"""Remove duplicates in a fasta file"""
import sys
from dp.associations import GeneAssociations
from dp.ontology import Ontology
from collections import Counter
from dp.utils import parseFasta
seqs = set()
names = set()
fastafile = open(sys.argv[1])
MIN_SEQ_LEN = 32
MAX_SEQ_UNK = 0.1
TAXONS_HOMMO_SAPIENS = {9606}
asoc = GeneAssociations.fromFile(sys.argv[2], taxons=TAXONS_HOMMO_SAPIENS)
ontology = Ontology(sys.argv[3])
ontology.setAssociations(asoc)
asoc.transitiveClosure()
associated = set()
for k, v in asoc.associations.items():
associated.update({g.upper() for g in v})
ss = dict(parseFasta("data/ss.txt"))
#print(associated)
for l in fastafile:
name, typ, *_ = l[1:].split(" ")
name = name.upper()
seq = next(fastafile)
sskey = "%s:secstr" % name.replace("_", ":")
if typ != 'mol:protein' \
help="TreeLiker jar binary.")
options, args = parser.parse_args()
if len(args) != 2:
parser.error("Incorect number of arguments!")
oboFileName, associationsFileName = args
dp.utils.verbosity = options.verbosity
if options.backgroundKnowledge:
with open(options.backgroundKnowledge) as bk:
Gene.backgroundKnowledge = bk.read().splitlines()
ontology = Ontology(oboFileName)
ontology.geneFactory.deserialize = True if options.deserialize is None else False
#associations = GeneAssociations(associationsFileName, TAXONS_SACCHAROMYCES_CEREVISIAE)
dataset = None
if options.dataset:
# FIXME: When dataset is changed, serialized associations need to be regenerated. This is serious bug if we don't seed random to a constant
dataset = [l.strip() for l in open(options.dataset)]
random.shuffle(dataset)
#assert options.reserve > 0.0
#if options.reserve < 1.0: # Use ratio
# splitIndex = int(options.reserve * len(dataset))
#else:
# splitIndex = int(options.reserve)
#reserved = set(dataset[:splitIndex])
#dataset = set(dataset[splitIndex:])