def vcf2pd_main(args):
"""
Run the program
"""
args = parse_args(args)
vcf_fn = args.vcf_fn
# We'll require regions to start
regions = []
if args.regions is None:
regions.append((None, None, None, 'all'))
else:
with open(args.regions, 'r') as fh:
for line in fh:
data = line.strip().split('\t')
data[1] = int(data[1])
data[2] = int(data[2])
regions.append(data[:4])
items = []
for chrom, start, end, name in regions:
if name is not None:
oname = "%s%s%s.jl" % (args.out, chrom, name)
else:
oname = args.out + ".jl"
items.append({"input": vcf_fn, "parser": args.parser, "output": oname,
"chrom": chrom, "start": start, "end": end})
# must iterate once to ensure we got the results made
[_ for _ in vpq.fchain([task], items, args.threads)]
def main():
# build an fchain
data = range(10)
pipe = [(f1, {"f1arg": 'a'}), (f2, {"f2arg": 'boop'}), f3]
threads = 2 if len(sys.argv) == 1 else int(sys.argv[1])
for i in vpq.fchain(pipe, data, workers=threads):
print("collected %r" % (i))
def __init__(self, vpds, threads):
super().__init__()
pipe = [vpq.jl_load,
vpq.sample_gt_count]
self.result = defaultdict(Counter)
# consolidate
for piece in vpq.fchain(pipe, vpds, threads):
for samp in piece["samples"]:
self.result[samp].update(piece["sample_gt_count"][samp])
def __init__(self, vpds, threads=1):
super().__init__()
pipe = [vpq.jl_load,
vpq.categorize_sv, # This allows for some pieces to still be around, but as zeros...
vpq.add_cnt_column,
vpq.add_sizebin_column,
(vpq.groupcnt, {"key": "sztype_count", "group": ["szbin", "svtype"]})]
# we have to make szbins a series so pd won't reorder it
self.result = pd.MultiIndex.from_product([pd.Series(vpq.SZBINS), range(len(vpq.SV))]).to_frame()
self.result["cnt"] = 0
for _ in vpq.fchain(pipe, vpds, threads):
self.result["cnt"] += _["sztype_count"].fillna(0)
def __init__(self, vpds, threads=1):
super().__init__()
pipe = [vpq.jl_load,
vpq.add_cnt_column,
(vpq.groupcnt, {"key": "type_count", "group": ["svtype"]})
]
# Consolidate
self.result = pd.DataFrame(np.zeros(len(vpq.SV), dtype=int),
columns=["cnt"],
index=range(len(vpq.SV)))
total = len(vpds)
for piece in vpq.fchain(pipe, vpds, threads):
self.result = self.result.add(piece["type_count"].fillna(0), axis=0)
self.result = self.result.fillna(0)
def __init__(self, vpds, threads=1):
super().__init__()
pipe = [vpq.jl_load,
vpq.add_cnt_column,
vpq.add_qualbin_column,
(vpq.groupcnt, {"key": "qualbin_count", "group": ["qualbin"]})
]
# consolidate
self.result = pd.DataFrame(np.zeros(len(vpq.QUALBINS), dtype=int),
columns=["cnt"],
index=vpq.QUALBINS)
for piece in vpq.fchain(pipe, vpds, threads):
self.result = self.result.add(piece["qualbin_count"].fillna(0), axis=0)
self.result = self.result.fillna(0)