def gc_content_calc(consensus, annotation, windowsize, region, outfile):
''' (str, str, int, str, str) -> None
uses the consensus sequence generated above to calculate GC content
at selectively unconstrained sites (intronic, intergenic, 4D).
writes count of GC nucleotides + Ns + all non-N sites to specified outfile.
'''
chromosome, start, end = parse_region(region)
with open(outfile, 'w') as f:
f.write('start end GC N total_sites\n')
seq_index = 0
for window in tqdm(range(start, end, windowsize)):
counter = OrderedDict.fromkeys(
['GC_count', 'N_count', 'total_sites'], 0)
window_start = window
if window + windowsize > end:
window_end = end
else:
window_end = window + windowsize
p = ant.Reader(annotation)
for record in p.fetch(chromosome, window_start, window_end):
seq_index = record.pos - start # fix offset for string indexing
if record.is_fold4:
if consensus[seq_index] in ['G', 'C']:
counter['GC_count'] += 1
counter['total_sites'] += 1
elif consensus[seq_index] in ['A', 'T']: # Ns not counted
counter['total_sites'] += 1
elif consensus[seq_index] == 'N':
counter['N_count'] += 1
window_out = ' '.join(
[str(num) for num in [window_start, window_end]])
line_out_counts = ' '.join(
[str(i) for i in list(counter.values())])
f.write(window_out + ' ' + line_out_counts + '\n')
def prep_header(table, outfile):
with open(outfile, 'w') as f_out:
p = ant.Reader(table)
for item in p.header:
f_out.write(item.strip() + '\n')
f_out.write(
'##ld_rho=LDhelmet recombination rate on Quebec dataset. p/bp [FLOAT]'
+ '\n')
# add column names
p.cols.append('ld_rho')
p.cols = [item.strip() for item in p.cols]
col_line = '#' + '\t'.join(p.cols) + '\n'
f_out.write(col_line)
def write_lines(table, ldhelmet_dir, outfile):
for i in range(1, 18):
with open(outfile, 'a') as f_out:
with open(ldhelmet_dir + 'chromosome_{}.txt'.format(i)) as f:
for line in tqdm(f):
if line.startswith(('#', 'ver')):
continue
else:
split = line.split(' ')
start, end, rho = int(split[0]), int(split[1]), float(
split[2])
p = ant.Reader(table).fetch('chromosome_{}'.format(i),
start - 1,
end - 1,
raw=True)
for record in p:
record = record + '\t' + str(rho) + '\n'
f_out.write(record)
'''
add_ld_rho.py but it takes into account methylation
the methylation bed file has been split by chromosome in data/methylation/bed_split - should make it quicker to parse them
python3.5 add_rho_meth.py > [output table]
AH - 02/2018
'''
import ant
from tqdm import tqdm
p = ant.Reader('data/annotation_table.txt.gz')
for item in p.header:
print(item.strip())
print('##ld_rho=rate of recombination measured by LDhelmet on Quebec dataset. p/bp [FLOAT]')
print('##methylation=beta value at CpG sites in three clones of CC2937 [FLOAT]')
# add column names
p.cols.append('ld_rho')
p.cols.append('methylation')
p.cols = [item.strip() for item in p.cols]
print('#', '\t'.join(p.cols), sep = '') # add new cols
def makelookup(chrom):
filename = 'data/methylation/bed_split/{}.bed'.format(chrom)
lookup = {}
with open(filename, 'r') as m:
for line in tqdm(m):
def checkrho(record, ld_dict):
currentrho = 0.0
for key in ld_dict.keys():
if record.pos in key:
currentrho = ld_dict[key]
continue
elif record.pos not in key:
pass
return currentrho
print('chrom type avgrho numrecords')
p = ant.Reader(annotation).fetch(chrom)
exonic_rho = [checkrho(r, ldh_dict) for r in p if r.is_exonic]
were_exonic = len(exonic_rho)
exonic_rho = sum(exonic_rho) / were_exonic # overwrites massive list
print(chrom, 'exonic', exonic_rho, were_exonic)
p = ant.Reader(annotation).fetch(chrom)
intronic_rho = [checkrho(r, ldh_dict) for r in p if r.is_intronic]
were_intronic = len(intronic_rho)
intronic_rho = sum(intronic_rho) / were_intronic
print(chrom, 'intronic', intronic_rho, were_intronic)
p = ant.Reader(annotation).fetch(chrom)
genic_rho = [checkrho(r, ldh_dict) for r in p if r.is_genic]
were_genic = len(genic_rho)
genic_rho = sum(genic_rho) / were_genic