def get_gff3_features(gff3_file, assemblies=None):
'''
Parses the passed GFF3 file and returns two dicts, loaded with biocode.biothings objects:
1. The first dict are the Assembly objects, keyed on assembly ID. Each Assembly has all of the
children populated, so you can fully recover gene, RNA, exon and CDS features iterating on
the assembly.
2. The second dist is a flat structure of all the descendent feature objects of the Assemblies
keyed by the feature IDs.
See the documentation for each feature type in biocode.biothings for more info
'''
if assemblies is None:
assemblies = dict()
features = dict()
# these are related to parsing any embedded FASTA
in_fasta_section = False
is_assembly_fasta = False
current_fasta_id = None
lnum = 0
FASTA_RE = re.compile(r'^\#\#FASTA\s*$')
for line in open(gff3_file):
lnum = lnum + 1
if in_fasta_section == True:
m = re.search('>(\S+)\s*(.*)', line)
if m:
current_fasta_id = m.group(1)
if current_fasta_id in assemblies:
is_assembly_fasta = True
else:
is_assembly_fasta = False
else:
if current_fasta_id is None:
if (len(str(line.rstrip())) > 0):
raise Exception("FASTA parse error - sequence appears without preceding fasta id at line " + str(lnum))
if is_assembly_fasta == True:
# must be a sequence line for an assembly
# python 2.6+ makes string concatenation amortized O(n)
# http://stackoverflow.com/a/4435752/1368079
assemblies[current_fasta_id].residues += str(line.rstrip())
assemblies[current_fasta_id].length = len( assemblies[current_fasta_id].residues )
continue
elif FASTA_RE.match(line):
# all data to the end of the file must be FASTA
in_fasta_section = True
continue
# ignore all other comments
if line.startswith('#'):
continue
cols = line.split("\t")
if len(cols) != 9:
continue
mol_id = cols[0]
# initialize this assembly if we haven't seen it yet
if mol_id not in assemblies:
assemblies[mol_id] = biothings.Assembly( id=mol_id, residues='' )
current_assembly = assemblies[mol_id]
rfmin = int(cols[3]) - 1
rfmax = int(cols[4])
rstrand = None
atts = column_9_dict(cols[8])
feat_id = atts.get('ID')
parent_id = atts.get('Parent')
parent_feat = None
# sanity check
if rfmin > rfmax:
raise Exception("ERROR: Coordinates in GFF for feature id {0} appear to be reversed and violate GFF3 specification: {1} > {2}".format(feat_id, cols[3], cols[4]))
if 'locus_tag' in atts:
locus_tag = atts['locus_tag']
else:
locus_tag = None
# shared features are not yet supported
if isinstance(parent_id, list):
raise Exception("This line contains a shared feature with multiple parents. This isn't yet supported:\n{0}".format(line))
if parent_id is not None:
if parent_id in features:
parent_feat = features[parent_id]
else:
raise Exception("Error in GFF3: Parent {0} referenced by a child feature before it was defined".format(parent_id) )
if cols[6] == '-':
rstrand = -1
elif cols[6] == '+':
rstrand = 1
else:
rstrand = 0
phase = cols[7]
if cols[2] == 'gene':
gene = biothings.Gene(id=feat_id, locus_tag=locus_tag)
gene.locate_on(target=current_assembly, fmin=rfmin, fmax=rfmax, strand=rstrand)
features[feat_id] = gene
current_assembly.add_gene(gene)
elif cols[2] == 'mRNA':
mRNA = biothings.mRNA(id=feat_id, parent=parent_feat, locus_tag=locus_tag)
mRNA.locate_on(target=current_assembly, fmin=rfmin, fmax=rfmax, strand=rstrand)
parent_feat.add_mRNA(mRNA)
features[feat_id] = mRNA
elif cols[2] == 'rRNA':
rRNA = biothings.rRNA(id=feat_id, parent=parent_feat, locus_tag=locus_tag)
rRNA.locate_on(target=current_assembly, fmin=rfmin, fmax=rfmax, strand=rstrand)
parent_feat.add_rRNA(rRNA)
features[feat_id] = rRNA
elif cols[2] == 'tRNA':
tRNA = biothings.tRNA(id=feat_id, parent=parent_feat, locus_tag=locus_tag)
tRNA.locate_on(target=current_assembly, fmin=rfmin, fmax=rfmax, strand=rstrand)
parent_feat.add_tRNA(tRNA)
features[feat_id] = tRNA
elif cols[2] == 'exon':
exon = biothings.Exon(id=feat_id, parent=parent_feat)
exon.locate_on(target=current_assembly, fmin=rfmin, fmax=rfmax, strand=rstrand)
parent_feat.add_exon(exon)
features[feat_id] = exon
elif cols[2] == 'CDS':
if phase == '.':
phase = 0
else:
phase = int(phase)
CDS = biothings.CDS(id=feat_id, parent=parent_feat, phase=phase)
CDS.locate_on(target=current_assembly, fmin=rfmin, fmax=rfmax, strand=rstrand, phase=phase)
parent_feat.add_CDS(CDS)
features[feat_id] = CDS
elif cols[2] == 'polypeptide':
polypeptide = biothings.Polypeptide(id=feat_id, parent=parent_feat)
polypeptide.locate_on(target=current_assembly, fmin=rfmin, fmax=rfmax, strand=rstrand)
parent_feat.add_polypeptide(polypeptide)
polypeptide.annotation = parse_annotation_from_column_9(cols[8])
features[feat_id] = polypeptide
elif cols[2] == 'five_prime_UTR':
utr = biothings.FivePrimeUTR(id=feat_id, parent=parent_feat)
utr.locate_on(target=current_assembly, fmin=rfmin, fmax=rfmax, strand=rstrand)
parent_feat.add_five_prime_UTR(utr)
features[feat_id] = utr
elif cols[2] == 'three_prime_UTR':
utr = biothings.ThreePrimeUTR(id=feat_id, parent=parent_feat)
utr.locate_on(target=current_assembly, fmin=rfmin, fmax=rfmax, strand=rstrand)
parent_feat.add_three_prime_UTR(utr)
features[feat_id] = utr
else:
sys.stderr.write( "Skipping feature {0} with type {1}\n".format(feat_id, cols[2]) )
continue
features[feat_id].length = rfmax - rfmin
return (assemblies, features)
def get_gff3_features(gff3_file):
'''
Parses the passed GFF3 file and returns two dicts, loaded with biocode.biothings objects:
1. The first dict are the Assembly objects, keyed on assembly ID. Each Assembly has all of the
children populated, so you can fully recover gene, RNA, exon and CDS features iterating on
the assembly.
2. The second dist is a flat structure of all the descendent feature objects of the Assemblies
keyed by the feature IDs.
See the documentation for each feature type in biocode.biothings for more info
'''
assemblies = dict()
features = dict()
# these are related to parsing any embedded FASTA
in_fasta_section = False
is_assembly_fasta = False
current_fasta_id = None
for line in open(gff3_file):
if in_fasta_section == True:
m = re.search('>(\S+)\s*(.*)', line)
if m:
current_fasta_id = m.group(1)
if current_fasta_id in assemblies:
is_assembly_fasta = True
else:
is_assembly_fasta = False
else:
if is_assembly_fasta == True:
# must be a sequence line for an assembly
# python 2.6+ makes string concatenation amortized O(n)
# http://stackoverflow.com/a/4435752/1368079
assemblies[current_fasta_id].residues += str(line.rstrip())
assemblies[current_fasta_id].length = len( assemblies[current_fasta_id].residues )
continue
elif line.startswith("##FASTA"):
# all data to the end of the file must be FASTA
in_fasta_section = True
continue
cols = line.split("\t")
if len(cols) != 9:
continue
mol_id = cols[0]
# initialize this assembly if we haven't seen it yet
if mol_id not in assemblies:
assemblies[mol_id] = biothings.Assembly( id=mol_id, residues='' )
current_assembly = assemblies[mol_id]
rfmin = int(cols[3]) - 1
rfmax = int(cols[4])
rstrand = None
feat_id = column_9_value(cols[8], 'ID')
parent_id = column_9_value(cols[8], 'Parent')
parent_feat = None
# shared features is not yet supported
if isinstance(parent_id, list):
raise Exception("This line contains a shared feature with multiple parents. This isn't yet supported:\n{0}".format(line))
if parent_id is not None:
if parent_id in features:
parent_feat = features[parent_id]
else:
raise Exception("Error in GFF3: Parent {0} referenced by a child feature before it was defined".format(parent_id) )
if cols[6] == '-':
rstrand = -1
elif cols[6] == '+':
rstrand = 1
else:
rstrand = 0
if cols[2] == 'gene':
gene = biothings.Gene(id=feat_id)
gene.locate_on(target=current_assembly, fmin=rfmin, fmax=rfmax, strand=rstrand)
features[feat_id] = gene
current_assembly.add_gene(gene)
elif cols[2] == 'mRNA':
mRNA = biothings.mRNA(id=feat_id, parent=parent_feat)
mRNA.locate_on(target=current_assembly, fmin=rfmin, fmax=rfmax, strand=rstrand)
parent_feat.add_mRNA(mRNA)
features[feat_id] = mRNA
elif cols[2] == 'rRNA':
rRNA = biothings.rRNA(id=feat_id, parent=parent_feat)
rRNA.locate_on(target=current_assembly, fmin=rfmin, fmax=rfmax, strand=rstrand)
parent_feat.add_rRNA(rRNA)
features[feat_id] = rRNA
elif cols[2] == 'tRNA':
tRNA = biothings.tRNA(id=feat_id, parent=parent_feat)
tRNA.locate_on(target=current_assembly, fmin=rfmin, fmax=rfmax, strand=rstrand)
parent_feat.add_tRNA(tRNA)
features[feat_id] = tRNA
elif cols[2] == 'exon':
exon = biothings.Exon(id=feat_id, parent=parent_feat)
exon.locate_on(target=current_assembly, fmin=rfmin, fmax=rfmax, strand=rstrand)
parent_feat.add_exon(exon)
features[feat_id] = exon
elif cols[2] == 'CDS':
CDS = biothings.CDS(id=feat_id, parent=parent_feat)
CDS.locate_on(target=current_assembly, fmin=rfmin, fmax=rfmax, strand=rstrand)
parent_feat.add_CDS(CDS)
features[feat_id] = CDS
return (assemblies, features)
def main():
parser = argparse.ArgumentParser( description='Convert GenBank flat files to GFF3 format')
## output file to be written
parser.add_argument('-i', '--input_file', type=str, required=True, help='Path to an input GBK file' )
parser.add_argument('-o', '--output_file', type=str, required=False, help='Path to an output GFF file to be created' )
parser.add_argument('--with_fasta', dest='fasta', action='store_true', help='Include the FASTA section with genomic sequence at end of file. (default)' )
parser.add_argument('--no_fasta', dest='fasta', action='store_false' )
parser.set_defaults(fasta=True)
args = parser.parse_args()
## output will either be a file or STDOUT
ofh = sys.stdout
if args.output_file is not None:
ofh = open(args.output_file, 'wt')
ofh.write("##gff-version 3\n")
assemblies = dict()
current_assembly = None
current_gene = None
current_RNA = None
rna_count_by_gene = defaultdict(int)
exon_count_by_RNA = defaultdict(int)
seqs_pending_writes = False
features_skipped_count = 0
# each gb_record is a SeqRecord object
for gb_record in SeqIO.parse(open(args.input_file, "r"), "genbank"):
mol_id = gb_record.name
if mol_id not in assemblies:
assemblies[mol_id] = biothings.Assembly( id=mol_id )
if len(str(gb_record.seq)) > 0:
seqs_pending_writes = True
assemblies[mol_id].residues = str(gb_record.seq)
assemblies[mol_id].length = len(str(gb_record.seq))
current_assembly = assemblies[mol_id]
# each feat is a SeqFeature object
for feat in gb_record.features:
#print(feat)
fmin = int(feat.location.start)
fmax = int(feat.location.end)
if feat.location.strand == 1:
strand = '+'
elif feat.location.strand == -1:
strand = '-'
else:
raise Exception("ERROR: unstranded feature encountered: {0}".format(feat))
#print("{0} located at {1}-{2} strand:{3}".format( locus_tag, fmin, fmax, strand ) )
if feat.type == 'source':
continue
if feat.type == 'gene':
# print the previous gene (if there is one)
if current_gene is not None:
gene.print_as(fh=ofh, source='GenBank', format='gff3')
locus_tag = feat.qualifiers['locus_tag'][0]
gene = biothings.Gene( id=locus_tag )
gene.locate_on( target=current_assembly, fmin=fmin, fmax=fmax, strand=strand )
current_gene = gene
elif feat.type == 'mRNA':
locus_tag = feat.qualifiers['locus_tag'][0]
rna_count_by_gene[locus_tag] += 1
feat_id = "{0}.mRNA.{1}".format( locus_tag, rna_count_by_gene[locus_tag] )
mRNA = biothings.mRNA( id=feat_id, parent=current_gene )
mRNA.locate_on( target=current_assembly, fmin=fmin, fmax=fmax, strand=strand )
gene.add_mRNA(mRNA)
current_RNA = mRNA
if feat_id in exon_count_by_RNA:
raise Exception( "ERROR: two different RNAs found with same ID: {0}".format(feat_id) )
else:
exon_count_by_RNA[feat_id] = 0
elif feat.type == 'tRNA':
locus_tag = feat.qualifiers['locus_tag'][0]
rna_count_by_gene[locus_tag] += 1
feat_id = "{0}.tRNA.{1}".format( locus_tag, rna_count_by_gene[locus_tag] )
tRNA = biothings.tRNA( id=feat_id, parent=current_gene )
tRNA.locate_on( target=current_assembly, fmin=fmin, fmax=fmax, strand=strand )
gene.add_tRNA(tRNA)
current_RNA = tRNA
if feat_id in exon_count_by_RNA:
raise Exception( "ERROR: two different RNAs found with same ID: {0}".format(feat_id) )
else:
exon_count_by_RNA[feat_id] = 0
elif feat.type == 'rRNA':
locus_tag = feat.qualifiers['locus_tag'][0]
rna_count_by_gene[locus_tag] += 1
feat_id = "{0}.rRNA.{1}".format( locus_tag, rna_count_by_gene[locus_tag] )
rRNA = biothings.rRNA( id=feat_id, parent=current_gene )
rRNA.locate_on( target=current_assembly, fmin=fmin, fmax=fmax, strand=strand )
gene.add_rRNA(rRNA)
current_RNA = rRNA
if feat_id in exon_count_by_RNA:
raise Exception( "ERROR: two different RNAs found with same ID: {0}".format(feat_id) )
else:
exon_count_by_RNA[feat_id] = 0
elif feat.type == 'CDS':
locus_tag = feat.qualifiers['locus_tag'][0]
exon_count_by_RNA[current_RNA.id] += 1
cds_id = "{0}.CDS.{1}".format( current_RNA.id, exon_count_by_RNA[current_RNA.id] )
current_CDS_phase = 0
for loc in feat.location.parts:
subfmin = int(loc.start)
subfmax = int(loc.end)
CDS = biothings.CDS( id=cds_id, parent=current_RNA )
CDS.locate_on( target=current_assembly, fmin=subfmin, fmax=subfmax, strand=strand, phase=current_CDS_phase )
current_RNA.add_CDS(CDS)
# calculate the starting phase for the next CDS feature (in case there is one)
# 0 + 6 = 0 TTGCAT
# 0 + 7 = 2 TTGCATG
# 1 + 6 = 1 TTGCAT
# 2 + 7 = 1 TTGCATG
# general: 3 - ((length - previous phase) % 3)
current_CDS_phase = 3 - (((subfmax - subfmin) - current_CDS_phase) % 3)
if current_CDS_phase == 3:
current_CDS_phase = 0
exon_id = "{0}.exon.{1}".format( current_RNA.id, exon_count_by_RNA[current_RNA.id] )
exon = biothings.Exon( id=exon_id, parent=current_RNA )
exon.locate_on( target=current_assembly, fmin=subfmin, fmax=subfmax, strand=strand )
current_RNA.add_exon(exon)
exon_count_by_RNA[current_RNA.id] += 1
product = feat.qualifiers['product'][0]
else:
print("WARNING: The following feature was skipped:\n{0}".format(feat))
features_skipped_count += 1
# don't forget to do the last gene, if there were any
if current_gene is not None:
gene.print_as(fh=ofh, source='GenBank', format='gff3')
if args.fasta is True:
if seqs_pending_writes is True:
ofh.write("##FASTA\n")
for assembly_id in assemblies:
ofh.write(">{0}\n".format(assembly_id))
ofh.write("{0}\n".format(biocodeutils.wrapped_fasta(assemblies[assembly_id].residues)))
if features_skipped_count > 0:
print("Warning: {0} unsupported feature types were skipped".format(features_skipped_count))
def get_gff3_features(gff3_file, assemblies=None):
'''
Parses the passed GFF3 file and returns two dicts, loaded with biocode.biothings objects:
1. The first dict are the Assembly objects, keyed on assembly ID. Each Assembly has all of the
children populated, so you can fully recover gene, RNA, exon and CDS features iterating on
the assembly.
2. The second dist is a flat structure of all the descendent feature objects of the Assemblies
keyed by the feature IDs.
See the documentation for each feature type in biocode.biothings for more info
'''
if assemblies is None:
assemblies = dict()
features = dict()
# these are related to parsing any embedded FASTA
in_fasta_section = False
is_assembly_fasta = False
current_fasta_id = None
for line in open(gff3_file):
#print("INFO: processing line: {0}".format(line))
if in_fasta_section == True:
m = re.search('>(\S+)\s*(.*)', line)
if m:
current_fasta_id = m.group(1)
if current_fasta_id in assemblies:
is_assembly_fasta = True
else:
is_assembly_fasta = False
else:
if is_assembly_fasta == True:
# must be a sequence line for an assembly
# python 2.6+ makes string concatenation amortized O(n)
# http://stackoverflow.com/a/4435752/1368079
assemblies[current_fasta_id].residues += str(line.rstrip())
assemblies[current_fasta_id].length = len(
assemblies[current_fasta_id].residues)
continue
elif line.startswith("##FASTA"):
# all data to the end of the file must be FASTA
in_fasta_section = True
continue
# ignore all other comments
if line.startswith('#'):
continue
cols = line.split("\t")
if len(cols) != 9:
continue
mol_id = cols[0]
# initialize this assembly if we haven't seen it yet
if mol_id not in assemblies:
assemblies[mol_id] = biothings.Assembly(id=mol_id, residues='')
current_assembly = assemblies[mol_id]
rfmin = int(cols[3]) - 1
rfmax = int(cols[4])
rstrand = None
atts = column_9_dict(cols[8])
feat_id = atts.get('ID')
parent_id = atts.get('Parent')
parent_feat = None
# sanity check
if rfmin > rfmax:
raise Exception(
"ERROR: Coordinates in GFF for feature id {0} appear to be reversed and violate GFF3 specification: {1} > {2}"
.format(feat_id, cols[3], cols[4]))
if 'locus_tag' in atts:
locus_tag = atts['locus_tag']
else:
locus_tag = None
# shared features is not yet supported
if isinstance(parent_id, list):
raise Exception(
"This line contains a shared feature with multiple parents. This isn't yet supported:\n{0}"
.format(line))
if parent_id is not None:
if parent_id in features:
parent_feat = features[parent_id]
else:
raise Exception(
"Error in GFF3: Parent {0} referenced by a child feature before it was defined"
.format(parent_id))
if cols[6] == '-':
rstrand = -1
elif cols[6] == '+':
rstrand = 1
else:
rstrand = 0
if cols[2] == 'gene':
gene = biothings.Gene(id=feat_id, locus_tag=locus_tag)
gene.locate_on(target=current_assembly,
fmin=rfmin,
fmax=rfmax,
strand=rstrand)
features[feat_id] = gene
current_assembly.add_gene(gene)
elif cols[2] == 'mRNA':
mRNA = biothings.mRNA(id=feat_id,
parent=parent_feat,
locus_tag=locus_tag)
mRNA.locate_on(target=current_assembly,
fmin=rfmin,
fmax=rfmax,
strand=rstrand)
parent_feat.add_mRNA(mRNA)
features[feat_id] = mRNA
elif cols[2] == 'rRNA':
rRNA = biothings.rRNA(id=feat_id,
parent=parent_feat,
locus_tag=locus_tag)
rRNA.locate_on(target=current_assembly,
fmin=rfmin,
fmax=rfmax,
strand=rstrand)
parent_feat.add_rRNA(rRNA)
features[feat_id] = rRNA
elif cols[2] == 'tRNA':
tRNA = biothings.tRNA(id=feat_id,
parent=parent_feat,
locus_tag=locus_tag)
tRNA.locate_on(target=current_assembly,
fmin=rfmin,
fmax=rfmax,
strand=rstrand)
parent_feat.add_tRNA(tRNA)
features[feat_id] = tRNA
elif cols[2] == 'exon':
exon = biothings.Exon(id=feat_id, parent=parent_feat)
exon.locate_on(target=current_assembly,
fmin=rfmin,
fmax=rfmax,
strand=rstrand)
parent_feat.add_exon(exon)
features[feat_id] = exon
elif cols[2] == 'CDS':
CDS = biothings.CDS(id=feat_id, parent=parent_feat)
CDS.locate_on(target=current_assembly,
fmin=rfmin,
fmax=rfmax,
strand=rstrand)
parent_feat.add_CDS(CDS)
features[feat_id] = CDS
elif cols[2] == 'polypeptide':
polypeptide = biothings.Polypeptide(id=feat_id, parent=parent_feat)
polypeptide.locate_on(target=current_assembly,
fmin=rfmin,
fmax=rfmax,
strand=rstrand)
parent_feat.add_polypeptide(polypeptide)
polypeptide.annotation = parse_annotation_from_column_9(cols[8])
features[feat_id] = polypeptide
elif cols[2] == 'five_prime_UTR':
utr = biothings.FivePrimeUTR(id=feat_id, parent=parent_feat)
utr.locate_on(target=current_assembly,
fmin=rfmin,
fmax=rfmax,
strand=rstrand)
parent_feat.add_five_prime_UTR(utr)
features[feat_id] = utr
elif cols[2] == 'three_prime_UTR':
utr = biothings.ThreePrimeUTR(id=feat_id, parent=parent_feat)
utr.locate_on(target=current_assembly,
fmin=rfmin,
fmax=rfmax,
strand=rstrand)
parent_feat.add_three_prime_UTR(utr)
features[feat_id] = utr
else:
sys.stderr.write("Skipping feature {0} with type {1}\n".format(
feat_id, cols[2]))
continue
features[feat_id].length = rfmax - rfmin
return (assemblies, features)
def main():
parser = argparse.ArgumentParser( description='Convert GenBank flat files to GFF3 format')
## output file to be written
parser.add_argument('-i', '--input_file', type=str, required=True, help='Path to an input GBK file' )
parser.add_argument('-o', '--output_file', type=str, required=False, help='Path to an output GFF file to be created' )
parser.add_argument('--with_fasta', dest='fasta', action='store_true', help='Include the FASTA section with genomic sequence at end of file. (default)' )
parser.add_argument('--no_fasta', dest='fasta', action='store_false' )
parser.set_defaults(fasta=True)
args = parser.parse_args()
## output will either be a file or STDOUT
ofh = sys.stdout
if args.output_file is not None:
ofh = open(args.output_file, 'wt')
ofh.write("##gff-version 3\n")
assemblies = dict()
current_assembly = None
current_gene = None
current_RNA = None
rna_count_by_gene = defaultdict(int)
exon_count_by_RNA = defaultdict(int)
seqs_pending_writes = False
features_skipped_count = 0
# each gb_record is a SeqRecord object
for gb_record in SeqIO.parse(open(args.input_file, "r"), "genbank"):
mol_id = gb_record.name
if mol_id not in assemblies:
assemblies[mol_id] = biothings.Assembly( id=mol_id )
if len(str(gb_record.seq)) > 0:
seqs_pending_writes = True
assemblies[mol_id].residues = str(gb_record.seq)
assemblies[mol_id].length = len(str(gb_record.seq))
current_assembly = assemblies[mol_id]
# each feat is a SeqFeature object
for feat in gb_record.features:
#print(feat)
fmin = int(feat.location.start)
fmax = int(feat.location.end)
if feat.location.strand == 1:
strand = '+'
elif feat.location.strand == -1:
strand = '-'
else:
raise Exception("ERROR: unstranded feature encountered: {0}".format(feat))
#print("{0} located at {1}-{2} strand:{3}".format( locus_tag, fmin, fmax, strand ) )
if feat.type == 'source':
continue
if feat.type == 'gene':
# print the previous gene (if there is one)
if current_gene is not None:
gene.print_as(fh=ofh, source='GenBank', format='gff3')
locus_tag = feat.qualifiers['locus_tag'][0]
gene = biothings.Gene( id=locus_tag )
gene.locate_on( target=current_assembly, fmin=fmin, fmax=fmax, strand=strand )
current_gene = gene
current_RNA = None
elif feat.type == 'mRNA':
locus_tag = feat.qualifiers['locus_tag'][0]
rna_count_by_gene[locus_tag] += 1
feat_id = "{0}.mRNA.{1}".format( locus_tag, rna_count_by_gene[locus_tag] )
mRNA = biothings.mRNA( id=feat_id, parent=current_gene )
mRNA.locate_on( target=current_assembly, fmin=fmin, fmax=fmax, strand=strand )
gene.add_mRNA(mRNA)
current_RNA = mRNA
if feat_id in exon_count_by_RNA:
raise Exception( "ERROR: two different RNAs found with same ID: {0}".format(feat_id) )
else:
exon_count_by_RNA[feat_id] = 0
elif feat.type == 'tRNA':
locus_tag = feat.qualifiers['locus_tag'][0]
rna_count_by_gene[locus_tag] += 1
feat_id = "{0}.tRNA.{1}".format( locus_tag, rna_count_by_gene[locus_tag] )
tRNA = biothings.tRNA( id=feat_id, parent=current_gene )
tRNA.locate_on( target=current_assembly, fmin=fmin, fmax=fmax, strand=strand )
gene.add_tRNA(tRNA)
current_RNA = tRNA
if feat_id in exon_count_by_RNA:
raise Exception( "ERROR: two different RNAs found with same ID: {0}".format(feat_id) )
else:
exon_count_by_RNA[feat_id] = 0
elif feat.type == 'rRNA':
locus_tag = feat.qualifiers['locus_tag'][0]
rna_count_by_gene[locus_tag] += 1
feat_id = "{0}.rRNA.{1}".format( locus_tag, rna_count_by_gene[locus_tag] )
rRNA = biothings.rRNA( id=feat_id, parent=current_gene )
rRNA.locate_on( target=current_assembly, fmin=fmin, fmax=fmax, strand=strand )
gene.add_rRNA(rRNA)
current_RNA = rRNA
if feat_id in exon_count_by_RNA:
raise Exception( "ERROR: two different RNAs found with same ID: {0}".format(feat_id) )
else:
exon_count_by_RNA[feat_id] = 0
elif feat.type == 'CDS':
locus_tag = feat.qualifiers['locus_tag'][0]
# If processing a prokaryotic GBK, we'll encounter CDS before mRNA, so we have to
# manually make one
if current_RNA is None:
feat_id = "{0}.mRNA.{1}".format( locus_tag, rna_count_by_gene[locus_tag] )
mRNA = biothings.mRNA( id=feat_id, parent=current_gene )
mRNA.locate_on( target=current_assembly, fmin=fmin, fmax=fmax, strand=strand )
gene.add_mRNA(mRNA)
current_RNA = mRNA
exon_count_by_RNA[current_RNA.id] += 1
cds_id = "{0}.CDS.{1}".format( current_RNA.id, exon_count_by_RNA[current_RNA.id] )
current_CDS_phase = 0
for loc in feat.location.parts:
subfmin = int(loc.start)
subfmax = int(loc.end)
CDS = biothings.CDS( id=cds_id, parent=current_RNA )
CDS.locate_on( target=current_assembly, fmin=subfmin, fmax=subfmax, strand=strand, phase=current_CDS_phase )
current_RNA.add_CDS(CDS)
# calculate the starting phase for the next CDS feature (in case there is one)
# 0 + 6 = 0 TTGCAT
# 0 + 7 = 2 TTGCATG
# 1 + 6 = 1 TTGCAT
# 2 + 7 = 1 TTGCATG
# general: 3 - ((length - previous phase) % 3)
current_CDS_phase = 3 - (((subfmax - subfmin) - current_CDS_phase) % 3)
if current_CDS_phase == 3:
current_CDS_phase = 0
exon_id = "{0}.exon.{1}".format( current_RNA.id, exon_count_by_RNA[current_RNA.id] )
exon = biothings.Exon( id=exon_id, parent=current_RNA )
exon.locate_on( target=current_assembly, fmin=subfmin, fmax=subfmax, strand=strand )
current_RNA.add_exon(exon)
exon_count_by_RNA[current_RNA.id] += 1
else:
print("WARNING: The following feature was skipped:\n{0}".format(feat))
features_skipped_count += 1
# don't forget to do the last gene, if there were any
if current_gene is not None:
gene.print_as(fh=ofh, source='GenBank', format='gff3')
if args.fasta is True:
if seqs_pending_writes is True:
ofh.write("##FASTA\n")
for assembly_id in assemblies:
ofh.write(">{0}\n".format(assembly_id))
ofh.write("{0}\n".format(biocodeutils.wrapped_fasta(assemblies[assembly_id].residues)))
if features_skipped_count > 0:
print("Warning: {0} unsupported feature types were skipped".format(features_skipped_count))
gene.locate_on(target=current_assembly, fmin=rfmin, fmax=rfmax, strand=rstrand)
print("DEBUG: locating gene {0} on {1} at coordinates fmin:{2}-fmax:{3} strand:{4}".format(feat_id, mol_id, rfmin, rfmax, rstrand) )
features[feat_id] = gene
current_assembly.add_gene(gene)
last_gene = gene
elif cols[2] == 'mRNA':
mRNA = biothings.mRNA(id=feat_id, parent=parent_feat)
mRNA.locate_on(target=current_assembly, fmin=rfmin, fmax=rfmax, strand=rstrand)
print("DEBUG: attaching mRNA:{0} to parent gene:{1}".format(feat_id, parent_feat.id) )
parent_feat.add_mRNA(mRNA)
features[feat_id] = mRNA
elif cols[2] == 'rRNA':
rRNA = biothings.rRNA(id=feat_id, parent=parent_feat)
rRNA.locate_on(target=current_assembly, fmin=rfmin, fmax=rfmax, strand=rstrand)
parent_feat.add_rRNA(rRNA)
features[feat_id] = rRNA
=======
## then mark any that align to it (except self)
for sbj_gene in things:
if qry_gene.id == sbj_gene.id:
continue
elif qry_gene.overlaps_with(sbj_gene):
handled_ids[sbj_gene.id] = 1
>>>>>>> .r101
return nonoverlapping_set