def MatchVcfPed(vcfs, peds):
res = []
for vcf in vcfs:
for ped in peds:
if utils.GetBaseName(vcf) == utils.GetBaseName(ped):
print utils.GetBaseName(vcf), utils.GetBaseName(ped)
res.append((vcf, ped))
break
print "Finded vcf-ped pairs:"
for v, p in res:
print v, p
print '\n'
return res
def OutputSHMsForVGenes(shm_df, output_config):
gene_type_dict = dict()
gene_len = dict()
num_aligned = dict()
for it in shm_df:
if it.segment not in gene_type_dict:
gene_type_dict[it.segment] = dict()
segment_dict = gene_type_dict[it.segment]
gene_name = utils.GetBaseName(it.gene_name)
if gene_name not in gene_type_dict[it.segment]:
gene_type_dict[it.segment][gene_name] = []
gene_len[gene_name] = 0
num_aligned[gene_name] = 0
gene_type_dict[it.segment][gene_name].extend(shm_df[it])
gene_len[gene_name] = max(gene_len[gene_name], it.gene_len)
num_aligned[gene_name] += 1
for segment in gene_type_dict:
segment_dict = gene_type_dict[segment]
for gene_name in segment_dict:
num_aligned_seq = num_aligned[gene_name]
if num_aligned_seq < 10:
continue
output_fname = os.path.join(
output_config.GetSHMDirBySegment(segment), gene_name)
nucl_pos_dict = OutputGeneSHMPlot(segment_dict[gene_name],
gene_name, gene_len[gene_name],
num_aligned[gene_name],
output_fname,
output_config.Log())
output_config.AddSHMFileForSegment(segment, output_fname)
OutputGeneSHMsToTxt(
nucl_pos_dict, num_aligned[gene_name],
os.path.join(output_config.GetSHMDirBySegment(segment),
gene_name) + '.txt')
def _CreateVJDicts(self):
self.vj_dict = dict()
self.v_dict = dict()
self.j_dict = dict()
for i in range(len(self.vj_df)):
base_v = utils.GetBaseName(self.vj_df['V_hit'][i])
base_j = utils.GetBaseName(self.vj_df['J_hit'][i])
if (base_v, base_j) not in self.vj_dict:
self.vj_dict[(base_v, base_j)] = 0
self.vj_dict[(base_v, base_j)] += 1
if base_v not in self.v_dict:
self.v_dict[base_v] = 0
self.v_dict[base_v] += 1
if base_j not in self.j_dict:
self.j_dict[base_j] = 0
self.j_dict[base_j] += 1
self.sorted_vs = sorted(self.v_dict.keys())
self.sorted_js = sorted(self.j_dict.keys())
def OutputVJGenesMutability(shm_df, output_config):
v_gene_mutability = dict()
j_gene_mutability = dict()
for it in shm_df:
cur_dict = v_gene_mutability
if not it.is_variable():
cur_dict = j_gene_mutability
gene_name = utils.GetBaseName(it.gene_name)
if gene_name not in cur_dict:
cur_dict[gene_name] = []
mutability = float(len(shm_df[it])) / it.gene_len
cur_dict[gene_name].append(mutability)
OutputGeneMutability(v_gene_mutability, output_config.v_mutability, 'V',
output_config.Log())
OutputGeneMutability(j_gene_mutability, output_config.j_mutability, 'J',
output_config.Log())
def SplitInheritencePattern(InpDir, VCF, PED):
vcfs = utils.get_files(InpDir, '.tsv')
peds = utils.get_files(InpDir, '.ped')
vcf_peds = MatchVcfPed(vcfs, peds)
#vcf_peds = [[VCF,PED]]
for vcf, ped in vcf_peds:
Ped = Pedigree(ped)
print "Processing vcf: %s\tpedigree: %s" % (vcf, ped)
fout_AR = open(utils.GetBaseName(vcf) + '_AR.tsv', 'wb')
fout_AR.write('Autsomal Recessive Variants\n')
fout_AD = open(utils.GetBaseName(vcf) + '_AD.tsv', 'wb')
fout_AD.write('Autsomal Dominant Variants\n')
fout_XL = open(utils.GetBaseName(vcf) + '_XL.tsv', 'wb')
fout_XL.write('X-linked Variants\n')
fout_CH = open(utils.GetBaseName(vcf) + '_CH.tsv', 'wb')
fout_CH.write('Compound Heterozygote\n')
fin = open(vcf, 'rb')
header = fin.readline()
fout_AR.write(header)
fout_AD.write(header)
fout_XL.write(header)
fout_CH.write(header)
headerList = header.strip().split('\t')
RecordLen = len(headerList)
Gene_idx = headerList.index('Annotation')
Model_idx = headerList.index('GeneticModels')
Format_idx = headerList.index('FORMAT')
CH_buffer = []
LastGene = None
for l in fin:
llist = l.strip().split('\t')
Gene = llist[Gene_idx]
Model = llist[Model_idx].split(':')[-1]
if 'AR' in Model and selectAR(llist, headerList, Ped, Model,
Format_idx):
fout_AR.write(l)
if 'AD' in Model and selectAD_DN(llist, headerList, Ped, Model,
Format_idx):
fout_AD.write(l)
if 'X' in Model:
fout_XL.write(l)
if Gene != LastGene:
if LastGene != None:
LookCompoundHeter(CH_buffer, fout_CH, RecordLen,
Format_idx, headerList, Ped)
CH_buffer = [l]
LastGene = Gene
else:
CH_buffer.append(l)
LookCompoundHeter(CH_buffer, fout_CH, RecordLen, Format_idx,
headerList, Ped)
fout_AR.close()
fout_AD.close()
fout_XL.close()
fout_CH.close()
new_dir = os.path.join(InpDir, utils.GetBaseName(vcf))
print new_dir
if not os.path.exists(new_dir):
os.mkdir(new_dir)
shutil.copy(vcf, new_dir)
shutil.copy(ped, new_dir)
for dest in [
utils.GetBaseName(vcf) + '_AR.tsv',
utils.GetBaseName(vcf) + '_AD.tsv',
utils.GetBaseName(vcf) + '_XL.tsv',
utils.GetBaseName(vcf) + '_CH.tsv'
]:
if os.path.exists(os.path.join(new_dir, dest)):
os.remove(os.path.join(new_dir, dest))
shutil.move(dest, new_dir)