def read_dir(self, d):
lookup = dict()
scores_by_what = dict()
if not os.path.exists(d):
for p in DATA_SEARCH_PATH:
if os.path.exists(os.path.join(p, d)):
d = os.path.join(p, d)
break
for file in os.listdir(d):
if file.endswith(".match"):
for line in (open(os.path.join(d, file))):
if line.startswith("#"): continue
line = line.strip()
f = line.split()
if len(f) >= 4:
chr, start, end, what = f[0:4]
else:
continue
if not chr in lookup: lookup[chr] = intervals.Intersecter()
lookup[chr].add_interval(
intervals.Interval(int(start), int(end), what))
else:
k = file.split('.')[0]
if not k in scores_by_what:
scores_by_what[k] = FileBinnedArray(
open(os.path.join(d, file)))
if lookup == {}:
self.lookup = None
self.scores = scores_by_what
#return None, scores_by_what
else:
self.lookup = lookup
self.scores = scores_by_what
def __init__(self,
qualfiles=None,
qualspecies=None,
minqual=None,
mask="?",
cache=100):
if not qualfiles:
raise Exception("No quality files.")
if not qualspecies:
raise Exception("No species dictionary.")
if not minqual:
raise Exception("No minimum quality specified.")
self.mask = "?"
self.minqual = minqual
self.mask = mask
self.total = 0
self.masked = 0
self.qualfiles = qualfiles
self.qualspecies = qualspecies
self.cache = cache * 2 # typical bin size is 512K
# load quality files into FileBinnedArray
self.qualities = {}
for species, qualfile in self.qualfiles.items():
specdict = {}
for chrom in self.qualspecies[species]:
specdict[chrom] = FileBinnedArray( \
open(qualfile + "." + chrom + ".bqv", "rb"), \
cache = self.cache/len(qualfiles) )
self.qualities[species] = specdict
def __getitem__(self, key):
value = None
if key in self.cache:
value = self.cache[key]
else:
fname = os.path.join(self.dir, "%s.ba" % key)
if os.path.exists(fname):
value = FileBinnedArray(open(fname))
self.cache[key] = value
if value is None:
raise KeyError("File does not exist: " + fname)
return value