def main():
(elutionF, refF, outD) = sys.argv[1:]
reference, elutionData, scoreCalc = calcS.loadData(refF, elutionF)
iexFractions = range(1, 49)
out = np.array([[-1.00] * 48] * 48)
for removeLeft in range(1, 49):
tmpFracs = copy.copy(iexFractions)
for i in range(1, removeLeft):
if i in tmpFracs:
tmpFracs.remove(i)
for removeRight in reversed(range(removeLeft + 1, 50)):
if removeRight in tmpFracs:
tmpFracs.remove(removeRight)
print tmpFracs
fractions = getIEXFracs(tmpFracs)
tmpElution = copy.copy(elutionData)
tmpElution.getSubset(fractions)
scoreCalc = calcS.CalculateCoElutionScores(tmpElution)
scoreCalc.calculateAllScores([calcS.Euclidiean()], reference)
data, targets = scoreCalc.toSklearnData()
clf = calcS.RandomForest(data, targets)
scores = clf.getValScores()
out[removeLeft - 1][49 - removeRight] = scores[1]
print "%i\t%i\t%.2f" % (removeLeft - 1, 49 - removeRight, scores[1])
outFH = open(outD + ".iex.dat", "w")
print >> outFH, "\t" + "\t".join(map(str, range(48)))
for i in range(48):
print >> outFH, "%i\t%s" % (i, "\t".join(map("{0:.2f}".format, out[i])))
outFH.close()
def main():
(scoreF, refF, elutionF, geneNameF, outF) = sys.argv[1:]
geneNameFH = open(geneNameF)
geneName= {}
species = {}
for line in geneNameFH:
line = line.rstrip()
ida, idb, spec = line.split("\t")
if ida not in geneName: geneName[ida] = set([])
geneName[ida].add(idb)
species[ida] = spec
species[idb] = spec
geneNameFH.close()
toLearn, toPred = calcS.loadScoreData(scoreF, refF)
rfc = calcS.trainML(toLearn)
print rfc.getValScores()
ref, eluD, calc = calcS.loadData(refF, elutionF)
calc.calculate2DScores(ref)
outFH = open(outF + ".arff", "w")
outFH.write(calc.toArffData())
outFH.close()
print "Calculated scores"
rfc2 = calcS.trainML(calc)
print rfc2.getValScores()
data, targets = toPred.toSklearnData()
dataL, targetsL = toLearn.toSklearnData()
preds = rfc.predict(data)
prots = []
for protA, protB, label in toPred.scores:
prots.append((protA, protB))
outFH = open(outF, "w")
for i in range(len(preds)):
protA, protB = prots[i]
if protA in geneName: geneA = ",".join(geneName[protA])
if protB in geneName: geneB = ",".join(geneName[protB])
spec = species[protA]
if preds[i][1]>0.5:
print >> outFH, "%s\t%s\t%s\t%s\t%s\t%f" % (protA, protB, geneA, geneB, spec, preds[i][1])
outFH.close()
def main():
(elutionF, refF, windowSize, outF) = sys.argv[1:]
windowSize = int(windowSize)
outData = ['']*3
reference, elutionData, scoreCalc = calcS.loadData(refF, elutionF)
j = 0
name = elutionF.split("Ce_")[1].split(".")[0]
for resultScore in getFracEvals(elutionData.elutionMat):
data_lines = entropyVSprecision(elutionData, reference, resultScore, windowSize)
for i in range(len(data_lines)):
outData[j] += "\n%s\t%i\t%s" % (name, windowSize, data_lines[i])
j += 1
if len(outData[0]) != 0:
printTable("%s_%s_Entropy_%i.dat" % (outF, name, windowSize), "Entropy", outData[0])
printTable("%s_%s_Prot-prob_%i.dat" % (outF, name, windowSize), "Prot-prob", outData[1])
printTable("%s_%s_Num-prots_%i.dat" % (outF, name, windowSize), "Num-prots", outData[2])
def main():
(elutionFiles, refF, direction, outF) = sys.argv[1:]
elutionFilesFH = open(elutionFiles)
outData = {}
maxSize = 0
for line in elutionFilesFH:
line = line.rstrip()
reference, elutionData, scoreCalc = calcS.loadData(refF, line)
scores = removeFracs(elutionData, reference, scoreCalc, direction)
name = line.split("Ce_")[2].split(".")[0]
outData[name] = scores
maxSize = max(len(scores), maxSize)
elutionFilesFH.close()
outFH = open(outF, "w")
print >> outFH, "Experiment_name\tFraction_%s" % ("\tFraction_".join(map(str,range(1, maxSize+1))))
for dataset in outData:
scores = outData[dataset]
numFracs = len(scores)
outline = "%s\t%s" % (dataset, "\t".join(map(str, scores)))
if maxSize-numFracs > 0:
outline = "%s\t%s" % (outline, "\t".join(["NA"]*(maxSize-numFracs)))
print >> outFH, outline
outFH.close()
