def get_systems_by_name(self, sysnames, keep_data=False):
co_list = {}
db_list = []
output = collections.OrderedDict()
# Weed out fake systems first
for s in sysnames:
coords_data = util.parse_coords(s)
if coords_data is not None:
cx, cy, cz, name = coords_data
co_list[s] = system.System(cx, cy, cz, name)
else:
db_list.append(s)
# Now query for the real ones
db_result = {}
if any(db_list):
result = self._backend.get_systems_by_name(db_list)
db_result = {
r.name.lower(): r
for r in [_make_known_system(t, keep_data) for t in result]
}
for s in sysnames:
if s.lower() in db_result:
output[s] = db_result[s.lower()]
elif s in co_list:
output[s] = co_list[s]
else:
output[s] = None
return output
def main():
import argparse
parser = argparse.ArgumentParser()
parser.add_argument('identifier', help='Identifier name e.g. cne.100899.FST.')
parser.add_argument('-f', '--file', nargs='?', type=argparse.FileType('r'), default=sys.stdin,
help='Input file e.g. d2z.dat.')
parser.add_argument('--extra_data', type=argparse.FileType('r'), default='hg18.toDanRer5.seqs.txt',
help='Extra data file e.g. hg18.toDanRer5.seqs.txt.')
parser.add_argument('--valid', type=int, default=9, help='Field number of valid test coordinates.')
OPTS = parser.parse_args()
cne_dict = parse_dat(read_fields(f=OPTS.file))
line_tups = read_fields(f=OPTS.extra_data)
cne_valids = {}
for l in line_tups:
danrer_co, valid_co = parse_coords(l[4]), parse_coords(l[OPTS.valid-1])
valid_indices = (valid_co['start'] - danrer_co['start'], valid_co['end'] - danrer_co['start'])
cne_valids[l[0]] = valid_indices
plot_cne(OPTS.identifier, cne_dict[OPTS.identifier], cne_valids[OPTS.identifier])
def get_system_by_name(self, sysname, keep_data=False):
# Check the input against the "fake" system format of "[123.4,56.7,-89.0]"...
coords_data = util.parse_coords(sysname)
if coords_data is not None:
cx, cy, cz, name = coords_data
return system.System(cx, cy, cz, name)
else:
result = self._backend.get_system_by_name(sysname)
if result is not None:
return _make_known_system(result, keep_data)
else:
return None
def row_search(OPTS, m, line_tups):
"""Writes the scan results to a pickle file.
The pickled object is a dictionary with cne names as keys and list as value.
Each list contains tuples of (coord, index) where coord is a coord dict
and index is int index starting from 0.
"""
rows = {}
for i, l in enumerate(line_tups):
name, a, b, c, = l[0], l[OPTS.a-1], l[OPTS.b-1], parse_coords(l[OPTS.c-1])
m.fit([a])
hits = m.scan([b], n=None, reverse_complement=True, sort=False)[0]
scores = [score for index, score in hits]
rows[name] = scores
progress(50, (float(i)+1)/len(line_tups)*100, pre="Processing genes")
pkl_filename = OPTS.model + '.pkl'
sys.stdout.write("Writing pickle file %s. Do not exit!\n" % pkl_filename)
pkl = open(OPTS.model + '.pkl', 'wb')
pickle.dump(rows, pkl, pickle.HIGHEST_PROTOCOL)
