def computeSimilarity(dataArray):
start = time()
sampleIDList = dataArray[8]
n = len(sampleIDList)
matrix = np.zeros((n,n))
for i in range(n):
for j in range(i,n):
#Total ratio computation
common,in1,in2,_,_,_,_,_ = compute(dataArray[7],[sampleIDList[i]],[sampleIDList[j]])
commonA = countAssignmentsInCommon(common,[sampleIDList[i]],[sampleIDList[j]])
numberA1 = countAssignments(in1,[sampleIDList[i]])
numberA2 = countAssignments(in2,[sampleIDList[j]])
tratio = totalRatio(commonA,numberA1,numberA2)
#Pattern ratio computation
commonPatternsList = enumerateCommonPatterns(dataArray[7],[sampleIDList[i]],[sampleIDList[j]])
specificPatternsList1 = enumerateSpecificPatterns(dataArray[7],[sampleIDList[i]],[sampleIDList[j]])
specificPatternsList2 = enumerateSpecificPatterns(dataArray[7],[sampleIDList[i]],[sampleIDList[j]])
pRatio = patternRatio(commonPatternsList,specificPatternsList1,specificPatternsList2)
#Diversity coefficient
dRatio1,_ = computeDiversityCoefficient(dataArray[5],[sampleIDList[i]],dataArray)
dRatio2,_ = computeDiversityCoefficient(dataArray[5],[sampleIDList[j]],dataArray)
subdRatio = abs(dRatio1 - dRatio2)
if subdRatio:
s = sumOpInf(pRatio,tratio) - subdRatio
else:
s = sumOpInf(pRatio,tratio)
matrix[i][j] = s
matrix[j][i] = s
end = time()
print "TIME:",(end-start)
return matrix
def totalDiffRatioAct(dataArray):
print "First list of samples."
sampleNameList1,metadataList1,interval1List1,interval2List1 = createSampleNameList(dataArray)
print "Second list of samples."
sampleNameList2,metadataList2,interval1List2,interval2List2 = createSampleNameList(dataArray)
common,in1,in2,numberA1,numberA2,_,_,_ = compute(dataArray[7],sampleNameList1,sampleNameList2)
commonA = countAssignmentsInCommon(common,sampleNameList1,sampleNameList2)
tratio = totalRatio(commonA,numberA1,numberA2)
ntRatio = totalRatioNormalized(commonA,numberA1,numberA2)
dratio = diffRatio(commonA)
ndRatio = diffRatioNormalized(commonA,numberA1,numberA2)
print "\nTotal Ratio Distance is: " + str(tratio)
print "normalized Total Ratio is: " + str(ntRatio) + "\n[The more it is close to 1, the more the two groups are alike]\n"
print "Diff Ratio Distance is: " + str(dratio)
print "normalized Diff Ratio is: " + str(ndRatio) + "\n[The more it is close to 0, the more the two groups are alike]\n"
print "[If you have obtained +inf (resp. -inf), it could mean you have selected no sample.]\n"
answer = raw_input("Save the results? Y/N\n")
if (answer == "Y"):
data = "Total Ratio Results ****\n for " + str(sampleNameList1) + "\n"
if metadataList1:
data += "selected on metadata: " + str(metadataList1) + " with extremum values: " + str(interval1List1) + " (lower bounds) and " + str(interval2List1) + " (upper bounds) \n"
data += " and " + str(sampleNameList2) + "\n"
if metadataList2:
data += "selected on metadata: " + str(metadataList2) + " with extremum values: " + str(interval1List2) + " (lower bounds) and " + str(interval2List2) + " (upper bounds) \n"
data += "\nTotal Ratio Distance is: " + str(tratio) + "\n normalized Total Ratio is: " + str(ntRatio) + "\nDiff Ratio Distance is: " + str(dratio) + "\n normalized Diff Ratio is: " + str(ndRatio) +"\n\nEND OF FILE ****"
writeFile(data,"","text")
elif not (answer == "N"):
print "/!\ You should answer 'Y' or 'N'!"