def parse():
list_obj = []
for line in open(ifile,'r'):
line = line.strip()
if len(line) > 0 and not line.startswith('#'):
obj = classGene.GFF3(line)
if obj.types() == 'mRNA':
obj_gene = classGene.LongestCodingIsoform(line, obj)
list_obj.append(obj_gene)
if obj.types() == 'exon':
obj_gene.addExon(line, obj)
if obj.types() == 'CDS':
obj_gene.addCDS(line, obj)
hash_geneID = {}
for obj in list_obj:
id, parent, exon_length, cds_length = obj.getLongestIsoform()
if parent in hash_geneID:
if hash_geneID[parent][1] < cds_length:
hash_geneID[parent][1] = cds_length
hash_geneID[parent][2] = exon_length
hash_geneID[parent][0] = id
elif hash_geneID[parent][1] == cds_length:
if hash_geneID[parent][2] < exon_length:
hash_geneID[parent][2] = exon_length
hash_geneID[parent][0] = id
else:
hash_geneID[parent] = [id, cds_length, exon_length]
hash_tid = {}
for item in hash_geneID:
hash_tid[hash_geneID[item][0]] = ''
### process the GFF3 file
print_flag = False
for line in open(ifile,'r'):
line = line.strip()
if len(line) > 0 and not line.startswith('#'):
obj = classGene.GFF3(line)
if obj.types() == 'gene':
print line
elif obj.types() == 'mRNA':
if str(obj) in hash_tid:
print line
print_flag = True
else:
print_flag = False
elif print_flag == True:
print line
def hash_combined(hash_combine_ID):
hash_combine = {}
count = 0
for line in open(combined_gff3, 'r'):
line = line.strip()
count += 1
if re.search('ID=', line):
if len(line) > 0 and not line.startswith('#'):
if count % 100000 == 0:
print 'Hashing combined: ', '{:9,.0f}'.format(count)
obj = classGene.GFF3(line)
if obj.types() == 'mRNA':
ID = str(obj)
if ID in hash_combine_ID:
hash_combine[ID] = {}
hash_combine[ID]['mRNA'] = line
hash_combine[ID]['exon'] = ''
hash_combine[ID]['CDS'] = ''
hash_lines = True
else:
hash_lines = False
elif obj.types() == 'exon':
if hash_lines == True:
hash_combine[ID]['exon'] += (',' + line)
elif obj.types() == 'CDS':
if hash_lines == True:
hash_combine[ID]['CDS'] += (',' + line)
return hash_combine
def hash_last_gff3_CDS(chromosome, start_min, start_max):
CDS = {}
count = 0
exons = {}
hash_flag = False
for line in open(last_gff3, 'r'):
line = line.strip()
if len(line) > 0 and not line.startswith('#'):
### print the lines processed
count += 1
if count % 100000 == 0:
print 'Hashing CDS: ', '{:9,.0f}'.format(count)
### check if it is gff3 format line
if re.search('ID=', line):
obj = classGene.GFF3(line)
if obj.types() == 'mRNA':
if obj.seqids() == chromosome and (start_min <= int(
obj.starts()) < start_max):
ID = str(obj)
CDS[ID] = 0
exons[ID] = []
hash_flag = True
else:
hash_flag = False
if obj.types() == 'exon' and obj.seqids(
) == chromosome and hash_flag == True:
exons[ID].append((int(obj.starts()), int(obj.ends())))
if obj.types() == 'CDS' and obj.seqids(
) == chromosome and hash_flag == True:
if ID in CDS:
CDS[ID] += (int(obj.ends()) - int(obj.starts()) + 1)
return exons, CDS
def hash_last_gff3_CDS(chromosome):
CDS = {}
count = 0
exons = {}
last_gff3_temp = last_gff3 + '.temp'
os.system('grep ' + chromosome + "'\t' " + last_gff3 + ' > ' +
last_gff3_temp)
for line in open(last_gff3_temp, 'r'):
line = line.strip()
if len(line) > 0 and not line.startswith('#'):
### print the lines processed
count += 1
if count % 100000 == 0:
print 'Hashing CDS: ', '{:9,.0f}'.format(count)
### check if it is gff3 format line
if re.search('ID=', line):
obj = classGene.GFF3(line)
if obj.types() == 'mRNA' and obj.seqids() == chromosome:
ID = str(obj)
CDS[ID] = 0
exons[ID] = []
if obj.types() == 'exon' and obj.seqids() == chromosome:
exons[ID].append((int(obj.starts()), int(obj.ends())))
if obj.types() == 'CDS' and obj.seqids() == chromosome:
if ID in CDS:
CDS[ID] += (int(obj.ends()) - int(obj.starts()) + 1)
os.system('rm ' + last_gff3_temp)
return exons, CDS
def printOut(Hash):
for line in open(gff3, 'r'):
if len(line) > 0 and not line.startswith('#'):
obj = classGene.GFF3(line)
if obj.types() == FeatureType:
if obj.seqids() in Hash:
print obj
def parseGFF3(EC, Function, Name, GeneComment, ProductType, GO, SYNONYM,
header, pf_out, g):
### Attributes
#attributes = ['ID', 'NAME', 'STARTBASE', 'ENDBASE', 'PRODUCT-TYPE','SYNONYM','GENE-COMMENT','FUNCTION','EC','GO','DBLINK','//']
genetic_elem_write = True
for line in open(gff3, 'r'):
line = line.strip()
if len(line) > 0 and not line.startswith('#'):
obj = classGene.GFF3(line)
if obj.types() == 'mRNA' and obj.seqids() == header:
t_id = str(obj)
if genetic_elem_write == True:
g.write('ID\t' + str(header) + '\n')
g.write('Name\t' + str(header) + '\n')
g.write('TYPE\t:CHRSM\n')
g.write('CIRCULAR?\tN\n')
g.write('ANNOT-FILE\t' + header + '.pf\n')
g.write('SEQ-FILE\t' + header + '.fa\n')
g.write('//' + '\n')
genetic_elem_write = False
pf_out.write('ID\t' + str(t_id) + '\n')
pf_out.write(attributes[1] + '\t' +
'.'.join(t_id.split('.')[:-1]) + '\n')
pf_out.write(attributes[2] + '\t' + str(obj.starts()) + '\n')
pf_out.write(attributes[3] + '\t' + str(obj.ends()) + '\n')
if t_id in ProductType:
pf_out.write(attributes[4] + '\t' +
str(ProductType[t_id]) + '\n')
else:
pf_out.write(attributes[4] + '\t' + str('P') + '\n')
if t_id in SYNONYM:
pf_out.write(attributes[5] + '\t' + str(SYNONYM[t_id]) +
'\n')
if t_id in Name:
if t_id in GeneComment:
pf_out.write(attributes[6] + '\t' +
str(GeneComment[t_id]) + '\n')
if t_id in Function:
pf_out.write(attributes[7] + '\t' +
str(Function[t_id]) + '\n')
if t_id in EC:
for j in range(len(EC[t_id])):
ec_no = EC[t_id][j].replace('EC-', '')
if len(ec_no.split('.')) == 3:
ec_no += '.-'
pf_out.write(attributes[8] + '\t' +
str(ec_no + '\n'))
if t_id in GO:
for j in range(len(GO[t_id])):
pf_out.write(attributes[9] + '\t' + str(GO[t_id][j]) +
'\n')
pf_out.write(attributes[10] + '\tGO:' +
str(GO[t_id][j]).split('|')[1] + '\n')
pf_out.write(attributes[11] + '\n')
pf_out.close()
def hashGFF3(chrom):
cds_coords = {}
for line in open(ifile, 'r'):
line = line.strip()
if len(line) > 1 and line.startswith(chrom):
obj = classGene.GFF3(line)
if obj.types() == 'CDS':
for i in range(int(obj.starts()), int(obj.ends()) + 1, 1):
cds_coords[i] = ''
return cds_coords
def parse(hash_anno):
for line in open(ifile,'r'):
line = line.strip()
if len(line)>0 and not line.startswith('#'):
obj = classGene.GFF3(line)
if obj.types() == 'mRNA':
print line + ';'+'Annotation="'+hash_anno[str(obj)].replace(',','')+'"'
else:
print line
def parse():
print_flag = False
for line in open(infile, 'r'):
line = line.strip()
obj = classGene.GFF3(line)
if obj.types() == 'gene':
if re.search("Name=CUFF", line):
print_flag = True
else:
print_flag = False
if print_flag == True:
print line
def parse():
print_flag = False
for line in open(infile, 'r'):
line = line.strip()
obj = classGene.GFF3(line)
if obj.types() == 'gene':
token = line.split('\t')
if token[1] == "CUFFLINKS":
print_flag = True
else:
print_flag = False
if print_flag == True:
print line
def hash_evidences(chromosome, exons, CDS, start_min, start_max):
evidences = {} ### hash exonic co-ordinates by evidence
CDS_transcript = {}
exons_transcript = {}
count = 0
hash_flag = False
for line in open(evidences_gff3, 'r'):
line = line.strip()
count += 1
if len(line) > 0 and not line.startswith('#'):
if count % 100000 == 0:
print 'Hashing evidence: ', '{:9,.0f}'.format(count)
### check if it is gff3 format line
if re.search('ID=', line):
obj = classGene.GFF3(line)
if obj.types() == 'mRNA':
if obj.seqids() == chromosome and (start_min <= int(
obj.starts()) < start_max):
ID = str(obj)
CDS_transcript[ID] = 0
exons_transcript[ID] = []
hash_flag = True
### print the lines processed
if obj.sources() not in evidences:
evidences[obj.sources()] = {}
for i in range(
int(obj.starts()) - int(gene_size_difference) -
1,
int(obj.starts()) + int(gene_size_difference) +
1):
evidences[obj.sources()][i] = str(obj)
for i in range(
int(obj.ends()) - int(gene_size_difference) -
1,
int(obj.ends()) + int(gene_size_difference) +
1):
evidences[obj.sources()][i] = str(obj)
else:
hash_flag = False
if obj.types() == 'exon' and obj.seqids() == chromosome:
if hash_flag == True:
exons_transcript[ID].append(
(int(obj.starts()), int(obj.ends())))
if obj.types() == 'CDS' and obj.seqids() == chromosome:
if hash_flag == True:
if ID in CDS_transcript:
CDS_transcript[ID] += (int(obj.ends()) -
int(obj.starts()) + 1)
find_gene_overlaps(evidences, exons_transcript, CDS_transcript, exons, CDS,
chromosome, start_min, start_max)
def print_intron(file):
last_parent_ID = ''
last_end = ''
for line in open(file,'r'):
line = line.strip()
if len(line) > 1 and not line.startswith('#'):
obj = classGene.GFF3(line)
if obj.types() == 'exon':
if last_parent_ID == get_PARENT(line):
print abs(last_end - int(obj.starts()))
last_parent_ID = get_PARENT(line)
last_end = int(obj.ends())
def hash_gff3():
HASH_GFF3 = {}
for line in open(GFF3, 'r'):
line = line.strip()
obj = classGene.GFF3(line)
if obj.types() == 'mRNA':
try:
anno = re.search(r'Annotation=".+"',
line).group(0).split('"')[1]
g_id = (get_PARENT(line)).replace('clover_', 'occidentale_')
HASH_GFF3[g_id] = anno.split('|')[4]
except:
continue
return HASH_GFF3
def hash_gff3(chromosome):
HASH_GFF3 = {}
for line in open(GFF3, 'r'):
line = line.strip()
obj = classGene.GFF3(line)
if obj.types() == 'mRNA' and obj.seqids() == chromosome:
try:
anno = re.search(r'Annotation=".+"',
line).group(0).split('"')[1]
for i in range(int(obj.starts()), int(obj.ends())):
if i not in HASH_GFF3:
HASH_GFF3[i] = anno
except:
continue
return HASH_GFF3
def change_gff3(prior_n):
for line in open(infile, 'r'):
line = line.strip()
if len(line) > 0 and not line.startswith('#'):
obj = classGene.GFF3(line)
string = obj.seqids() + '\t' + \
str(prior_n) + '\t' + \
obj.types() + '\t' + \
obj.starts() + '\t' + \
obj.ends() + '\t' + \
obj.scores() + '\t' + \
obj.strands() + '\t' + \
obj.phases() + '\t' + \
obj.attributes()
print string
def parse(gff3):
gene_objs = []
for line in open(gff3,'r'):
line = line.strip()
obj = classGene.GFF3(line)
if obj.types()=="gene":
LIobj = classGene.LongestIsoform(obj)
gene_objs.append(LIobj)
elif obj.types()=="mRNA":
LIobj.add_mRNA(line, obj)
else:
LIobj.add_feature(line)
for obj in gene_objs:
print str(obj)
print obj.features
def get_exon_fraction(chrom, hash_call):
HEADER = 'Lj30_ID\tExonLength\tCDSLength\tCallableExon\tCallableCDS'
obj_list = []
for line in open(ifile, 'r'):
if len(line) > 1 and not line.startswith('#'):
line = line.strip()
obj = classGene.GFF3(line)
if obj.types() == "mRNA":
obj_mRNA = classmRNA.mRNA(line, obj)
obj_list.append(obj_mRNA)
if obj.types() == "mRNA" or obj.types() == "exon" or obj.types(
) == "CDS":
obj_mRNA.AddData(line, obj)
print HEADER
for obj_mRNA in obj_list:
print str(obj_mRNA), obj_mRNA.GetExonLength(), obj_mRNA.GetCDSLength(
), obj_mRNA.GetExonicOverlap(hash_call), obj_mRNA.GetCDSOverlap(
hash_call)
def find_high_gene_density(chromosome, count, avg_gd):
o_frag = open(infile + '.' + chromosome + '.frags.temp', 'w')
last_start = -100000
new_block = True
correct_chro = False
for line in open(infile, 'r'):
line = line.strip()
if len(line) > 0 and not line.startswith('#'):
obj = classGene.GFF3(line)
if obj.seqids() == chromosome:
correct_chro = True
if obj.types() == 'gene':
start = int(obj.starts())
if start - last_start < max_dist:
if max_size > int(obj.ends()) - int(obj.starts()):
region += int(obj.ends()) - int(obj.starts())
gene_count += 1
gene_id.append(str(obj))
new_block = False
else:
if new_block == False:
if gene_count * 1000 / float(
gene_count
) >= 5 * avg_gd and gene_count > 2:
print chromosome, block_start
for g_id in gene_id:
o_frag.write(g_id + '\n')
gene_count = 0
region = int(obj.ends()) - int(obj.starts())
new_block = True
gene_id = []
if max_size > int(obj.ends()) - int(obj.starts()):
gene_id.append(str(obj))
gene_count += 1
block_start = int(obj.starts())
last_start = start
if correct_chro == True and obj.seqids() != chromosome:
break
o_frag.close()
def hash_evidences(chromosome, exons, CDS):
evidences = {} ### hash exonic co-ordinates by evidence
CDS_transcript = {}
exons_transcript = {}
count = 0
evidences_gff3_temp = evidences_gff3 + '.temp'
os.system('grep ' + chromosome + "'\t' " + evidences_gff3 + ' > ' +
evidences_gff3_temp)
for line in open(evidences_gff3_temp, 'r'):
line = line.strip()
count += 1
if len(line) > 0 and not line.startswith('#'):
if count % 100000 == 0:
print 'Hashing evidence: ', '{:9,.0f}'.format(count)
### check if it is gff3 format line
if re.search('ID=', line):
obj = classGene.GFF3(line)
if obj.types() == 'mRNA' and obj.seqids() == chromosome:
ID = str(obj)
CDS_transcript[ID] = 0
exons_transcript[ID] = []
### print the lines processed
if obj.sources() not in evidences:
evidences[obj.sources()] = {}
for i in range(
int(obj.starts()) - int(gene_size_difference) - 1,
int(obj.starts()) + int(gene_size_difference) + 1):
evidences[obj.sources()][i] = str(obj)
for i in range(
int(obj.ends()) - int(gene_size_difference) - 1,
int(obj.ends()) + int(gene_size_difference) + 1):
evidences[obj.sources()][i] = str(obj)
if obj.types() == 'exon' and obj.seqids() == chromosome:
exons_transcript[ID].append(
(int(obj.starts()), int(obj.ends())))
if obj.types() == 'CDS' and obj.seqids() == chromosome:
if ID in CDS_transcript:
CDS_transcript[ID] += (int(obj.ends()) -
int(obj.starts()) + 1)
find_gene_overlaps(evidences, exons_transcript, CDS_transcript, exons, CDS,
chromosome)
os.system('rm ' + evidences_gff3_temp)
def hash_GFF3(chromosome):
count = 0
coords = {}
for line in open(gff3, 'r'):
line = line.strip()
if len(line) > 0 and not line.startswith('#'):
obj = classGene.GFF3(line)
if obj.seqids() == chromosome:
count += 1
if count % 1000 == 0:
print 'Number of lines processed: ', chromosome, '{:9,.0f}'.format(
count)
for i in range(int(obj.starts()), int(obj.ends()) + 1):
if obj.types() == 'mRNA':
coords[i] = 'mRNA'
if obj.types() == 'exon':
coords[i] = 'exon'
if obj.types() == 'CDS':
coords[i] = 'CDS'
return coords
def find_avg_gene_density(chromosome, count, gene_count_hash):
correct_chro = False
gene_count = 0
for line in open(infile, 'r'):
line = line.strip()
if len(line) > 0 and not line.startswith('#'):
obj = classGene.GFF3(line)
count += 1
if count % 10000 == 0:
print 'Number of lines processed: ', chromosome, '{:9,.0f}'.format(
count)
if obj.seqids() == chromosome:
correct_chro = True
if obj.types() == 'gene':
gene_count += 1
if correct_chro == True and obj.seqids() != chromosome:
break
gene_count_hash[chromosome] = gene_count
def parseGFF3(EC, Function, Name, GeneComment, ProductType, GO, SYNONYM, pf_out, g):
### Attributes
#attributes = ['ID', 'NAME', 'STARTBASE', 'ENDBASE', 'PRODUCT-TYPE','SYNONYM','GENE-COMMENT','FUNCTION','EC','GO','DBLINK','//']
seq = {}
for line in open(gff3, 'r'):
line = line.strip()
if len(line)>0 and not line.startswith('#'):
obj = classGene.GFF3(line)
if obj.seqids() not in seq:
header = obj.seqids()
seq[obj.seqids()] = ''
t_id = str(obj)
g.write('ID\t'+str(header)+'\n')
g.write('Name\t'+str(header)+'\n')
g.write('TYPE\t:CHRSM\n')
g.write('CIRCULAR?\tN\n')
g.write('ANNOT-FILE\t'+header+'.pf\n')
g.write('SEQ-FILE\t'+header+'.fsa\n')
g.write('//'+'\n')
'''
def hashGFF3(chrom):
cds_coords = {}
cds_bound = {}
for line in open(ifile,'r'):
line = line.strip()
if len(line) > 1 and line.startswith(chrom):
obj = classGene.GFF3(line)
if obj.types() == 'CDS' and obj.get_parent() == candidate:
cds_bound[int(obj.starts())] = ''
cds_bound[int(obj.ends())] = ''
for i in range(int(obj.starts()), int(obj.ends())+1, 1):
cds_coords[i] = ''
if len(cds_coords) != 0:
if obj.strands() == '+':
start = min(cds_bound)
else:
start = max(cds_bound)
hashAlignment(chrom, cds_coords, start)
def get_exon_fraction(chrom, hash_call):
first_transcript = True
hash_exon = {}
hash_cds = {}
HEADER = 'Lj30_ID\tExonLength\tCDSLength\tCallableExon\tCallableCDS'
print HEADER
for line in open(ifile, 'r'):
if len(line) > 1 and not line.startswith('#'):
line = line.strip()
obj = classGene.GFF3(line)
if obj.seqids() == chrom:
if obj.types() == "mRNA":
if first_transcript == False:
exon_len = len(hash_exon)
cds_len = len(hash_cds)
exon_call_len = 0
cds_call_len = 0
for i in hash_exon:
if i in hash_call:
exon_call_len += 1
for i in hash_cds:
if i in hash_call:
cds_call_len += 1
print id + '\t' + str(exon_len) + '\t' + str(
exon_call_len) + '\t' + str(cds_len) + '\t' + str(
cds_call_len)
first_transcript = False
hash_exon = {}
hash_cds = {}
id = str(obj)
elif obj.types() == "exon":
for i in range(int(obj.starts()), int(obj.ends())):
hash_exon[i] = ''
elif obj.types() == "CDS":
for i in range(int(obj.starts()), int(obj.ends())):
hash_cds[i] = ''
print id + '\t' + str(exon_len) + '\t' + str(exon_call_len) + '\t' + str(
cds_len) + '\t' + str(cds_call_len)
def hash_coords(file, chr):
coords_dis={}
coords_mRNA_len = {}
exons = []
first_gene = True
for line in open(file, 'r'):
if len(line) > 1 and not line.startswith('#'):
obj = classGene.GFF3(line)
if obj.seqids() == chr:
if obj.types() == "mRNA":
strand = obj.strands()
if first_gene == False:
coords_dis, coords_mRNA_len = hash_coords_mRNA(coords_dis, exons, strand, coords_mRNA_len)
exons = []
first_gene = False
if obj.types() == "exon":
exons.append(int(obj.starts()))
exons.append(int(obj.ends()))
### for last mRNA
coords_dis, coords_mRNA_len = hash_coords_mRNA(coords_dis, exons, strand, coords_mRNA_len)
return coords_dis, coords_mRNA_len
def hashGFF3(chrom, align_hash):
cds_coords = {}
cds_bound = {}
first_transcript = True
for line in open(ifile,'r'):
line = line.strip()
if len(line) > 1 and line.startswith(chrom):
obj = classGene.GFF3(line)
if obj.types() == 'CDS':
for i in range(int(obj.starts()), int(obj.ends())+1, 1):
cds_coords[i] = ''
if obj.types() == 'mRNA':
if first_transcript == False:
processTranscript(chrom, cds_coords, tid, align_hash)
tid = str(obj)
cds_coords = {}
first_transcript = False
if len(line) > 1 and line.startswith(chrom):
processTranscript(chrom, cds_coords, tid, align_hash)
def hash_annotations(gff3, chro):
exons = {}
cds = {}
utr = {}
intron = {}
inter = {}
mRNA_type = {}
first_line = True
last_end = 0
last_exon_end = 0
parent_id = ''
for line in open(gff3, 'r'):
line = line.strip()
if len(line) > 1 and not line.startswith('#'):
obj = classGene.GFF3(line)
if obj.seqids() == chro:
if obj.types() == 'mRNA':
if first_line == False:
exons, cds, utr, intron, mRNA_type = addCoords(
gs, exons, cds, utr, intron, mRNA_type)
first_line = False
gs = classGeneStructure.GeneStructure(obj)
gs.addmRNA(obj)
if obj.types() == 'exon':
gs.addexon(obj)
if obj.types() == 'CDS':
gs.addcds(obj)
exons, cds, utr, intron, mRNA_type = addCoords(gs, exons, cds, utr, intron,
mRNA_type)
return exons, cds, utr, intron, mRNA_type
def find_gene_overlaps(evidences, exons_transcript, CDS_transcript, exons, CDS,
chromosome, start_min, start_max):
count = 0
for line in open(last_gff3, 'r'):
line = line.strip()
if len(line) > 0 and not line.startswith('#'):
count += 1
if count % 10000 == 0:
print 'find_gene_overlaps: ', '{:9,.0f}'.format(count)
obj = classGene.GFF3(line)
if obj.types() == 'mRNA' and obj.seqids() == chromosome and (
start_min <= int(obj.starts()) < start_max):
ID = str(obj)
### find the best replacement
delta_CDS = []
evi_list = []
larger_found = False
for i in evidences:
if int(obj.starts()) in evidences[i] and int(
obj.ends()) in evidences[i]:
if ID != evidences[i][int(
obj.starts())] and ID != evidences[i][int(
obj.ends())]:
evidence_id = evidences[i][int(obj.starts())]
if CDS[ID] + float(min_coding_differences) * CDS[
ID] < CDS_transcript[evidence_id] and CDS[
ID] >= 0:
### make sure that two IDs are isoformorms
if '.'.join(ID.split('.')[:-1]) != '.'.join(
evidence_id.split('.')[:-1]):
### check the exonic difference
set_exons_transcript = {}
for (start,
end) in exons_transcript[evidence_id]:
for j in range(start, end + 1):
set_exons_transcript[j] = ''
set_exons_transcript = set(
set_exons_transcript)
set_exons = {}
for start, end in exons[ID]:
for j in range(start, end + 1):
set_exons[j] = ''
set_exons = set(set_exons)
if abs(
int(
len(set_exons_transcript -
set_exons))
) < int(max_exonic_differences):
delta_CDS.append(
CDS_transcript[evidence_id] -
CDS[ID])
evi_list.append(
ID + '\t' + evidence_id + '\t' +
i + '\t' + str(obj.seqids()) +
'\t' + str(obj.starts()) + '\t' +
str(CDS_transcript[evidence_id] -
CDS[ID]) + '\t' + str(
len(set_exons_transcript -
set_exons)))
larger_found = True
if larger_found == True:
outfile.write(evi_list[delta_CDS.index(min(delta_CDS))] +
'\n')
def FixBoundries():
first_gene = True
gene = []
for line in open(infile, 'r'):
line = line.strip()
if len(line) > 0 and not line.startswith('#'):
obj = classGene.GFF3(line)
if obj.types() == 'gene':
if first_gene == False:
obj = classGene.GFF3(gene[0])
print obj.seqids() + '\t' + \
obj.sources() + '\t' + \
obj.types() + '\t' + \
str(gene_start) + '\t' + \
str(gene_end) + '\t' + \
obj.scores() + '\t' + \
obj.strands() + '\t' + \
str(obj.phases()) + '\t' + \
obj.attributes()
for l in gene[1:]:
print l
gene = []
obj = classGene.GFF3(line)
gene.append(line)
gene_start = int(obj.starts())
gene_end = int(obj.ends())
first_gene = False
elif obj.types() == 'mRNA':
gene.append(line)
if gene_start > int(obj.starts()):
gene_start = int(obj.starts())
if int(obj.ends()) > gene_end:
gene_end = int(obj.ends())
mRNA_start = int(obj.starts())
mRNA_end = int(obj.ends())
mRNA_coord = {}
for i in range(int(obj.starts()), int(obj.ends()) + 1):
mRNA_coord[i] = ''
else:
if int(obj.starts()) not in mRNA_coord:
line = obj.seqids() + '\t' + \
obj.sources() + '\t' + \
obj.types() + '\t' + \
str(int(mRNA_start)) + '\t' + \
obj.ends() + '\t' + \
obj.scores() + '\t' + \
obj.strands() + '\t' + \
str(obj.phases()) + '\t' + \
obj.attributes()
if int(obj.ends()) not in mRNA_coord:
line = obj.seqids() + '\t' + \
obj.sources() + '\t' + \
obj.types() + '\t' + \
obj.starts() + '\t' + \
str(mRNA_end) + '\t' + \
obj.scores() + '\t' + \
obj.strands() + '\t' + \
str(obj.phases()) + '\t' + \
obj.attributes()
gene.append(line)
obj = classGene.GFF3(gene[0])
print obj.seqids() + '\t' + \
obj.sources() + '\t' + \
obj.types() + '\t' + \
str(gene_start) + '\t' + \
str(gene_end) + '\t' + \
obj.scores() + '\t' + \
obj.strands() + '\t' + \
str(obj.phases()) + '\t' + \
obj.attributes()
for l in gene[1:]:
print l
def make_gff3(reaplcement_IDs, hash_combine):
out = open(last_gff3 + '.replaced', 'w')
count = 0
for line in open(last_gff3, 'r'):
line = line.strip()
count += 1
if len(line) > 0 and not line.startswith('#'):
if count % 100000 == 0:
print 'Printing final GFF3: ', '{:9,.0f}'.format(count)
obj = classGene.GFF3(line)
if obj.types() == "gene":
out.write(line + '\n')
source = obj.sources()
g_id = str(obj)
elif obj.types() == "mRNA":
ID = str(obj)
if ID in reaplcement_IDs:
token = hash_combine[reaplcement_IDs[ID]]['mRNA'].split(
'\t')
if obj.seqids() == token[0]:
print_flag = False
### print new mRNA
out.write(token[0] + '\t' + source + '\t' + token[2] +
'\t' + token[3] + '\t' + token[4] + '\t' +
token[5] + '\t' + token[6] + '\t' +
token[7] + '\t' +
(token[8].split("Parent=")[0]
).replace(reaplcement_IDs[ID], ID) +
"Parent=" + g_id + ";Name=" + ID + '\n')
### print exon lines of the lines
for i in hash_combine[
reaplcement_IDs[ID]]['exon'].split(',')[1:]:
i = i.replace(reaplcement_IDs[ID], ID)
token = i.split('\t')
out.write(token[0]+'\t'+ \
source + '\t' + \
'\t'.join(token[2:])+'\n')
### print CDS lines of the lines
CDS_count = 0
for i in hash_combine[
reaplcement_IDs[ID]]['CDS'].split(',')[1:]:
CDS_count += 1
i = i.replace(reaplcement_IDs[ID], ID)
token = i.split('\t')
out.write(
token[0] + '\t' + source + '\t' + token[2] +
'\t' + token[3] + '\t' + token[4] + '\t' +
token[5] + '\t' + token[6] + '\t' + token[7] +
'\t' +
(token[8].replace(reaplcement_IDs[ID], ID)
).split("ID=")[0] + "ID=" + ID + '.CDS.' +
str(CDS_count) + ";Parent=" + ID + '\n')
else:
print_flag = True
out.write(line + '\n')
else:
if print_flag == True:
out.write(line + '\n')
out.close()
