savesort.write("\t" + str(used))
savesort.write("\t" + str(skipped))
savesort.write("\t" + str(split))
savesort.write("\n")
writeTrees(trees, label)
savesort.close()
codeToSampleMap = getCodeToSampleMap()
#Read in the 'optim' input files, and parse them to figure out where the splits are.
allsets, labelSamples = readOptimInputs(codeToSampleMap)
inputSplits = getOptimInputSplits(allsets)
#Read in the deletions/CNVs
(patientSampleMap, samplePatientMap) = lps.getPatientSampleMap()
deletions, CNVs = lps.loadDeletionsAndCNVs(samplePatientMap)
#Read in the mutations
mutations = readMutations()
sortMutations(mutations, allsets, inputSplits, deletions, CNVs, labelSamples)
#writeAllSampleVAFs(mutations, patientSampleMap, deletions)
count = 0
countByCall = {}
for patient in mutations:
for chrom in mutations[patient]:
for pos in mutations[patient][chrom]:
if len(list(mutations[patient][chrom][pos])) > 1:
localmin = val
elif direction == "down":
if val < localmin:
localmin = val
distance = 0
else:
distance += 1
if distance >= 40 and val - localmin > 0.25:
#print("Switched directions: going up at", key, distance, localmax, val)
direction = "up"
localmax = val
maxkey = key
return ret
(patientSampleMap, samplePatientMap) = lsl.getPatientSampleMap(
dipvtet_file="calling_evidence_odds.tsv")
deletions, CNVs = lsl.loadDeletionsAndCNVs(samplePatientMap)
summary = open("summary_smoothed_and_fit.tsv", "w")
summary.write("Patient")
summary.write("\tSample")
summary.write("\tnPoints")
summary.write("\tCall")
summary.write("\tGroup")
#summary.write("\tMean x2")
#summary.write("\tStdev x2")
summary.write("\tHistMax x2")
#summary.write("\tHistMax height")
summary.write("\tFitNormal x2")
summary.write("\tFitNormal weight")
summary.write("\tHistMax x2")
False if not.
"""
if patient not in deletions:
return False
if sample not in deletions[patient]:
return False
if chrom not in deletions[patient][sample]:
return False
for (start, end) in deletions[patient][sample][chrom]:
if start <= pos and end >= pos:
return True
return False
mutations = {}
(__, samplePatientMap) = lsl.getPatientSampleMap()
patientSampleMap = {}
with open(mutation_file, 'r') as csvfile:
for lvec in csv.reader(csvfile):
if "DNANum" in lvec[0]:
continue
(sample, __, __, chr, pos, ref, alt, is_snv, is_2p) = lvec[0:9]
if (is_snv == "f"):
continue
if (is_2p == "f"):
continue
# if ("N" in sample):
# continue
refcnt = int(lvec[-2])
bafcnt = int(lvec[-1])
)
for patient in groupdata:
for samples in groupdata[patient]:
outfile.write(patient)
outfile.write("\t" + groupdata[patient][samples]["matches_tree"])
outfile.write("\t" + str(groupdata[patient][samples]["count"]))
outfile.write("\t" + str(groupdata[patient][samples]["percentage"]))
outfile.write("\t" + str(groupdata[patient][samples]["cnv_count"]))
outfile.write("\t" +
str(groupdata[patient][samples]["cnv_percentage"]))
for sample in samples:
outfile.write("\t" + sample)
outfile.write("\n")
outfile.close()
patientSampleMap, __ = lsl.getPatientSampleMap()
#Now do some analysis
types = ["Singleton", "Root", "Grouped", "Ungrouped"]
outfile = open(groupdir + outfilename, "w")
outfile.write("Patient\tSubclone SNV Threshhold\tnSNVmax\tGD Samples")
for type in types:
outfile.write("\t" + type + " counts")
outfile.write("\t" + type + " total")
outfile.write(
"\tUngrouped potential subclone counts\tUngrouped potential subclone total\n"
)
for patient in groupdata:
smallestSNVcount = 100000
maxSNVcount = 0
GDsamples = set()
for sample in patientSampleMap[patient]:
print("Invalid match grid for patient", patient, "samples",
str(samples), "at chr", str(chr), str(segpair))
return (Nvec, Svec, allBalanced)
#Main routine:
CNlist = []
for (_, _, f) in walk(CN_input):
CNlist += f
BAFlist = []
for bafin in BAF_input:
for (_, _, f) in walk(bafin):
BAFlist += f
(s2p, p2s) = lsl.getPatientSampleMap(dipvtet_file)
for patient in p2s:
if somepatientsonly and patient not in somepatients:
continue
samples = p2s[patient]
samples.sort()
segments = getSegmentCalls(patient, samples, s2p, CNlist)
allA = open(BEAST_output + patient + "_allA.txt", "w")
allB = open(BEAST_output + patient + "_allB.txt", "w")
writeHeader(allA, samples)
writeHeader(allB, samples)
chrs = list(segments.keys())
chrs.sort()
for chr in chrs:
shouldSwitch = True
prevN = [-1] * len(samples)