def import_blast(filename):
fp = file(filename)
total_lines = sum(1 for row in fp)
print >>sys.stderr, \
"Read BLAST file %(filename)s (total %(total_lines)d lines)" % locals()
fp.seek(0)
j = 0
for row in fp:
j += 1
if j % 100000 == 0: print >>sys.stderr, j, "read..."
atoms = row.split()
a, b, bitscore = atoms[0], atoms[1], float(atoms[-1])
a, b = gene_name(a), gene_name(b)
if a not in tandem_map or b not in tandem_map:
continue
a, b = tandem_map[a], tandem_map[b]
if a == b: continue
# keep the best blast hit
if a not in blast_pool:
blast_pool[a] = Hit(b, bitscore)
else:
blast_pool[a].update(b, bitscore)
fp.close()
def import_blast(filename):
fp = file(filename)
total_lines = sum(1 for row in fp)
print >>sys.stderr, \
"Read BLAST file %(filename)s (total %(total_lines)d lines)" % locals()
fp.seek(0)
j = 0
for row in fp:
j += 1
if j % 100000 == 0: print >> sys.stderr, j, "read..."
atoms = row.split()
a, b, bitscore = atoms[0], atoms[1], float(atoms[-1])
a, b = gene_name(a), gene_name(b)
if a not in tandem_map or b not in tandem_map:
continue
a, b = tandem_map[a], tandem_map[b]
if a == b: continue
# keep the best blast hit
if a not in blast_pool:
blast_pool[a] = Hit(b, bitscore)
else:
blast_pool[a].update(b, bitscore)
fp.close()
def process_tandems(fp_blast, sizes, ranks, tandem):
fp_blast.seek(0)
total_lines = sum(1 for row in fp_blast)
print >>sys.stderr, "Read self BLAST file (total %d lines)" % total_lines
fp_blast.seek(0)
j = 0
for row in fp_blast:
j += 1
if j % 100000 == 0:
print >>sys.stderr, j, "read..."
atoms = row.split()
a, b = atoms[:2]
a, b = gene_name(a), gene_name(b)
if a not in ranks or b not in ranks:
continue
chr_a, rank_a = ranks[a]
chr_b, rank_b = ranks[b]
if chr_a == chr_b and abs(rank_a - rank_b) <= Tandem_Nmax:
tandem.join(a, b)
tandem_removed = set() # the filtered gene set
tandem_map = {} # unfiltered => filtered
for tandem_group in tandem:
longest_gene, longest_size = "", 0
for gene in tandem_group:
if gene in sizes:
gene_size = sizes[gene]
if gene_size > longest_size:
longest_gene, longest_size = gene, gene_size
for gene in tandem_group:
tandem_map[gene] = longest_gene
tandem_removed.add(longest_gene)
print >>sys.stderr, len(tandem_removed), "genes after tandem removal"
return tandem_map, tandem_removed
def process_tandems(fp_blast, sizes, ranks, tandem):
fp_blast.seek(0)
total_lines = sum(1 for row in fp_blast)
print >>sys.stderr, \
"Read self BLAST file (total %d lines)" % total_lines
fp_blast.seek(0)
j = 0
for row in fp_blast:
j += 1
if j % 100000 == 0: print >> sys.stderr, j, "read..."
atoms = row.split()
a, b = atoms[:2]
a, b = gene_name(a), gene_name(b)
if a not in ranks or b not in ranks: continue
chr_a, rank_a = ranks[a]
chr_b, rank_b = ranks[b]
if chr_a == chr_b and abs(rank_a - rank_b) <= Tandem_Nmax:
tandem.join(a, b)
tandem_removed = set() # the filtered gene set
tandem_map = {} # unfiltered => filtered
for tandem_group in tandem:
longest_gene, longest_size = "", 0
for gene in tandem_group:
if gene in sizes:
gene_size = sizes[gene]
if gene_size > longest_size:
longest_gene, longest_size = gene, gene_size
for gene in tandem_group:
tandem_map[gene] = longest_gene
tandem_removed.add(longest_gene)
print >> sys.stderr, len(tandem_removed), "genes after tandem removal"
return tandem_map, tandem_removed
def load_sizes(fp_sizes):
# load the gene size info (for keeping longest gene in a tandem group)
fp_sizes.seek(0)
sizes = {} # gene => size
print >>sys.stderr, "Read .sizes file"
for row in fp_sizes:
gene, size = row.split()
# sizes are calculated for the transcripts, and we keep longest transcript
gene, size = gene_name(gene), int(size)
if gene not in sizes or size > sizes[gene]:
sizes[gene] = size
return sizes
def load_sizes(fp_sizes):
# load the gene size info (for keeping longest gene in a tandem group)
fp_sizes.seek(0)
sizes = {} # gene => size
print >> sys.stderr, "Read .sizes file"
for row in fp_sizes:
gene, size = row.split()
# sizes are calculated for the transcripts, and we keep longest transcript
gene, size = gene_name(gene), int(size)
if gene not in sizes or size > sizes[gene]:
sizes[gene] = size
return sizes
