def get_annotated_cds(cls, protein_sequence_handler, seq_region,
protein_id, cds_list):
cds_strand = cds_list[0]['cds_strand']
protein_id = cds_list[0]['protein_id']
(stable_id, stable_id_version) = protein_id.split(".")
(translation_start,
translation_end) = cls.get_translation_loc(cds_list)
translation = {}
translation['assembly_id'] = cls.ASSEMBLY_ID
translation['assembly_name'] = cls.ASSEMBLY_NAME
translation['stable_id'] = stable_id
translation['stable_id_version'] = stable_id_version
translation['loc_start'] = translation_start
translation['loc_end'] = translation_end
translation['loc_strand'] = cds_strand
translation['loc_region'] = seq_region
translation[
'translation_seq'] = protein_sequence_handler.get_fasta_seq_by_id(
protein_id)
translation['seq_checksum'] = ChecksumHandler.get_seq_checksum(
translation, 'translation_seq')
translation['session_id'] = None
translation['loc_checksum'] = ChecksumHandler.get_location_checksum(
translation)
translation[
'translation_checksum'] = ChecksumHandler.get_translation_checksum(
translation)
return translation
def get_annotated_transcript(cls, sequence_handler, chrom, mRNA_feature):
transcript = {}
# Note we have shifted one base here
transcript['assembly_id'] = cls.ASSEMBLY_ID
transcript['assembly_name'] = cls.ASSEMBLY_NAME
transcript['loc_start'] = str(mRNA_feature.location.start + 1)
transcript['loc_end'] = str(mRNA_feature.location.end)
transcript['loc_strand'] = str(mRNA_feature.location.strand)
transcript['loc_region'] = str(chrom)
stable_id = mRNA_feature.qualifiers['transcript_id'][0]
(transcript_stable_id, transcript_stable_id_version) = stable_id.split(".")
transcript['stable_id'] = transcript_stable_id
transcript['stable_id_version'] = transcript_stable_id_version
# populate biotype for transcript
if transcript_stable_id[0:3] == "NM_":
transcript['biotype'] = "protein_coding"
elif transcript_stable_id[0:3] == "NR_":
transcript['biotype'] = "non_protein_coding"
elif transcript_stable_id[0:3] == "XM_":
transcript['biotype'] = "predicted_protein_coding"
elif transcript_stable_id[0:3] == "XR_":
transcript['biotype'] = "predicted_non_protein_coding"
else:
transcript['biotype'] = None
transcript['session_id'] = None
transcript['transcript_checksum'] = None
transcript['exon_set_checksum'] = None
transcript['loc_checksum'] = ChecksumHandler.get_location_checksum(transcript)
transcript['sequence'] = sequence_handler.get_sequence_by_id(mRNA_feature.qualifiers['transcript_id'][0])
transcript['seq_checksum'] = ChecksumHandler.get_seq_checksum(transcript, 'sequence')
return transcript
def load_release_set(self, assembly_id, session_id, data_release_set=None):
if data_release_set is None:
today = datetime.now().date()
default_config = ConfigHandler().getInstance().get_section_config()
data_release_set = collections.OrderedDict()
data_release_set["shortname"] = default_config["shortname"]
data_release_set["description"] = default_config["description"]
data_release_set["assembly_id"] = str(assembly_id)
data_release_set["release_date"] = str(today)
data_release_set["session_id"] = str(session_id)
data_release_set["source_id"] = default_config["source"]
release_set_checksum = ChecksumHandler.checksum_list(
list(data_release_set.values()))
data_release_set["release_checksum"] = release_set_checksum
insert_release_set = (
"INSERT INTO release_set (shortname, description, assembly_id, release_date, session_id, \
release_checksum, source_id) VALUES \
(%(shortname)s, %(description)s, %(assembly_id)s, %(release_date)s, %(session_id)s, \
X%(release_checksum)s, %(source_id)s)\
ON DUPLICATE KEY UPDATE release_id=LAST_INSERT_ID(release_id)"
)
release_id = self.insert_data(insert_release_set, data_release_set)
return release_id
def get_annotated_exon(cls, seq_region, exon_feature, exon_sequence):
exon = {}
exon['assembly_id'] = cls.ASSEMBLY_ID
exon['assembly_name'] = cls.ASSEMBLY_NAME
exon['loc_start'] = exon_feature["exon_start"]
exon['loc_end'] = exon_feature["exon_end"]
exon['loc_strand'] = exon_feature["exon_strand"]
exon['loc_region'] = str(seq_region)
exon['loc_checksum'] = ChecksumHandler.get_location_checksum(exon)
exon['exon_order'] = exon_feature["exon_order"]
exon['stable_id'] = exon_feature["exon_stable_id"]
exon['stable_id_version'] = exon_feature["exon_stable_id_version"]
exon['session_id'] = None
exon['exon_seq'] = exon_sequence
exon['seq_checksum'] = ChecksumHandler.get_seq_checksum(exon, 'exon_seq')
exon['exon_checksum'] = ChecksumHandler.get_exon_checksum(exon)
return exon
def get_annotated_gene(cls, chrom, gene_feature):
gene = {}
gene['loc_start'] = str(gene_feature.location.start)
gene['loc_end'] = str(gene_feature.location.end)
gene['loc_strand'] = str(gene_feature.location.strand)
gene['loc_region'] = str(chrom)
gene['stable_id'] = cls.parse_qualifiers(gene_feature.qualifiers,
"Dbxref", "GeneID")
gene['stable_id_version'] = 1
gene['assembly_id'] = cls.ASSEMBLY_ID
gene['assembly_name'] = cls.ASSEMBLY_NAME
# make it none for the moment, otherwise you will get integrity exception
hgnc_id = cls.parse_qualifiers(gene_feature.qualifiers, "Dbxref",
"HGNC:HGNC")
if hgnc_id is not None:
hgnc_id = "HGNC:" + hgnc_id
gene['hgnc_id'] = hgnc_id
gene['session_id'] = None
gene['loc_checksum'] = ChecksumHandler.get_location_checksum(gene)
gene['gene_checksum'] = ChecksumHandler.get_gene_checksum(gene)
return gene
def get_annotated_transcript(cls, sequence_handler, chrom, mRNA_feature):
transcript = {}
# Note we have shifted one base here
transcript['assembly_id'] = cls.ASSEMBLY_ID
transcript['assembly_name'] = cls.ASSEMBLY_NAME
transcript['loc_start'] = str(mRNA_feature.location.start + 1)
transcript['loc_end'] = str(mRNA_feature.location.end)
transcript['loc_strand'] = str(mRNA_feature.location.strand)
transcript['loc_region'] = str(chrom)
stable_id = mRNA_feature.qualifiers['transcript_id'][0]
(transcript_stable_id,
transcript_stable_id_version) = stable_id.split(".")
transcript['stable_id'] = transcript_stable_id
transcript['stable_id_version'] = transcript_stable_id_version
transcript['session_id'] = None
transcript['transcript_checksum'] = None
transcript['exon_set_checksum'] = None
transcript['loc_checksum'] = ChecksumHandler.get_location_checksum(
transcript)
transcript['sequence'] = sequence_handler.get_sequence_by_id(
mRNA_feature.qualifiers['transcript_id'][0])
transcript['seq_checksum'] = ChecksumHandler.get_seq_checksum(
transcript, 'sequence')
return transcript
def run(self):
mydb_config = ConfigHandler().getInstance().get_section_config(section_name="DATABASE")
dbh = DatabaseHandler(db_config=mydb_config,
mypool_name="mypool_" + str(self.seq_region))
dbc = dbh.get_connection()
sequence_handler = FastaHandler(self.downloaded_files['fasta'])
print("Loading protein.....")
print(self.downloaded_files['protein'])
protein_sequence_handler = FastaHandler(self.downloaded_files['protein'])
print("Working on Seq region limit " + str(self.seq_region))
gff_handle = open(self.downloaded_files['gff'])
# Chromosome seq level
for rec in GFF.parse(gff_handle, limit_info=self.limits, target_lines=1000):
for gene_feature in rec.features:
# skip regions
if gene_feature.type == "region":
continue
annotated_gene = AnnotationHandler.get_annotated_gene(self.seq_region, gene_feature)
# gene level
annotated_transcripts = []
for mRNA_feature in gene_feature.sub_features:
if 'transcript_id' in mRNA_feature.qualifiers:
transcript_id = mRNA_feature.qualifiers['transcript_id'][0]
else:
continue
refseq_exon_list = []
refseq_exon_order = 1
refseq_cds_list = []
refseq_cds_order = 1
for mRNA_sub_feature in mRNA_feature.sub_features:
refseq_exon_dict = {}
if 'exon' in mRNA_sub_feature.type:
# print("Transcript Has exons" + str(mRNA_sub_feature.id))
refseq_exon_dict['exon_stable_id'] = str(mRNA_sub_feature.id)
refseq_exon_dict['exon_stable_id_version'] = 1 # dummmy version
refseq_exon_dict['exon_order'] = refseq_exon_order
# note that we are shifting one base here
refseq_exon_dict['exon_start'] = str(mRNA_sub_feature.location.start + 1)
refseq_exon_dict['exon_end'] = str(mRNA_sub_feature.location.end)
refseq_exon_dict['exon_strand'] = str(mRNA_sub_feature.location.strand)
refseq_exon_list.append(refseq_exon_dict)
refseq_exon_order += 1
refseq_cds_dict = {}
if 'CDS' in mRNA_sub_feature.type:
refseq_cds_dict['cds_order'] = refseq_cds_order
# note that we are shifting one base here
refseq_cds_dict['cds_start'] = str(mRNA_sub_feature.location.start + 1)
refseq_cds_dict['cds_end'] = str(mRNA_sub_feature.location.end)
refseq_cds_dict['cds_strand'] = str(mRNA_sub_feature.location.strand)
refseq_cds_dict['cds_id'] = str(mRNA_sub_feature.id)
refseq_cds_dict['protein_id'] = str(mRNA_sub_feature.qualifiers['protein_id'][0]) # @IgnorePep8
refseq_cds_list.append(refseq_cds_dict)
refseq_cds_order += 1
annotated_transcript = AnnotationHandler.get_annotated_transcript(sequence_handler,
self.seq_region,
mRNA_feature)
# add sequence and other annotations
annotated_exons = []
if len(refseq_exon_list) > 0:
annotated_exons = AnnotationHandler.get_annotated_exons(sequence_handler, self.seq_region,
transcript_id,
refseq_exon_list)
if annotated_exons is not None and len(annotated_exons) > 0:
exon_set_checksum = ChecksumHandler.get_exon_set_checksum(annotated_exons)
annotated_transcript['exon_set_checksum'] = exon_set_checksum
annotated_transcript['exons'] = annotated_exons
else:
annotated_transcript['exons'] = []
annotated_translation = []
if len(refseq_cds_list) > 0:
protein_id = refseq_cds_list[0]['protein_id']
annotated_translation = AnnotationHandler.get_annotated_cds(protein_sequence_handler,
self.seq_region,
protein_id,
refseq_cds_list)
annotated_transcript['translation'] = annotated_translation
else:
annotated_transcript['translation'] = []
annotated_transcript['transcript_checksum'] = ChecksumHandler.get_transcript_checksum(annotated_transcript) # @IgnorePep8
annotated_transcripts.append(annotated_transcript)
annotated_gene['transcripts'] = annotated_transcripts
feature_object_to_save = {}
feature_object_to_save["gene"] = annotated_gene
if not self.dryrun and annotated_gene is not None and annotated_gene['stable_id'] is not None:
print("About to load gene => " + str(annotated_gene['stable_id']))
feature_handler = FeatureHandler(parent_ids=self.parent_ids, dbc=dbc)
feature_handler.save_features_to_database(feature_object_to_save)
dbc.close()
gff_handle.close()
print("About to write to the status file")
status_dir = self.download_dir + '/' + 'status_logs'
if not os.path.exists(status_dir):
os.makedirs(status_dir)
self.status_file = status_dir + '/' + 'status_file_chr' + str(self.seq_region)
status_handle = open(self.status_file, "w")
status_handle.write("Done")
status_handle.close()