def Commit(self):
if not self.dbc:
self.Connect()
allTables = []
import time
start = time.time()
pb = OWGUI.ProgressBar(self, iterations=1000)
table = None
ids = []
for item in self.experimentsWidget.selectedItems():
ids += str(item.text(5)).split(",")
table = self.dbc.get_single_data(
ids=ids,
callback=pb.advance,
exclude_constant_labels=self.excludeconstant)
end = int(time.time() - start)
pb.finish()
#self.send("Example table", None)
# table.taxid = "352472"
# table.genesinrows = False
from Orange.orng.orngDataCaching import data_hints
data_hints.set_hint(table, "taxid", "352472", 10.0)
data_hints.set_hint(table, "genesinrows", False, 10.0)
self.send("Example table", table)
def chipdata(self, data):
self.data = []
if data:
self.infob.setText("")
numFiles = reduce(lambda a, b: a + len(b[1]), data, 0)
lenSD = len(data)
self.infoa.setText("%d set%s, total of %d data file%s." %
(lenSD, ["", "s"][lenSD != 1], numFiles,
["", "s"][numFiles != 1]))
numExamplesList = []
# construct a list of ExampleTable lengths and a list of attribute names
for (name, etList) in data:
for et in etList:
setattr(et, "dirname", name)
setattr(et, "strain", name)
self.data.append(et)
numExamplesList.append(len(et))
if len(self.data) > 1:
# test that files contain the same attributes and equal number of examples
attrSorted = self.data[0].domain.attributes
attrSorted.sort()
numEx = len(self.data[0])
for et in self.data[1:]:
attrSorted2 = et.domain.attributes
attrSorted2.sort()
if map(lambda x: x.name, attrSorted) != map(
lambda x: x.name, attrSorted2):
self.data = []
self.infob.setText(
"Error: data files contain different attributes, aborting distance computation."
)
return
if len(et) != numEx:
self.data = []
self.infob.setText(
"Error: data files contain unequal number of examples, aborting distance computation."
)
return
# compute distances
pb = OWGUI.ProgressBar(self, iterations=len(self.data))
self.computeMatrix()
pb.finish()
else:
self.data = []
self.infob.setText(
'Error: not enough data, aborting distance computation.')
else:
self.infoa.setText('No data on input.')
def update_distances(self, base_indices=()):
"""Recompute the experiment distances.
"""
distance = self.selected_distance()
if base_indices == ():
base_group_index = self.selected_base_group_index()
base_indices = [ind[base_group_index] \
for _, ind in self.groups]
assert (len(base_indices) == len(self.groups))
base_distances = []
attributes = self.data.domain.attributes
pb = OWGUI.ProgressBar(self, len(self.groups) * \
len(attributes))
cached_distances, filled_set = self.get_cached_distances(distance)
for (group, indices), base_index in zip(self.groups, base_indices):
# Base column of the group
if base_index is not None:
base_vec = exp.linearize(self.data, [base_index])
distances = []
# Compute the distances between base column
# and all the rest data columns.
for i in range(len(attributes)):
if i == base_index:
distances.append(0.0)
elif self.get_cached_distance(distance, i,
base_index) is not None:
distances.append(
self.get_cached_distance(distance, i, base_index))
else:
vec_i = exp.linearize(self.data, [i])
dist = distance(base_vec, vec_i)
self.store_distance(distance, i, base_index, dist)
distances.append(dist)
pb.advance()
base_distances.append(distances)
else:
base_distances.append(None)
pb.finish()
self.distances = base_distances
def compute(self, res=None, dm=None):
collectionNames = [ self.geneSel[a] for a in self.gridSel ]
organism = self.organismTaxids[self.organismIndex]
if self.gsgo:
collectionNames.append((("GO",),organism))
if self.gskegg:
collectionNames.append((("KEGG",),organism))
self.geneSets = obiGeneSets.collections(*collectionNames)
self.resultsOut(None)
qApp.processEvents()
self.res = res
self.dm = dm
clearListView(self.listView)
self.addComment("Computing...")
qApp.processEvents()
self.phenVar = self.phenCands[self.selectedPhenVar][0]
self.geneVar = self.geneCands[self.selectedGeneVar]
if self.res == None and self.data:
self.setSelMode(False)
pb = OWGUI.ProgressBar(self, iterations=self.perms+2)
if hasattr(self, "btnApply"):
self.btnApply.setFocus()
kwargs = {}
dkwargs = {}
dkwargs["phenVar"] = self.phenVar
dkwargs["geneVar"] = self.geneVar
if not obiGsea.already_have_correlations(self.data):
selectedClasses = self.psel.getSelection()
fc = "Phenotype group empty. Stopped."
if len(selectedClasses[0]) == 0:
self.addComment(fc)
return
elif len(selectedClasses[1]) == 0:
self.addComment(fc)
return
dkwargs["classValues"] = selectedClasses
dkwargs["atLeast"] = self.atLeast
permtype = self.permutationTypes[self.ptype][1]
kwargs["permutation"] = "class" if permtype == "p" else "genes"
def ifr(case, t, f):
if case: return t
else: return f
kwargs["minSize"] = \
ifr(self.minSubsetSizeC, self.minSubsetSize, 1)
kwargs["maxSize"] = \
ifr(self.maxSubsetSizeC, self.maxSubsetSize, 1000000)
kwargs["minPart"] = \
ifr(self.minSubsetPartC, self.minSubsetPart/100.0, 0.0)
#create gene matcher
genematcher = obiGene.matcher([[obiGene.GMKEGG(organism)] + ([obiGene.GMDicty()] if organism == "352472" else [])])
#dkwargs["caseSensitive"] = self.csgm
gso = obiGsea.GSEA(self.data, matcher=genematcher, **dkwargs)
for gs in self.geneSets:
gso.addGenesets([gs])
qApp.processEvents()
self.res = gso.compute(n=self.perms, callback=pb.advance, **kwargs)
pb.finish()
def update_scores(self):
""" Compute the scores and update the histogram.
"""
self.clear_plot()
self.error(0)
label, values = self.current_target_selection
if not self.data or label is None:
return
_, score_func, _, two_sample_test = self.score_methods[
self.method_index]
if two_sample_test:
target = self.targets
score_target = set(target)
ind1, ind2 = score_func(
self.data, self.genes_in_columns).test_indices(score_target)
if not len(ind1) or not len(ind2):
self.error(
0,
"Target labels most exclude/include at least one value.")
return
else: # ANOVA should use all labels.
target = dict(self.data_labels)[label]
if self.genes_in_columns:
target = [(label, t) for t in target]
score_target = target
# indices = score_func(self.data, self.genes_in_columns).test_indices(score_target)
# TODO: Check that each label has more than one measurement, raise warning otherwise.
pb = OWGUI.ProgressBar(
self, 4 + self.permutations_count if self.compute_null else 3)
self.scores = dict(
self.compute_scores(self.data,
score_func,
self.genes_in_columns,
score_target,
advance=pb.advance))
pb.advance()
if self.compute_null:
self.null_dist = self.compute_null_distribution(
self.data,
score_func,
self.genes_in_columns,
score_target,
self.permutations_count,
advance=pb.advance)
else:
self.null_dist = []
pb.advance()
htype = self.histType[self.score_methods[self.method_index][2]]
score_type = self.score_methods[self.method_index][0]
self.histogram.type = htype
if self.scores:
self.histogram.setValues(self.scores.values())
low, high = self.thresholds.get(score_type,
(float("-inf"), float("inf")))
minx, maxx = self.histogram.minx, self.histogram.maxx
low, high = max(low, minx), min(high, maxx)
if htype == "hiTail":
low = high
if htype == "lowTail":
high = low
self.histogram.setBoundary(low, high)
if self.compute_null and self.null_dist:
nullY, nullX = numpy.histogram(self.null_dist,
bins=self.histogram.xData)
nullY = nullY / self.permutations_count
self.histogram.nullCurve = self.histogram.addCurve(
"nullCurve",
Qt.black,
Qt.black,
6,
symbol=QwtSymbol.NoSymbol,
style=QwtPlotCurve.Steps,
xData=nullX,
yData=nullY)
minx = min(min(nullX), minx)
maxx = max(max(nullX), maxx)
miny = min(min(nullY), self.histogram.miny)
maxy = max(max(nullY), self.histogram.maxy)
spanx, spany = maxx - minx, maxy - miny
self.histogram.setAxisScale(QwtPlot.xBottom,
minx - 0.05 * spanx,
maxx + 0.05 * spanx)
self.histogram.setAxisScale(QwtPlot.yLeft, miny - 0.05 * spany,
maxy + 0.05 * spany)
state = dict(hiTail=(False, True),
lowTail=(True, False),
twoTail=(True, True))
for spin, visible in zip(
(self.upperBoundarySpin, self.lowerBoundarySpin),
state[self.histogram.type]):
spin.setVisible(visible)
# If this is a two sample test add markers to the left and right
# plot indicating which target group is over-expressed in that
# part
if self.method_index in [0, 2, 6]:
if self.method_index == 0: ## fold change is centered on 1.0
x1, y1 = (self.histogram.minx + 1) / 2, self.histogram.maxy
x2, y2 = (self.histogram.maxx + 1) / 2, self.histogram.maxy
else:
x1, y1 = (self.histogram.minx) / 2, self.histogram.maxy
x2, y2 = (self.histogram.maxx) / 2, self.histogram.maxy
if self.genes_in_columns:
label = target[0][0]
target_values = [t[1] for t in target]
values = dict(self.data_labels)[label]
else:
target_values = target
values = self.data_labels[0][1]
left = ", ".join(v for v in values if v not in target_values)
right = ", ".join(v for v in values if v in target_values)
self.histogram.addMarker(left, x1, y1)
self.histogram.addMarker(right, x2, y2)
self.warning(0)
else:
self.warning(0, "No scores obtained.")
self.histogram.replot()
pb.advance()
pb.finish()
self.update_data_info_label()
