def compareTools(reg1, reg2) :
regs = lc.getFragments(reg1, reg2)
comp = lc.doComparison2(regs, reg1, reg2)
mat = ls.doContingency(comp)
jcc = ls.jaccardIndex(comp)
num = ls.regionNumber(regs)
base = ls.baseSimilarity(regs, reg1, reg2)
regions = ls.regSimilarity(regs, reg1, reg2)
jcca = getJaccard(comp, "A")
jccn = getJaccard(comp, "N")
jccl = getJaccard(comp, "L")
jccd = getJaccard(comp, "D")
st = "{nr}\t{bs}\t{rs}\t{mcca}\t{mccn}\t{mccl}\t{mccd}\t{jcc}\t{jcca}\t{jccn}\t{jccl}\t{jccd}\t{pur1}\t{pur2}\t{plo1}\t{plo2}".format(
nr = num, bs = base, rs = regions, mcca = mat["A"]["MCC"], mccn = mat["N"]["MCC"], mccl = mat["L"]["MCC"], mccd = mat["D"]["MCC"], jcc = jcc, jcca = jcca, jccn = jccn, jccl = jccl, jccd = jccd, pur1 = reg1["purity"], pur2 = reg2["purity"], plo1 = reg1["ploidy"], plo2 = reg2["ploidy"])
return st
def launchAnalysis(folder, array, ascat, facets):
pythonpath = os.path.dirname(os.path.realpath(__file__))
#Go to the working directory to run the comparison analysis
print("INFO: Analysing {}".format(folder))
os.chdir(folder)
ascatReg = None
facetsReg = None
arrayReg = None
arrayPath = "../{}".format(array)
id = folder.split("/")[-1]
if ascat != "":
ascatReg = comp.convert2region(ascat, "ascat")
if facets != "":
facetsReg = comp.convert2region(facets, "facets")
arrayReg = comp.convert2region(arrayPath, "array")
#if ascatReg != None and facetsReg != None : #Calculate logR between AscatNGS and FACETS
#regA_F = comp.getFragments(ascatReg, facetsReg)
#st.logRcomp(regA_F, ascatReg, facetsReg, "ascatNGS", "FACETS")
if facetsReg != None and arrayReg != None:
regAr_F = comp.getFragments(arrayReg, facetsReg)
mt1 = comp.doComparison2(regAr_F, facetsReg, arrayReg)
jc1 = st.jaccardIndex(mt1, ["A", "D"])
mt1 = comp.doComparison(regAr_F, facetsReg, arrayReg)
cm1 = st.doContingency(mt1, ["A", "D"])
if ascatReg != None and arrayReg != None:
regAr_A = comp.getFragments(arrayReg, ascatReg)
mt2 = comp.doComparison2(regAr_F, ascatReg, arrayReg)
jc2 = st.jaccardIndex(mt2, ["A", "D"])
mt2 = comp.doComparison(regAr_F, ascatReg, arrayReg)
cm2 = st.doContingency(mt2, ["A", "D"])
#Return to current python path
os.chdir(pythonpath)
#Store the summary data in the corresponding files
if ascatReg != None:
with open("ascatPurities.txt", "a") as fi:
fi.write("{}\t{}\n".format(id, ascatReg["purity"]))
with open("ascatPloidies.txt", "a") as fi:
fi.write("{}\t{}\n".format(id, ascatReg["ploidy"]))
with open("ascatJaccard.txt", "a") as fi:
fi.write("{}\t{}\n".format(id, jc2))
with open("ascatACCamplification.txt", "a") as fi:
fi.write("{}\t{}\n".format(id, cm2["A"]["ACC"]))
with open("ascatTPRamplification.txt", "a") as fi:
fi.write("{}\t{}\n".format(id, cm2["A"]["TPR"]))
with open("ascatTNRamplification.txt", "a") as fi:
fi.write("{}\t{}\n".format(id, cm2["A"]["TNR"]))
with open("ascatPPVamplification.txt", "a") as fi:
fi.write("{}\t{}\n".format(id, cm2["A"]["PPV"]))
with open("ascatFDRamplification.txt", "a") as fi:
fi.write("{}\t{}\n".format(id, cm2["A"]["ACC"]))
with open("ascatACCdeletion.txt", "a") as fi:
fi.write("{}\t{}\n".format(id, cm2["D"]["ACC"]))
with open("ascatTPRdeletion.txt", "a") as fi:
fi.write("{}\t{}\n".format(id, cm2["D"]["TPR"]))
with open("ascatTNRdeletion.txt", "a") as fi:
fi.write("{}\t{}\n".format(id, cm2["D"]["TNR"]))
with open("ascatPPVdeletion.txt", "a") as fi:
fi.write("{}\t{}\n".format(id, cm2["D"]["PPV"]))
with open("ascatFDRdeletion.txt", "a") as fi:
fi.write("{}\t{}\n".format(id, cm2["D"]["ACC"]))
if facetsReg != None:
with open("facetsPurities.txt", "a") as fi:
fi.write("{}\t{}\n".format(id, facetsReg["purity"]))
with open("facetsPloidies.txt", "a") as fi:
fi.write("{}\t{}\n".format(id, facetsReg["ploidy"]))
with open("facetsJaccard.txt", "a") as fi:
fi.write("{}\t{}\n".format(id, jc1))
with open("facetsACCamplification.txt", "a") as fi:
fi.write("{}\t{}\n".format(id, cm1["A"]["ACC"]))
with open("facetsTPRamplification.txt", "a") as fi:
fi.write("{}\t{}\n".format(id, cm1["A"]["TPR"]))
with open("facetsTNRamplification.txt", "a") as fi:
fi.write("{}\t{}\n".format(id, cm1["A"]["TNR"]))
with open("facetsPPVamplification.txt", "a") as fi:
fi.write("{}\t{}\n".format(id, cm1["A"]["PPV"]))
with open("facetsFDRamplification.txt", "a") as fi:
fi.write("{}\t{}\n".format(id, cm1["A"]["ACC"]))
with open("facetsACCdeletion.txt", "a") as fi:
fi.write("{}\t{}\n".format(id, cm1["D"]["ACC"]))
with open("facetsTPRdeletion.txt", "a") as fi:
fi.write("{}\t{}\n".format(id, cm1["D"]["TPR"]))
with open("facetsTNRdeletion.txt", "a") as fi:
fi.write("{}\t{}\n".format(id, cm1["D"]["TNR"]))
with open("facetsPPVdeletion.txt", "a") as fi:
fi.write("{}\t{}\n".format(id, cm1["D"]["PPV"]))
with open("facetsFDRdeletion.txt", "a") as fi:
fi.write("{}\t{}\n".format(id, cm1["D"]["ACC"]))
import libcomparison as compi
import libgetters as ge
import libstatistics as sts
import libconstants as cts
print(
"INFO: Test unitario para comparar el output de un ejemplo de FACETS con los datos del array descargado desde TCGA"
)
print("TEST 1) Extraer datos")
ar = compi.convert2region(
"/g/strcombio/fsupek_cancer2/TCGA_bam/OV/TCGA-25-1315/1948ef01_VS_d57f7ca3_PURPLE/TUMOR.purple.cnv.somatic.tsv",
"PURPLE")
fa = compi.convert2region(
"/g/strcombio/fsupek_cancer2/TCGA_bam/OV/TCGA-25-1315/1948ef01_VS_d57f7ca3_FACETS/facets_comp_cncf.tsv",
"FACETS")
print("TEST 2) Buscando las regiones en comun para estudio")
regs = compi.getFragments(ar, fa)
print("TEST 3) Crear la tabla comparativa 4x4")
dc = compi.doComparison(regs, ar, fa)
print(sts.printTable(dc, "Array", "FACETS", False))
print("TEST 4) Resultados de la matriz de confusion para cada aberracion")
c1, c2 = sts.calculateCounts(dc)
dicContingency = sts.doContingency(dc, ["A", "D", "N"])
print("\tAmplificacion\n\t{}\n\n".format(dicContingency["A"]))
print("\tDelecion\n\t{}\n\n".format(dicContingency["D"]))
print("\tNormal\n\t{}\n".format(dicContingency["N"]))
print("TEST 5) Jaccard index de cada una de las aberraciones")
jci2 = compi.doComparison2(regs, ar, fa)
jaccard = sts.jaccardIndex(jci2, ["A", "D", "N"])
print("\tJaccard index\n\t{}\n\n".format(jaccard))
def main():
fvaFi = "facetsVSascatngs.tsv"
fvsFi = "facetsVSsequenza.tsv"
avsFi = "ascatVSsequenza.tsv"
# Write the output files' header
with open(fvaFi, "w") as fi:
fi.write(
"Case\tregSim\tbaseSim\tMCCA\tMCCN\tMCCL\tMCCD\tjcc\tFpurity\tApurity\tFploidy\tAploidy\n"
)
with open(fvsFi, "w") as fi:
fi.write(
"Case\tregSim\tbaseSim\tMCCA\tMCCN\tMCCL\tMCCD\tjcc\tFpurity\tSpurity\tFploidy\tSploidy\n"
)
with open(avsFi, "w") as fi:
fi.write(
"Case\tregSim\tbaseSim\tMCCA\tMCCN\tMCCL\tMCCD\tjcc\tApurity\tSpurity\tAploidy\tSploidy\n"
)
table = []
with dbcon:
cur = dbcon.cursor()
q = cur.execute("SELECT submitter FROM patient WHERE cancer='OV'")
cases = q.fetchall()
for c in cases:
# Recollir la informacio dels bams i el sexe que te el cas registrats
with dbcon:
cur = dbcon.cursor()
q = cur.execute(
"SELECT uuid FROM sample WHERE submitter='{}' AND tumor LIKE '%Tumor%'"
.format(c[0]))
tumors = q.fetchall()
q = cur.execute(
"SELECT uuid FROM sample WHERE submitter='{}' AND tumor LIKE '%Normal%'"
.format(c[0]))
controls = q.fetchall()
for tm in tumors:
for cn in controls:
fva = []
fvs = []
avs = []
# Get the absolute path the and the prefix for the tool output
tf = "{wd}/{sub}/{tm}_VS_{cn}".format(wd=wd,
sub=c[0],
tm=tm[0].split("-")[0],
cn=cn[0].split("-")[0])
fva.append(tf.split("/")[-1])
fvs.append(tf.split("/")[-1])
avs.append(tf.split("/")[-1])
facets = "{}_FACETS/facets_comp_cncf.tsv".format(tf)
ascat = mm.findAscatName("{}_ASCAT/".format(tf))
sequenza = "{}_Sequenza/{}_segments.txt".format(tf, c[0])
if os.path.isfile(facets):
outf = lc.convert2region(facets, "facets")
if os.path.isfile(ascat):
outa = lc.convert2region(ascat, "ascatngs")
if os.path.isfile(sequenza):
outs = lc.convert2region(sequenza, "sequenza")
# Compare FACETS vs ascatNGS
if os.path.isfile(facets) and os.path.isfile(ascat):
regs = lc.getFragments(outf, outa)
c1 = lc.doComparison(regs, outf, outa)
c2 = lc.doComparison2(regs, outf, outa)
sts = ls.doContingency(
c2) # Get the MCC for all the aberrations
jcc = ls.jaccardIndex(c2)
fva.append(ls.regSimilarity(regs, outf, outa))
fva.append(ls.baseSimilarity(regs, outf, outa))
for ab in cte.aberrations:
fva.append(sts[ab]["MCC"])
fva.append(jcc)
fva.append(outf["purity"])
fva.append(outs["purity"])
fva.append(outf["ploidy"])
fva.append(outs["ploidy"])
else:
fva.append("NA")
fva.append("NA")
for ab in cte.aberrations:
fva.append("NA")
for ab in cte.aberrations:
fva.append("NA")
fva.append("NA")
fva.append("NA")
fva.append("NA")
fva.append("NA")
# Compare FACETS VS Sequenza
if os.path.isfile(facets) and os.path.isfile(sequenza):
regs = lc.getFragments(outf, outs)
c1 = lc.doComparison(regs, outf, outs)
c2 = lc.doComparison2(regs, outf, outs)
sts = ls.doContingency(
c2) # Get the MCC for all the aberrations
jcc = ls.jaccardIndex(
c2) # Get the Jaccard index for all the aberrations
fvs.append(ls.regSimilarity(regs, outf, outs))
fvs.append(ls.baseSimilarity(regs, outf, outs))
for ab in cte.aberrations:
fvs.append(sts[ab]["MCC"])
fvs.append(jcc)
fvs.append(outf["purity"])
fvs.append(outs["purity"])
fvs.append(outf["ploidy"])
fvs.append(outs["ploidy"])
else:
fvs.append("NA")
fvs.append("NA")
for ab in cte.aberrations:
fvs.append("NA")
for ab in cte.aberrations:
fvs.append("NA")
fvs.append("NA")
fvs.append("NA")
fvs.append("NA")
fvs.append("NA")
# Compare ascatNGS VS Sequenza
if os.path.isfile(ascat) and os.path.isfile(sequenza):
regs = lc.getFragments(outa, outs)
c1 = lc.doComparison(regs, outa, outs)
c2 = lc.doComparison2(regs, outa, outs)
sts = ls.doContingency(
c2) # Get the MCC for all the aberrations
jcc = ls.jaccardIndex(
c2) # Get the Jaccard index for all the aberrations
avs.append(ls.regSimilarity(regs, outa, outs))
avs.append(ls.baseSimilarity(regs, outa, outs))
for ab in cte.aberrations:
avs.append(sts[ab]["MCC"])
avs.append(jcc)
avs.append(outf["purity"])
avs.append(outs["purity"])
avs.append(outf["ploidy"])
avs.append(outs["ploidy"])
else:
avs.append("NA")
avs.append("NA")
for ab in cte.aberrations:
avs.append("NA")
for ab in cte.aberrations:
avs.append("NA")
avs.append("NA")
avs.append("NA")
avs.append("NA")
avs.append("NA")
# Write the output in the corresponding files for each comparison
with open(fvaFi, "a") as fi:
fi.write(
"{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\n".
format(fva[0], fva[1], fva[2], fva[3], fva[4], fva[5],
fva[6], fva[7], fva[8], fva[9], fva[10],
fva[11]))
with open(fvsFi, "a") as fi:
fi.write(
"{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\n".
format(fvs[0], fvs[1], fvs[2], fvs[3], fvs[4], fvs[5],
fvs[6], fvs[7], fvs[8], fvs[9], fvs[10],
fvs[11]))
with open(avsFi, "a") as fi:
fi.write(
"{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\n".
format(avs[0], avs[1], avs[2], avs[3], avs[4], avs[5],
avs[6], avs[7], avs[8], avs[9], avs[10],
avs[11]))
print("Array : {}".format(lg.getCopyNumber(brca1[1:3], brca1[0], array)))
print("ascatNGS: {}".format(lg.getCopyNumber(brca1[1:3], brca1[0], ascatngs)))
print("FACETS : {}".format(lg.getCopyNumber(brca1[1:3], brca1[0], facets)))
print("Sequenza: {}".format(lg.getCopyNumber(brca1[1:3], brca1[0], sequenza)))
print("----------------------------------------")
print("INFO: Copy number reported in BRCA2")
print("ASCAT2 : {}".format(lg.getCopyNumber(brca2[1:3], brca2[0], ascat)))
print("Array : {}".format(lg.getCopyNumber(brca2[1:3], brca2[0], array)))
print("ascatNGS: {}".format(lg.getCopyNumber(brca2[1:3], brca2[0], ascatngs)))
print("FACETS : {}".format(lg.getCopyNumber(brca2[1:3], brca2[0], facets)))
print("Sequenza: {}".format(lg.getCopyNumber(brca2[1:3], brca2[0], sequenza)))
print("----------------------------------------")
print("INFO: Comparing ASCAT vs Array")
regs = lc.getFragments(ascat, array)
print("\tGenome divided in {} regions".format(ls.regionNumber(regs)))
comp = lc.doComparison2(regs, ascat, array)
sts = ls.doContingency(comp)
print(ls.printTable(comp, "ASCAT2", "Array", False))
print("\n\tMatthews Correlation Coefficient")
for s in sts:
print("\t\t{} - {}".format(s, sts[s]["MCC"]))
print("\n\tJaccard Index")
print("\t\t{}".format(ls.jaccardIndex(comp)))
print("\n\tBase ACC")
print("\t\t{}".format(ls.baseSimilarity(regs, ascat, array)))
print("\n\tRegion ACC")
print("\t\t{}".format(ls.regSimilarity(regs, ascat, array)))
print("----------------------------------------")
print("INFO: Comparison numbers")
print(comp)