def main():
"""Select desired genes from a gff based on gene IDs
If a given gene has even a single transcript in the to-select list,
then all isoforms will be collected
Usage: select-gene-records.py list-of-transcript-ids.txt gene-records.gff
Output: gene-records-selected.gff
"""
idlist = sys.argv[1]
gff = sys.argv[2]
outnam = re.sub(".gff", "-selected.gff", gff)
ids = set()
with open(idlist, "r") as file_object:
for line in file_object:
ids.add(line.strip("\n"))
for i in range(1, 99):
deriv_geneID = re.sub("\Z", "-mRNA-" + str(i),
line.strip("\n"))
ids.add(deriv_geneID)
outfile = open(outnam, "w")
with open(gff, "r") as file_object:
for line in file_object:
if line[0] == "#":
# outfile.write(line)
continue
feature = gfftools.GFF(line)
if feature.id in ids:
outfile.write(line)
else:
for papa in feature.parent:
if papa in ids:
outfile.write(line)
break
outfile.close()
def main():
gene_loci = 'version3-gene-locus-tags.tsv'
mrna_loci = 'version3-transcript-locus-tags.tsv'
gff = '/projects/bullfrog_assembly/genome/ARCS/annotation/fresh-Oct2018/maker-second-round/analysis/third-round/gag_output/genome.gff'
blastp = '/projects/bullfrog_assembly/genome/ARCS/annotation/fresh-Oct2018/submission_prep/blastp-annotated2.tsv'
pfam = 'pfam-hits.tsv'
genevalidator = 'genevalidator-90.txt'
POS = 25.0
QCOV = 50.0
EVALUE = 0.00001
mrna_tags = load_tags(mrna_loci)
gene_tags = load_tags(gene_loci)
blast_aln = load_blast(blastp, POS, QCOV)
pfam_aln = load_pfam(pfam, EVALUE)
gv_good = load_gv(genevalidator)
with open(gff, 'r') as infile:
for line in infile:
if line[0] == '#':
sys.stdout.write(line)
else:
rec = gfftools.GFF(line)
tag = get_tag(rec, mrna_tags, gene_tags)
if rec.type == 'mRNA':
if rec.id in blast_aln:
hit = blast_aln[rec.id]
prot = hit.get_best_blast()
else:
prot = ''
if rec.id in pfam_aln:
hit = pfam_aln[rec.id]
dom = hit.get_best_pfam()
else:
dom = ''
if prot:
hit_name = prot.sname
if prot.qid in gv_good:
rec.add_prot_desc(hit_name, flag=True)
else:
rec.add_prot_desc(hit_name)
elif dom:
pfam_name = dom.sname
rec.add_pfam(pfam_name)
else:
rec.add_prot_desc('hypothetical protein', flag=True)
if tag:
rec.add_locus_tag(tag)
# process attributes and print out
rec.extend_attr()
print(rec.print_record())
elif rec.type == 'gene':
# genes only get locus_tags
if tag:
rec.add_locus_tag(tag)
rec.extend_attr()
print(rec.print_record())
else:
# non-gene or non-mRNA record
rec.extend_attr()
print(rec.print_record())
def main():
"""Extract CDS coordinates for each mRNA in a gff file,
then print the corresponding sequence.
The subfeatures MUST be sorted by their start coordinate
Usage: extract-cds.py genes.gff genome.fa > genes-cds.fa
"""
if len(sys.argv) is not 3:
print main.__doc__
sys.exit(1)
gff = sys.argv[1]
fasta = sys.argv[2]
cds_coords = {}
with open(gff, "r") as file_object:
for line in file_object:
if line[0] == "#":
sys.stdout.write(line)
continue
feature = gfftools.GFF(line)
if feature.type == "CDS":
for parent in feature.parent:
if parent in cds_coords:
if feature.strand == "+":
cds_coords[parent]['coords'].append(
[feature.start, feature.end])
else:
cds_coords[parent]['coords'].insert(
0, [feature.start, feature.end])
else:
cds_coords[parent] = {
'coords': [[feature.start, feature.end]],
'scaf': feature.seqid,
'strand': feature.strand
}
sequences = {}
with open(fasta, "r") as file_object:
for rec in fastatools.fasta_iter(file_object):
sequences[rec[0]] = rec[1]
for record in cds_coords:
whole = ""
for coords in cds_coords[record]['coords']:
seqid = cds_coords[record]['scaf']
if cds_coords[record]['strand'] == "-":
piece = fastatools.revcomp(
sequences[seqid][int(coords[0]) -
1:int(coords[1])]) ### watch for OBO
else:
piece = sequences[seqid][int(coords[0]) -
1:int(coords[1])] ### watch for OBO
whole += piece
else:
sys.stdout.write(">" + record + "_CDS\n" + whole + "\n")
import re
import gfftools
# n.b. s.argv[0] == script name
gff = sys.argv[1]
# dicts to hold the exon and cds records
edic = {}
cdic = {}
with open(gff, "r") as infile:
for line in infile:
if line[0] != "#":
rec = gfftools.GFF(line)
if rec.type == "mRNA" or rec.type == "ncRNA":
mid = rec.id
fstart = int(rec.start) - 1
fend = int(rec.end) - 1
edic[mid] = [
rec.seqid, fstart, fend, mid, 0, rec.strand, fstart, fend,
[255, 0, 0]
]
# if rec.type != "ncRNA":
# cdic[mid]=[rec.seqid,fstart,fend,mid,0,rec.strand,fstart,fend,[0,0,255]]
elif rec.type == "exon":
mid = re.sub(":exon", "", rec.id)
fstart = int(rec.start) - 1 - edic[mid][1]
fend = int(rec.end) - 1 - edic[mid][1]
#gffout=open(re.sub(".gff","-prepared.gff",gff),"w")
tblout=open(re.sub(".gff","-prepared.tbl",gff),"w")
lokey=open(re.sub(".gff","-locus_tag-conversion-key.tsv",gff),"w")
###################
# MAIN
genes = {}
# read in the entries from the gff file
with open(gff,"r") as infile:
print "Reading the annotations"
for line in infile:
if line[0] == "#" or line[0] == "-":
continue
feature = gfftools.GFF(line)
if feature.type == "gene":
genes[feature.id] = gfftools.Gene(feature)
elif feature.type == "mRNA" or feature.type == "ncRNA":
# transcripts can only have one parent, so access it explicitly
genes[feature.parent[0]].add_transcript(feature)
elif feature.type == "exon":
if feature.name:
genes[feature.name].transcript[feature.parent[0]].add_exon(feature)
continue
for parent in feature.parent:
for gene in genes:
if re.findall(genes[gene].id,parent):