def test_basic_construction(self):
db = SequenceFileDB(self.dbfile)
try:
assert str(db.get('seq1')).startswith('atggtgtca')
assert str(db.get('seq2')).startswith('GTGTTGAA')
finally:
db.close()
def test_build_seqLenDict_with_reader(self):
"Test that building things works properly when specifying a reader."
class InfoBag(object):
def __init__(self, **kw):
self.__dict__.update(kw)
# first, load the db & save the sequence info in a list
l = []
db = SequenceFileDB(self.dbfile)
try:
for k, v in db.items():
info = InfoBag(id=k, length=len(v), sequence=str(v))
l.append(info)
finally:
# now, erase the existing files, and recreate the db.
db.close()
self.trash_intermediate_files()
# create a fake reader with access to the saved info
def my_fake_reader(fp, filename, info_list=l):
return info_list
# now try creating with the fake reader
db = SequenceFileDB(self.dbfile, reader=my_fake_reader)
# did it work?
try:
assert str(db.get('seq1')).startswith('atggtgtca')
assert str(db.get('seq2')).startswith('GTGTTGAA')
finally:
db.close()
def test_basic_construction(self):
db = SequenceFileDB(self.dbfile)
try:
assert str(db.get('seq1')).startswith('atggtgtca')
assert str(db.get('seq2')).startswith('GTGTTGAA')
finally:
db.close()
def test_inverse_add_behavior(self):
dnaseq = testutil.datafile('dnaseq.fasta')
seqdb = SequenceFileDB(dnaseq)
try:
seq = seqdb['seq1']
name = (~self.db)[seq]
finally:
seqdb.close() # only need to close if exception occurs
def test_inverse_add_behavior(self):
dnaseq = testutil.datafile('dnaseq.fasta')
seqdb = SequenceFileDB(dnaseq)
try:
seq = seqdb['seq1']
name = (~self.db)[seq]
finally:
seqdb.close() # only need to close if exception occurs
def test_funny_key2(self):
"check handling of ID containing multiple separators"
dnaseq = testutil.datafile('funnyseq.fasta')
seqdb = SequenceFileDB(dnaseq) # contains 'seq1', 'seq2'
try:
pudb = PrefixUnionDict({'prefix': seqdb})
seq = pudb['prefix.seq.2.even.longer']
finally:
seqdb.close()
def test_funny_key2(self):
"check handling of ID containing multiple separators"
dnaseq = testutil.datafile('funnyseq.fasta')
seqdb = SequenceFileDB(dnaseq) # contains 'seq1', 'seq2'
try:
pudb = PrefixUnionDict({'prefix': seqdb})
seq = pudb['prefix.seq.2.even.longer']
finally:
seqdb.close()
def test_cache(self):
"Sequence slice cache mechanics."
dnaseq = testutil.datafile('dnaseq.fasta')
db = SequenceFileDB(dnaseq)
try:
# create cache components
cacheDict = {}
cacheHint = db.cacheHint
# get seq1
seq1 = db['seq1']
# _cache is only created on first cache attempt
assert not hasattr(db, '_cache')
# build an 'owner' object
class AnonymousOwner(object):
pass
owner = AnonymousOwner()
# save seq1 in cache
cacheDict['seq1'] = (seq1.start, seq1.stop)
cacheHint(cacheDict, owner)
del cacheDict # 'owner' now holds reference
# peek into _cache and assert that only the ival coordinates
# are stored
v = db._cache.values()[0]
assert len(v['seq1']) == 2
del v
# force a cache access & check that now we've stored actual string
ival = str(seq1[5:10])
v = db._cache.values()[0]
# ...check that we've stored actual string
assert len(v['seq1']) == 3
# again force cache access, this time to the stored sequence string
ival = str(seq1[5:10])
# now, eliminate all references to the cache proxy dict
del owner
# trash unused objects - not strictly necessary, because there are
# no islands of circular references & so all objects are already
# deallocated, but that's implementation dependent.
gc.collect()
# ok, cached values should now be gone.
v = db._cache.values()
assert len(v) == 0
finally:
db.close()
def test_cache(self):
"Sequence slice cache mechanics."
dnaseq = testutil.datafile('dnaseq.fasta')
db = SequenceFileDB(dnaseq)
try:
# create cache components
cacheDict = {}
cacheHint = db.cacheHint
# get seq1
seq1 = db['seq1']
# _cache is only created on first cache attempt
assert not hasattr(db, '_cache')
# build an 'owner' object
class AnonymousOwner(object):
pass
owner = AnonymousOwner()
# save seq1 in cache
cacheDict['seq1'] = (seq1.start, seq1.stop)
cacheHint(cacheDict, owner)
del cacheDict # 'owner' now holds reference
# peek into _cache and assert that only the ival coordinates
# are stored
v = db._cache.values()[0]
assert len(v['seq1']) == 2
del v
# force a cache access & check that now we've stored actual string
ival = str(seq1[5:10])
v = db._cache.values()[0]
# ...check that we've stored actual string
assert len(v['seq1']) == 3
# again force cache access, this time to the stored sequence string
ival = str(seq1[5:10])
# now, eliminate all references to the cache proxy dict
del owner
# trash unused objects - not strictly necessary, because there are
# no islands of circular references & so all objects are already
# deallocated, but that's implementation dependent.
gc.collect()
# ok, cached values should now be gone.
v = db._cache.values()
assert len(v) == 0
finally:
db.close()
def test_basic_iadd(self):
dnaseq = testutil.datafile('dnaseq.fasta')
seqdb = SequenceFileDB(dnaseq)
try:
new_seq = seqdb['seq1']
self.db += new_seq
assert new_seq in self.db
name = (~self.db)[new_seq]
assert name == 'dnaseq.seq1', name
###
seqdb2 = SequenceFileDB(dnaseq)
try:
# Munge the filepath for testing.
seqdb2.filepath = 'foo'
new_seq2 = seqdb2['seq1']
self.db += new_seq2
name2 = (~self.db)[new_seq2]
assert name2 == 'foo.seq1', name2
finally:
seqdb2.close()
finally:
seqdb.close()
def test_build_seqLenDict_with_bad_reader(self):
"Test that building things fails properly with a bad reader."
class InfoBag(object):
def __init__(self, **kw):
self.__dict__.update(kw)
# first, load the db & save the sequence info in a list
l = []
db = SequenceFileDB(self.dbfile)
try:
for k, v in db.items():
info = InfoBag(id=k, length=0, sequence=str(v))
l.append(info)
finally:
# now, erase the existing files, and recreate the db.
db.close()
self.trash_intermediate_files()
# create a fake reader with access to the saved info
def my_fake_reader(fp, filename, info_list=l):
return info_list
# now try creating with the fake reader
try:
db = SequenceFileDB(self.dbfile, reader=my_fake_reader)
try:
assert 0, "should not reach here; db construction should fail!"
finally:
db.close()
except ValueError:
pass # ValueError is expected
def test_nlmsaslice_cache(self):
"NLMSASlice sequence caching & removal"
# set up sequences
dnaseq = testutil.datafile('dnaseq.fasta')
db = SequenceFileDB(dnaseq, autoGC=-1) # use pure WeakValueDict...
try:
gc.collect()
assert len(
db._weakValueDict) == 0, '_weakValueDict should be empty'
seq1, seq2 = db['seq1'], db['seq2']
assert len(db._weakValueDict)==2, \
'_weakValueDict should have 2 seqs'
# build referencing NLMSA
mymap = NLMSA('test', 'memory', db, pairwiseMode=True)
mymap += seq1
mymap[seq1] += seq2
mymap.build()
# check: no cache
assert not hasattr(db, '_cache'), 'should be no cache yet'
seq1, seq2 = db['seq1'], db['seq2'] # re-retrieve
# now retrieve a NLMSASlice, forcing entry of seq into cache
ival = seq1[5:10]
x = mymap[ival]
assert len(db._cache.values()) != 0
n1 = len(db._cache)
assert n1 == 1, "should be exactly one cache entry, not %d" % \
(n1, )
# ok, now trash referencing arguments & make sure of cleanup
del x
gc.collect()
assert len(db._cache.values()) == 0
n2 = len(db._cache)
assert n2 == 0, '%d objects remain; cache memory leak!' % n2
# FAIL because of __dealloc__ error in cnestedlist.NLMSASlice.
# Drop our references, the cache should empty.
del mymap, ival, seq1, seq2
gc.collect()
# check that db._weakValueDict cache is empty
assert len(
db._weakValueDict) == 0, '_weakValueDict should be empty'
finally:
db.close()
def test_headerfile_create_conflict(self):
"test non-empty prefixDict with a passed in PUD header file: conflict"
subdb = SequenceFileDB(self.dbfile)
try:
header = testutil.datafile('prefixUnionDict-1.txt')
try:
db = PrefixUnionDict(filename=header, prefixDict={ 'foo' : subdb })
assert 0, "should not get here"
except TypeError:
pass
finally:
subdb.close()
def load_resource(self):
#load genome sequence
print 'loading the genome sequence [%s] for HGVS...' % self.genome_fn
self.genome = SequenceFileDB(self.genome_fn)
print 'done.'
#load refseq into dic
print 'loading the refseq transcript [%s] for HGVS...' % self.refseq_fn
fp = open(self.refseq_fn, 'r')
self.refseq = pyhgvs.utils.read_transcripts(fp)
fp.close()
print 'done.'
def test_nlmsaslice_cache(self):
"NLMSASlice sequence caching & removal"
# set up sequences
dnaseq = testutil.datafile('dnaseq.fasta')
db = SequenceFileDB(dnaseq, autoGC=-1) # use pure WeakValueDict...
try:
gc.collect()
assert len(db._weakValueDict)==0, '_weakValueDict should be empty'
seq1, seq2 = db['seq1'], db['seq2']
assert len(db._weakValueDict)==2, \
'_weakValueDict should have 2 seqs'
# build referencing NLMSA
mymap = NLMSA('test', 'memory', db, pairwiseMode=True)
mymap += seq1
mymap[seq1] += seq2
mymap.build()
# check: no cache
assert not hasattr(db, '_cache'), 'should be no cache yet'
seq1, seq2 = db['seq1'], db['seq2'] # re-retrieve
# now retrieve a NLMSASlice, forcing entry of seq into cache
ival = seq1[5:10]
x = mymap[ival]
assert len(db._cache.values()) != 0
n1 = len(db._cache)
assert n1 == 1, "should be exactly one cache entry, not %d" % \
(n1, )
# ok, now trash referencing arguments & make sure of cleanup
del x
gc.collect()
assert len(db._cache.values()) == 0
n2 = len(db._cache)
assert n2 == 0, '%d objects remain; cache memory leak!' % n2
# FAIL because of __dealloc__ error in cnestedlist.NLMSASlice.
# Drop our references, the cache should empty.
del mymap, ival, seq1, seq2
gc.collect()
# check that db._weakValueDict cache is empty
assert len(db._weakValueDict)==0, '_weakValueDict should be empty'
finally:
db.close()
def test_headerfile_create_conflict(self):
"test non-empty prefixDict with a passed in PUD header file: conflict"
subdb = SequenceFileDB(self.dbfile)
try:
header = testutil.datafile('prefixUnionDict-1.txt')
try:
db = PrefixUnionDict(filename=header,
prefixDict={'foo': subdb})
assert 0, "should not get here"
except TypeError:
pass
finally:
subdb.close()
def test_iadd_db_twice(self):
dnaseq = testutil.datafile('dnaseq.fasta')
seqdb = SequenceFileDB(dnaseq)
try:
new_seq = seqdb['seq1']
self.db += new_seq
name1 = (~self.db)[new_seq]
self.db += new_seq # should do nothing...
name2 = (~self.db)[new_seq]
assert name1 == name2 # ...leaving seq with same name.
finally:
seqdb.close()
def test_build_seqLenDict_with_reader(self):
"Test that building things works properly when specifying a reader."
class InfoBag(object):
def __init__(self, **kw):
self.__dict__.update(kw)
# first, load the db & save the sequence info in a list
l = []
db = SequenceFileDB(self.dbfile)
try:
for k, v in db.items():
info = InfoBag(id=k, length=len(v), sequence=str(v))
l.append(info)
finally:
# now, erase the existing files, and recreate the db.
db.close()
self.trash_intermediate_files()
# create a fake reader with access to the saved info
def my_fake_reader(fp, filename, info_list=l):
return info_list
# now try creating with the fake reader
db = SequenceFileDB(self.dbfile, reader=my_fake_reader)
# did it work?
try:
assert str(db.get('seq1')).startswith('atggtgtca')
assert str(db.get('seq2')).startswith('GTGTTGAA')
finally:
db.close()
def test_iadd_db_twice(self):
dnaseq = testutil.datafile('dnaseq.fasta')
seqdb = SequenceFileDB(dnaseq)
try:
new_seq = seqdb['seq1']
self.db += new_seq
name1 = (~self.db)[new_seq]
self.db += new_seq # should do nothing...
name2 = (~self.db)[new_seq]
assert name1 == name2 # ...leaving seq with same name.
finally:
seqdb.close()
def test_no_db_info(self):
dnaseq = testutil.datafile('dnaseq.fasta')
seqdb = SequenceFileDB(dnaseq)
try:
new_seq = seqdb['seq1']
assert getattr(seqdb, '_persistent_id', None) is None
del seqdb.filepath
self.db += new_seq
name = (~self.db)[new_seq]
assert name == 'noname0.seq1'
finally:
seqdb.close()
def test_no_db_info(self):
dnaseq = testutil.datafile('dnaseq.fasta')
seqdb = SequenceFileDB(dnaseq)
try:
new_seq = seqdb['seq1']
assert getattr(seqdb, '_persistent_id', None) is None
del seqdb.filepath
self.db += new_seq
name = (~self.db)[new_seq]
assert name == 'noname0.seq1'
finally:
seqdb.close()
def test_inverse_noadd_behavior(self):
# compare with test_inverse_add_behavior...
db = SeqPrefixUnionDict(addAll=False)
dnaseq = testutil.datafile('dnaseq.fasta')
seqdb = SequenceFileDB(dnaseq)
try:
seq = seqdb['seq1']
try:
name = (~db)[seq]
assert 0, "should not get here"
except KeyError:
pass
finally:
seqdb.close()
def test_inverse_noadd_behavior(self):
# compare with test_inverse_add_behavior...
db = SeqPrefixUnionDict(addAll=False)
dnaseq = testutil.datafile('dnaseq.fasta')
seqdb = SequenceFileDB(dnaseq)
try:
seq = seqdb['seq1']
try:
name = (~db)[seq]
assert 0, "should not get here"
except KeyError:
pass
finally:
seqdb.close()
def test_no_file_given(self):
"Make sure that a TypeError is raised when no file is available"
try:
db = SequenceFileDB()
assert 0, "should not reach this point"
except TypeError:
pass
def test_headerfile_write_fail(self):
subdb = SequenceFileDB(self.dbfile)
try:
del subdb.filepath # remove 'filepath' attribute for test
db = PrefixUnionDict({'prefix': subdb})
assert len(db) == 2
assert 'prefix.seq1' in db
output = testutil.tempdatafile('prefixUnionDict-write-fail.txt')
try:
db.writeHeaderFile(output)
except AttributeError:
pass
finally:
subdb.close() # closes both db and subdb
def translate(variant,transcripts,get_transcript):
genome = SequenceFileDB('hg19.fa') #pip install bsddb3 is required
try:
chrom, offset, ref, alt = hgvs.parse_hgvs_name(variant, genome, get_transcript=get_transcript)
except:
return 1
return chrom, offset, ref, alt
def __init__(self,
lookup=None,
filename=None,
db_filename=None,
default_seq=None):
"""
A mock genome object that provides a pygr compatible interface.
lookup: a list of ((chrom, start, end), seq) values that define
a lookup table for genome sequence requests.
filename: a stream or filename containing a lookup table.
db_filename: a fasta file to use for genome sequence requests. All
requests are recorded and can be writen to a lookup table file
using the `write` method.
default_seq: if given, this base will always be returned if
region is unavailable.
"""
self._chroms = {}
self._lookup = lookup if lookup is not None else {}
self._genome = None
self._default_seq = default_seq
if db_filename:
# Use a real genome database.
if SequenceFileDB is None:
raise ValueError('pygr is not available.')
self._genome = SequenceFileDB(db_filename)
elif filename:
# Read genome sequence from lookup table.
self.read(filename)
def test_headerfile_write_fail(self):
subdb = SequenceFileDB(self.dbfile)
try:
del subdb.filepath # remove 'filepath' attribute for test
db = PrefixUnionDict({'prefix': subdb})
assert len(db) == 2
assert 'prefix.seq1' in db
output = testutil.tempdatafile('prefixUnionDict-write-fail.txt')
try:
db.writeHeaderFile(output)
except AttributeError:
pass
finally:
subdb.close() # closes both db and subdb
def codetest():
"Test the code here before adding to doctest @CTB"
import pygr
from pygr.seqdb import SequenceFileDB
db = SequenceFileDB(os.path.join('data', 'partial-yeast.fasta'))
chr02 = db['chr02']
start, stop = (87787, 86719)
x = chr02[start:stop]
def main():
global REFGENE
parser = argparse.ArgumentParser()
parser.add_argument('-i', '--readable_input',
help='readable input file for conversion.')
parser.add_argument('-o', '--writable_output',
help='writable output file for conversion.')
parser.add_argument('-g', '--genome_path', help='Link to hg38.fa.')
parser.add_argument('-r', '--reference_genome', default='./hg38.BRCA.refGene.txt',
help='Link to hg38.BRCA.refgene.txt.')
args = parser.parse_args()
GENOME = SequenceFileDB(args.genome_path)
REFGENE = args.reference_genome
f_in = open(args.readable_input, "r")
f_out = open(args.writable_output, "w")
f_out.write("\t".join(OUTPUT_COLUMNS) + "\n")
for index, line in enumerate(f_in):
#
# Clean the line by removing leading or trailing spaces adjacent to tabs.
#
line = re.sub("( )*\t( )*", "\t", line)
items = np.array(line.rstrip().split("\t"))
if index == 0:
# Handle column names
columns = np.array([i.replace(" ", "_") for i in items])
index_to_save = [np.where(columns == i)[0][0] for i in COLUMNS_TO_SAVE]
column_idx = dict(zip(COLUMNS_TO_SAVE, index_to_save))
continue
#
# In the date last evaluated field, delete the time last evaluated if provided.
#
date_last_evaluated_idx = column_idx["Date_last_evaluated"]
items[date_last_evaluated_idx] = items[date_last_evaluated_idx].split(' ')[0]
OMIM_id_index = column_idx["Condition_ID_value"]
items[OMIM_id_index] = convert_OMIM_id(items[OMIM_id_index])
items[column_idx["HGVS"]] = cleanup_HGVS(items[column_idx["Reference_sequence"]],
items[column_idx["HGVS"]], HP, EVM)
HGVS_cDNA = items[column_idx["Reference_sequence"]] + ":" + items[column_idx["HGVS"]]
print items[column_idx["Reference_sequence"]], items[column_idx["HGVS"]], HGVS_cDNA
try:
genome_coor, HGVS_p = convert_HGVS(HGVS_cDNA, GENOME)
except:
if (items[column_idx["HGVS"]]).find(";") > -1:
genome_coor, HGVS_p = create_None_filler()
aa_abrev_index = column_idx["Abbrev_AA_change"]
if HGVS_p not in ["p.?", "p.(=)", "None"]:
if items[aa_abrev_index] == '':
items[aa_abrev_index] = HGVS_p_to_AA_abrev(HGVS_p)
final_items = list(items[index_to_save])
final_items.insert(1, genome_coor)
final_items.append(HGVS_p)
new_line = "\t".join(list(final_items)) + "\n"
f_out.write(new_line)
f_in.close()
f_out.close()
def test_build_seqLenDict_with_bad_reader(self):
"Test that building things fails properly with a bad reader."
class InfoBag(object):
def __init__(self, **kw):
self.__dict__.update(kw)
# first, load the db & save the sequence info in a list
l = []
db = SequenceFileDB(self.dbfile)
try:
for k, v in db.items():
info = InfoBag(id=k, length=0, sequence=str(v))
l.append(info)
finally:
# now, erase the existing files, and recreate the db.
db.close()
self.trash_intermediate_files()
# create a fake reader with access to the saved info
def my_fake_reader(fp, filename, info_list=l):
return info_list
# now try creating with the fake reader
try:
db = SequenceFileDB(self.dbfile, reader=my_fake_reader)
try:
assert 0, "should not reach here; db construction should fail!"
finally:
db.close()
except ValueError:
pass # ValueError is expected
def test_basic_iadd(self):
dnaseq = testutil.datafile('dnaseq.fasta')
seqdb = SequenceFileDB(dnaseq)
try:
new_seq = seqdb['seq1']
self.db += new_seq
assert new_seq in self.db
name = (~self.db)[new_seq]
assert name == 'dnaseq.seq1', name
###
seqdb2 = SequenceFileDB(dnaseq)
try:
# Munge the filepath for testing.
seqdb2.filepath = 'foo'
new_seq2 = seqdb2['seq1']
self.db += new_seq2
name2 = (~self.db)[new_seq2]
assert name2 == 'foo.seq1', name2
finally:
seqdb2.close()
finally:
seqdb.close()
def initialize(hg_fasta, snpeff_predictor_bin):
''' Load required databases:
* human genome reference
* refGene.txt'''
global __GENOME__, __TRANSCRIPTS__
__GENOME__ = SequenceFileDB(hg_fasta)
with gzip.open(snpeff_predictor_bin) as f:
__TRANSCRIPTS__ = read_snpeff_transcripts(f)
def test_invalid_coordinates():
"""
Regression test for 17
"""
if not SequenceFileDB:
raise nose.SkipTest
genome = SequenceFileDB('pyhgvs/tests/data/test_refseqs.fa')
hgvs_name = 'NC_000005.10:g.177421339_177421327delACTCGAGTGCTCC'
parse_hgvs_name(hgvs_name, genome, get_transcript=get_transcript)
def main():
global REFGENE
parser = argparse.ArgumentParser()
parser.add_argument('-i',
'--readable_input',
type=argparse.FileType('r'),
help='Opened readable input file for conversion.')
parser.add_argument('-o',
'--writable_output',
type=argparse.FileType('w'),
help='Opened writable output file for conversion.')
parser.add_argument('-g', '--genome_path', help='Link to hg38.fa.')
parser.add_argument('-r',
'--reference_genome',
default='./hg38.BRCA.refGene.txt',
help='Link to hg38.BRCA.refgene.txt.')
args = parser.parse_args()
GENOME = SequenceFileDB(args.genome_path)
REFGENE = args.reference_genome
f_out = args.writable_output
f_out.write("\t".join(OUTPUT_COLUMNS) + "\n")
f_in = args.readable_input
for index, line in enumerate(f_in):
items = np.array(line.rstrip().split("\t"))
if index == 0:
# Handle column names
columns = np.array([i.replace(" ", "_") for i in items])
index_to_save = [
np.where(columns == i)[0][0] for i in COLUMNS_TO_SAVE
]
column_idx = dict(zip(COLUMNS_TO_SAVE, index_to_save))
continue
OMIM_id_index = column_idx["Condition_ID_value"]
items[OMIM_id_index] = convert_OMIM_id(items[OMIM_id_index])
HGVS_cDNA = (items[column_idx["Reference_sequence"]] + ":" +
items[column_idx["HGVS"]])
try:
genome_coor, HGVS_p = convert_HGVS(HGVS_cDNA, GENOME)
except:
if (items[column_idx["HGVS"]]).find(";") > -1:
genome_coor, HGVS_p = create_None_filler()
aa_abrev_index = column_idx["Abbrev_AA_change"]
if HGVS_p not in ["p.?", "p.(=)", "None"]:
if items[aa_abrev_index] == '':
items[aa_abrev_index] = HGVS_p_to_AA_abrev(HGVS_p)
final_items = list(items[index_to_save])
final_items.insert(1, genome_coor)
final_items.append(HGVS_p)
new_line = "\t".join(list(final_items)) + "\n"
f_out.write(new_line)
f_in.close()
f_out.close()
def __init__(self,
inFile,
genome,
vkey=False,
verbose=False,
log=sys.stderr):
self.inFile = inFile
self.verbose = verbose
self.log = log
self.vkey = vkey
self.genome = SequenceFileDB(genome)
self.infoHeader = "[" + self.__class__.__name__ + "]"
def get_genome_coor(hgvs_c):
genome = SequenceFileDB('data/hg19.fa')
refGene = "/Users/Molly/Desktop/web-dev/hgvs_counsyl/hgvs/pyhgvs/data/genes.refGene"
with open(refGene) as infile:
transcripts = pyhgvs_utils.read_transcripts(infile)
def get_transcript(name):
return transcripts.get(name)
chrom, offset, ref, alt = pyhgvs.parse_hgvs_name(
hgvs_c, genome, get_transcript=get_transcript)
return chrom + ":" + str(offset) + ":" + ref + ">" + alt
def HGVS_to_GenomeCoor(HGVS):
"""use counsyl pyhgvs for this"""
genome = SequenceFileDB('../data/hg19.fa')
refGene = "../data/BRCA12.refGene.txt"
with open(refGene) as infile:
transcripts = pyhgvs_utils.read_transcripts(infile)
def get_transcript(name):
return transcripts.get(name)
chrom, offset, ref, alt = pyhgvs.parse_hgvs_name(
HGVS, genome, get_transcript=get_transcript)
genome_coordinate = chrom + ":" + str(offset) + ":" + ref + ">" + alt
return genome_coordinate
def __init__(self):
"""
Initializes hg19 reference and reference transcripts
"""
genome_path = os.path.join(os.path.dirname(__file__), 'resources', 'hg19.fa')
refseq_path = os.path.join(os.path.dirname(__file__), 'resources', 'genes.refGene')
# Read genome sequence using pygr.
self.genome = SequenceFileDB(genome_path)
# Read RefSeq transcripts into a python dict.
with open(refseq_path) as infile:
self.transcripts = hgvs.utils.read_transcripts(infile)
def main(argv=None):
parser = make_parser()
args = parser.parse_args(argv)
genome = SequenceFileDB(args.genome)
pwm = [IUPAC_SCORES[l] for l in args.consensus]
pwm.extend([REQUIRED_SCORES[l] for l in args.required_3p_seq])
pwm = motility.PWM(pwm)
# find all matches
with open(args.outfile, 'w') as outfile:
for chrom in genome.keys():
chromseq = str(genome[chrom])
print "searching ", chrom, "of length", len(chromseq)
if len(chromseq) < len(pwm):
print 'chromosome/fragment', chrom, 'is too short'
continue
matches = pwm.find(chromseq, -args.mismatches)
for start, stop, strand, seq in matches:
score = pwm.calc_score(seq)
outfile.write('\t'.join(
[chrom, str(start), str(stop), seq, str(score),
'+' if strand == 1 else '-']) + '\n')
def test_iadd_duplicate_seqdb(self):
dnaseq = testutil.datafile('dnaseq.fasta')
seqdb = SequenceFileDB(dnaseq)
try:
seqdb2 = SequenceFileDB(dnaseq)
try:
new_seq = seqdb['seq1']
new_seq2 = seqdb2['seq1']
self.db += new_seq
try:
self.db += new_seq2
assert 0, "should never reach this point"
except ValueError:
pass
finally:
seqdb2.close()
finally:
seqdb.close()
def test_name_to_variant_refseqs():
"""
Convert HGVS names to variant coordinates using refseqs directly.
"""
if not SequenceFileDB:
print 'skip test_name_to_variant_refseqs'
return
genome = SequenceFileDB('pyhgvs/tests/data/test_refseqs.fa')
for hgvs_name, variant, name_canonical, var_canonical in _name_variants:
if not var_canonical or 'NM_' not in hgvs_name:
# Only test transcript HGVS names.
continue
hgvs_variant = parse_hgvs_name(hgvs_name,
genome,
get_transcript=get_transcript)
nose.tools.assert_equal(hgvs_variant, variant,
repr([hgvs_name, variant, hgvs_variant]))
def test_pyhgvs_cdna_coordinate_correct(self):
for i in self.data:
pyhgvs_coord = i['pyhgvs_Genomic_Coordinate_38']
pyhgvs_cDNA = i['pyhgvs_cDNA']
genome = SequenceFileDB('../reference_genome/hg38/hg38.fa')
def get_transcript(name):
REFGENE = "../refgene38_brca.txt"
with open(REFGENE) as infile:
TRANSCRIPTS = pyhgvs_utils.read_transcripts(infile)
return TRANSCRIPTS.get(name)
chrom, offset, ref, alt = pyhgvs.parse_hgvs_name(
pyhgvs_cDNA, genome, get_transcript=get_transcript)
test_coord = chrom + ":" + "g." + str(
offset) + ":" + ref + ">" + alt
self.assertEqual(pyhgvs_coord, test_coord)
def __init__(self):
"""
Initializes hg19 reference and reference transcripts
"""
#genome_path = '/Users/charlesmarkello/leidenv1.0/resources/hg38.fa'
#refseq_path = '/Users/charlesmarkello/leidenv1.0/resources/refGene.txt'
#genome_path = '/Users/charlesmarkello/leidenv1.0/resources/hg19.fa'
#refseq_path = '/Users/charlesmarkello/leidenv1.0/resources/genes.refGene'
genome_path = os.path.join(os.path.dirname(__file__), 'resources',
'hg19.fa')
refseq_path = os.path.join(os.path.dirname(__file__), 'resources',
'genes.refGene')
print 'genome_path: ', genome_path
# Read genome sequence using pygr.
self.genome = SequenceFileDB(genome_path)
# Read RefSeq transcripts into a python dict.
with open(refseq_path) as infile:
self.transcripts = pyhgvs.utils.read_transcripts(infile)
def test_iadd_duplicate_seqdb(self):
dnaseq = testutil.datafile('dnaseq.fasta')
seqdb = SequenceFileDB(dnaseq)
try:
seqdb2 = SequenceFileDB(dnaseq)
try:
new_seq = seqdb['seq1']
new_seq2 = seqdb2['seq1']
self.db += new_seq
try:
self.db += new_seq2
assert 0, "should never reach this point"
except ValueError:
pass
finally:
seqdb2.close()
finally:
seqdb.close()
class SequenceFileDB_Test(unittest.TestCase):
"""
Test for all of the basic dictionary functions on 'SequenceFileDB',
among other things.
"""
def setUp(self):
"Test setup"
dnaseq = testutil.datafile('dnaseq.fasta')
self.db = SequenceFileDB(dnaseq) # contains 'seq1', 'seq2'
self.db._weakValueDict.clear() # clear the cache
def tearDown(self):
self.db.close() # must close SequenceFileDB!
def test_len(self):
assert len(self.db) == 2
def test_seqInfoDict_len(self):
assert len(self.db.seqInfoDict) == 2
def test_no_file_given(self):
"Make sure that a TypeError is raised when no file is available"
try:
db = SequenceFileDB()
assert 0, "should not reach this point"
except TypeError:
pass
def test_seq_descriptor(self):
"Check the '.seq' attribute (tied to a descriptor)"
s = self.db['seq1']
assert str(s) == str(s.seq)
def test_cache(self):
"SequenceDB cache test"
assert len(self.db._weakValueDict) == 0
seq1 = self.db['seq1']
# cache populated?
assert len(self.db._weakValueDict) == 1
assert 'seq1' in self.db._weakValueDict
# cache functions?
seq1_try2 = self.db['seq1']
assert seq1 is seq1_try2
def test_clear_cache(self):
"SequenceDB clear_cache test"
assert len(self.db._weakValueDict) == 0
seq1 = self.db['seq1']
# cache populated?
assert len(self.db._weakValueDict) == 1
assert 'seq1' in self.db._weakValueDict
# clear_cache functions?
self.db.clear_cache()
seq1_try3 = self.db['seq1']
assert seq1 is not seq1_try3
def test_keys(self):
"SequenceFileDB keys"
k = self.db.keys()
k.sort()
assert k == ['seq1', 'seq2']
def test_contains(self):
"SequenceFileDB contains"
assert 'seq1' in self.db, self.db.keys()
assert 'seq2' in self.db
assert 'foo' not in self.db
def test_invert_class(self):
"SequenceFileDB __invert__"
seq = self.db['seq1']
inversedb = ~self.db
assert inversedb[seq] == 'seq1'
assert seq in inversedb
assert 'foo' not in inversedb
def test_keys_info(self):
"SequenceFileDB keys info"
k = self.db.seqInfoDict.keys()
k.sort()
assert k == ['seq1', 'seq2']
def test_contains_info(self):
"SequenceFileDB contains info"
assert 'seq1' in self.db.seqInfoDict
assert 'seq2' in self.db.seqInfoDict
assert 'foo' not in self.db.seqInfoDict
def test_has_key(self):
"SequenceFileDB has key"
assert 'seq1' in self.db
assert 'seq2' in self.db
assert 'foo' not in self.db
def test_get(self):
"SequenceFileDB get"
assert self.db.get('foo') is None
assert self.db.get('seq1') is not None
assert str(self.db.get('seq1')).startswith('atggtgtca')
assert self.db.get('seq2') is not None
assert str(self.db.get('seq2')).startswith('GTGTTGAA')
def test_items(self):
"SequenceFileDB items"
i = [k for (k, v) in self.db.items()]
i.sort()
assert i == ['seq1', 'seq2']
def test_iterkeys(self):
"SequenceFileDB iterkeys"
kk = self.db.keys()
kk.sort()
ik = list(self.db.iterkeys())
ik.sort()
assert kk == ik
def test_itervalues(self):
"SequenceFileDB itervalues"
kv = self.db.values()
kv.sort()
iv = list(self.db.itervalues())
iv.sort()
assert kv == iv
def test_iteritems(self):
"SequenceFileDB iteritems"
ki = self.db.items()
ki.sort()
ii = list(self.db.iteritems())
ii.sort()
assert ki == ii
def test_readonly(self):
"SequenceFileDB readonly"
try:
self.db.copy() # what should 'copy' do on SequenceFileDB?
assert 0, 'this method should raise NotImplementedError'
except NotImplementedError:
pass
try:
self.db.clear()
assert 0, 'this method should raise NotImplementedError'
except NotImplementedError:
pass
try:
self.db.setdefault('foo')
assert 0, 'this method should raise NotImplementedError'
except NotImplementedError:
pass
try:
self.db.pop()
assert 0, 'this method should raise NotImplementedError'
except NotImplementedError:
pass
try:
self.db.popitem()
assert 0, 'this method should raise NotImplementedError'
except NotImplementedError:
pass
try:
self.db.update({})
assert 0, 'this method should raise NotImplementedError'
except NotImplementedError:
pass
# test some things other than dict behavior
def test_keyerror(self):
"""SequenceFileDB keyerror.
Make sure that the SequenceFileDB KeyError is informative."""
try:
self.db['foo']
except KeyError, e:
assert "no key 'foo' in database <SequenceFileDB" in str(e), str(e)
def setUp(self):
dnaseq = testutil.datafile('dnaseq.fasta')
self.seqdb = SequenceFileDB(dnaseq) # contains 'seq1', 'seq2'
self.db = SeqPrefixUnionDict({'prefix': self.seqdb})
class SeqPrefixUnionDict_Test(unittest.TestCase):
"""
Test SeqPrefixUnionDict.
"""
def setUp(self):
dnaseq = testutil.datafile('dnaseq.fasta')
self.seqdb = SequenceFileDB(dnaseq) # contains 'seq1', 'seq2'
self.db = SeqPrefixUnionDict({'prefix': self.seqdb})
def tearDown(self):
self.seqdb.close()
def test_basic_iadd(self):
dnaseq = testutil.datafile('dnaseq.fasta')
seqdb = SequenceFileDB(dnaseq)
try:
new_seq = seqdb['seq1']
self.db += new_seq
assert new_seq in self.db
name = (~self.db)[new_seq]
assert name == 'dnaseq.seq1', name
###
seqdb2 = SequenceFileDB(dnaseq)
try:
# Munge the filepath for testing.
seqdb2.filepath = 'foo'
new_seq2 = seqdb2['seq1']
self.db += new_seq2
name2 = (~self.db)[new_seq2]
assert name2 == 'foo.seq1', name2
finally:
seqdb2.close()
finally:
seqdb.close()
# NOTE, the important thing here is less the specific names that
# are given (which are based on filepath) but that different names
# are created for the various sequences when they are added.
def test_iadd_db_twice(self):
dnaseq = testutil.datafile('dnaseq.fasta')
seqdb = SequenceFileDB(dnaseq)
try:
new_seq = seqdb['seq1']
self.db += new_seq
name1 = (~self.db)[new_seq]
self.db += new_seq # should do nothing...
name2 = (~self.db)[new_seq]
assert name1 == name2 # ...leaving seq with same name.
finally:
seqdb.close()
def test_iadd_user_seq(self):
seq = Sequence('ATGGCAGG', 'foo')
self.db += seq
name = (~self.db)[seq]
assert name == 'user.foo' # created a new 'user' db.
# ok, make sure it doesn't wipe out the old 'user' db...
seq2 = Sequence('ATGGCAGG', 'foo2')
self.db += seq2
name = (~self.db)[seq2]
assert name == 'user.foo2'
first_name = (~self.db)[seq]
assert first_name == 'user.foo'
def test_iadd_duplicate_seqdb(self):
dnaseq = testutil.datafile('dnaseq.fasta')
seqdb = SequenceFileDB(dnaseq)
try:
seqdb2 = SequenceFileDB(dnaseq)
try:
new_seq = seqdb['seq1']
new_seq2 = seqdb2['seq1']
self.db += new_seq
try:
self.db += new_seq2
assert 0, "should never reach this point"
except ValueError:
pass
finally:
seqdb2.close()
finally:
seqdb.close()
def test_no_db_info(self):
dnaseq = testutil.datafile('dnaseq.fasta')
seqdb = SequenceFileDB(dnaseq)
try:
new_seq = seqdb['seq1']
assert getattr(seqdb, '_persistent_id', None) is None
del seqdb.filepath
self.db += new_seq
name = (~self.db)[new_seq]
assert name == 'noname0.seq1'
finally:
seqdb.close()
def test_inverse_add_behavior(self):
dnaseq = testutil.datafile('dnaseq.fasta')
seqdb = SequenceFileDB(dnaseq)
try:
seq = seqdb['seq1']
name = (~self.db)[seq]
finally:
seqdb.close() # only need to close if exception occurs
def test_inverse_noadd_behavior(self):
# compare with test_inverse_add_behavior...
db = SeqPrefixUnionDict(addAll=False)
dnaseq = testutil.datafile('dnaseq.fasta')
seqdb = SequenceFileDB(dnaseq)
try:
seq = seqdb['seq1']
try:
name = (~db)[seq]
assert 0, "should not get here"
except KeyError:
pass
finally:
seqdb.close()
def setUp(self):
"Test setup"
dnaseq = testutil.datafile('dnaseq.fasta')
self.db = SequenceFileDB(dnaseq) # contains 'seq1', 'seq2'
self.db._weakValueDict.clear() # clear the cache
def setup(self):
self.db = SequenceFileDB('dnaseq') # contains 'seq1', 'seq2'
class SequenceFileDB_Test(object):
"""
Test for all of the basic dictionary functions on 'SequenceFileDB'.
"""
def setup(self):
self.db = SequenceFileDB('dnaseq') # contains 'seq1', 'seq2'
def keys_test(self):
k = self.db.keys()
k.sort()
assert k == ['seq1', 'seq2']
def contains_test(self):
assert 'seq1' in self.db, self.db.keys()
assert 'seq2' in self.db
assert 'foo' not in self.db
def keys_info_test(self):
k = self.db.seqInfoDict.keys()
k.sort()
assert k == ['seq1', 'seq2']
def contains_info_test(self):
assert 'seq1' in self.db.seqInfoDict
assert 'seq2' in self.db.seqInfoDict
assert 'foo' not in self.db.seqInfoDict
def has_key_test(self):
assert self.db.has_key('seq1')
assert self.db.has_key('seq2')
assert not self.db.has_key('foo')
def get_test(self):
assert self.db.get('foo') is None
assert self.db.get('seq1') is not None
assert str(self.db.get('seq1')).startswith('atggtgtca')
assert self.db.get('seq2') is not None
assert str(self.db.get('seq2')).startswith('GTGTTGAA')
def items_test(self):
i = [ k for (k,v) in self.db.items() ]
i.sort()
assert i == ['seq1', 'seq2']
def iterkeys_test(self):
kk = self.db.keys()
kk.sort()
ik = list(self.db.iterkeys())
ik.sort()
assert kk == ik
def itervalues_test(self):
kv = self.db.values()
kv.sort()
iv = list(self.db.itervalues())
iv.sort()
assert kv == iv
def iteritems_test(self):
ki = self.db.items()
ki.sort()
ii = list(self.db.iteritems())
ii.sort()
assert ki == ii
def readonly_test(self):
try:
self.db.copy() # what should 'copy' do on SequenceFileDB?
assert 0, 'this method should raise NotImplementedError'
except NotImplementedError:
pass
try:
self.db.clear()
assert 0, 'this method should raise NotImplementedError'
except NotImplementedError:
pass
try:
self.db.setdefault('foo')
assert 0, 'this method should raise NotImplementedError'
except NotImplementedError:
pass
try:
self.db.pop()
assert 0, 'this method should raise NotImplementedError'
except NotImplementedError:
pass
try:
self.db.popitem()
assert 0, 'this method should raise NotImplementedError'
except NotImplementedError:
pass
try:
self.db.update({})
assert 0, 'this method should raise NotImplementedError'
except NotImplementedError:
pass
# test some things other than dict behavior
def keyerror_test(self):
"Make sure that the SequenceFileDB KeyError is informative."
try:
self.db['foo']
except KeyError, e:
assert "no key 'foo' in database <SequenceFileDB" in str(e), str(e)
