def main(input, truth_set, aaf_cutoff, chunk_size):
iterator = JimFile(input)
iterable = windower(iterator, chunker(chunk_size))
upton(iterable, truth_set, aaf_cutoff)
""" group by transcript and split at gaps > 50 bases"""
return inext.transcript != grp[0].transcript \
or inext.start - grp[-1].end > 50
def smallchunk(grp, inext):
return len(grp) > 50 or inext.transcript != grp[0].transcript \
or inext.start - grp[-1].end > 30
def rescale(vals):
minv, maxv = min(vals), max(vals)
return [float(v - minv) / ((maxv - minv) or 1) for v in vals]
# allow a gap of at most 50 bases.
from collections import defaultdict
saved = defaultdict(list)
for chunk in windower(iterator, smallchunk):
frv = FRV(chunk)
iafi = IAFI(chunk)
values = dnds_ratio(chunk)
saved["chrom"].append(chunk[0].chrom)
saved["start"].append(chunk[0].start)
saved["end"].append(chunk[-1].end)
saved["frv"].append(frv)
saved["iafi"].append(iafi)
saved["dnds"].append(values["dnds"])
saved["dn"].append(values["dn"])
saved["ds"].append(values["ds"])
saved["na"].append(values["na"])
saved["trans"].append(chunk[0].transcript)