def initialize_polypeptides( log_fh, fasta_file ):
'''
Reads a FASTA file of (presumably) polypeptide sequences and creates a dict of Polypeptide
objects, keyed by ID, with bioannotation.FunctionalAnnotation objects attached.
'''
seqs = utils.fasta_dict_from_file(fasta_file)
polypeptides = dict()
for seq_id in seqs:
polypeptide = things.Polypeptide(id=seq_id, length=len(seqs[seq_id]['s']), residues=seqs[seq_id]['s'])
annot = annotation.FunctionalAnnotation(product_name=DEFAULT_PRODUCT_NAME)
log_fh.write("INFO: {0}: Set initial product name to '{1}'\n".format(seq_id, DEFAULT_PRODUCT_NAME))
polypeptide.annotation = annot
polypeptides[seq_id] = polypeptide
return polypeptides
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] = things.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 = things.Gene(id=locus_tag, locus_tag=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 = things.mRNA(id=feat_id,
parent=current_gene,
locus_tag=locus_tag)
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])
if 'product' in feat.qualifiers:
anticodon = feat.qualifiers['product'][0]
else:
anticodon = None
tRNA = things.tRNA(id=feat_id,
parent=current_gene,
anticodon=anticodon)
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])
if 'product' in feat.qualifiers:
product = feat.qualifiers['product'][0]
else:
product = None
annot = annotation.FunctionalAnnotation(product_name=product)
rRNA = things.rRNA(id=feat_id,
parent=current_gene,
annotation=annot)
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 = things.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 'product' in feat.qualifiers:
product = feat.qualifiers['product'][0]
else:
product = None
if 'gene' in feat.qualifiers:
gene_symbol = feat.qualifiers['gene'][0]
else:
gene_symbol = None
annot = annotation.FunctionalAnnotation(
product_name=product, gene_symbol=gene_symbol)
if 'db_xref' in feat.qualifiers:
for dbxref in feat.qualifiers['db_xref']:
annot.add_dbxref(dbxref)
polypeptide_id = "{0}.polypeptide.{1}".format(
locus_tag, rna_count_by_gene[locus_tag])
polypeptide = things.Polypeptide(id=polypeptide_id,
parent=mRNA,
annotation=annot)
mRNA.add_polypeptide(polypeptide)
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 = things.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 = things.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(
utils.wrapped_fasta(assemblies[assembly_id].residues)))
if features_skipped_count > 0:
print("Warning: {0} unsupported feature types were skipped".format(
features_skipped_count))
def main():
flawed_gff_file = 'canonical.flawed.gff3'
ilri_gff = 'Theileria-all-Theileria1_ourids.gff'
source = 'GenBank'
out_gff = 'canonical.corrected.gff3'
fout = open(out_gff, mode='wt', encoding='utf-8')
fout.write("##gff-version 3\n")
(assemblies, features) = gff.get_gff3_features(flawed_gff_file)
print("INFO: loaded {0} assemblies and {1} features".format(
len(assemblies), len(features)))
polypeptides = dict()
for line in open(ilri_gff):
cols = line.split("\t")
if len(cols) != 9 or cols[2] != 'polypeptide':
continue
id = gff.column_9_value(cols[8], 'ID')
parent = gff.column_9_value(cols[8], 'Parent')
polypeptides[parent] = things.Polypeptide(id=id, parent=parent)
polypeptides[parent].locate_on(target=assemblies[cols[0]],
fmin=int(cols[3]) - 1,
fmax=int(cols[4]),
strand=cols[6])
print("DEBUG: loaded {0} polypeptides from ILRI file".format(
len(polypeptides)))
for assembly_id in assemblies:
for gene in assemblies[assembly_id].genes():
for mRNA in gene.mRNAs():
if mRNA.id not in polypeptides:
print(
"DEBUG: {0} not found as a parent to any polypeptide".
format(mRNA.id))
else:
polypeptide = polypeptides[mRNA.id]
# pull this outside of the iteration since iterating might delete some
CDSs = mRNA.CDSs()
for CDS in CDSs:
keep = True
if CDS < polypeptide:
mRNA.delete_CDS(CDS)
elif CDS <= polypeptide:
CDS.location().fmin = polypeptide.location().fmin
if CDS > polypeptide:
mRNA.delete_CDS(CDS)
elif CDS >= polypeptide:
CDS.location().fmax = polypeptide.location().fmax
#print("WARN: found a CDS {0}:{1}-{2} outside the range of the polypeptide {3}:{4}-{5}".format( \
# CDS.id, CDS.locations[0].fmin, CDS.locations[0].fmax, \
# polypeptide.id, polypeptide.locations[0].fmin, polypeptide.locations[0].fmax))
gene.print_as(fh=fout, source=source, format='gff3')
def main():
parser = argparse.ArgumentParser( description='Metagenemark GFF -> GFF3 conversion script')
## output file to be written
parser.add_argument('-i', '--input', type=str, required=True, help='Path to a GFF file created by Metagenemark' )
parser.add_argument('-o', '--output', type=str, required=True, help='Path to an output file to be created' )
parser.add_argument('-p', '--prefix', type=str, required=True, help='Prefix to use in ID generation')
parser.add_argument('-pf', '--protein_fasta', type=str, required=False, help='Optional protein FASTA to be written')
args = parser.parse_args()
assemblies = dict()
current_assembly = None
# key like 2 = SRS014890.polypeptide.2
polypeptide_lookup = dict()
writing_protein = False
gene = None
mRNAs = dict()
current_sequence = None
current_gene_comment_lines = list()
fout = open(args.output, mode='wt', encoding='utf-8')
fout.write("##gff-version 3\n")
if args.protein_fasta is not None:
protein_out = open(args.protein_fasta, mode='wt', encoding='utf-8')
for line in open(args.input):
if line.startswith("#"):
if line.startswith("##FASTA"):
current_gene_comment_lines.append("#{0}".format(line))
elif line.startswith("##end-Protein"):
writing_protein = False
current_gene_comment_lines.append(line)
# since we're already doing our own header, don't duplicate the old one
elif line.startswith("##gff-version"):
continue
else:
if line.startswith("##Protein "):
m = re.match("##Protein (\d+)", line)
if m:
writing_protein = True
protein_out.write(">{0}\n".format(polypeptide_lookup[m.group(1)]))
else:
raise Exception("ERROR: Expected line to match: ##Protein N")
elif writing_protein == True:
protein_out.write(line[2:])
current_gene_comment_lines.append(line)
else:
cols = line.split("\t")
if len(cols) != 9:
continue
mol_id = cols[0]
mol_id_m = re.match('^(\S+) ', mol_id)
if mol_id_m:
print("MATCH!")
mol_id = mol_id_m.group(1)
feat_type = cols[2]
## we expect only gene types here
if feat_type not in ['gene', 'CDS']:
raise Exception("ERROR: expected only 'gene' or 'CDS' feature types as input (depending on metagenemark version).")
m_gene = re.match('gene_id[ =](\d+)', cols[8])
if m_gene:
gene_num = m_gene.group(1)
else:
raise Exception("ERROR: expected 9th column to have gene ids like: gene_id 5")
## initialize this assembly if we haven't seen it yet
if mol_id not in assemblies:
assemblies[mol_id] = things.Assembly(id=mol_id)
current_assembly = assemblies[mol_id]
gene = things.Gene(id="{0}.gene.{1}".format(args.prefix, gene_num))
gene.locate_on( target=current_assembly, fmin=int(cols[3]) - 1, fmax=int(cols[4]), strand=cols[6] )
mRNA = things.mRNA(id="{0}.mRNA.{1}".format(args.prefix, gene_num), parent=gene.id)
mRNA.locate_on( target=current_assembly, fmin=int(cols[3]) - 1, fmax=int(cols[4]), strand=cols[6] )
gene.add_mRNA(mRNA)
CDS = things.CDS(id="{0}.CDS.{1}".format(args.prefix, gene_num), parent=mRNA.id)
CDS.locate_on( target=current_assembly, fmin=int(cols[3]) - 1, fmax=int(cols[4]), strand=cols[6], phase=int(cols[7]) )
mRNA.add_CDS(CDS)
exon = things.Exon(id="{0}.exon.{1}".format(args.prefix, gene_num), parent=mRNA.id)
exon.locate_on( target=current_assembly, fmin=int(cols[3]) - 1, fmax=int(cols[4]), strand=cols[6] )
mRNA.add_exon(exon)
polypeptide_id = "{0}.polypeptide.{1}".format(args.prefix, gene_num)
polypeptide = things.Polypeptide(id=polypeptide_id, parent=mRNA.id)
polypeptide.locate_on( target=current_assembly, fmin=int(cols[3]) - 1, fmax=int(cols[4]), strand=cols[6] )
mRNA.add_polypeptide(polypeptide)
polypeptide_lookup[gene_num] = polypeptide_id
gene.print_as(fh=fout, source='GeneMark.hmm', format='gff3')
fout.write( "".join(current_gene_comment_lines) )
current_gene_comment_lines = list()
def check_and_add_polypeptide(mRNA):
polypeptides = mRNA.polypeptides()
if len(polypeptides) == 0:
polypeptide_id = "{0}.polypeptide".format(mRNA.id)
polypeptide = things.Polypeptide(id=polypeptide_id, parent=mRNA.id)
mRNA.add_polypeptide(polypeptide)