m1 = []
m2 = []
percentTotalVariation = []
for line in file(inFilePath):
lineItems = line.rstrip().split("\t")
pProbe = float(lineItems[2])
m1Probe = float(lineItems[4])
m2Probe = float(lineItems[5])
normalized.append(float(lineItems[1]))
probCodes.append(pProbe)
raw.append(float(lineItems[3]))
m1.append(m1Probe)
m2.append(m2Probe)
percentTotalVariation.append(((1 - pProbe) * m1Probe) + (pProbe * m2Probe))
meanProbCode = utilities.calculateMean(probCodes)
r2BackgroundActual = math.pow(utilities.calculateCorrelationCoefficient(raw, m1), 2)
r2PercentTotalVariation = math.pow(utilities.calculateCorrelationCoefficient(raw, percentTotalVariation), 2)
percentVariationBackground = r2BackgroundActual / r2PercentTotalVariation
preSignalNoise = (r2PercentTotalVariation - r2BackgroundActual) / r2BackgroundActual
postSignalNoise = preSignalNoise / (1 - r2PercentTotalVariation)
postPreSignalNoise = postSignalNoise / preSignalNoise
outFile.write("\t".join([str(x) for x in [fileName, meanProbCode, r2BackgroundActual, r2PercentTotalVariation, percentVariationBackground, preSignalNoise, postSignalNoise, postPreSignalNoise]]) + "\n")
outFile.flush()
outFile.close()
def standardScore(x):
mean = utilities.calculateMean([float(y) for y in x])
sd = utilities.calculateStandardDeviation([float(y) for y in x])
return [(float(y) - mean) / sd for y in x]
def mean(x):
return [utilities.calculateMean([float(y) for y in x])]
def calculateMean(values):
return utilities.calculateMean(values)
uniqueSamples = set([x[sampleIndex] for x in data])
print "%i samples" % len(uniqueSamples)
uniqueGenes = set([x[geneIndex] for x in data])
print "%i genes with at least one variant" % len(uniqueGenes)
uniquePathways = set()
for row in data:
pathways = row[pathwaysIndex].split(",")
for pathway in pathways:
if pathway != "":
uniquePathways.add(pathway)
print "%i pathways with at least one variant" % len(uniquePathways)
numGenesMutatedPerSample = [float(len(set([x[geneIndex] for x in data if x[sampleIndex]==sample]))) for sample in uniqueSamples]
print "%.1f genes with variant per sample" % utilities.calculateMean(numGenesMutatedPerSample)
numSamplesMutatedPerGene = [float(len(set([x[sampleIndex] for x in data if x[geneIndex]==gene]))) for gene in uniqueGenes]
print "%.1f samples with variant per gene" % utilities.calculateMean(numSamplesMutatedPerGene)
numMutatedPerGenePerSample = []
for x in sampleGeneVariantDict.keys():
for geneID in sampleGeneVariantDict[x]:
numMutatedPerGenePerSample.append(float(len(sampleGeneVariantDict[x][geneID])))
print "%.3f variants per gene when sample had at least one variant in gene" % utilities.calculateMean(numMutatedPerGenePerSample)
numMutatedPerGenePerSample = [float(len(set([x[geneIndex] for x in data if x[sampleIndex]==sample]))) for sample in uniqueSamples]
print "%.1f variants per genes with variant per sample" % utilities.calculateMean(numGenesMutatedPerSample)
numSamplesPerPathway = []
for pathway in uniquePathways:
def calculateMean(values):
return utilities.calculateMean(values)
print sampleID
sampleDataDict = {}
sampleData = getData(sampleSection, "ID_REF")
sampleData.pop(0)
for row in sampleData:
geneID = probeGeneDict[row[0]]
value = row[1]
if geneID != '':
sampleDataDict[geneID] = sampleDataDict.setdefault(geneID, []) + [value]
for geneID in sampleDataDict.keys():
values = [float(x) for x in sampleDataDict[geneID]]
sampleDataDict[geneID] = utilities.calculateMean(values)
allSamplesDataDict[sampleID] = sampleDataDict
sampleIDs = sorted(allSamplesDataDict.keys())
geneIDs = sorted(allSamplesDataDict[sampleIDs[0]])
outData = []
outData.append(["Description"] + sampleIDs)
for geneID in geneIDs:
outData.append([geneID] + [str(allSamplesDataDict[sampleID][geneID]) for sampleID in sampleIDs])
utilities.writeMatrixToFile(outData, outFilePath)
inFile = open(inFilePath)
header = None
if hasHeader:
header = inFile.readline().rstrip()
for line in inFile:
lineItems = line.rstrip().split("\t")
meta = lineItems[0]
values = [float(x) for x in lineItems[1:]]
outDict[meta] = outDict.setdefault(meta, []) + [values]
inFile.close()
outData = []
if header != None:
outData.append([header])
for meta in outDict.keys():
outValues = [meta]
for i in range(len(outDict[meta][0])):
iValues = [x[i] for x in outDict[meta]]
outValues.append(utilities.calculateMean(iValues))
outData.append(outValues)
utilities.writeMatrixToFile(outData, outFilePath)
sampleDataDict = {}
sampleData = getData(sampleSection, "ID_REF")
sampleData.pop(0)
for row in sampleData:
geneID = probeGeneDict[row[0]]
value = row[1]
if geneID != '':
sampleDataDict[geneID] = sampleDataDict.setdefault(geneID,
[]) + [value]
for geneID in sampleDataDict.keys():
values = [float(x) for x in sampleDataDict[geneID]]
sampleDataDict[geneID] = utilities.calculateMean(values)
allSamplesDataDict[sampleID] = sampleDataDict
sampleIDs = sorted(allSamplesDataDict.keys())
geneIDs = sorted(allSamplesDataDict[sampleIDs[0]])
outData = []
outData.append(["Description"] + sampleIDs)
for geneID in geneIDs:
outData.append(
[geneID] +
[str(allSamplesDataDict[sampleID][geneID]) for sampleID in sampleIDs])
utilities.writeMatrixToFile(outData, outFilePath)
def averageHighestGenes(values):
median = utilities.calculateMedian(values)
return utilities.calculateMean([x for x in values if x > median])