def test_call(self):
human_chrom = CoordSystem('chromosome',
core_db=human.CoreDb,
species='human')
human_contig = CoordSystem(1, species='human')
self.assertEqual(human_chrom.coord_system_id, 4)
self.assertEqual(human_contig.name, 'contig')
self.assertEqual(human_contig.attr, 'default_version, sequence_level')
def _set_species_feature_levels(self, species, core_db, feature_types,
var_db, otherfeature_db):
if species not in self._species_feature_levels:
self._species_feature_levels[species] = {}
self._species_feature_dbs[species] = []
coord_system = CoordSystem(core_db=core_db)
if set(feature_types).intersection(set(['cpg', 'repeat', 'gene'])):
if 'core_db' not in self._species_feature_dbs[species]:
self._species_feature_dbs[species].append('core_db')
records = self._get_meta_coord_records(core_db)
self._add_species_feature_levels(species, records, 'core',
coord_system)
if 'variation' in feature_types:
if 'var_db' not in self._species_feature_dbs[species]:
self._species_feature_dbs[species].append('var_db')
assert var_db is not None
records = self._get_meta_coord_records(var_db)
self._add_species_feature_levels(species, records, 'var',
coord_system)
if 'est' in feature_types:
if 'otherfeature_db' not in self._species_feature_dbs[species]:
self._species_feature_dbs[species].append('otherfeature_db')
assert otherfeature_db is not None
records = self._get_meta_coord_records(otherfeature_db)
self._add_species_feature_levels(species, records,
'otherfeature', coord_system)
def _get_seq_region_id(self, CoordName):
"""returns the seq_region_id for the provided CoordName"""
seq_region_table = self.CoreDb.getTable('seq_region')
coord_systems = CoordSystem(core_db=self.CoreDb)
coord_system_ids = [k for k in coord_systems if not isinstance(k, str)]
record = sql.select(
[seq_region_table.c.seq_region_id],
sql.and_(seq_region_table.c.name == CoordName,
seq_region_table.c.coord_system_id.in_(coord_system_ids)))
record = asserted_one(record.execute().fetchall())
return record['seq_region_id']
def get_lower_coord_conversion(coord, species, core_db):
coord_system = CoordSystem(species=species, core_db=core_db)
seq_level_coord_type = CoordSystem(species=species,core_db=core_db,
seq_level=True)
query_rank = coord_system[coord.CoordType].rank
seq_level_rank = coord_system[seq_level_coord_type].rank
assemblies = None
for rank in range(query_rank+1, seq_level_rank):
coord_type = None
for key in coord_system.keys():
if coord_system[key].rank == rank:
coord_type = coord_system[key].name
break
if coord_type is None:
continue
assemblies = get_coord_conversion(coord, coord_type, core_db)
if assemblies:
break
return assemblies
def get_lower_coord_conversion(coord, species, core_db):
coord_system = CoordSystem(species=species, core_db=core_db)
seq_level_coord_type = CoordSystem(species=species,
core_db=core_db,
seq_level=True)
query_rank = coord_system[coord.CoordType].rank
seq_level_rank = coord_system[seq_level_coord_type].rank
assemblies = None
for rank in range(query_rank + 1, seq_level_rank):
coord_type = None
for key in coord_system.keys():
if coord_system[key].rank == rank:
coord_type = coord_system[key].name
break
if coord_type is None:
continue
assemblies = get_coord_conversion(coord, coord_type, core_db)
if assemblies:
break
return assemblies
def _get_sequence_from_direct_assembly(coord=None, DEBUG=False):
# TODO clean up use of a coord
genome = coord.genome
# no matter what strand user provide, we get the + sequence first
coord.Strand = 1
species = genome.Species
coord_type = CoordSystem(species=species,
core_db=genome.CoreDb,
seq_level=True)
if DEBUG:
print('Created Coordinate:', coord, coord.EnsemblStart,
coord.EnsemblEnd)
print(coord.CoordType, coord_type)
assemblies = get_coord_conversion(coord, coord_type, genome.CoreDb)
if not assemblies:
raise NoItemError('no assembly for %s' % coord)
dna = genome.CoreDb.getTable('dna')
seqs, positions = [], []
for q_loc, t_loc in assemblies:
assert q_loc.Strand == 1
length = len(t_loc)
# get MySQL to do the string slicing via substr function
query = sql.select([
substr(dna.c.sequence, t_loc.EnsemblStart,
length).label('sequence')
], dna.c.seq_region_id == t_loc.seq_region_id)
record = asserted_one(query.execute().fetchall())
seq = record['sequence']
seq = DNA.makeSequence(seq)
if t_loc.Strand == -1:
seq = seq.rc()
seqs.append(str(seq))
positions.append((q_loc.Start, q_loc.End))
sequence = _assemble_seq(seqs, coord.Start, coord.End, positions)
return sequence
