def getSFSfromSLiM(inp):
polymorphisms = []
fixations = 0
if len(inp) == 0: return None, None
for l in inp.split('\n'):
if len(l) == 0: continue
if l[0] == 'm' or l[0] == 'g':
continue
x = l.split(' ')
if int(x[7]) > int(x[6]):
if int(x[6]) > 10000: # Assuming Ne = 1000, burn-in is 10,000
fixations += 1
elif int(x[7]) < int(x[6]):
polymorphisms.append(int(x[7]))
return fixations, SFS.SFS_from_all_frequencies(polymorphisms, 20)
def orgPolyDict(organ_mutations, N):
sfsDict = {}
for i in organ_mutations.keys():
orgDict = {}
orgDictRaw = {}
for q in organ_mutations[i]:
if q[1] not in orgDictRaw.keys():
orgDictRaw[q[1]] = [int(q[7])]
else:
orgDictRaw[q[1]].append(int(q[7]))
for o in orgDictRaw.keys():
orgDict[o] = sfs_tools.SFS_from_all_frequencies(orgDictRaw[o], N)
# print orgDict[o]
sfsDict[i] = orgDict
return sfsDict
def get_sfs_dict_from_sample(slim_input):
data = [i.strip() for i in gzip.open(slim_input).readlines()]
x = ts.slim(data, fixed=True, give_genomes=True)
if not x.sanity:
return [None, None]
# print x.name
genomes = x.genome_dict()
mutations = x.mutations_dict()
lengthDict = parseLengths(x.organ_lengths())
individuals = [random.choice(genomes.keys()) for i in range(20)]
# print individuals
# if x.name == '/exports/csce/eddie/biology/groups/eddie_biology_ieb_keightley/toms_simulations/updated_DFE/longRuns/full_usfs/configs/3381.temp.slim':
# individuals = ['p1:1398', 'p1:1646', 'p1:297', 'p1:165', 'p1:999', 'p1:1451', 'p1:982', 'p1:973', 'p1:615', 'p1:832', 'p1:12', 'p1:1109', 'p1:1137', 'p1:496', 'p1:164', 'p1:412', 'p1:1687', 'p1:1373', 'p1:72', 'p1:39']
muts_by_organ = x.organ_mutations()
new_muts = Counter()
for g in individuals:
for m in genomes[g]:
new_muts[m] += 1
polyDict = {}
for h in muts_by_organ.keys():
mTypeDict = {}
for m in muts_by_organ[h]:
if new_muts[m[0]] == 0: continue
if m[1] not in mTypeDict.keys():
mTypeDict[m[1]] = [new_muts[m[0]]]
else:
mTypeDict[m[1]].append(new_muts[m[0]])
# print h, mTypeDict
mPoly = {}
for k in mTypeDict.keys():
mPoly[k] = sfs_tools.SFS_from_all_frequencies(mTypeDict[k], 20)
polyDict[h] = mPoly
# print '!', x.name
fixedDict = x.organ_fixed(threshold=int(x.N) * 20)
fixD = orgFixDict(fixedDict)
polyfix = combinePolyFix(polyDict, fixD)
elDict = combineElements(polyfix, lengthDict)
print 'processed ' + x.name
return [x.name, elDict]
def get_sfs_dict_from_sample(slim_input):
x = ts.slim(slim_input, fixed=True, give_genomes=True)
if not x.sanity:
return [None, None]
genomes = x.genome_dict()
mutations = x.mutations_dict()
lengthDict = parseLengths(x.organ_lengths())
individuals = [random.choice(genomes.keys()) for i in range(20)]
muts_by_organ = x.organ_mutations()
new_muts = Counter()
for g in individuals:
for m in genomes[g]:
new_muts[m] += 1
polyDict = {}
for h in muts_by_organ.keys():
mTypeDict = {}
for m in muts_by_organ[h]:
if new_muts[m[0]] == 0: continue
if m[1] not in mTypeDict.keys():
mTypeDict[m[1]] = [new_muts[m[0]]]
else:
mTypeDict[m[1]].append(new_muts[m[0]])
# print h, mTypeDict
mPoly = {}
for k in mTypeDict.keys():
mPoly[k] = sfs_tools.SFS_from_all_frequencies(mTypeDict[k], 20)
polyDict[h] = mPoly
thresh = x.N * 10
fixedDict = x.organ_fixed(threshold=int(x.N) * 10)
fixD = orgFixDict(fixedDict)
polyfix = combinePolyFix(polyDict, fixD)
elDict = combineElements(polyfix, lengthDict)
print 'processed ' + x.name
return [x.name, elDict]
def sfsFromFreq(chunk, minCoverage=0):
sfs_all = []
sfs_ncpg = []
rat_div_all = 0
fam_div_all = 0
rat_div_ncpg = 0
fam_div_ncpg = 0
for i in chunk:
freq = i.strip().split()
if int(freq[2]) <= minCoverage: continue
if freq[3] == '.':
pass ## Site is not a variant, so no need to look at HWE
elif freq[3] != '.':
if float(freq[3]) < 0.0002:
continue ## Site is a variant, so need to check for HWE
cast_alleles = freq[5].split(',')
if cast_alleles[0] == '.':
continue
cast_alleles = map(int, cast_alleles)
fam_alleles = freq[7].split(',')
if fam_alleles[0] == '.':
continue
fam_alleles = map(int, fam_alleles)
rat_alleles = freq[9].split(',')
if rat_alleles[0] == '.':
continue
rat_alleles = map(int, rat_alleles)
alleleFreq = getAlleleFreq(cast_alleles)
sfs_all.append(alleleFreq)
cast_cpg = freq[4]
fam_cpg = freq[6]
rat_cpg = freq[8]
rat_div_temp = divergent(cast_alleles, rat_alleles, out_alleles=1)
fam_div_temp = divergent(cast_alleles, fam_alleles, out_alleles=2)
if rat_div_temp:
rat_div_all += rat_div_temp
if fam_div_temp:
fam_div_all += fam_div_temp
if '1' not in [cast_cpg, fam_cpg, rat_cpg]:
sfs_ncpg.append(alleleFreq)
rat_div_temp2 = divergent(cast_alleles, rat_alleles, out_alleles=1)
fam_div_temp2 = divergent(cast_alleles, fam_alleles, out_alleles=2)
if rat_div_temp2:
rat_div_ncpg += rat_div_temp2
if fam_div_temp2:
fam_div_ncpg += fam_div_temp2
divList = [fam_div_all, rat_div_all, fam_div_ncpg, rat_div_ncpg]
allSitesSFS = site_frequency_spectrum.SFS_from_all_frequencies(sfs_all, 20)
ncpgSitesSFS = site_frequency_spectrum.SFS_from_all_frequencies(
sfs_ncpg, 20)
return allSitesSFS, ncpgSitesSFS, divList