def _dump_indel_lens(self):
run_name = self._run_file()
if not os.path.exists(run_name):
raise Exception("Run must be run before attempting dump")
spats = Spats()
spats.load(run_name)
countinfo = spats.counters.counts_dict()
ilen_cnt = []
for lc in sorted([int(k.split('_')[-1]) for k in countinfo.keys() if k.startswith('mapped_indel_len_')]):
ilen_cnt.append((lc, countinfo["mapped_indel_len_{}".format(lc)]))
output_path = os.path.join(self.path, 'mapped_indel_len_counts.csv')
self._write_csv(output_path, [ "Indel Length", "Reads" ], ilen_cnt)
def validate(self):
"""Validate the results of a previous 'process' run against a second (slower) algorithm.
"""
run_name = self._run_file()
if not os.path.exists(run_name):
raise Exception("Run must be performed before validating")
spats = Spats()
spats.load(run_name)
if spats.validate_results(self.r1, self.r2):
self._add_note("Validation pass")
else:
self._add_note("Validation FAILURE")
def _dump_mut_counts(self):
run_name = self._run_file()
if not os.path.exists(run_name):
raise Exception("Run must be run before attempting dump")
spats = Spats()
spats.load(run_name)
countinfo = spats.counters.counts_dict()
mut_cnts = []
for muts in sorted([int(k.split('_')[-1]) for k in countinfo.keys() if k.startswith('mut_count_')]):
mut_cnts.append((muts, countinfo["mut_count_{}".format(muts)]))
output_path = os.path.join(self.path, 'mut_counts.csv')
self._write_csv(output_path, [ "Mutation Count", "Reads" ], mut_cnts)
mut_cnts = []
for muts in sorted([int(k.split('_')[-1]) for k in countinfo.keys() if k.startswith('mapped_mut_count_')]):
mut_cnts.append((muts, countinfo["mapped_mut_count_{}".format(muts)]))
output_path = os.path.join(self.path, 'mapped_mut_counts.csv')
self._write_csv(output_path, [ "Mutation Count", "Reads" ], mut_cnts)
def _dump_prefixes(self):
run_name = self._run_file()
if not os.path.exists(run_name):
raise Exception("Run must be run before attempting dump")
spats = Spats()
spats.load(run_name)
countinfo = spats.counters.counts_dict()
total = float(countinfo['total_pairs']) / 100.0
for mask in spats.run.masks:
prefixes = []
keyprefix = "prefix_{}_".format(mask)
for key in sorted([k for k in countinfo.keys() if k.startswith(keyprefix)], key = lambda k : countinfo[k], reverse = True):
prefixes.append((key[len(keyprefix):], float(countinfo[key]) / total, countinfo[key]))
output_path = os.path.join(self.path, 'prefixes_{}.csv'.format(mask))
self._write_csv(output_path, [ "Tag", "Percentage", "Count" ], prefixes)
total = float(countinfo['registered_pairs']) / 100.0
for mask in spats.run.masks:
prefixes = []
keyprefix = "mapped_prefix_{}_".format(mask)
for key in sorted([k for k in countinfo.keys() if k.startswith(keyprefix)], key = lambda k : countinfo[k], reverse = True):
prefixes.append((key[len(keyprefix):], float(countinfo[key]) / total, countinfo[key]))
output_path = os.path.join(self.path, 'mapped_prefixes_{}.csv'.format(mask))
self._write_csv(output_path, [ "Tag", "Percentage", "Count" ], prefixes)
def _dump_run(self):
run_name = self._run_file()
if not os.path.exists(run_name):
raise Exception("Run must be run before attempting dump")
spats = Spats()
spats.load(run_name)
profiles = spats.compute_profiles()
mutations = spats.run.count_mutations
indels = spats.run.handle_indels
headers = [ "L", "site", "nt", "f+", "f-" ]
if indels:
headers += [ "ins+", "ins-", "del+", "del-" ]
if mutations:
headers += [ "mut+", "mut-", "beta", "mu", "r" ]
else:
headers += [ "beta", "theta", "rho" ]
headers += [ "c", "c alt" ]
data = []
if self.cotrans:
tgt = spats.targets.targets[0]
tseq = tgt.seq
for key in profiles.cotrans_keys():
end = int(key.split('_')[-1])
prof = profiles.profilesForTargetAndEnd(tgt.name, end)
for i in xrange(end + 1):
datapt = [ end, i, tseq[i - 1] if i else '*', prof.treated[i], prof.untreated[i] ]
if indels:
datapt += [ prof.treated_inserts[i], prof.untreated_inserts[i], prof.treated_deletes[i], prof.untreated_deletes[i] ]
if mutations:
datapt += [ prof.treated_muts[i], prof.untreated_muts[i], prof.beta[i], prof.mu[i], prof.r_mut[i] ]
else:
datapt += [ prof.beta[i], prof.theta[i], prof.rho[i] ]
datapt += [ prof.c, prof.c_alt ]
data.append(datapt)
output_path = os.path.join(self.path, '{}.csv'.format(tgt.name))
self._write_csv(output_path, headers, data)
empty_cell = ''
keys = [ 'treated', 'untreated' ]
if indels:
keys += [ 'treated_inserts', 'untreated_inserts', 'treated_deletes', 'untreated_deletes' ]
if mutations:
keys += [ 'treated_mut', 'untreated_mut', 'beta', 'mu', 'r' ]
else:
keys += [ 'beta', 'theta', 'rho' ]
cotrans_keys = profiles.cotrans_keys()
for key in keys:
ncols = 0
mat = []
for pkey in cotrans_keys:
end = int(pkey.split('_')[-1])
prof = profiles.profilesForTargetAndEnd(tgt.name, end)
vals = getattr(prof, key)
if not ncols:
ncols = len(cotrans_keys) + len(vals)
if len(vals) < ncols:
vals += ([empty_cell] * (ncols - len(vals)))
mat.append(vals)
self._write_csv('{}_{}_mat.csv'.format(tgt.name, key), None, mat)
else:
for tgt in spats.targets.targets:
tseq = tgt.seq
end = len(tgt.seq)
prof = profiles.profilesForTarget(tgt)
data = []
for i in xrange(end + 1):
datapt = [ end, i, tseq[i - 1] if i else '*', prof.treated[i], prof.untreated[i] ]
if indels:
datapt += [ prof.treated_inserts[i], prof.untreated_inserts[i], prof.treated_deletes[i], prof.untreated_deletes[i] ]
if mutations:
datapt += [ prof.treated_muts[i], prof.untreated_muts[i], prof.beta[i], prof.mu[i], prof.r_mut[i] ]
else:
datapt += [ prof.beta[i], prof.theta[i], prof.rho[i] ]
datapt += [ prof.c, prof.c_alt ]
data.append(datapt)
output_path = os.path.join(self.path, '{}.csv'.format(tgt.name))
self._write_csv(output_path, headers, data)