def doF(studyName):
myStudy = study.Study(studyName)
myStudy.configStudy()
files = os.listdir(".")
fCases = []
for fName in files:
if fName.startswith("Fstat") and fName.endswith(".txt"):
fCases.append(int(fName[5:fName.find("-")]))
fCases = set(fCases)
for fCase in fCases:
doZip(None, "f%d.zip" % fCase, "Fstat%d-*.txt" % fCase)
vals = []
posChro = {}
valsChro = {}
for chro in karyo.groups:
posChro[chro] = []
valsChro[chro] = []
try:
f = open("Fstat%d-%s.txt" % (fCase, chro))
except IOError:
continue # Chro is missing, OK (for sex chros)
f.readline()
for l in f:
toks = l.rstrip().split("\t")
if toks[2] == "None":
continue
pos = int(toks[0].split("/")[2])
val = float(toks[2]) # FST
posChro[chro].append(pos)
valsChro[chro].append(val)
vals.append(float(toks[2]))
f.close()
vals.sort()
w = open("FST-summary-%d" % fCase, "w")
w.write("mean\tmin\t0.005\t0.025\t0.5\t0.975\t0.995\tmax\n")
myStats = (
sum(vals) / len(vals),
min(vals),
vals[int(len(vals) * 0.005)],
vals[int(len(vals) * 0.025)],
vals[len(vals) // 2],
vals[int(len(vals) * 0.975)],
vals[int(len(vals) * 0.995)],
max(vals))
w.write("%f\t%f\t%f\t%f\t%f\t%f\t%f\t%f\n" % myStats)
w.close()
w = open("FST-window-%d" % fCase, "w")
fsts = []
ggp = plot.GridGenomePlot(karyo.genome, 4,
fig=plt.figure(figsize=(16, 9)))
for chro in karyo.groups:
xsnp = posChro[chro]
ysnp = valsChro[chro]
ysAll, buckPos = doSlidingWindow(xsnp, ysnp, 200000, 200000)
xs = [(x[0] + x[1]) // 2 for x in buckPos]
ys = []
p = 0
for lst in ysAll:
if lst == []:
ys.append(None)
w.write("%s\t%d\tNone\t\n" % (chro, xs[p]))
fsts.append((chro, xs[p], None, None))
else:
ys.append(max(lst))
# fstLStr = "/".join(map(lambda x:str(x),lst))
# above01 = float(len(filter(lambda x:x>0.1,lst)))/len(lst)
above01 = float(len([x for x in lst if x > 0.1]))
w.write("%s\t%d\t%f\t%s\f" % (
chro, xs[p], max(lst), above01))
fsts.append((chro, xs[p], max(lst), above01))
p += 1
data = list(zip(xsnp, ysnp))
ax = ggp.features[chro]
plot.plot_percentile(ax, data, 200000, 200000,
funcs=[(max, '.'), (numpy.mean, '#0099FF')])
bot, top = ax.get_ylim()
ax.set_ylim(0, 1)
ggp.fig.savefig("Fst-%d.png" % fCase)
w.close()
sortWrite("FST-window-%s" % fCase, fsts)
print(freq, means[freq], stds[freq], file=STAT)
#print freq, means[freq], stds[freq]
STAT.close()
chroCalc={}
myBins = {}
for k in range(1, maxChro+1):
w=open("%d.iHS" %(k,), "w")
pos, val = [], []
f=open("%d.uiHS" %(k,))
for res in standartizeIHS(f, means, stds):
pos.append(res["pos"])
val.append(res["iHS"])
#print res["freq"], "Y", res["iHS"], "X", res["uiHS"], means[res["freq"]],stds[res["freq"]]
w.write("%s\t%d\t%f\n" % (res["snp"], res["pos"], res["iHS"]))
f.close()
buckets, buckPos = doSlidingWindow(pos, val, slide, size)
posResult = {}
chroCalc[k] = posResult
for i in range(len(buckPos)):
myMin, myMax = buckPos[i]
vals = buckets[i]
above = [x for x in vals if abs(x)>iHSThr]
absVals = [abs(x) for x in vals]
#print vals
if len(vals)>0:
bin = len(vals)/20
iHSCut = float(len(above))/len(vals)
myBins.setdefault(bin,[]).append(iHSCut)
xpEHHCut =max(absVals)
posResult[(myMin+myMax)/2] = len(vals), iHSCut, sum(absVals)/len(vals), bin
#print len(vals), iHSCut, sum(absVals)/len(vals), bin