def testIntronsSNPs(self):
length = 0
t = 0
exon_id = 0
for x in range(self.mOffset, self.contigSize - self.mIncrement - self.mOffset, self.mIncrement):
# exons
for y in range(0, self.mExonSize):
base = x + y
base_x, base_y = self.toRange(base, base + 1)
result = snp2counts.getCDSPosition(
self.mExons, base_x, base_y, lcontig=self.contigSize, fasta=self.mFasta)
self.assertEqual(result.strand, self.strand)
self.assertEqual(result.cds_start, t)
self.assertEqual(result.cds_end, t + 1)
self.assertEqual(result.intron_start, None)
self.assertEqual(result.intron_end, None)
self.assertEqual(len(result.cds_seq) % 3, 0)
self.assertEqual(result.nc_seq, None)
self.assertEqual(result.nc_start, None)
self.assertEqual(result.nc_end, None)
self.assertEqual(result.exon_id, exon_id)
self.assertEqual(result.intron_id, None)
t += 1
exon_id += 1
# introns
for y in range(self.mExonSize, self.mExonSize + self.mIntronSize):
base = x + y
base_x, base_y = self.toRange(base, base + 1)
result = snp2counts.getCDSPosition(
self.mExons, base_x, base_y, lcontig=self.contigSize, fasta=self.mFasta)
self.assertEqual(result.strand, self.strand)
self.assertEqual(result.cds_start, None)
self.assertEqual(result.cds_end, None)
self.assertEqual(result.cds_phase, None)
self.assertEqual(result.intron_start, x + self.mExonSize)
self.assertEqual(
result.intron_end, x + self.mIntronSize + self.mExonSize)
self.assertEqual(result.cds_seq, None)
self.assertEqual(result.cds_seq_start, None)
self.assertEqual(result.cds_seq_end, None)
self.assertEqual(len(result.nc_seq), 1)
self.assert_(result.nc_seq not in "abc")
self.assertEqual(result.nc_start, base)
self.assertEqual(result.nc_end, base + 1)
self.assertEqual(result.exon_id, exon_id)
self.assertEqual(result.intron_id, exon_id - 1)
def testIndels( self ):
'''test with segments of size 5'''
size = 5
length = 0
framed_length = (3 - self.frame ) % 3
phase = (3 - self.frame ) % 3
if self.strand == "+":
motif = "123"
else:
motif = "321"
for x in range( self.mOffset, self.contigSize - self.mIncrement - self.mOffset, self.mIncrement):
for y in range( -2*size, self.mExonSize + 2*size):
base = x+y
if base < self.mOffset: continue
base_x, base_y = self.toRange( base, base+size )
result = snp2counts.getCDSPosition( self.mExons, base_x, base_y, lcontig = self.contigSize, fasta = self.mFasta )
if -size < y < self.mExonSize:
# overlap with coding sequence
self.assertEqual( len(result.cds_seq) % 3, 0 )
self.assertEqual( result.cds_start, length )
if y < 0:
self.assertEqual( result.cds_end, length + size + y )
else:
self.assertEqual( result.cds_end, length + min( size, self.mExonSize - y ) )
self.assertEqual( result.cds_phase, phase)
self.assertEqual( result.strand, self.strand )
ncodons = int(math.ceil( (result.cds_phase + result.cds_end - result.cds_start) / 3.0))
if self.frame == 0:
self.assertEqual( result.cds_seq, motif * ncodons )
self.assertEqual( result.cds_seq_start, framed_length % 3 )
self.assertEqual( result.cds_seq_end, framed_length % 3 + min( size, size + y, self.mExonSize - y) )
if result.nc_end != None:
self.assertEqual( result.cds_end - result.cds_start + (result.nc_end - result.nc_start), size )
self.assertEqual( len(result.nc_seq), (result.nc_end - result.nc_start))
else:
self.assertEqual( result.cds_start, None )
self.assertEqual( result.cds_end, None )
self.assertEqual( result.cds_phase, None )
if y > self.mExonSize - size:
self.assertEqual( result.intron_start, x + self.mExonSize)
self.assertEqual( result.intron_end, x + self.mIntronSize + self.mExonSize )
elif y < 0:
self.assertEqual( result.intron_start, x - self.mIntronSize)
self.assertEqual( result.intron_end, x )
if 0 <= y < self.mExonSize:
length += 1
framed_length += 1
phase += 1
if phase >= 3: phase = 0
def testCodingSNPs(self):
length = 0
framed_length = (3 - self.frame) % 3
phase = (3 - self.frame) % 3
if self.strand == "+":
motif = "123"
else:
motif = "321"
for x in range(self.mOffset, self.contigSize - self.mOffset, self.mIncrement):
for y in range(0, self.mExonSize):
base = x + y
rangex, rangey = self.toRange(base, base + 1)
result = snp2counts.getCDSPosition(self.mExons, rangex, rangey,
lcontig=self.contigSize,
fasta=self.mFasta)
self.assertEqual(result.strand, self.strand)
self.assertEqual(result.cds_start, length)
self.assertEqual(result.cds_end, length + 1)
self.assertEqual(result.cds_phase, phase)
self.assertEqual(result.intron_start, None)
self.assertEqual(result.intron_end, None)
self.assertEqual(len(result.cds_seq), 3)
# print x, y, base, str(result)
if self.frame == 0:
self.assertEqual(result.cds_seq, motif)
self.assertEqual(result.cds_seq_start, framed_length % 3)
self.assertEqual(result.cds_seq_end, (framed_length % 3) + 1)
self.assertEqual(result.nc_seq, None)
self.assertEqual(result.nc_start, None)
self.assertEqual(result.nc_end, None)
if base in self.mSplitCodonsPrev:
self.assertEqual(
result.prev_exon_end, self.mSplitCodonsPrev[base])
else:
self.assertEqual(result.prev_exon_end, None)
if base in self.mSplitCodonsNext:
self.assertEqual(
result.next_exon_start, self.mSplitCodonsNext[base])
else:
self.assertEqual(result.next_exon_start, None)
length += 1
framed_length += 1
phase += 1
if phase >= 3:
phase = 0
def testIndels(self):
'''test with segments of size 5'''
size = 5
length = 0
framed_length = (3 - self.frame) % 3
phase = (3 - self.frame) % 3
if self.strand == "+":
motif = "123"
else:
motif = "321"
for x in range(self.mOffset, self.contigSize - self.mIncrement - self.mOffset, self.mIncrement):
for y in range(-2 * size, self.mExonSize + 2 * size):
base = x + y
if base < self.mOffset:
continue
base_x, base_y = self.toRange(base, base + size)
result = snp2counts.getCDSPosition(
self.mExons, base_x, base_y, lcontig=self.contigSize, fasta=self.mFasta)
if -size < y < self.mExonSize:
# overlap with coding sequence
self.assertEqual(len(result.cds_seq) % 3, 0)
self.assertEqual(result.cds_start, length)
if y < 0:
self.assertEqual(result.cds_end, length + size + y)
else:
self.assertEqual(
result.cds_end, length + min(size, self.mExonSize - y))
self.assertEqual(result.cds_phase, phase)
self.assertEqual(result.strand, self.strand)
ncodons = int(
math.ceil((result.cds_phase + result.cds_end - result.cds_start) / 3.0))
if self.frame == 0:
self.assertEqual(result.cds_seq, motif * ncodons)
self.assertEqual(result.cds_seq_start, framed_length % 3)
self.assertEqual(
result.cds_seq_end, framed_length % 3 + min(size, size + y, self.mExonSize - y))
if result.nc_end is not None:
self.assertEqual(
result.cds_end - result.cds_start + (result.nc_end - result.nc_start), size)
self.assertEqual(
len(result.nc_seq), (result.nc_end - result.nc_start))
else:
self.assertEqual(result.cds_start, None)
self.assertEqual(result.cds_end, None)
self.assertEqual(result.cds_phase, None)
if y > self.mExonSize - size:
self.assertEqual(result.intron_start, x + self.mExonSize)
self.assertEqual(
result.intron_end, x + self.mIntronSize + self.mExonSize)
elif y < 0:
self.assertEqual(result.intron_start, x - self.mIntronSize)
self.assertEqual(result.intron_end, x)
if 0 <= y < self.mExonSize:
length += 1
framed_length += 1
phase += 1
if phase >= 3:
phase = 0