min_gene_len = int(argv[6])
min_rpkm = float(argv[7])
min_diff = log(
float(argv[8]), 2
) # minimum fold difference between the highest and lowest to be considered (filter)
# make cdt file
labels = rpkm_file.readline().rstrip('\n').split('\t')[2::]
my_str = "UID\tNAME\tGWEIGHT"
my_str2 = "EWEIGHT\t\t"
for label in labels:
my_str += "\t" + label
my_str2 += "\t1"
cdt_file.write(my_str + '\n' + my_str2)
counts = E.Counter()
for line in rpkm_file:
data = line[:-1].split("\t")
counts.input += 1
if int(data[1]) < min_gene_len: # exclude genes that are too short
counts.skipped_length += 1
continue
name = data[0]
la = map(float, data[2:])
if max(la) < min_rpkm:
counts.skipped_rpkm += 1
continue
background_file.write(name + "\n")
