        self.progressBarFinished()
        try:
            keys, genome = self._genomeTask.result()
        except Exception as err:
            self.error(0, str(err))
            return

        entries = [genome[key] for key in keys]
        items = [entry.definition for entry in entries]
        codes = [entry.organism_code for entry in entries]

        self.organismCodes = codes
        self.organismComboBox.clear()
        self.organismComboBox.addItems(items)
        self.organismComboBox.setCurrentIndex(self.organismIndex)

        self.setEnabled(True)
        self.setBlocking(False)
        self.infoLabel.setText("No data on input\n")

    def Clear(self):
        """
        Clear the widget state.
        """
        self.queryGenes = []
        self.referenceGenes = []
        self.genes = {}
        self.uniqueGenesDict = {}
        self.revUniqueGenesDict = {}
        self.pathways = {}
        self.org = None
        self.geneAttrCandidates[:] = []

        self.infoLabel.setText("No data on input\n")
        self.listView.clear()
        self.pathwayView.SetPathway(None)

        self.send("Selected Examples", None)
        self.send("Unselected Examples", None)

    def SetData(self, data=None):
        self.closeContext()
        self.data = data
        self.warning(0)
        self.error(0)
        self.information(0)

        if data is not None:
            vars = data.domain.variables + data.domain.getmetas().values()
            vars = [var for var in vars
                    if isinstance(var, (Orange.feature.String,
                                        Orange.feature.Discrete))]
            self.geneAttrCandidates[:] = vars

            # Try to guess the gene name variable
            names_lower = [v.name.lower() for v in vars]
            scores = [(name == "gene", "gene" in name)
                      for name in names_lower]
            imax, _ = max(enumerate(scores), key=itemgetter(1))
            self.geneAttrIndex = imax

            taxid = data_hints.get_hint(data, "taxid", None)
            if taxid:
                try:
                    code = kegg.from_taxid(taxid)
                    self.organismIndex = self.organismCodes.index(code)
                except Exception as ex:
                    print ex, taxid

            self.useAttrNames = data_hints.get_hint(data, "genesinrows",
                                                    self.useAttrNames)

            self.openContext("", data)
        else:
            self.Clear()

        self.has_new_data = True

    def SetRefData(self, data=None):
        self.refData = data
        self.information(1)

        self.has_new_reference_set = True

    def handleNewSignals(self):
        if self.has_new_data or (self.has_new_reference_set and \
                                 self.useReference):
            self.Update()

            self.has_new_data = False
            self.has_new_reference_set = False

    def UpdateListView(self):
        self.bestPValueItem = None
        self.listView.clear()
        if not self.data:
            return

        allPathways = self.org.pathways()
        allRefPathways = kegg.pathways("map")

        items = []
        kegg_pathways = kegg.KEGGPathways()

        org_code = self.organismCodes[min(self.organismIndex,
                                          len(self.organismCodes) - 1)]

        if self.showOrthology:
            self.koOrthology = kegg.KEGGBrite("ko00001")
            self.listView.setRootIsDecorated(True)
            path_ids = set([s[-5:] for s in self.pathways.keys()])

            def _walkCollect(koEntry):
                num = koEntry.title[:5] if koEntry.title else None
                if num in path_ids:
                    return ([koEntry] +
                            reduce(lambda li, c: li + _walkCollect(c),
                                   [child for child in koEntry.entries],
                                   []))
                else:
                    c = reduce(lambda li, c: li + _walkCollect(c),
                               [child for child in koEntry.entries],
                               [])
                    return c + (c and [koEntry] or [])

            allClasses = reduce(lambda li1, li2: li1 + li2,
                                [_walkCollect(c) for c in self.koOrthology],
                                [])

            def _walkCreate(koEntry, lvItem):
                item = QTreeWidgetItem(lvItem)
                id = "path:" + org_code + koEntry.title[:5]

                if koEntry.title[:5] in path_ids:
                    p = kegg_pathways.get_entry(id)
                    if p is None:
                        # In case the genesets still have obsolete entries
                        name = koEntry.title
                    else:
                        name = p.name
                    genes, p_value, ref = self.pathways[id]
                    item.setText(0, name)
                    item.setText(1, "%.5f" % p_value)
                    item.setText(2, "%i of %i" % (len(genes), len(self.genes)))
                    item.setText(3, "%i of %i" % (ref, len(self.referenceGenes)))
                    item.pathway_id = id if p is not None else None
                else:
                    if id in allPathways:
                        text = kegg_pathways.get_entry(id).name
                    else:
                        text = koEntry.title
                    item.setText(0, text)

                    if id in allPathways:
                        item.pathway_id = id
                    elif "path:map" + koEntry.title[:5] in allRefPathways:
                        item.pathway_id = "path:map" + koEntry.title[:5]
                    else:
                        item.pathway_id = None

                for child in koEntry.entries:
                    if child in allClasses:
                        _walkCreate(child, item)

            for koEntry in self.koOrthology:
                if koEntry in allClasses:
                    _walkCreate(koEntry, self.listView)

            self.listView.update()
        else:
            self.listView.setRootIsDecorated(False)
            pathways = self.pathways.items()
            pathways.sort(lambda a, b: cmp(a[1][1], b[1][1]))

            for id, (genes, p_value, ref) in pathways:
                item = QTreeWidgetItem(self.listView)
                item.setText(0, kegg_pathways.get_entry(id).name)
                item.setText(1, "%.5f" % p_value)
                item.setText(2, "%i of %i" % (len(genes), len(self.genes)))
                item.setText(3, "%i of %i" % (ref, len(self.referenceGenes)))
                item.pathway_id = id
                items.append(item)

        self.bestPValueItem = items and items[0] or None
        self.listView.expandAll()
        for i in range(4):
            self.listView.resizeColumnToContents(i)

        if self.bestPValueItem:
            index = self.listView.indexFromItem(self.bestPValueItem)
            self.listView.selectionModel().select(
                index, QItemSelectionModel.ClearAndSelect
            )

    def UpdatePathwayView(self):
        items = self.listView.selectedItems()

        if len(items) > 0:
            item = items[0]
        else:
            item = None

        self.Commit()
        item = item or self.bestPValueItem
        if not item or not item.pathway_id:
            self.pathwayView.SetPathway(None)
            return

        def get_kgml_and_image(pathway_id):
            """Return an initialized KEGGPathway with pre-cached data"""
            p = kegg.KEGGPathway(pathway_id)
            p._get_kgml()  # makes sure the kgml file is downloaded
            p._get_image_filename()  # makes sure the image is downloaded
            return (pathway_id, p)

        self.setEnabled(False)
        self._pathwayTask = Task(
            function=lambda: get_kgml_and_image(item.pathway_id)
        )
        self._pathwayTask.finished.connect(self._onPathwayTaskFinshed)
        self._executor.submit(self._pathwayTask)

    def _onPathwayTaskFinshed(self):
        self.setEnabled(True)
        pathway_id, self.pathway = self._pathwayTask.result()
        self.pathwayView.SetPathway(
            self.pathway,
            self.pathways.get(pathway_id, [[]])[0]
        )

    def UpdatePathwayViewTransform(self):
        self.pathwayView.updateTransform()

    def Update(self):
        """
        Update (recompute enriched pathways) the widget state.
        """
        if not self.data:
            return

        self.error(0)
        self.information(0)

        # XXX: Check data in setData, do not even alow this to be executed if
        # data has no genes
        try:
            genes = self.GeneNamesFromData(self.data)
        except ValueError:
            self.error(0, "Cannot extract gene names from input.")
            genes = []

        if not self.useAttrNames and any("," in gene for gene in genes):
            genes = reduce(add, (split_and_strip(gene, ",")
                                 for gene in genes),
                           [])
            self.information(0,
                             "Separators detected in input gene names. "
                             "Assuming multiple genes per instance.")
        self.queryGenes = genes

        self.information(1)
        reference = None
        if self.useReference and self.refData:
            reference = self.GeneNamesFromData(self.refData)
            if not self.useAttrNames \
                    and any("," in gene for gene in reference):
                reference = reduce(add, (split_and_strip(gene, ",")
                                         for gene in reference),
                                   [])
                self.information(1,
                                 "Separators detected in reference gene "
                                 "names. Assuming multiple genes per "
                                 "instance.")

        org_code = self.SelectedOrganismCode()

        def run_enrichment(org_code, genes, reference=None, progress=None):
            org = kegg.KEGGOrganism(org_code)
            if reference is None:
                reference = org.get_genes()

            # Map 'genes' and 'reference' sets to unique KEGG identifiers
            unique_genes, _, _ = org.get_unique_gene_ids(set(genes))
            unique_ref_genes, _, _ = org.get_unique_gene_ids(set(reference))

            taxid = kegg.to_taxid(org.org_code)
            # Map the taxid back to standard 'common' taxids
            # (as used by 'geneset') if applicable
            r_tax_map = dict((v, k) for k, v in
                             kegg.KEGGGenome.TAXID_MAP.items())
            if taxid in r_tax_map:
                taxid = r_tax_map[taxid]

            # We use the kegg pathway gene sets provided by 'geneset' for
            # the enrichment calculation.

            # Ensure we are using the latest genesets
            # TODO: ?? Is updating the index enough?
            serverfiles.update(geneset.sfdomain, "index.pck")
            kegg_gs_collections = geneset.collections(
                (("KEGG", "pathways"), taxid)
            )

            pathways = pathway_enrichment(
                kegg_gs_collections, unique_genes.keys(),
                unique_ref_genes.keys(),
                callback=progress
            )
            # Ensure that pathway entries are pre-cached for later use in the
            # list/tree view
            kegg_pathways = kegg.KEGGPathways()
            kegg_pathways.pre_cache(
                pathways.keys(), progress_callback=progress
            )

            return pathways, org, unique_genes, unique_ref_genes

        self.progressBarInit()
        self.setEnabled(False)
        self.infoLabel.setText("Retrieving...\n")

        progress = methodinvoke(self, "setProgress", (float,))
        self._enrichTask = Task(
            function=lambda:
                run_enrichment(org_code, genes, reference, progress)
        )
        self._enrichTask.finished.connect(self._onEnrichTaskFinished)
        self._executor.submit(self._enrichTask)

    def _onEnrichTaskFinished(self):
        self.setEnabled(True)
        self.setBlocking(False)

        self.progressBarFinished()
        try:
            pathways, org, unique_genes, unique_ref_genes = \
                self._enrichTask.result()
        except Exception:
            raise

        self.org = org
        self.genes = unique_genes.keys()
        self.uniqueGenesDict = unique_genes
        self.revUniqueGenesDict = dict([(val, key) for key, val in
                                        self.uniqueGenesDict.items()])
        self.referenceGenes = unique_ref_genes.keys()
        self.pathways = pathways

        if not self.pathways:
            self.warning(0, "No enriched pathways found.")
        else:
            self.warning(0)

        count = len(set(self.queryGenes))
        self.infoLabel.setText(
            "%i unique gene names on input\n"
            "%i (%.1f%%) genes names matched" %
            (count, len(unique_genes),
             100.0 * len(unique_genes) / count if count else 0.0)
        )

        self.UpdateListView()

    @pyqtSlot(float)
    def setProgress(self, value):
        self.progressBarValue = value

    def GeneNamesFromData(self, data):
        """
        Extract and return gene names from `data`.
        """
        if self.useAttrNames:
            genes = [str(v.name).strip() for v in data.domain.attributes]
        elif self.geneAttrCandidates:
            index = min(self.geneAttrIndex, len(self.geneAttrCandidates) - 1)
            geneAttr = self.geneAttrCandidates[index]
            genes = [str(e[geneAttr]) for e in data
                     if not e[geneAttr].isSpecial()]
        else:
            raise ValueError("No gene names in data.")
        return genes

    def SelectedOrganismCode(self):
        """
        Return the selected organism code.
        """
        return self.organismCodes[min(self.organismIndex,
                                      len(self.organismCodes) - 1)]

    def selectAll(self):
        """
        Select all items in the pathway view.
        """
        changed = False
        scene = self.pathwayView.scene()
        with disconnected(scene.selectionChanged, self._onSelectionChanged):
            for item in scene.items():
                if item.flags() & QGraphicsItem.ItemIsSelectable and \
                        not item.isSelected():
                    item.setSelected(True)
                    changed = True
        if changed:
            self._onSelectionChanged()

    def _onSelectionChanged(self):
        # Item selection in the pathwayView/scene has changed
        if self.autoCommit:
            self.Commit()
        else:
            self._changedFlag = True

    def Commit(self):
        if self.data:
            selectedItems = self.pathwayView.scene().selectedItems()
            selectedGenes = reduce(set.union, [item.marked_objects
                                               for item in selectedItems],
                                   set())

            if self.useAttrNames:
                selectedVars = [self.data.domain[self.uniqueGenesDict[gene]]
                                for gene in selectedGenes]
                newDomain = Orange.data.Domain(selectedVars, 0)
                data = Orange.data.Table(newDomain, self.data)
                self.send("Selected Examples", data)
            elif self.geneAttrCandidates:
                geneAttr = self.geneAttrCandidates[min(self.geneAttrIndex,
                                                       len(self.geneAttrCandidates) - 1)]
                selectedExamples = []
                otherExamples = []
                for ex in self.data:
                    names = [self.revUniqueGenesDict.get(name, None)
                             for name in split_and_strip(str(ex[geneAttr]), ",")]
                    if any(name and name in selectedGenes for name in names):
                        selectedExamples.append(ex)
                    else:
                        otherExamples.append(ex)

                if selectedExamples:
                    selectedExamples = Orange.data.Table(selectedExamples)
                else:
                    selectedExamples = None

                if otherExamples:
                    otherExamples = Orange.data.Table(otherExamples)
                else:
                    otherExamples = None

                self.send("Selected Examples", selectedExamples)
                self.send("Unselected Examples", otherExamples)
        else:
            self.send("Selected Examples", None)
            self.send("Unselected Examples", None)

    def ClearCache(self):
        from ..kegg import caching
        try:
            caching.clear_cache()
        except Exception, ex:
            QMessageBox.warning(self, "Cache clear", ex.args[0])