def get_ec_gbk_table_dict(connection):
"""Load the Genbanks_Ecs table into memory and compile a dict.
The the result dict is keyed by EC IDS.
The table contains the current Genbank and EC number relationships in
the local CAZyme db.
:param connection: open sqlalchemy connection to an SQLite db
Return dict {ec_id: {gbk ids}}
"""
with Session(bind=connection) as session:
all_gbk_ec_records = session.query(Genbank, Ec).\
join(Ec, Genbank.ecs).\
all()
ec_gbk_table_dict = {}
for record in all_gbk_ec_records:
genbank_id = record[0].genbank_id
ec_id = record[1].ec_id
try:
ec_gbk_table_dict[ec_id].add(genbank_id)
except KeyError:
ec_gbk_table_dict[ec_id] = {genbank_id}
return ec_gbk_table_dict
def get_gbk_fam_table_dict(connection):
"""Build dict representing the records present in the Genbanks_CazyFamilies table
If a GenBank accession is in the db but not has not CazyFamilies instances related to it,
the GenBank accession is not returned when quering the db.
:param connection: open sqlalchemy connection to an SQLite3 db engine
Return
- dict: {gbk_acc: {'families': {'fam subfam': fam_id}}, 'gbk_id': gbk_id }
- set of tuples: (gbk_id, fam_id), each representing one row in the table
"""
with Session(bind=connection) as session:
all_gbk_fam_records = session.query(Genbank, CazyFamily).\
join(CazyFamily, Genbank.families).\
all()
existing_rel_tuples = set() # set of tuples (gbk_id, fam_id)
gbk_fam_table_dict = {}
# {gbk_acc: {'families': {'fam subfam': fam_id}}, 'gbk_id': gbk_id }
for record in tqdm(all_gbk_fam_records, ' Retreving existing gbk-fam relationships from db'):
gbk_accession = record[0].genbank_accession
gbk_id = record[0].genbank_id
family = record[1].family
if record[1].subfamily is None:
subfamily = '_'
else:
subfamily = record[1].subfamily
fam_id = record[1].family_id
existing_rel_tuples.add( (gbk_id, fam_id) )
try:
gbk_fam_table_dict[gbk_accession]
try:
gbk_fam_table_dict[gbk_accession][f'{family} {subfamily}']
except KeyError:
gbk_fam_table_dict[gbk_accession][f'{family} {subfamily}'] = fam_id
except KeyError:
gbk_fam_table_dict[gbk_accession] = {
'families': {f'{family} {subfamily}': fam_id},
'gbk_id': gbk_id,
}
return gbk_fam_table_dict, existing_rel_tuples
def get_ec_table_dict(connection):
"""Create dict of objects present in the CazyFamilies table.
:param connection: open sqlalchemy db engine connection
Return dict {ec_number: ec_id}
"""
with Session(bind=connection) as session:
db_ec_records = session.query(Ec).all()
ec_table_dict = {} # {ec_number: ec_id}
for record in tqdm(db_ec_records, desc="Retrieving existing EC# records"):
ec_table_dict[record.ec_number] = record.ec_id
return ec_table_dict
def get_pdb_table_dict(connection):
"""Create dict of objects present in the Pdbs table.
:param connection: open sqlalchemy db engine connection
Return dict {pdb_accession: pdb_db_id}
"""
with Session(bind=connection) as session:
db_pdb_records = session.query(Pdb).all()
pdb_table_dict = {} # {pdb_accession: pdb_db_id}
for record in tqdm(db_pdb_records, desc="Loading existing PDB db records"):
pdb_table_dict[record.pdb_accession] = record.pdb_id
return pdb_table_dict
def get_kingdom_table_dict(connection):
"""Load and parse the Kingdoms table from the db and compile a dict {kgnd: id}
:param connection:
Return dict {kingdom: kindom_db_id}
"""
with Session(bind=connection) as session:
kingdom_table = session.query(Kingdom).all()
kingdom_dict = {} # {kingdom: kindom_db_id}
for kingdom_obj in kingdom_table:
kingdom_dict[kingdom_obj.kingdom] = kingdom_obj.kingdom_id
return kingdom_dict
def get_gbk_seq(gbk_dict, query_data, connection):
"""Retrieve GenBank protein sequences for the provided Gbks.
:param gbk_dict: dict of selected GenBank accessions {acc: id}
:param query_data: dict containing all data retrieved from the db
:param connection: open sqlaclchemy connection for an SQLite db
Return query_data: dict containing all data retrieved from the db
"""
logger = logging.getLogger(__name__)
gbk_accessions = list(gbk_dict.keys())
# retrieve the data from the Taxonomy and Kingdom tables
with Session(bind=connection) as session:
gbk_query = session.query(Genbank).\
filter(Genbank.genbank_accession.in_(gbk_accessions)).\
all()
if len(gbk_query) == 0:
logger.warning("No GenBank records retrieved for any of the selected GenBank accessions.")
return query_data
for record in tqdm(gbk_query, desc="Getting GenBank protein sequences"):
gbk_acc = record[0].genbank_accession
seq = record[0].sequence
seq_date = record[0].seq_update_date
try:
query_data[gbk_acc]
try:
logger.warning(
f"Multiple GBK records found for GBK acc {gbk_acc}\n"
"Retreiving only one gbk sequence."
)
query_data[gbk_acc]['gbk_sequence'] = seq
query_data[gbk_acc]['gbk_sequence_date'] = seq_date
except KeyError:
query_data[gbk_acc]['gbk_sequence'] = seq
query_data[gbk_acc]['gbk_sequence_date'] = seq_date
except KeyError:
query_data[gbk_acc] = {'gbk_sequence': seq, 'gbk_sequence_date': seq_date}
return query_data
def get_gbk_kingdom_dict(connection):
"""Compile dict of Genbank, Taxonomy and Kingdom records
:param connection: open sqlalchemy db connection
Return dict {kingdom: {genus: {species: {protein_accessions}}}
"""
with Session(bind=connection) as session:
query_results = session.query(Genbank, Taxonomy, Kingdom).\
join(Taxonomy, (Taxonomy.kingdom_id == Kingdom.kingdom_id)).\
join(Genbank, (Genbank.taxonomy_id == Taxonomy.taxonomy_id)).\
all()
genbank_kingdom_dict = {} # kingdom: {genus: {species: {protein_accessions}}}
for result in tqdm(query_results, desc="Retreving GenBank accessions and taxonomy"):
genbank_accession = result[0].genbank_accession
genus = result[1].genus
species = result[1].species
kingdom = result[2].kingdom
try:
genbank_kingdom_dict[kingdom]
try:
genbank_kingdom_dict[kingdom][genus]
try:
genbank_kingdom_dict[kingdom][genus][species].add(genbank_accession)
except KeyError:
genbank_kingdom_dict[kingdom][genus][species] = {genbank_accession}
except KeyError:
genbank_kingdom_dict[kingdom][genus] = {species: {genbank_accession}}
except KeyError:
genbank_kingdom_dict[kingdom] = {
genus: {
species: {genbank_accession},
},
}
return genbank_kingdom_dict
def get_gbk_table_seq_dict(connection):
"""Compile a dict of the data in the Genbanks table
:param connection: open connection to an SQLite3 database
Return dict {genbank_accession: 'sequence': str, 'seq_date': str}
"""
with Session(bind=connection) as session:
all_genbank = session.query(Genbank).all()
db_gbk_dict = {} # {genbank_accession: 'sequence': str, 'seq_date': str}
for gbk in all_genbank:
db_gbk_dict[f"{gbk.genbank_accession}"] = {
'sequence': gbk.sequence,
'seq_date': gbk.seq_update_date
}
return db_gbk_dict
def get_gbk_table_dict(connection):
"""Compile a dict of the data in the Genbanks table
:param connection: open connection to an SQLite3 database
Return dict {genbank_accession: 'taxa_id': int, 'gbk_id': int}
"""
with Session(bind=connection) as session:
all_genbank = session.query(Genbank).all()
db_gbk_dict = {} # {genbank_accession: 'taxa_id': str, 'id': int}
for gbk in all_genbank:
db_gbk_dict[f"{gbk.genbank_accession}"] = {
'taxa_id': gbk.taxonomy_id,
'gbk_id': gbk.genbank_id
}
return db_gbk_dict
def apply_ec_filters(
current_gbk_objs,
ec_filters,
connection,
):
"""Apply EC number filter to the retrieved Genbank records.
:param current_gbk_objs: list of db Genbank objs retrieved from the db
:param ec_filters: set of EC numbers to limit the retrieval of data to
:param connection: open sqlaclchemy connection for an SQLite db
Return set of db Genbank objects.
"""
logger = logging.getLogger(__name__)
ec_gbk_ids = set()
# Retrieve all Genbank.genbank_ids for each EC number
for ec in tqdm(ec_filters, desc="Retrieving gbks for EC# filters"):
with Session(bind=connection) as session:
gbk_query = session.query(Genbank.genbank_id).\
join(Ec, Genbank.ecs).\
filter(Ec.ec_number == ec).\
all()
for gbk_id in gbk_query:
ec_gbk_ids.add(gbk_id)
if len(ec_gbk_ids) == 0:
logger.error(
"Retrieved NO proteins matching the provided EC numbers\n"
"Check the local CAZyme db contains the EC numbers provided\n"
"Terminating program")
sys.exit(1)
ec_filtered_gbks = set()
for gbk_record in tqdm(current_gbk_objs,
desc="Checking gbk records against EC filters"):
if (gbk_record.genbank_id, ) in ec_gbk_ids:
ec_filtered_gbks.add(gbk_record)
return ec_filtered_gbks
def get_ids(genbank_accessions, connection):
"""Get the local CAZyme database IDs for the list of provided GenBank accessions.
:param genbank_accessions: set of GenBank accessions
:param connection: open sqlalchemy engine connection
Return dict, keyed by GenBank accession and valued by database record ID.
"""
gbk_dict = {}
for accession in tqdm(genbank_accessions,
desc="Getting local db record IDs"):
with Session(bind=connection) as session:
gbk_query = session.query(Genbank).\
filter(Genbank.genbank_accession == accession).\
first()
gbk_dict[accession] = gbk_query.genbank_id
return gbk_dict
def get_ec_annotations(gbk_dict, query_data, connection):
"""Retrieve EC number annotations for the provided Gbks.
:param gbk_dict: dict of selected GenBank accessions {acc: id}
:param query_data: dict containing all data retrieved from the db
:param connection: open sqlaclchemy connection for an SQLite db
Return query_data: dict containing all data retrieved from the db
"""
logger = logging.getLogger(__name__)
gbk_accessions = list(gbk_dict.keys())
# retrieve the data from the Taxonomy and Kingdom tables
with Session(bind=connection) as session:
ec_query = session.query(Genbank, Ec).\
join(Ec, Genbank.ecs).\
filter(Genbank.genbank_accession.in_(gbk_accessions)).\
all()
if len(ec_query) == 0:
logger.warning("No EC annotations retrieved for any of the selected GenBank accessions.")
return query_data
for record in tqdm(ec_query, desc="Getting EC number annotations"):
gbk_acc = record[0].genbank_accession
ec_number = record[1].ec_number
try:
query_data[gbk_acc]
try:
query_data[gbk_acc]['ec_numbers'].add(ec_number)
except KeyError:
query_data[gbk_acc]['ec_numbers'] = {ec_number}
except KeyError:
query_data[gbk_acc] = {'ec_numbers': {ec_number}}
return query_data
def get_pdb_accessions(gbk_dict, query_data, connection):
"""Retrieve PDB accessions for the provided Gbks.
:param gbk_dict: dict of selected GenBank accessions {acc: id}
:param query_data: dict containing all data retrieved from the db
:param connection: open sqlaclchemy connection for an SQLite db
Return query_data: dict containing all data retrieved from the db
"""
logger = logging.getLogger(__name__)
gbk_accessions = list(gbk_dict.keys())
# retrieve the data from the Taxonomy and Kingdom tables
with Session(bind=connection) as session:
pdb_query = session.query(Genbank, Pdb).\
join(Pdb, Genbank.pdbs).\
filter(Genbank.genbank_accession.in_(gbk_accessions)).\
all()
if len(pdb_query) == 0:
logger.warning("No PDB accessions retrieved for any of the selected GenBank accessions.")
return query_data
for record in tqdm(pdb_query, desc="Getting PDB accessions"):
gbk_acc = record[0].genbank_accession
pdb_accession = record[1].pdb_accession
try:
query_data[gbk_acc]
try:
query_data[gbk_acc]['pdb_accessions'].add(pdb_accession)
except KeyError:
query_data[gbk_acc]['pdb_accessions'] = {pdb_accession}
except KeyError:
query_data[gbk_acc] = {'pdb_accessions': {pdb_accession}}
return query_data
def get_fams_table_dict(connection):
"""Create dict of objects present in the CazyFamilies table.
:param connection: open sqlalchemy db engine connection
Return dict {family subfamily: db_family_id}
"""
with Session(bind=connection) as session:
all_families = session.query(CazyFamily).all()
db_fam_dict = {}
for fam in all_families:
if fam.subfamily is None:
subfam = '_'
else:
subfam = fam.subfamily
db_fam_dict[f"{fam.family} {subfam}"] = fam.family_id
return db_fam_dict
def get_uniprot_table_dict(connection):
"""Create dict of objects present in the Uniprots table.
:param connection: open sqlalchemy db engine connection
Return dict {acc: {name: str, gbk_id: int, seq: str, seq_date:str } }
"""
with Session(bind=connection) as session:
db_uniprot_records = session.query(Uniprot).all()
uniprot_table_dict = {} # {acc: {name: str, gbk_id: int, seq: str, seq_date:str } }
for record in tqdm(db_uniprot_records, desc="Retrieving existing UniProt records from db"):
uniprot_table_dict[record.uniprot_accession] = {
"name": record.uniprot_name,
"genbank_id": record.genbank_id,
"seq": record.sequence,
"seq_date": record.seq_update_date,
}
return uniprot_table_dict
def get_taxs_table_dict(connection):
"""Create dict of objects present in the Taxs table.
:param connection: open sqlalchemy db engine connection
Return dict {genus species: {'tax_id': db_tax_id, 'kingdom_id': kingdom_id}
"""
with Session(bind=connection) as session:
all_taxa = session.query(Taxonomy).all()
db_tax_dict = {}
for taxa in all_taxa:
if len(taxa.species) == 0:
db_tax_dict[f"{taxa.genus}"] = {
'tax_id': taxa.taxonomy_id,
'kingdom_id': taxa.kingdom_id,
}
else:
db_tax_dict[f"{taxa.genus} {taxa.species}"] = {
'tax_id': taxa.taxonomy_id,
'kingdom_id': taxa.kingdom_id,
}
return db_tax_dict
def get_gbk_pdb_table_dict(connection):
"""Create dict of objects present in the Genbanks_Pdbs table.
:param connection: open sqlalchemy db engine connection
Return dict {gbk_db_id: {pdb_db_id} }
"""
with Session(bind=connection) as session:
all_gbk_pdb_records = session.query(Genbank, Pdb).\
join(Pdb, Genbank.pdbs).\
all()
gbk_pdb_table_dict = {} # {pdb_accession: pdb_db_id}
for record in tqdm(all_gbk_pdb_records, desc="Loading existing Genbank_Pdbs db records"):
genbank_id = record[0].genbank_id
pdb_id = record[1].pdb_id
try:
gbk_pdb_table_dict[genbank_id].add(pdb_id)
except KeyError:
gbk_pdb_table_dict[genbank_id] = {pdb_id}
return gbk_pdb_table_dict
def get_tax_annotations(gbk_dict, query_data, connection, args):
"""Retrieve kingdom, genus and/or scientific name of the source organism for the provided Gbks.
:param gbk_dict: dict of selected GenBank accessions {acc: id}
:param query_data: dict containing all data retrieved from the db
:param connection: open sqlaclchemy connection for an SQLite db
:param args: cmd-line args parser
Return query_data: dict containing all data retrieved from the db
"""
logger = logging.getLogger(__name__)
gbk_accessions = list(gbk_dict.keys())
# retrieve the data from the Taxonomy and Kingdom tables
with Session(bind=connection) as session:
tax_query = session.query(Genbank, Taxonomy, Kingdom).\
join(Taxonomy, (Taxonomy.kingdom_id == Kingdom.kingdom_id)).\
join(Genbank, (Genbank.taxonomy_id == Taxonomy.taxonomy_id)).\
filter(Genbank.genbank_accession.in_(gbk_accessions)).\
all()
if len(tax_query) == 0:
logger.warning("No taxonomy data retrieved for any of the selected GenBank accessions.")
return query_data
for record in tqdm(tax_query, desc="Getting taxonomy data"):
gbk_acc = record[0].genbank_accession
if 'kingdom' in args.include:
kingdom = record[2].kingdom
try:
query_data[gbk_acc]
try:
query_data[gbk_acc]['kingdom']
logger.warning(
f"Multiple taxa found for {gbk_acc}\n"
"Retreiving only one record."
)
query_data[gbk_acc]['kingdom'] = kingdom
except KeyError:
query_data[gbk_acc]['kingdom'] = kingdom
except KeyError:
query_data[gbk_acc] = {'kingdom': kingdom}
if 'genus' in args.include:
genus = record[1].genus
try:
query_data[gbk_acc]
try:
query_data[gbk_acc]['genus']
logger.warning(
f"Multiple taxa found for {gbk_acc}\n"
"Retreiving only one record."
)
query_data[gbk_acc]['genus'] = genus
except KeyError:
query_data[gbk_acc]['genus'] = genus
except KeyError:
query_data[gbk_acc] = {'genus': genus}
if 'organism' in args.include:
genus = record[1].genus
species = record[1].species
organism = f"{genus} {species}"
try:
query_data[gbk_acc]
try:
query_data[gbk_acc]['organism']
logger.warning(
f"Multiple taxa found for {gbk_acc}\n"
"Retreiving only one record."
)
query_data[gbk_acc]['organism'] = organism
except KeyError:
query_data[gbk_acc]['organism'] = organism
except KeyError:
query_data[gbk_acc] = {'organism': organism}
return query_data
def get_class_fam_genbank_accessions(
class_filters,
family_filters,
connection,
):
"""Retrieve the GenBank accessions of proteins from user selected CAZy classes and (sub)families
:param class_filters: set of CAZy classes to retrieve data for
:param family_filters: set of CAZy families to retrieve data for
:param connection: open sqlaclchemy connection for an SQLite db
Return list of db objects containing a Genbank obj, Taxonomy obj and Kingdom obj.
"""
logger = logging.getLogger(__name__)
initially_selected_gbk = []
if len(class_filters) == 0 and len(family_filters) == 0:
logger.warning("No class or family filters applied")
# could retrieve all GenBank accessions
with Session(bind=connection) as session:
gbk_query = session.query(Genbank, Taxonomy, Kingdom).\
join(Taxonomy, (Taxonomy.kingdom_id == Kingdom.kingdom_id)).\
join(Genbank, (Genbank.taxonomy_id == Taxonomy.taxonomy_id)).\
join(CazyFamily, Genbank.families).\
all()
initially_selected_gbk = gbk_query
return initially_selected_gbk
if len(class_filters) != 0:
logger.warning("Applying CAZy class filter(s)")
for cazy_class in tqdm(
class_filters,
desc="Retrieving GenBank accessions for selected CAZy classes"):
class_abbrev = CLASS_ABBREVIATIONS[cazy_class]
# perform a subquery to retrieve all CAZy families in the CAZy class
inner_stmt = select(CazyFamily.family).where(
CazyFamily.family.like(f'{class_abbrev}%'))
subq = inner_stmt.subquery()
aliased_families = aliased(CazyFamily, subq)
stmt = select(aliased_families)
# perform query to retrieve proteins in the CAZy families
with Session(bind=connection) as session:
gbk_query = session.query(Genbank, Taxonomy, Kingdom).\
join(Taxonomy, (Taxonomy.kingdom_id == Kingdom.kingdom_id)).\
join(Genbank, (Genbank.taxonomy_id == Taxonomy.taxonomy_id)).\
join(CazyFamily, Genbank.families).\
filter(CazyFamily.family.in_(stmt)).\
all()
initially_selected_gbk += gbk_query
if len(family_filters) != 0:
logger.warning("Applying CAZy family filter(s)")
for cazy_family in tqdm(
family_filters,
desc="Retrieving GenBank accessions for selected CAZy families"):
inner_stmt = select(
CazyFamily.family).where(CazyFamily.family == cazy_family)
subq = inner_stmt.subquery()
aliased_families = aliased(CazyFamily, subq)
stmt = select(aliased_families)
if cazy_family.find('_') != -1: # subfamily
with Session(bind=connection) as session:
gbk_query = session.query(Genbank, Taxonomy, Kingdom).\
join(Taxonomy, (Taxonomy.kingdom_id == Kingdom.kingdom_id)).\
join(Genbank, (Genbank.taxonomy_id == Taxonomy.taxonomy_id)).\
join(CazyFamily, Genbank.families).\
filter(CazyFamily.subfamily.in_(stmt)).\
all()
else:
with Session(bind=connection) as session:
gbk_query = session.query(Genbank, Taxonomy, Kingdom).\
join(Taxonomy, (Taxonomy.kingdom_id == Kingdom.kingdom_id)).\
join(Genbank, (Genbank.taxonomy_id == Taxonomy.taxonomy_id)).\
join(CazyFamily, Genbank.families).\
filter(CazyFamily.family.in_(stmt)).\
all()
initially_selected_gbk += gbk_query
return list(set(initially_selected_gbk))
def get_uniprot_data(gbk_dict, query_data, connection, args):
"""Retrieve UniProt data for the provided Gbks.
:param gbk_dict: dict of selected GenBank accessions {acc: id}
:param query_data: dict containing all data retrieved from the db
:param connection: open sqlaclchemy connection for an SQLite db
:param args: cmd-line args parser
Return query_data: dict containing all data retrieved from the db
"""
logger = logging.getLogger(__name__)
gbk_accessions = list(gbk_dict.keys())
# retrieve the data from the Taxonomy and Kingdom tables
with Session(bind=connection) as session:
uniprot_query = session.query(Genbank, Uniprot).\
join(Uniprot, (Uniprot.genbank_id == Genbank.genbank_id)).\
filter(Genbank.genbank_accession.in_(gbk_accessions)).\
all()
if len(uniprot_query) == 0:
logger.warning("No UniProt records retrieved for any of the selected GenBank accessions.")
return query_data
for record in tqdm(uniprot_query, desc="Getting UniProt data"):
gbk_acc = record[0].genbank_accession
if 'uniprot_acc' in args.include:
uniprot_accession = record[1].uniprot_accession
try:
query_data[gbk_acc]
try:
query_data[gbk_acc]['uniprot_accession']
logger.warning(
f"Multiple UniProt records found for GBK acc {gbk_acc}\n"
"Retreiving only one."
)
query_data[gbk_acc]['uniprot_accession'] = uniprot_accession
except KeyError:
query_data[gbk_acc]['uniprot_accession'] = uniprot_accession
except KeyError:
query_data[gbk_acc] = {
'uniprot_accession': uniprot_accession,
}
if 'uniprot_name' in args.include:
uniprot_name = record[1].uniprot_name
try:
query_data[gbk_acc]
try:
query_data[gbk_acc]['uniprot_name']
logger.warning(
f"Multiple UniProt records found for GBK acc {gbk_acc}\n"
"Retreiving only one."
)
query_data[gbk_acc]['uniprot_name'] = uniprot_name
except KeyError:
query_data[gbk_acc]['uniprot_name'] = uniprot_name
except KeyError:
query_data[gbk_acc] = {
'uniprot_name': uniprot_name,
}
if 'uniprot_seq' in args.include:
seq = record[1].sequence
seq_date = record[1].seq_update_date
try:
query_data[gbk_acc]
try:
query_data[gbk_acc]['uniprot_sequence']
logger.warning(
f"Multiple UniProt records found for GBK acc {gbk_acc}\n"
"Retreiving only one record."
)
query_data[gbk_acc]['uniprot_sequence'] = seq
query_data[gbk_acc]['uniprot_sequence_date'] = seq_date
except KeyError:
query_data[gbk_acc]['uniprot_sequence'] = seq
query_data[gbk_acc]['uniprot_sequence_date'] = seq_date
except KeyError:
query_data[gbk_acc] = {'sequence': seq, 'sequence_date': seq_date}
return query_data
def get_class_fam_annotations(gbk_dict, query_data, connection, args):
"""Retrieve CAZy class and/or family annotations for the provided Gbks.
:param gbk_dict: dict of selected GenBank accessions {acc: id}
:param query_data: dict containing all data retrieved from the db
:param connection: open sqlaclchemy connection for an SQLite db
:param args: cmd-line args parser
Return query_data: dict containing all data retrieved from the db
"""
logger = logging.getLogger(__name__)
gbk_accessions = list(gbk_dict.keys())
# retrieve the data from the CAZy Family table
with Session(bind=connection) as session:
fam_table_query = session.query(Genbank, CazyFamily).\
join(CazyFamily, Genbank.families).\
filter(Genbank.genbank_accession.in_(gbk_accessions)).\
all()
if len(fam_table_query) == 0:
logger.warning(
"No CAZy class/family annotations retrieved for any of the selected "
"GenBank accessions."
)
return query_data
for record in tqdm(fam_table_query, desc="Getting CAZy class/family annotations"):
gbk_acc = record[0].genbank_accession
if 'class' in args.include:
fam = record[1].family
cazy_class = re.match(r"\D{2,3}\d", fam).group()[:-1]
try:
query_data[gbk_acc]
try:
query_data[gbk_acc]['class'].add(cazy_class)
except KeyError:
query_data[gbk_acc]['class'] = {cazy_class}
except KeyError:
query_data[gbk_acc] = {'class': {cazy_class}}
if 'family' in args.include:
fam = record[1].family
try:
query_data[gbk_acc]
try:
query_data[gbk_acc]['family'].add(fam)
except KeyError:
query_data[gbk_acc]['family'] = {fam}
except KeyError:
query_data[gbk_acc] = {'family': {fam}}
if 'subfamily' in args.include:
subfam = record[1].subfamily
try:
query_data[gbk_acc]
try:
query_data[gbk_acc]['subfamily'].add(subfam)
except KeyError:
query_data[gbk_acc]['subfamily'] = {subfam}
except KeyError:
query_data[gbk_acc] = {'subfamily': {subfam}}
return query_data
