def cor(ent1, ent2):
ls1 = []
ls2 = []
for elm in utils_distance.get_elements(ent1, ent2):
if elm in ent2 and elm in ent1:
ls1.append(ent1[elm])
ls2.append(ent2[elm])
#elif elm in ent1:
# ls1.append(ent1[elm])
# ls2.append(float(0))
#else:
# ls1.append(float(0))
# ls2.append(ent2[elm])
return utils_stats.pcc(ls1, ls2)
human_host_file = "working/Jun29/elmdict_H_sapiens.init"
chicken_host_file = "working/Jun29/elmdict_Gallus_gallus.init"
human_host_freqs = get_host_freqs(human_host_file)
chicken_host_freqs = get_host_freqs(chicken_host_file)
human_freqs = get_freqs(freq_file_human)
chicken_freqs = get_freqs(freq_file_chicken)
v1 = []
v2 = []
for key in human_freqs:
if key in uniq and key in chicken_freqs:
hf = human_freqs[key]
cf = chicken_freqs[key]
if abs(hf - cf) > 5:
sp = key.split(":")
elmseq = sp[1] + ":" + sp[2]
hhf = 0
chf = 0
if elmseq in human_host_freqs:
hhf = human_host_freqs[elmseq]
if elmseq in chicken_host_freqs:
chf = chicken_host_freqs[elmseq]
if chf != 0 and hhf != 0:
v1.append(hf / cf)
v2.append(float(hhf) / float(chf))
print key
print utils_stats.pcc(v1, v2)
with open("working/vals", "w") as f:
for a, b in zip(v1, v2):
f.write(str(a) + "\t" + str(b) + "\n")