def no_args():
species2dict = {}
for g in GENOMES:
d = utils.get_seq2count_dict('results/elmdict_'
+ g + '.txt',
float(0))
entropy = utils.get_species_entropy(d)
with open(os.path.join(RESULTSDIR, g + '.elm_entropy'), 'w') as f:
for elm in entropy:
f.write(elm + '\t' + str(entropy[elm]) + '\n')
for flu in FLU_NAMES:
d = utils.get_seq2count_dict('results/flu_elmdict_'
+ flu,
float(0))
entropy = utils.get_species_entropy(d)
with open(os.path.join(RESULTSDIR, 'flu_' + flu + '.elm_entropy'), 'w') as f:
for elm in entropy:
f.write(elm + '\t' + str(entropy[elm]) + '\n')
def main(args):
suffix = sys.argv[1]
elms = sys.argv[2:]
file_species_pairs = []
for g in global_settings.GENOMES:
file_species_pairs.append(('results/elmdict_'
+ g + suffix, g))
plot_dir = 'plots/for_aydin/'
species2elms = {}
for file, species in file_species_pairs:
species2elms[species] = utils.get_seq2count_dict(file, float(0))
for elm in elms:
utils_plot.elm_host_barplot(species2elms, elm,
os.path.join(plot_dir,
elm + '.hosts'
+ suffix + '.png'))
def main(args):
file_species_pairs = []
i = 1
while i < len(args)-2:
file_species_pairs.append([args[i], args[i+1]])
i += 2
cutoff = float(sys.argv[-2])
plot_dir = sys.argv[-1]
species2elms = {}
for file, species in file_species_pairs:
species2elms[species] = utils.get_seq2count_dict(file, cutoff)
elms = {}
for species in species2elms:
for elm in species2elms[species]:
elms[elm] = True
for elm in elms:
if utils.check_ones(species2elms, elm) and (elm in species2elms['swine'] or elm in species2elms['human'] or elm in species2elms['chicken']) and elm in species2elms['H_sapiens'] :
utils_plot.elm_host_barplot(species2elms, elm,
os.path.join(plot_dir,
elm + '.hosts.png'))
def main(args):
file_species_pairs = []
i = 1
while i < len(args)-2:
file_species_pairs.append([args[i], args[i+1]])
i += 2
cutoff = float(sys.argv[-2])
plot_dir = sys.argv[-1]
species2elms = {}
virus2elms = {}
# first grab virus ELMs
for file, species in file_species_pairs:
if file.find('flu') != -1:
virus2elms[species] = utils.get_seq2count_dict(file, cutoff)
else:
species2elms[species] = True
elms = {}
for species in virus2elms:
for elm in virus2elms[species]:
elms[elm] = True
for species in species2elms:
species2elms[species] = utils.get_seq2count_dict_for_seqs(file,
cutoff,
virus2elms)
for virus in virus2elms:
species2elms[virus] = virus2elms[virus]
for elm in elms:
if utils.check_ones(species2elms, elm):
if utils_distance.distance_elms(species2elms['Sus_scrofa'][elm],
species2elms['H_sapiens'][elm]) > float(-1) or utils_distance.distance_elms(species2elms['Sus_scrofa'][elm],
species2elms['Gallus_gallus'][elm]) > float(0):
utils_plot.elm_host_barplot(species2elms, elm,
os.path.join(plot_dir,
elm + '.virus_hosts.png'))
#'HCV':('all',)}
virus2conservedELMs = {}
all_elms = {}
for virus in viruses:
virus2conservedELMs[virus] = getConservedELMs(virus, subtypes)
for elm in virus2conservedELMs[virus]:
all_elms[elm] = True
# load ELM seq fractions
host2elmFreqs = {}
virus2elmFreqs = {}
use_seqs = {}
for host in hosts:
host2elmFreqs[host] = utils.get_seq2count_dict(os.path.join(local_settings.RESULTSDIR,
'elmdict_' + host + suffix),
float(0))
for elm in host2elmFreqs[host]:
if elm not in use_seqs:
use_seqs[elm] = {}
for seq in host2elmFreqs[host][elm]:
if seq not in use_seqs[elm]:
use_seqs[elm][seq] = 0
use_seqs[elm][seq] += 1
for elm in use_seqs:
rm_ls = []
for seq in use_seqs[elm]:
if use_seqs[elm][seq] != len(hosts.keys()):
rm_ls.append(seq)
for seq in rm_ls:
""" I need to find ELM sequences that differ across
species hosts and viruses.
"""
import sys, utils, utils_distance, itertools, utils_plot
from global_settings import *
virus2dict = {}
virus2dict['chicken'] = utils.get_seq2count_dict('results/flu_elmdict_swine',
float(0.05))
virus2dict['swine'] = utils.get_seq2count_dict('results/flu_elmdict_chicken',
float(0.05))
virus2dict['human'] = utils.get_seq2count_dict('results/flu_elmdict_human',
float(0.05))
genomes = ('H_sapiens', 'Gallus_gallus', 'Sus_scrofa')
species2dict = {}
for g in genomes:
species2dict[g] = utils.get_seq2count_dict('results/elmdict_'
+ g + '.txt',
float(0.01))
use_elms = {}
for elm in virus2dict['human']:
if elm in virus2dict['chicken'] and elm in virus2dict['swine']:
distance_is_0 = False
for v1, v2 in itertools.combinations(virus2dict.keys(), 2):
distance = utils_distance.distance_elms(virus2dict[v1][elm],
virus2dict[v2][elm])
if distance == float(0):
distance_is_0 = True
if not distance_is_0:
import utils
d = {}
for g in ('human', 'chicken', 'swine', 'equine'):
d[g] = utils.get_seq2count_dict('results/' + g + '.elms.90.freq.redo', float(0))
top_seqs = {}
for g in d:
for elm in d[g]:
ls = []
for seq in d[g][elm]:
ls.append([d[g][elm][seq], seq])
ls.sort()
if not g in top_seqs:
top_seqs[g] = {}
top_seqs[g][elm] = ls[-1][1]
for elm in top_seqs['human']:
if elm in top_seqs['swine']:
if top_seqs['human'][elm] != top_seqs['swine'][elm] and top_seqs['human'][elm] != top_seqs['chicken'][elm] and top_seqs['human'][elm] != top_seqs['equine'][elm]:
print elm + '\t' + top_seqs['human'][elm] + '\t' + top_seqs['swine'][elm] + '\t' + top_seqs['chicken'][elm] + '\t' + top_seqs['equine'][elm]
'equine':('H3N8',),
'chicken':('H5N1', 'H9N2'),
'duck':('H5N1', 'H9N2'),
'HIV':('all',),
'HCV':('all',)}
virus2conservedELMs = {}
for virus in viruses:
virus2conservedELMs[virus] = getConservedELMs(virus, subtypes)
# load ELM seq fractions
host2elmFreqs = {}
virus2elmFreqs = {}
for host in hosts:
host2elmFreqs[host] = utils.get_seq2count_dict(os.path.join(local_settings.RESULTSDIR,
'elmdict_' + host + '.txt'),
float(0))
#tmp_input = 'tmp_input' + str(random.randint(0,100))
tmp_input = 'plots/for_aydin/cos_host_host.tab'
with open(tmp_input, 'w') as f:
f.write('Host_Host\tELM\tDistance\n')
for elm in virus2conservedELMs[virus]:
counter = 0
for host1,host2 in (('H_sapiens', 'Macaca_mulatta'),
('H_sapiens', 'M_musculus'),
('H_sapiens', 'R_norvegicus'),
('H_sapiens', 'Sus_scrofa'),
('H_sapiens', 'Equus_caballus'),
('H_sapiens', 'Canis_familiaris'),
('H_sapiens', 'Bos_taurus'),
('H_sapiens', 'Gallus_gallus'),
def one_arg(afile):
d = utils.get_seq2count_dict(afile, float(0))
entropy = utils.get_species_entropy(d)
for elm in entropy:
print elm + '\t' + str(entropy[elm])
# float(.05))
# mouse = elm_hists.get_seq2count_dict('results/elmdict_M_musculus.txt',
# float(.05))
# monkey = elm_hists.get_seq2count_dict('results/elmdict_Macaca_mulatta.txt',
# float(.05))
# print utils_distance.distance_species(human,
# monkey)
# print utils_distance.distance_species(human,
# mouse)
# print utils_distance.distance_species(monkey,
# mouse)
species2dict = {}
virus2dict = {}
virus2dict['swineFlu'] = utils.get_seq2count_dict('results/flu_elmdict_swine',
float(.4))
virus2dict['chickenFlu'] = utils.get_seq2count_dict('results/flu_elmdict_chicken',
float(.4))
virus2dict['humanFlu'] = utils.get_seq2count_dict('results/flu_elmdict_human',
float(.4))
for g in ('H_sapiens', 'Gallus_gallus', 'Sus_scrofa'):
species2dict[g] = utils.get_seq2count_dict_for_seqs('results/elmdict_'
+ g + '.txt',
float(0),
virus2dict)
for v in virus2dict:
species2dict[v] = virus2dict[v]
d = utils_distance.distance_matrix(species2dict)
elm_d = utils_distance.elm_distance_matrix(species2dict)
#for elm in elm_d:
import utils, utils_plot, utils_distance
d = {'web':utils.get_seq2count_dict('results/human.website.elm.elmdict',
float(.01)),
'regex':utils.get_seq2count_dict('results/hprd_new.regex.elms.elmdict',
float(.01))}
elms = utils_distance.get_elements(d['web'], d['regex'])
for elm in elms:
utils_plot.elm_host_barplot(d, elm, 'plots/hprd/' + elm + '.png')
""" Each human ELM/seq pair has a normalized fraction
(normalized to that ELM). I'll split the ELM/seq
pairs into virus and non-virus, and use the
one-sided wilcoxon test to see if virus
ELM/seq pairs have lower fractions.
"""
import utils, os, utils_stats
from global_settings import *
from local_settings import *
species2dict = {}
flu2dict = {}
for g in ['H_sapiens']:#GENOMES:
species2dict[g] = utils.get_seq2count_dict('results/elmdict_'
+ g + '.redo',
float(0))
for flu in ['HIV']:#FLU_NAMES:
flu2dict[flu] = utils.get_seq2count_dict('results/hiv_freq',
float(0.1))
virus_like = []
non_virus = []
host = 'H_sapiens'
virus = 'HIV'
for elm in species2dict[host]:
if elm in flu2dict[virus]:
for seq in species2dict[host][elm]:
if seq in flu2dict[virus][elm]:
#if flu2dict[virus][elm][seq] > float(.05):
