# for elmseq in seq_percents_bird:
# if elmseq in uniq:
# print elmseq + '\t' + str(sum(seq_percents_bird[elmseq])/float(total_bird)) + '\t' + str(sum(seq_percents_human[elmseq])/float(total_human))
human_host_file = 'working/Jun29/elmdict_H_sapiens.simple'
chicken_host_file = 'working/Jun29/elmdict_Gallus_gallus.simple'
human_host_freqs = get_host_freqs(human_host_file)
chicken_host_freqs = get_host_freqs(chicken_host_file)
with open('working/hi', 'w') as hi:
with open('working/low', 'w') as low:
for elmseq in uniq:
if 'X' not in elmseq and 'J' not in elmseq and 'B' not in elmseq:
per = sp[elmseq]
dels = elmseq.split(':')
new_seq = utils.mk_sub(dels[2])
key = dels[1] + ':' + new_seq
human_host = 0
chicken_host = 0
if key in chicken_host_freqs:
chicken_host = chicken_host_freqs[key]
if key in human_host_freqs:
human_host = human_host_freqs[key]
bird_flu_frac = sum(seq_percents_bird[elmseq])/float(total_bird)
human_flu_frac = sum(seq_percents_human[elmseq])/float(total_human)
if bird_flu_frac >= float(80):
hi.write(str(chicken_host) + '\n')
else:
low.write(str(chicken_host) + '\n')
"""Convert raw sequences to simplified versions
for use in scanning with simple patterns
(those made with utils.mk_sub)"""
import sys, global_settings, os, utils
in_file = sys.argv[1]
out_file = sys.argv[2]
with open(out_file, 'w') as f:
for ID, seq in utils.fasta_iter(in_file):
f.write('>' + ID + '\n')
f.write(utils.mk_sub(seq) + '\n')
"""There is little overlap between
host and flu sequences. To solve
this problem, I'm substiting some
residues with symbols for their
properties"""
import sys, global_settings, utils
from collections import defaultdict
elmdict_file = sys.argv[1]
elm2seq2count = defaultdict(utils.init_zero)
elm2count = defaultdict(utils.init_zero)
with open(elmdict_file) as f:
for line in f:
(elm, seq, count, frq) = line.strip().split('\t')
if 'X' not in seq:
new_seq = utils.mk_sub(seq)
elm2seq2count[elm+':'+new_seq] += int(count)
elm2count[elm] += int(count)
for elmSeq in elm2seq2count:
elm, seq = elmSeq.split(':')
count = elm2seq2count[elmSeq]
freq = float(count)/float(elm2count[elm])
print('%s\t%s\t%d\t%.10f' %
(elm, seq, count, freq))
"""Substitute residues for properties
in flu ELMs file"""
import sys, global_settings, utils
from collections import defaultdict
with open(sys.argv[1]) as f:
for line in f:
(protein, st, stp, elm, seq, ELM) = line.strip().split('\t')
if 'X' not in seq and 'B' not in seq and 'Z' not in seq and 'J' not in seq:
print('\t'.join((protein, st, stp,
elm, utils.mk_sub(seq), ELM)))