chicken_all_vals) = plot_fore_control(outfile, protein,
chicken_fore,
chicken_control,
chicken_all,
chicken_host_freqs,
human_host_freqs)
human_control_pval = utils.my_wil_rank_sum_gtr(human_fore_vals,
human_control_vals,
human_all_vals)
chicken_control_pval = utils.my_wil_rank_sum_gtr(chicken_fore_vals,
chicken_control_vals,
chicken_all_vals)
if protein in ('polymerase PB1', 'neuraminidase', 'polymerase PB2',
'nonstructural protein 2', 'polymerase PA',
'matrix protein 2'):
human_pval = 'fore gtr ' + str(len(human_fore_vals)) + ' ' + str(len(human_control_vals) )+ ' ' + str(utils_stats.wilcox_gtr(human_fore_vals, human_control_vals))
p = utils_stats.wilcox_gtr(chicken_fore_vals, chicken_control_vals)
print 'pval', p, 'yo'
chicken_pval = 'fore gtr ' + str(len(chicken_fore_vals)) + ' ' + str(len(chicken_control_vals)) + ' ' + str(p)
else:
human_pval = 'ctrl gtr ' + str(len(human_control_vals)) + ' ' + str(len(human_fore_vals)) + ' ' + str(utils_stats.wilcox_gtr(human_control_vals, human_fore_vals))
chicken_pval = 'ctrl gtr ' + str(len(chicken_control_vals)) + ' ' + str(len(chicken_fore_vals)) + ' ' + str(utils_stats.wilcox_gtr(chicken_control_vals, chicken_fore_vals))
print 'human pval', protein, human_pval
print 'chicken pval', protein, chicken_pval
# print(protein, len(human_fore), len(human_control),
# len(chicken_fore), len(chicken_control),
# len(set(human_fore.keys()) & set(chicken_fore.keys())))
masters_human.append((human_fore, human_control))
masters_chicken.append((chicken_fore, chicken_control))
# for p1,p2 in itertools.combinations(masters_human, 2):
# print len(utils_graph.intersectLists([p1[0],p2[1]])), len(utils_graph.intersectLists([p2[0],p1[1]]))
import utils_stats
elm_ls = [[1, 'YIIK'], [1, 'YIVK'], [1, 'YLDK'], [1, 'YTIR'], [2, 'YLMA'], [3, 'YLLV'], [5, 'YIEG'], [10, 'YVNT'], [15, 'YTID'], [28, 'YVSM'], [129, 'YLLA'], [143, 'YLLT'], [260, 'YTLD'], [266, 'YINT'], [271, 'YVRT'], [273, 'YCVL'], [274, 'YLEK'], [275, 'YFTA'], [275, 'YIMK'], [277, 'YVDG']]
cut = 200
virus = []
nonvirus = []
found_seqs = {}
with open('results/elmdict_Gallus_gallus.txt') as f:
for line in f:
elm, seq, count, frac_st = line.strip().split('\t')
if elm == 'LIG_SH2_STAT5':
appended = False
for elm_count, elm_seq in elm_ls:
if seq == elm_seq:
found_seqs[seq] = True
if elm_count > cut:
virus.append(float(frac_st))
else:
nonvirus.append(float(frac_st))
appended = True
break
if not appended:
nonvirus.append(float(frac_st))
for count, seq in elm_ls:
if not seq in found_seqs and count > cut:
virus.append(float(0))
print utils_stats.wilcox_gtr(virus, nonvirus)
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):
# virus_like.append([elm+':'+seq,species2dict[host][elm][seq]])
virus_like.append(species2dict[host][elm][seq])
#else:
# non_virus.append(species2dict[host][elm][seq])
else:
non_virus.append(species2dict[host][elm][seq])
else:
for seq in species2dict[host][elm]:
non_virus.append(species2dict[host][elm][seq])
print utils_stats.wilcox_gtr(virus_like, non_virus)
with open('virus', 'w') as f:
for item in virus_like:
f.write('blank\t' + str(item) + '\n')
with open('nonvirus', 'w') as f:
for item in non_virus:
f.write('blank\t' + str(item) + '\n')
print len(virus_like), len(non_virus)
found_seqs = {}
for seq in elm_counts[protein][elm]:
if float(elm_counts[protein][elm][seq])/protein_count > float(.9):
if seq in host_freqs[elm]:
virus_freqs.append(host_freqs[elm][seq])
found_seqs[seq] = True
else:
virus_freqs.append(float(0))
else:
if seq in host_freqs[elm]:
non_virus_freqs.append(host_freqs[elm][seq])
found_seqs[seq] = True
else:
non_virus_freqs.append(float(0))
for seq in host_freqs[elm]:
if not seq in found_seqs:
non_virus_freqs.append(host_freqs[elm][seq])
#line = ''
if len(virus_freqs) > 2 and len(non_virus_freqs) > 2:
lines += protein + '\t'+ elm + '\t'+ str(utils_stats.wilcox_gtr(virus_freqs, non_virus_freqs)) + '\t' + str(utils_stats.wilcox_less(virus_freqs, non_virus_freqs)) + '\n'
#else:
# line = protein + '\t'+ elm + '\t'+ 'NO_DATA(' + str(len(virus_freqs)) + ',' + str(len(non_virus_freqs)) + ')'
#lines += line + '\n'
with open(ofile,'w') as f:
f.write(lines)
virus_freqs.append(host_freqs[elm][seq])
found_seqs[seq] = True
else:
virus_freqs.append(float(0))
else:
if seq in host_freqs[elm]:
non_virus_freqs.append(host_freqs[elm][seq])
found_seqs[seq] = True
else:
non_virus_freqs.append(float(0))
for seq in host_freqs[elm]:
if not seq in found_seqs:
non_virus_freqs.append(host_freqs[elm][seq])
# #line = ''
if len(virus_freqs) > 2 and len(non_virus_freqs) > 2:
lines += (
protein
+ "\t"
+ elm
+ "\t"
+ str(utils_stats.wilcox_gtr(virus_freqs, non_virus_freqs))
+ "\t"
+ str(utils_stats.wilcox_less(virus_freqs, non_virus_freqs))
+ "\n"
)
# #else:
# # line = protein + '\t'+ elm + '\t'+ 'NO_DATA(' + str(len(virus_freqs)) + ',' + str(len(non_virus_freqs)) + ')'
# #lines += line + '\n'
with open(ofile, "w") as f:
f.write(lines)