def _collect(self):
match_type_getter = attrgetter('match_type')
groups = collectiontools.groupby(self.spectrum_matches,
match_type_getter)
by_scan_groups = {}
for group, members in groups.items():
acc = []
for by_scan in collectiontools.groupby(
members, lambda x: x.scan.id).values():
scan = by_scan[0].scan
self.spectrum_ids.add(scan.scan_id)
ss = MultiScoreSpectrumSolutionSet(scan, by_scan)
ss.sort()
acc.append(ss)
acc.sort(key=lambda x: x.scan.id)
by_scan_groups[group] = acc
return by_scan_groups
def load(cls, fh, session=None, hypothesis_id=1, fuzzy=True):
site_models = GlycosylationSiteModel.load(fh)
if session is not None:
return cls.bind_to_hypothesis(session, site_models, hypothesis_id,
fuzzy)
by_protein_name = groupby(site_models, lambda x: x.protein_name)
result = []
for name, models in by_protein_name.items():
result.append(cls(ProteinStub(name), models))
return result
def sort_peak_match_pairs(pairs):
series = groupby(pairs, key_fn=lambda x: x.fragment.series)
segments = []
for key in sorted(series, key=lambda x: x.int_code):
segment = series[key]
if hasattr(segment[0].fragment, "position"):
segment.sort(key=lambda x: x.fragment.position)
else:
segment.sort(key=lambda x: x.fragment.mass)
segments.extend(segment)
return segments
def _by_scan_id(self):
acc = []
for by_scan in collectiontools.groupby(self.spectrum_matches,
lambda x: x.scan.id).values():
scan = by_scan[0].scan
self.spectrum_ids.add(scan.scan_id)
ss = MultiScoreSpectrumSolutionSet(scan, by_scan)
ss.sort()
acc.append(ss)
acc.sort(key=lambda x: x.scan.id)
return acc
def _exclusive(self, score_getter=None, min_value=0):
if score_getter is None:
score_getter = attrgetter('score')
groups = collectiontools.groupby(self.spectrum_matches,
lambda x: x.scan.id)
by_match_type = defaultdict(list)
for _scan_id, members in groups.items():
top_match = max(members, key=score_getter)
top_score = score_getter(top_match)
seen = set()
for match in members:
if isclose(top_score, score_getter(match)) and score_getter(
match) > 0 and match.match_type not in seen:
seen.add(match.match_type)
by_match_type[match.match_type].append(match)
for _group_label, matches in by_match_type.items():
matches.sort(key=lambda x: (x.scan.id, score_getter(x)))
return by_match_type