def doxpEHH(study, force, pop, supp):
logging.info("Starting xpEHH %s %s", pop, supp)
shutil.rmtree("xpEHH", True)
os.mkdir("xpEHH")
os.chdir("xpEHH")
setIndivs = study.getStatIndivs("xpEHH", pop)
suppSetIndivs = study.getStatIndivs("xpEHH", supp)
logging.info("numIndivs: %d %d", len(setIndivs), len(suppSetIndivs))
hash = MEGA.getHash(setIndivs)
hashSupp = MEGA.getHash(suppSetIndivs)
myDir = MEGA.cacheDB + "/sets/" + karyo.karyotype + "/" + hash
logging.info("xpEHH: %s hash: %s phase: %s", pop, hash,
study.getPhasePop("xpEHH", pop))
if not force:
if os.path.isfile(myDir + "/%s-22.uxpEHH" % (supp,)):
logging.info("Already done and not forced")
os.chdir("..")
return
name = "study/%s/xpEHH/%s" % (study.name, pop)
sql.addId(name, hash)
suppName = "study/%s/xpEHH/%s-support" % (study.name, supp)
sql.addId(suppName, hashSupp)
stats.doBasicInfoSet(setIndivs, force)
stats.doBasicInfoSet(suppSetIndivs, force)
w = open(tmpDir + "/xpEHH/inds", "w")
for famId, sampId in setIndivs:
w.write("%s\t%s\n" % (famId, sampId))
w.close()
w = open(tmpDir + "/xpEHH/suppinds", "w")
for famId, sampId in suppSetIndivs:
w.write("%s\t%s\n" % (famId, sampId))
w.close()
os.system("python3 %s/doHaploStats.py prepareData %s %s suppinds %s %s xpEHH" %
(MEGA.haploScripts, karyo.karyotype, study.name,
study.xpEHHConf["source"],
study.getPhasePop("xpEHH", supp)))
for k in range(1, 23 + 1):
shutil.copyfile("%d.hap" % (k,),
"s%d.hap" % (k,))
os.system("python3 %s/doHaploStats.py prepareData %s %s inds %s %s xpEHH" % (
MEGA.haploScripts, karyo.karyotype, study.name,
study.xpEHHConf["source"], study.getPhasePop("xpEHH", pop),))
os.system("python3 %s/doHaploStats.py XPEHH %s" % (
MEGA.haploScripts, karyo.karyotype))
for k in range(1, 23 + 1):
shutil.copyfile("%d.xpEHH" % (k,),
myDir + "/%s-%d.uxpEHH" % (supp, k,))
os.chdir("..")
logging.info("xpEHH Done")
return hash
def doiHS(study, force, pop):
logging.info("Starting iHS")
shutil.rmtree("iHS", True)
os.mkdir("iHS")
os.chdir("iHS")
setIndivs = study.getStatIndivs("iHS", pop)
logging.info("numIndivs: %d", len(setIndivs))
hash = MEGA.getHash(setIndivs)
myDir = MEGA.cacheDB + "/sets/" + karyo.karyotype + "/" + hash
logging.info("iHS: %s hash: %s phase: %s",
pop, hash, study.getPhasePop("iHS", pop))
if not force:
if os.path.isfile(myDir + "/22.uiHS"):
logging.info("Already done and not forced")
os.chdir("..")
return
name = "study/%s/iHS/%s" % (study.name, pop)
sql.addId(name, hash)
stats.doBasicInfoSet(setIndivs, force)
w = open(tmpDir + "/iHS/inds", "w")
for famId, sampId in setIndivs:
w.write("%s\t%s\n" % (famId, sampId))
w.close()
os.system("python3 %s/doHaploStats.py prepareData %s %s inds %s %s" %
(MEGA.haploScripts, karyo.karyotype, study.name,
study.iHSConf["source"], study.getPhasePop("iHS", pop)))
os.system("python3 %s/doHaploStats.py iHS %s %s" % (MEGA.haploScripts,
karyo.karyotype, study.name))
for k in range(1, 23 + 1):
try:
shutil.copyfile("%d.uiHS" % k, myDir + "/%d.uiHS" % k)
except IOError:
pass
os.chdir("..")
logging.info("iHS Done")
return hash
def doZoom(study, force, name):
logging.info("Starting Zoom - %s", name)
#XXX FORCE
try:
os.mkdir("zoom")
except OSError:
pass # Already exists
os.chdir("zoom")
chro, start, end, pop, ibd = study.zoom.general[name]
pops = pop.split("+")
indivs = []
for myPop in pops:
if ibd:
indivs.extend(study.ibd.getIndivsPop(myPop, ibd, False))
else:
indivs.extend(study.pops.getIndivs(myPop))
hash = MEGA.getHash(indivs)
posAlls = ensembl.getSNPs(chro)
poses = {}
snps = []
snpAtPos = {}
for rs, content in list(posAlls.items()):
pos = content[0]
if pos >= start and pos <= end:
snps.append(rs)
poses[rs] = pos
snpAtPos[pos] = rs
ancAlls = ensembl.getAncs(chro)
ancs = {}
for snp in snps:
ancs[snp] = ancAlls.get(snp, None)
gPoses = {}
f = open(MEGA.geneticMapDB + "/37-%d.map" % chro)
f.readline()
for l in f:
toks = l.rstrip().split("\t")
pos = int(toks[1])
if pos < start:
continue
if pos > end:
break
rs = snpAtPos.get(pos, "")
if rs == "":
continue
gPos = float(toks[3])
gPoses[rs] = gPos
f.close()
#assuming ihs phase.conf exists
source = study.iHSConf["source"]
refPop = study.getPhasePop("iHS", pop)
if refPop != "shapeIt":
phasedFile = "%s/%s/%s-%d.gz" % (MEGA.phaseDB, source, refPop, chro)
else:
phasedFile = "%s/%s/%d.gz" % (MEGA.phaseDB, source, chro)
inds = [x[1] for x in indivs]
project_beagle_phase(gzip.open("%d.gz" % chro, "w"),
gzip.open(phasedFile),
ind_retain=inds, snp_retain=snps,
want_phased=True, is_phased=True)
realSNPs = set()
f = gzip.open("%d.gz" % chro)
f.readline()
f.readline()
for l in f:
toks = l.rstrip().split("\t")
realSNPs.add(toks[1])
f.close()
stats = study.zoom.stats[name]
vals = {}
for stat, params in stats:
if stat == "iHS":
vals["iHS"] = {}
myHash = MEGA.getHash(study.getStatIndivs("iHS", pop))
f = open(MEGA.cacheDB + "/sets/" + karyo.karyotype + "/" +
myHash + "/" + str(chro) + ".iHS")
for l in f:
toks = l.rstrip().split("\t")
if toks[0] in realSNPs:
vals["iHS"][toks[0]] = float(toks[2])
elif stat == "xpEHH":
vals["xpEHH"] = {}
myHash = MEGA.getHash(study.getStatIndivs("xpEHH", pop))
f = open(MEGA.cacheDB + "/sets/" + karyo.karyotype + "/" +
myHash + "/" + params[0] + "-" + str(chro) + ".xpEHH")
for l in f:
toks = l.rstrip().split("\t")
if toks[0] in realSNPs:
vals["xpEHH"][toks[0]] = float(toks[2])
elif stat == "EHH":
rsId = params[0]
allele = params[1]
vals["EHH"] = calcEHH(chro, poses, rsId, allele)
statNames = list(vals.keys())
statNames.sort()
w = open("zoom.txt", "w")
f = gzip.open("%d.gz" % chro)
f.readline()
header = f.readline().rstrip().split("\t")[2:]
w.write("RS\tpos\tgPos\tanc\t")
for name in statNames:
w.write(name + "\t")
w.write("\t".join(header))
w.write("\n")
for l in f:
toks = l.rstrip().split("\t")
rs = toks[1]
haplos = toks[2:]
w.write("%s\t%d\t%s\t%s\t" % (rs, poses.get(rs, 0),
str(gPoses.get(rs, "")), ancs.get(rs, "")))
for name in statNames:
w.write(str(vals[name].get(rs, "")) + "\t")
w.write("\t".join(haplos))
w.write("\n")
w.close()
f.close()
shutil.copyfile("zoom.txt", MEGA.cacheDB + "/sets/" + karyo.karyotype +
"/" + hash + "/" +
"%s-%d-%d-%d.zoom" % (study.name, chro, start, end))