def get_sample_ids(vcf_path,
provided_t_name=None,
provided_n_name=None,
provided_r_name=None,
return_names=False):
t_ids, n_ids, r_ids = [], [], []
t_names, n_names, r_names = [], [], []
from cyvcf2 import VCF
vcf_samples = VCF(vcf_path).samples
if provided_t_name:
for sname in provided_t_name.split(','):
assert sname in vcf_samples,\
f'Tumor sample name {sname} is not in VCF {vcf_path}. Found: {vcf_samples}'
t_names.append(sname)
if provided_n_name:
for sname in provided_n_name.split(','):
assert sname in vcf_samples,\
f'Normal sample name {sname} is not in VCF {vcf_path}. Found: {vcf_samples}'
n_names.append(sname)
if provided_r_name:
for sname in provided_r_name.split(','):
assert sname in vcf_samples,\
f'RNA sample name {sname} is not in VCF {vcf_path}. Found: {vcf_samples}'
r_names.append(sname)
if len(vcf_samples) == 1:
t_names = [vcf_samples[0]]
t_ids = [0]
else:
guessed_t_name, guessed_n_name = guess_sample_names(vcf_path)
if not t_names:
if not guessed_t_name:
critical(
f'Can\'t guess tumor sample name from the VCF {vcf_path}')
t_names = [guessed_t_name]
if not n_names:
if guessed_n_name:
n_names = [guessed_n_name]
else:
if t_names:
n_names = [
s for s in vcf_samples
if s not in t_names and s not in r_names
]
if not n_names:
critical(
f'Can\'t guess normal sample name from the VCF {vcf_path}'
)
else:
critical(
f'Can\'t guess normal sample name from the VCF {vcf_path}'
)
if t_names:
assert set(t_names) & set(
vcf_samples), f't_names: {t_names}, vcf_samples: {vcf_samples}'
t_ids = [vcf_samples.index(tn) for tn in t_names]
if n_names:
assert set(n_names) & set(
vcf_samples), f'n_names: {n_names}, vcf_samples: {vcf_samples}'
n_ids = [vcf_samples.index(nn) for nn in n_names]
if r_names:
assert set(r_names) & set(
vcf_samples), f'r_names: {r_names}, vcf_samples: {vcf_samples}'
r_ids = [vcf_samples.index(rn) for rn in r_names]
if return_names:
ret = t_names[0] if len(t_names) == 1 else t_names, \
n_names[0] if len(n_names) == 1 else n_names
if r_names:
ret += (r_names[0] if len(r_names) == 1 else r_names, )
else:
ret = t_ids[0] if len(t_names) == 1 else t_ids, \
n_ids[0] if len(n_names) == 1 else n_ids
if r_names:
ret += (r_ids[0] if len(r_ids) == 1 else r_ids, )
return ret
