def run_tests(names, num_items, num_its, type_=int):
fns = dict([
('btrees', lambda: BTrees.OOBTree.OOBTree()),
('blist', lambda: blist.sorteddict()),
('bintrees', lambda: bintrees.FastRBTree()), ('dict', lambda: dict()),
('banyan_red_black_tree',
lambda: banyan.SortedDict(key_type=type_, alg=banyan.RED_BLACK_TREE)),
('banyan_red_black_tree_rank_updator',
lambda: banyan.SortedDict(key_type=type_,
alg=banyan.RED_BLACK_TREE,
updator=banyan.RankUpdator)),
('banyan_red_black_tree_min_max_updator',
lambda: banyan.SortedDict(key_type=type_,
alg=banyan.RED_BLACK_TREE,
updator=banyan.MinMaxUpdator)),
('banyan_splay_tree',
lambda: banyan.SortedDict(key_type=type_, alg=banyan.SPLAY_TREE)),
('banyan_sorted_list',
lambda: banyan.SortedDict(key_type=type_, alg=banyan.SORTED_LIST)),
('banyan_red_black_tree_gen',
lambda: banyan.SortedDict(alg=banyan.RED_BLACK_TREE)),
('banyan_splay_tree_gen',
lambda: banyan.SortedDict(alg=banyan.SPLAY_TREE)),
('banyan_sorted_list_gen',
lambda: banyan.SortedDict(alg=banyan.SORTED_LIST))
])
t = dict([])
for name in names:
t[name] = _run_test(fns[name], type_, num_items, num_its)
return t
def construct_intervals(mesh, zhat):
points, triangles = mesh
intervals = banyan.SortedDict([], key_type = (float,float), updator = banyan.OverlappingIntervalsUpdator)
for i, tri in enumerate(triangles):
a,b,c = tri
z = [np.dot(zhat, points[a]),np.dot(zhat, points[b]),np.dot(zhat, points[c])]
intervals[(min(z),max(z))] = i
return intervals
def run_tests(names, num_items, num_its):
# Tmp Ami - make key-type separate labels below
fns = dict([
('btrees', lambda es: BTrees.OOBTree.OOBTree([(e, 1) for e in es])),
('blist', lambda es: blist.sorteddict([(e, 1) for e in es])),
('bintrees', lambda es: bintrees.FastRBTree([(e, 1) for e in es])),
('set', lambda es: set([(e, 1) for e in es])),
('banyan_red_black_tree', lambda es: banyan.SortedDict(
[(e, 1) for e in es], alg=banyan.RED_BLACK_TREE)),
('banyan_splay_tree', lambda es: banyan.SortedDict(
[(e, 1) for e in es], alg=banyan.SPLAY_TREE)),
('banyan_sorted_list', lambda es: banyan.SortedDict(
[(e, 1) for e in es], alg=banyan.SORTED_LIST)),
('banyan_red_black_tree_gen', lambda es: banyan.SortedDict(
[(e, 1) for e in es], key_type=int, alg=banyan.RED_BLACK_TREE)),
('banyan_splay_tree_gen', lambda es: banyan.SortedDic(
[(e, 1) for e in es], key_type=int, alg=banyan.SPLAY_TREE)),
('banyan_sorted_list_gen', lambda es: banyan.SortedDict(
[(e, 1) for e in es], key_type=int, alg=banyan.SORTED_LIST))
])
t = dict([])
for name in names:
t[name] = _run_test(fns[name], int, num_items, num_its)
return t
if ((precise) and
(call['RV'] >= 2)) or ((not precise) and
(call['DV'] >= 2)):
gqAlt.append(call['GQ'])
genotypeRatio = float(len(gqAlt) + len(gqRef)) / float(
len(record.samples))
if genotypeRatio > ratioGeno:
if (len(gqRef)) and (len(gqAlt)) and (
numpy.median(gqRef) >=
gqRefCut) and (numpy.median(gqAlt) >= gqAltCut):
if (numpy.percentile(ratioRef, 99)
== 0) and (numpy.median(ratioAlt) >= altAF):
#print(record.INFO['END']-record.POS, len(gqRef), len(gqAlt), numpy.median(gqRef), numpy.median(gqAlt), numpy.percentile(ratioRef, 99), numpy.median(ratioAlt), genotypeRatio, sep="\t")
if not sv.has_key(record.CHROM):
sv[record.CHROM] = banyan.SortedDict(
key_type=(int, int),
alg=banyan.RED_BLACK_TREE,
updator=banyan.OverlappingIntervalsUpdator)
if (record.POS,
record.INFO['END']) not in sv[record.CHROM]:
sv[record.CHROM][(record.POS,
record.INFO['END'])] = (
record.ID, record.INFO['PE'],
record.INFO['CT'])
else:
svDups[(record.CHROM, record.POS,
record.INFO['END'])].append(
(record.ID, record.INFO['PE'],
record.INFO['CT']))
# Output vcf records
if args.vcfFile:
def __init__(self):
self.ports = collections.defaultdict(lambda: banyan.SortedDict(
key_type=(float, float),
updator=banyan.OverlappingIntervalsUpdator))
metavar='out.vcf',
required=True,
dest='outVCF',
help='output vcf file (required)')
args = parser.parse_args()
# Compute all stretches of Ns in the reference
nRun = dict()
refLen = dict()
f_in = gzip.open(args.ref) if args.ref.endswith('.gz') else open(args.ref)
for seqName, seqNuc, seqQuals in readfq(f_in):
refLen[seqName] = len(seqNuc)
print("Processing", seqName, refLen[seqName])
if not nRun.has_key(seqName):
nRun[seqName] = banyan.SortedDict(
key_type=(int, int),
alg=banyan.RED_BLACK_TREE,
updator=banyan.OverlappingIntervalsUpdator)
for m in re.compile("([Nn]+)").finditer(seqNuc):
nRun[seqName][m.span()] = 1
# Add read-depth control region to VCF file
if args.vcfFile:
vcf_reader = vcf.Reader(
open(args.vcfFile), 'r',
compressed=True) if args.vcfFile.endswith('.gz') else vcf.Reader(
open(args.vcfFile), 'r', compressed=False)
if 'CONTROL' not in vcf_reader.infos.keys():
vcf_reader.infos['CONTROL'] = vcf.parser._Info('CONTROL', 1, 'Integer',
'Control variant.')
vcf_writer = vcf.Writer(open(args.outVCF, 'w'),
vcf_reader,