#!/usr/bin/env python
import argparse
import os
from libkuleshov.stats import n50
##############################################################################
parser = argparse.ArgumentParser()
parser.add_argument('--contigs')
args = parser.parse_args()
##############################################################################
with os.popen("cat %s | grep '>'" % args.contigs) as contigs:
lengths = list()
for line in contigs:
name, length, unk = line[1:].strip().split()
lengths.append(int(length))
print 'N50:', n50(lengths)
# and ctg_lengths[o[0]] > 700 and ctg_lengths[o[1]] > 700}
overlaps = {e for e,c in edge_counts.iteritems() if c >= 3}
validated_overlaps = overlaps & true_overlaps
print len(overlaps), len(validated_overlaps), len(true_overlaps)
# validated_overlaps = true_overlaps
# compute connected components
G = nx.Graph()
G.add_nodes_from(fasta.references)
for i, ovl1 in enumerate(validated_overlaps):
if i % 1000 == 0: print '%d/%d' % (i, len(validated_overlaps))
ol = list(ovl1)
if len(ol) > 2:
assert False
elif len(ol) < 2:
continue
G.add_edge(ol[0],ol[1])
connected_components = nx.connected_components(G)
cc_list = [cc for cc in connected_components]
component_lengths = [sum([ctg_lengths[ctg] for ctg in component])
for component in cc_list]
print sum(component_lengths)
print n50(component_lengths)
import os
import sys
from libkuleshov.stats import n50
print "cat %s | tr -d '\\n' | sed -re 's/>ctg([0-9])+/\\n/g' | perl -nle 'print length'" % sys.argv[
1]
fasta_length_pipe = os.popen(
"cat %s | tr -d '\\n' | sed -re 's/>ctg([0-9])+/\\n/g' | perl -nle 'print length'"
% sys.argv[1])
fasta_lengths = [int(line) for line in fasta_length_pipe]
total_len = sum(fasta_lengths)
print 'N50: %d (%d total)' % (n50(fasta_lengths), total_len)
assert False
elif len(ol) < 2:
continue
G.add_edge(ol[0], ol[1])
connected_components = nx.connected_components(G)
cc_list = [cc for cc in connected_components]
pickle.dump(cc_list, open('connected_components.pkl', 'wb'))
component_lengths = [
sum([ctg_lengths[ctg] for ctg in component]) for component in cc_list
]
print sum(fasta.lengths)
print sum(component_lengths)
print n50(component_lengths)
# ----------------------------------------------------------------------------
# compute overlaps
def myalign(s1, s2):
return align.globalms(s1, s2, 2, -1, -2, -0.2, penalize_end_gaps=False)[0]
def ctgs_overlap(ctg1, ctg2):
head_seq1 = ctg_seq[ctg1][:100]
tail_seq1 = ctg_seq[ctg1][-100:]
head_seq2 = ctg_seq[ctg2][:100]
tail_seq2 = ctg_seq[ctg2][-100:]
import os
import sys
from libkuleshov.stats import n50
print "cat %s | tr -d '\\n' | sed -re 's/>ctg([0-9])+/\\n/g' | perl -nle 'print length'" % sys.argv[1]
fasta_length_pipe = os.popen("cat %s | tr -d '\\n' | sed -re 's/>ctg([0-9])+/\\n/g' | perl -nle 'print length'" % sys.argv[1])
fasta_lengths = [int(line) for line in fasta_length_pipe]
total_len = sum(fasta_lengths)
print 'N50: %d (%d total)' % (n50(fasta_lengths), total_len)
#!/usr/bin/env python
import argparse
from libkuleshov.stats import n50
from libkuleshov.fastx import read_bed
##############################################################################
parser = argparse.ArgumentParser()
parser.add_argument('--contigs')
args = parser.parse_args()
##############################################################################
bed = read_bed(args.contigs)
lengths = [(end - start + 1) for ctg in bed.keys()
for (start, end) in bed[ctg]]
print 'Contigs:', len(lengths)
print 'Average length:', float(sum(lengths)) / len(lengths)
print 'N50 length:', n50(lengths)
#!/usr/bin/env python
import argparse
import os
from libkuleshov.stats import n50
##############################################################################
parser = argparse.ArgumentParser()
parser.add_argument('--contigs')
args = parser.parse_args()
##############################################################################
with os.popen("cat %s | grep '>'" % args.contigs) as contigs:
lengths = list()
for line in contigs:
name, length, unk = line[1:].strip().split()
lengths.append(int(length))
print 'N50:', n50(lengths)
#!/usr/bin/env python
import argparse
from libkuleshov.stats import n50
from libkuleshov.fastx import read_bed
##############################################################################
parser = argparse.ArgumentParser()
parser.add_argument('--contigs')
args = parser.parse_args()
##############################################################################
bed = read_bed(args.contigs)
lengths = [(end - start + 1) for ctg in bed.keys() for (start, end) in bed[ctg]]
print 'Contigs:', len(lengths)
print 'Average length:', float(sum(lengths)) / len(lengths)
print 'N50 length:', n50(lengths)