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 * " " + "<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()
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 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 * " " + "<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()
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()
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))
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))
