#!/usr/bin/env python
"""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("_",":")
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:])
dataset = set(dataset)
associations = GeneAssociations.fromFile(associationsFileName,
taxons=TAXONS,
dataset=dataset)
#reservedAssociations = GeneAssociations.fromFile(associationsFileName+"_reserved", dataset = reserved)
ontology.setAssociations(associations)
#ontology.setAssociations(reservedAssociations, 'reserved')
if options.associationsDump:
associations.serialize(options.associationsDump)
#reservedAssociations.serialize(options.associationsDump+"_reserved")
sys.exit()
ontology.deleteSmallTerms(options.lb)
associations.shrink(options.max, options.lb)
ontology.overView()
ontology.dotExport()
#!/usr/bin/env python
"""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("_", ":")
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:])
dataset = set(dataset)
associations = GeneAssociations.fromFile(associationsFileName, taxons = TAXONS, dataset = dataset)
#reservedAssociations = GeneAssociations.fromFile(associationsFileName+"_reserved", dataset = reserved)
ontology.setAssociations(associations)
#ontology.setAssociations(reservedAssociations, 'reserved')
if options.associationsDump:
associations.serialize(options.associationsDump)
#reservedAssociations.serialize(options.associationsDump+"_reserved")
sys.exit()
ontology.deleteSmallTerms(options.lb)
associations.shrink(options.max, options.lb)
ontology.overView()
ontology.dotExport()
