def test_genome_index(index_dir, loc_and_seq):
from gimmemotifs.genome_index import GenomeIndex
g = GenomeIndex(index_dir)
loc, seq = loc_and_seq
assert seq == g.get_sequence(*loc)
def index_dir():
from gimmemotifs.genome_index import GenomeIndex
test_index_dir = 'tests/data/index/'
g = GenomeIndex()
g.create_index('tests/data/genome/', test_index_dir)
return test_index_dir
def index(args):
if not os.path.exists(args.indexdir):
print "Index_dir %s does not exist!" % (args.indexdir)
sys.exit(1)
fasta_dir = args.fastadir
index_dir = os.path.join(args.indexdir, args.indexname)
g = GenomeIndex()
g.create_index(fasta_dir, index_dir)
create_bedtools_fa(index_dir, fasta_dir)
def __init__(self,
session=None,
conn='mysql://pita:@localhost/pita',
new=False,
index=None):
self.logger = logging.getLogger("pita")
# initialize db session
if session:
self.session = session
else:
self._init_session(conn, new)
# index to retrieve sequence
self.index = None
if index:
self.index = GenomeIndex(index)
self.cache_splice_stats = {}
self.cache_feature_stats = {}
def test2_as_fasta(self):
""" convert bed, regions, etc to Fasta """
tmpdir = mkdtemp()
g = GenomeIndex()
g.create_index(self.genome_dir, tmpdir)
fafile = os.path.join(self.datadir, "test.fa")
fa = Fasta(fafile)
bedfile = os.path.join(self.datadir, "test.bed")
regionfile = os.path.join(self.datadir, "test.txt")
regions = [l.strip() for l in open(regionfile)]
self.assertTrue(isinstance(as_fasta(fa), Fasta))
self.assertTrue(isinstance(as_fasta(fafile), Fasta))
self.assertTrue(isinstance(as_fasta(bedfile, tmpdir), Fasta))
self.assertTrue(isinstance(as_fasta(regionfile, tmpdir), Fasta))
self.assertTrue(isinstance(as_fasta(regions, tmpdir), Fasta))
with self.assertRaises(ValueError):
as_fasta(bedfile)
rmtree(tmpdir)
def _scan_regions(self, regions, nreport, scan_rc, cutoff=0.95):
index_dir = self.index_dir
motif_file = self.motifs
motif_digest = self.checksum.get(motif_file, None)
# determine which regions are not in the cache
scan_regions = regions
if self.use_cache:
scan_regions = []
for region in regions:
key = str((region, index_dir, motif_digest, nreport, scan_rc,
cutoff))
ret = self.cache.get(key)
if ret == NO_VALUE:
scan_regions.append(region)
# scan the regions that are not in the cache
if len(scan_regions) > 0:
n = 12
genome_index = GenomeIndex(index_dir)
motifs = load_motifs(motif_file, cutoff)
scan_func = partial(scan_region_mult,
genome_index=genome_index,
motifs=motifs,
nreport=nreport,
scan_rc=scan_rc)
jobs = []
chunksize = len(scan_regions) / n + 1
for i in range(n):
job = pool.apply_async(
scan_func,
(scan_regions[i * chunksize:(i + 1) * chunksize], ))
jobs.append(job)
# return values or store values in cache
i = 0
for job in jobs:
for ret in job.get():
if self.use_cache:
# store values in cache
region = scan_regions[i]
key = str((region, index_dir, motif_digest, nreport,
scan_rc, cutoff))
self.cache.set(key, ret)
else:
#return values
yield ret
i += 1
if self.use_cache:
# return results from cache
for region in regions:
key = str((region, index_dir, motif_digest, nreport, scan_rc,
cutoff))
ret = self.cache.get(key)
if ret == NO_VALUE or ret is None:
raise Exception("cache is not big enough to hold all "
"results, try increasing the cache size "
"or disable cache")
yield ret
