def setUpClass(cls):
cls.cols = {
"intA":
numpy.random.randint(0, high=2**16, size=size),
"intB":
numpy.random.randint(-10, high=20, size=size),
"idxA":
numpy.arange(size),
"chrA":
numpy.array([chr(65 + (X % (91 - 65))) for X in range(size)]),
"strA":
numpy.array([
str(GenomicSegment("chrA", X, X + 500, "+"))
for X in range(size)
]),
"strB":
numpy.array([
str(GenomicSegment("chrB", X / 2, X / 2 + 500, "-"))
for X in range(size)
]),
"floatA":
10 * numpy.random.randn(size) + 500,
"floatB": (10**-5) * numpy.random.random(size),
"objA":
numpy.tile(None, 5000),
"objB":
numpy.array([
GenomicSegment("chrC", X, X + Y, "+") for X, Y in zip(
range(size), numpy.random.randint(2, high=1000, size=size))
]),
}
def setUpClass(cls):
cls.ivs = [
GenomicSegment("chrA", 100, 190, "+"),
GenomicSegment("chrA", 200, 203, "+"),
GenomicSegment("chrA", 200, 201, "+"),
GenomicSegment("chrA", 204, 206, "+"),
]
cls.common_attr = dict(common1="common",
common2="also common",
common3="still common")
cls.attrs = [
dict(common_diff_val="unique_f1", unique_f1_key="something"),
dict(common_diff_val="unique_f2",
unique_f2_key="something",
unique_f2f3="something else"),
dict(common_diff_val="unique_f3",
unique_f3_key="something",
unique_f2f3="something else",
unique_f3f4="f3 only"),
dict(common_diff_val="unique_f4",
unique_f4_key="something",
unique_f3f4="f4 only"),
]
for x in cls.attrs:
x.update(cls.common_attr)
def test_window_landmark():
# test cases: plus and minus-strand IVCs with splicing
flank_up = 50
flank_down = 100
my_segmentchains = [
SegmentChain(GenomicSegment("chrA", 50, 350, "+"),
GenomicSegment("chrA", 500, 900, "+")),
SegmentChain(GenomicSegment("chrA", 50, 350, "-"),
GenomicSegment("chrA", 500, 900, "-")),
]
for my_segmentchain in my_segmentchains:
for landmark in range(0, 700, 50):
yield check_window_landmark, my_segmentchain, landmark, flank_up, flank_down
def setUpClass(cls):
min_ = 25
max_ = 40
cls.strands = ("+","-")
cls.segs = { X : GenomicSegment("mock",0,2000,X) for X in cls.strands }
cls.reads = { Y : [cls.make_alignment(0,X,Y) for X in range(min_,max_)] for Y in ("+","-") }
cls.expected = {}
for mapping in ("fiveprime","threeprime","center"):
for param in (0,10):
for strand in cls.strands:
cls.expected[(mapping,param,strand)] = numpy.zeros(2000)
cls.expected[("fiveprime",0, "+")][0] = max_ - min_
cls.expected[("fiveprime",10,"+")][10] = max_ - min_
cls.expected[("fiveprime",0, "-")][min_-1:max_-1] = 1
cls.expected[("fiveprime",10,"-")][min_-11:max_-11] = 1
cls.expected[("threeprime",0, "-")][0] = max_ - min_
cls.expected[("threeprime",10,"-")][10] = max_ - min_
cls.expected[("threeprime",0, "+")][min_-1:max_-1] = 1
cls.expected[("threeprime",10,"+")][min_-11:max_-11] = 1
for my_len in range(min_,max_):
cls.expected[("center",0,"+")][:my_len] += 1.0/my_len
cls.expected[("center",0,"-")][:my_len] += 1.0/my_len
cls.expected[("center",10,"+")][10:my_len-10] += 1.0/(my_len-2*10)
cls.expected[("center",10,"-")][10:my_len-10] += 1.0/(my_len-2*10)
cls.map_factories = {
"fiveprime" : FivePrimeMapFactory,
"threeprime" : ThreePrimeMapFactory,
"fiveprime_variable" : VariableFivePrimeMapFactory,
"center" : CenterMapFactory
}
def check_random_windows_against_wig(self, strand):
chrdict = self.chrdict
chroms = list(self.chrdict)
chridx = numpy.random.randint(0, high=len(chroms), size=50)
ga = GenomeArray()
i = 0
with open(wigfile) as fin:
ga.add_from_wiggle(fin, strand)
while i < 50:
chrom = chroms[chridx[i]]
maxlength = chrdict[chrom]
start = numpy.random.randint(0, high=maxlength - 2000)
end = numpy.random.randint(start + 10000, high=start + 20000)
# make sure we don't go off chrom
while end > maxlength:
end = numpy.random.randint(start + 100, high=start + 10000)
seg = GenomicSegment(chrom, start, end, strand)
expected = ga[seg]
# make sure segment has counts in it
if expected.sum() > 0:
i += 1
found = self.bw[seg]
yield self.check_vals_against_wig, expected, found
def covered_by_repetitive(query_junc,minus_range,plus_range,cross_hash):
"""Determine whether one or both ends of a splice site overlap with
a repetitive area of the genome.
Parameters
----------
query_junc : |SegmentChain|
A two-exon fragment representing a query splice junction
minus_range : int <= 0
Maximum number of nucleotides splice junction could be moved
to the left without reducing sequence support for the junction
see :py:func:`find_match_range`
plus_range : int >= 0
Maximum number of nucleotides splice junction could be moved
to the right without reducing sequence support for the junction
see :py:func:`find_match_range`
cross_hash : |GenomeHash|
|GenomeHash| of 1-length features denoting repetitive regions of the genome
Returns
-------
bool
`True` if any of the genomic positions within `minus_range...plus_range`
of the 5' or 3' splice sites of `query_junc` overlap a repetitive
region of the genome as annotated by ``cross_hash``.
Otherwise, `False`
"""
chrom = query_junc.spanning_segment.chrom
strand = query_junc.spanning_segment.strand
qend = query_junc[0].end
qstart = query_junc[1].start
fiveprime_splice_area = GenomicSegment(chrom,
qend + minus_range,
qend + plus_range + 1,
strand)
threeprime_splice_area = GenomicSegment(chrom,
qstart + minus_range,
qstart + plus_range + 1,
strand)
support_region = SegmentChain(fiveprime_splice_area,threeprime_splice_area)
return len(cross_hash.get_overlapping_features(support_region)) > 0
def test_get_chromosome_counts_zero_fill(self):
ga = GenomeArray()
with open(wigfile) as fin:
ga.add_from_wiggle(fin, "+")
for chrom, length in self.chrdict.items():
seg = GenomicSegment(chrom, 0, length, "+")
expected = ga[seg]
found = self.bw.get_chromosome_counts(chrom)
yield self.check_vals_against_wig, expected, found
def test_variable_stratified_mapping_plus(self):
offsets = {
26 : 6,
27 : 22,
28 : 13,
29 : 4,
30 : 5
}
chains = {
"fw" : SegmentChain(GenomicSegment('chrII',392959,393180,'+'),
GenomicSegment('chrII',393510,394742,'+'),
GenomicSegment('chrII',394860,394901,'+'),
ID='YBR078W_mRNA'),
"rc" : SegmentChain(GenomicSegment('chrVIII',189061,189749,'-'),
GenomicSegment('chrVIII',189850,190017,'-'),
ID='YHR041C_mRNA')
}
expected = {
"fw" : numpy.loadtxt(resource_filename("plastid","test/data/stratmap/strat_fw_vec.txt"),delimiter="\t"),
"rc" : numpy.loadtxt(resource_filename("plastid","test/data/stratmap/strat_rc_vec.txt"),delimiter="\t"),
}
ga = BAMGenomeArray([resource_filename("plastid","test/data/stratmap/strat.bam")])
ga.set_mapping(StratifiedVariableFivePrimeMapFactory(offsets,26,30))
def filter(self,line):
"""Parse a read alignment as |SegmentChain| from a line of `bowtie`_ output"""
items = line.strip("\n").split("\t")
read_name = items[0]
strand = items[1]
ref_seq = items[2]
coord = int(items[3])
attr = { 'seq_as_aligned' : items[4],
'qualstr' : items[5],
'mismatch_str' : items[7],
'type' : "alignment",
'ID' : read_name,
}
iv = GenomicSegment(ref_seq,coord,coord+len(attr['seq_as_aligned']),strand)
feature = SegmentChain(iv,**attr)
return feature
def test_fill_val_present_chrom(self):
filldef = BigWigReader(bigwigfile)
fillnan = BigWigReader(bigwigfile, fill=numpy.nan)
fill0 = BigWigReader(bigwigfile, fill=0)
fill10 = BigWigReader(bigwigfile, fill=10)
# empty region
seg = GenomicSegment("chrIV", 5, 10, "+")
assert_equal(len(filldef[seg]), len(seg), "fetched wrong size")
# assert_true(numpy.isnan(filldef[seg]).all(),
# "default not nan")
#
# assert_true(numpy.isnan(fillnan[seg]).all(),
# "nanfill didn't work")
assert_true((fill0[seg] == 0).all(), "0-fill didn't work")
def test_fill_val_absent_chrom(self):
filldef = BigWigReader(bigwigfile)
fillnan = BigWigReader(bigwigfile, fill=numpy.nan)
fill0 = BigWigReader(bigwigfile, fill=0)
fill10 = BigWigReader(bigwigfile, fill=10)
# chrVI is not in dataset; this should be an empty array
seg = GenomicSegment("chrVI", 5, 1000, "+")
assert_equal(len(filldef[seg]), len(seg), "fetched wrong size")
# assert_true(numpy.isnan(filldef[seg]).all(),
# "default not nan")
#
# assert_true(numpy.isnan(fillnan[seg]).all(),
# "nanfill didn't work")
assert_true((fill0[seg] == 0).all(), "0-fill didn't work")
def revcomp_mask_chain(seg, k, offset=0):
"""Reverse-complement a single-interval mask, correcting for `offset`.
Parameters
----------
seg : |SegmentChain|
Plus-strand mask, including `offset`
k : int
Length of k-mers
offset : int, optional
Offset from 5' end of read at which to map mask (Default: `0`)
Returns
-------
|SegmentChain|
Mask on minus strand corresponding to `seg`
"""
# Algorithm note:
#
# Let
# FW = plus-strand coordinate
# RC = minus-strand coordinate
#
# Then
# RC = FW + k - 1 - offset
#
# But we are given FW + offset, so:
#
# RC + offset = (FW + offset) + k - 1 - offset
# RC = (FW + offset) + k - 1 - 2*offset
span = seg.spanning_segment
new_offset = k - 1 - 2 * offset
ivminus = GenomicSegment(span.chrom, span.start + new_offset,
span.end + new_offset, "-")
return SegmentChain(ivminus)
def test_search_fields_multivalue(self):
reader = BigBedReader(self.bb_indexed)
found = list(
reader.search("name", "should_have_no_match",
"should_also_have_no_match"))
self.assertEqual([], found)
found = list(reader.search("Name", "Sam-S-RE", "Sam-S-RK"))
expected = [
SegmentChain(GenomicSegment('2L', 106902, 107000, '+'),
GenomicSegment('2L', 107764, 107838, '+'),
GenomicSegment('2L', 108587, 108809, '+'),
GenomicSegment('2L', 110405, 110483, '+'),
GenomicSegment('2L', 110754, 110877, '+'),
GenomicSegment('2L', 111906, 112019, '+'),
GenomicSegment('2L', 112689, 113369, '+'),
GenomicSegment('2L', 113433, 114432, '+'),
Alias="'['M(2)21AB-RE', 'CG2674-RE']'",
ID='FBtr0089437',
Name='Sam-S-RE',
color='#000000',
gene_id='FBgn0005278',
score='0.0',
thickend='113542',
thickstart='108685',
type='exon'),
SegmentChain(GenomicSegment('2L', 107760, 107838, '+'),
GenomicSegment('2L', 108587, 108809, '+'),
GenomicSegment('2L', 110405, 110483, '+'),
GenomicSegment('2L', 110754, 111337, '+'),
Alias='na',
ID='FBtr0308091',
Name='Sam-S-RK',
color='#000000',
gene_id='FBgn0005278',
score='0.0',
thickend='110900',
thickstart='108685',
type='exon'),
]
self.assertEqual(expected, found)
def fa_to_bed(toomany_fh, k, offset=0):
"""Create a `BED`_ file indicating genomic origins of reads in a `bowtie`_ ``toomany`` file
Parameters
----------
toomany_fh : file-like
Open filehandle to fasta-formatted ``toomany`` file from `bowtie`_
k : int
Length of k-mers
offset : int, optional
Offset from 5' end of read at which to map read, if any (Default: `0`)
Yields
------
|SegmentChain|
Plus-strand |SegmentChain| representing a repetitive region
|SegmentChain|
Minus-strand |SegmentChain| representing a repetitive region
"""
last_chrom = None
last_pos = None
start_pos = None
reader = FastaNameReader(toomany_fh)
for n, read_name in enumerate(reader):
chrom, pos = namepat.search(read_name).groups()
pos = int(pos) + offset
if chrom != last_chrom:
if last_chrom is not None:
plus_chain = SegmentChain(
GenomicSegment(last_chrom, start_pos, last_pos + 1, "+"))
minus_chain = revcomp_mask_chain(plus_chain, k, offset)
last_chrom = chrom
start_pos = pos
last_pos = pos
yield plus_chain, minus_chain
else:
last_chrom = chrom
start_pos = pos
last_pos = pos
else:
delta = pos - last_pos
if delta > 1:
plus_chain = SegmentChain(
GenomicSegment(chrom, start_pos, last_pos + 1, "+"))
minus_chain = revcomp_mask_chain(plus_chain, k, offset)
last_pos = pos
start_pos = pos
yield plus_chain, minus_chain
elif delta == 1:
last_pos = pos
else:
msg = "k-mers are not sorted at read %s! Aborting." % read_name
raise MalformedFileError(toomany_fh, msg, line_num=n)
# export final feature
plus_chain = SegmentChain(
GenomicSegment(chrom, start_pos, last_pos + 1, "+"))
minus_chain = revcomp_mask_chain(plus_chain, k, offset)
yield plus_chain, minus_chain
],
# these below are all 1 nucleotide outside match range
'YBR215W_mRNA_0' : ['YBR215W_mRNA_0:0-105^189-2175(+)', 'YBR215W_mRNA_0:0-109^193-2175(+)'],
'YHL001W_mRNA_0' : ['YHL001W_mRNA_0:0-143^541-961(+)', 'YHL001W_mRNA_0:0-149^547-961(+)'],
'YIL018W_mRNA_0' : ['YIL018W_mRNA_0:0-28^428-1280(+)', 'YIL018W_mRNA_0:0-34^434-1280(+)'],
'YIL133C_mRNA_0' : ['YIL133C_mRNA_0:0-644^934-1007(-)', 'YIL133C_mRNA_0:0-650^940-1007(-)'],
'YIL156W_B_mRNA_0': ['YIL156W_B_mRNA_0:0-40^102-408(+)', 'YIL156W_B_mRNA_0:0-45^107-408(+)'],
'YKL006W_mRNA_0' : ['YKL006W_mRNA_0:0-154^552-954(+)', 'YKL006W_mRNA_0:0-160^558-954(+)'],
'YMR194C_B_mRNA_0': ['YMR194C_B_mRNA_0:0-324^396-729(-)', 'YMR194C_B_mRNA_0:0-328^400-729(-)'],
'YPL249C_A_mRNA_0': ['YPL249C_A_mRNA_0:0-410^648-697(-)', 'YPL249C_A_mRNA_0:0-417^655-697(-)']
}
unmatched_query_juncs = { K : [SegmentChain.from_str(X) for X in V] for K,V in unmatched_query_juncs.items() }
"""Query junctions with no known matches"""
unmatched_noncan_query_juncs = ["YNL130C:0-23^145-180(-)",
"YNL130C:0-53^165-180(-)",
"YNL130C:0-70^141-180(-)",
"YNL130C:0-49^121-180(-)",
]
unmatched_noncan_query_juncs = [SegmentChain.from_str(X) for X in unmatched_noncan_query_juncs]
"""Query junctions without canonical splice junctions in the match range"""
repetitive_regions = [
"YBR215W_mRNA_0:190-193(+)", # threeprime splice site plus
"YHL001W_mRNA_0:144-149(+)", # fiveprime splice site plus
"YIL133C_mRNA_0:935-940(-)", # threeprime splice site minus
"YMR194C_B_mRNA_0:325-328(-)", # fiveprime splice site minus
]
cross_hash = GenomeHash([SegmentChain(GenomicSegment.from_str(X)) for X in repetitive_regions])
cross_hash_seqs = { X.chrom for X in cross_hash.feature_dict.values() }
reader = BED_Reader(cStringIO.StringIO(_NARROW_PEAK_TEXT),
extra_columns=14)
with warnings.catch_warnings(record=True) as warns:
warnings.simplefilter("always")
ltmp = list(reader)
assert_greater_equal(len(warns), 0)
#===============================================================================
# INDEX: test data
#===============================================================================
# test dataset, constructed manually to include various edge cases
_TEST_SEGMENTCHAINS = [
# single-interval
SegmentChain(GenomicSegment("chrA", 100, 1100, "+"), ID="IVC1p"),
SegmentChain(GenomicSegment("chrA", 100, 1100, "-"), ID="IVC1m"),
# multi-interval
SegmentChain(GenomicSegment("chrA", 100, 1100, "+"),
GenomicSegment("chrA", 2100, 2600, "+"),
ID="IVC2p"),
SegmentChain(GenomicSegment("chrA", 100, 1100, "-"),
GenomicSegment("chrA", 2100, 2600, "-"),
ID="IVC2m"),
# multi-interval, with score
SegmentChain(GenomicSegment("chrA", 100, 1100, "+"),
GenomicSegment("chrA", 2100, 2600, "+"),
ID="IVC3p",
score=500),
SegmentChain(GenomicSegment("chrA", 100, 1100, "-"),
GenomicSegment("chrA", 2100, 2600, "-"),
def test_search_fields_singlevalue(self):
reader = BigBedReader(self.bb_indexed)
found = list(reader.search("name", "should_have_no_match"))
self.assertEqual([], found)
found = list(reader.search("Name", "Sam-S-RE"))
expected = [
SegmentChain(GenomicSegment('2L', 106902, 107000, '+'),
GenomicSegment('2L', 107764, 107838, '+'),
GenomicSegment('2L', 108587, 108809, '+'),
GenomicSegment('2L', 110405, 110483, '+'),
GenomicSegment('2L', 110754, 110877, '+'),
GenomicSegment('2L', 111906, 112019, '+'),
GenomicSegment('2L', 112689, 113369, '+'),
GenomicSegment('2L', 113433, 114432, '+'),
Alias="'['M(2)21AB-RE', 'CG2674-RE']'",
ID='FBtr0089437',
Name='Sam-S-RE',
color='#000000',
gene_id='FBgn0005278',
score='0.0',
thickend='113542',
thickstart='108685',
type='exon'),
]
self.assertEqual(expected, found)
found = list(reader.search("gene_id", "FBgn0005278"))
expected = [
SegmentChain(GenomicSegment('2L', 106902, 107000, '+'),
GenomicSegment('2L', 107764, 107838, '+'),
GenomicSegment('2L', 108587, 108809, '+'),
GenomicSegment('2L', 110405, 110483, '+'),
GenomicSegment('2L', 110754, 110877, '+'),
GenomicSegment('2L', 111906, 112019, '+'),
GenomicSegment('2L', 112689, 113369, '+'),
GenomicSegment('2L', 113433, 114432, '+'),
Alias="'['M(2)21AB-RE', 'CG2674-RE']'",
ID='FBtr0089437',
Name='Sam-S-RE',
color='#000000',
gene_id='FBgn0005278',
score='0.0',
thickend='113542',
thickstart='108685',
type='exon'),
SegmentChain(GenomicSegment('2L', 107760, 107838, '+'),
GenomicSegment('2L', 108587, 108809, '+'),
GenomicSegment('2L', 110405, 110483, '+'),
GenomicSegment('2L', 110754, 111337, '+'),
Alias='na',
ID='FBtr0308091',
Name='Sam-S-RK',
color='#000000',
gene_id='FBgn0005278',
score='0.0',
thickend='110900',
thickstart='108685',
type='exon'),
SegmentChain(GenomicSegment('2L', 107760, 107838, '+'),
GenomicSegment('2L', 108587, 108809, '+'),
GenomicSegment('2L', 110405, 110483, '+'),
GenomicSegment('2L', 110754, 110877, '+'),
GenomicSegment('2L', 111004, 111117, '+'),
GenomicSegment('2L', 111906, 112019, '+'),
GenomicSegment('2L', 112689, 113369, '+'),
GenomicSegment('2L', 113433, 114210, '+'),
Alias="'['M(2)21AB-RB', 'CG2674-RB']'",
ID='FBtr0089428',
Name='Sam-S-RB',
color='#000000',
gene_id='FBgn0005278',
score='0.0',
thickend='112741',
thickstart='108685',
type='exon'),
SegmentChain(GenomicSegment('2L', 107760, 107838, '+'),
GenomicSegment('2L', 108587, 108809, '+'),
GenomicSegment('2L', 110405, 110483, '+'),
GenomicSegment('2L', 110754, 110877, '+'),
GenomicSegment('2L', 111906, 112019, '+'),
GenomicSegment('2L', 112689, 113369, '+'),
GenomicSegment('2L', 113433, 114432, '+'),
Alias="'['M(2)21AB-RA', 'CG2674-RA']'",
ID='FBtr0089429',
Name='Sam-S-RA',
color='#000000',
gene_id='FBgn0005278',
score='0.0',
thickend='113542',
thickstart='108685',
type='exon'),
SegmentChain(GenomicSegment('2L', 107760, 107956, '+'),
GenomicSegment('2L', 108587, 108809, '+'),
GenomicSegment('2L', 110405, 110483, '+'),
GenomicSegment('2L', 110754, 110877, '+'),
GenomicSegment('2L', 112689, 113369, '+'),
GenomicSegment('2L', 113433, 114432, '+'),
Alias='na',
ID='FBtr0330656',
Name='Sam-S-RL',
color='#000000',
gene_id='FBgn0005278',
score='0.0',
thickend='112781',
thickstart='108685',
type='exon'),
SegmentChain(GenomicSegment('2L', 107936, 108226, '+'),
GenomicSegment('2L', 108587, 108809, '+'),
GenomicSegment('2L', 110405, 110483, '+'),
GenomicSegment('2L', 110754, 110877, '+'),
GenomicSegment('2L', 111906, 112019, '+'),
GenomicSegment('2L', 112689, 113369, '+'),
GenomicSegment('2L', 113433, 114210, '+'),
Alias="'['M(2)21AB-RH', 'CG2674-RH']'",
ID='FBtr0089432',
Name='Sam-S-RH',
color='#000000',
gene_id='FBgn0005278',
score='0.0',
thickend='113542',
thickstart='108685',
type='exon'),
SegmentChain(GenomicSegment('2L', 107936, 108101, '+'),
GenomicSegment('2L', 108587, 108809, '+'),
GenomicSegment('2L', 110405, 110483, '+'),
GenomicSegment('2L', 110754, 110877, '+'),
GenomicSegment('2L', 111906, 112019, '+'),
GenomicSegment('2L', 112689, 113369, '+'),
GenomicSegment('2L', 113433, 114432, '+'),
Alias="'['M(2)21AB-RD', 'CG2674-RD']'",
ID='FBtr0089430',
Name='Sam-S-RD',
color='#000000',
gene_id='FBgn0005278',
score='0.0',
thickend='113542',
thickstart='108685',
type='exon'),
SegmentChain(GenomicSegment('2L', 107936, 108101, '+'),
GenomicSegment('2L', 108587, 108809, '+'),
GenomicSegment('2L', 110405, 110483, '+'),
GenomicSegment('2L', 110754, 110877, '+'),
GenomicSegment('2L', 111004, 111117, '+'),
GenomicSegment('2L', 112689, 113369, '+'),
GenomicSegment('2L', 113433, 114432, '+'),
Alias="'['M(2)21AB-RC', 'CG2674-RC']'",
ID='FBtr0089431',
Name='Sam-S-RC',
color='#000000',
gene_id='FBgn0005278',
score='0.0',
thickend='113542',
thickstart='108685',
type='exon'),
SegmentChain(GenomicSegment('2L', 108088, 108226, '+'),
GenomicSegment('2L', 108587, 108809, '+'),
GenomicSegment('2L', 110405, 110483, '+'),
GenomicSegment('2L', 110754, 110877, '+'),
GenomicSegment('2L', 111906, 112019, '+'),
GenomicSegment('2L', 112689, 113369, '+'),
GenomicSegment('2L', 113433, 114432, '+'),
Alias="'['M(2)21AB-RF', 'CG2674-RF']'",
ID='FBtr0089433',
Name='Sam-S-RF',
color='#000000',
gene_id='FBgn0005278',
score='0.0',
thickend='113542',
thickstart='108685',
type='exon'),
SegmentChain(GenomicSegment('2L', 108132, 108346, '+'),
GenomicSegment('2L', 108587, 108809, '+'),
GenomicSegment('2L', 110405, 110483, '+'),
GenomicSegment('2L', 110754, 110877, '+'),
GenomicSegment('2L', 111906, 112019, '+'),
GenomicSegment('2L', 112689, 113369, '+'),
GenomicSegment('2L', 113433, 114432, '+'),
Alias="'['M(2)21AB-RI', 'CG2674-RI']'",
ID='FBtr0089434',
Name='Sam-S-RI',
color='#000000',
gene_id='FBgn0005278',
score='0.0',
thickend='113542',
thickstart='108685',
type='exon'),
SegmentChain(GenomicSegment('2L', 108132, 108226, '+'),
GenomicSegment('2L', 108587, 108809, '+'),
GenomicSegment('2L', 110405, 110483, '+'),
GenomicSegment('2L', 110754, 110877, '+'),
GenomicSegment('2L', 111004, 111117, '+'),
GenomicSegment('2L', 112689, 113369, '+'),
GenomicSegment('2L', 113433, 114432, '+'),
Alias="'['M(2)21AB-RJ', 'CG2674-RJ']'",
ID='FBtr0089435',
Name='Sam-S-RJ',
color='#000000',
gene_id='FBgn0005278',
score='0.0',
thickend='113542',
thickstart='108685',
type='exon'),
SegmentChain(GenomicSegment('2L', 109593, 109793, '+'),
GenomicSegment('2L', 110405, 110483, '+'),
GenomicSegment('2L', 110754, 110877, '+'),
GenomicSegment('2L', 111004, 111117, '+'),
GenomicSegment('2L', 112689, 113369, '+'),
GenomicSegment('2L', 113433, 114210, '+'),
Alias="'['M(2)21AB-RG', 'CG2674-RG']'",
ID='FBtr0089436',
Name='Sam-S-RG',
color='#000000',
gene_id='FBgn0005278',
score='0.0',
thickend='113542',
thickstart='109750',
type='exon'),
]
self.assertEqual(sorted(expected), sorted(found))
CCCTCCTTCCGCTGGCCCCGACTGC
>chr30b:1(+)
CCTCCTTCCGCTGGCCCCGACTGCC
>chr30b:2(+)
CTCCTTCCGCTGGCCCCGACTGCCC
>chr30b:3(+)
TCCTTCCGCTGGCCCCGACTGCCCC
>chr30b:4(+)
CCTTCCGCTGGCCCCGACTGCCCCA
>chr30b:5(+)
CTTCCGCTGGCCCCGACTGCCCCAG
"""
CROSSMAP1 = [
(
SegmentChain(GenomicSegment("chr50a", 1, 10, "+")),
SegmentChain(GenomicSegment("chr50a", 1 + 25 - 1, 10 + 25 - 1, "-")),
),
(
SegmentChain(GenomicSegment("chr50a", 19, 26, "+")),
SegmentChain(GenomicSegment("chr50a", 19 + 25 - 1, 26 + 25 - 1, "-")),
),
(
SegmentChain(GenomicSegment("chr30b", 0, 6, "+")),
SegmentChain(GenomicSegment("chr30b", 0 + 25 - 1, 6 + 25 - 1, "-")),
)
]
CROSSMAP2 = [
(
SegmentChain(GenomicSegment("chr50a", 1 + 1000, 10 + 1000, "+")),
'YMR194C_B_mRNA_0':
['YMR194C_B_mRNA_0:0-324^396-729(-)', 'YMR194C_B_mRNA_0:0-328^400-729(-)'],
'YPL249C_A_mRNA_0':
['YPL249C_A_mRNA_0:0-410^648-697(-)', 'YPL249C_A_mRNA_0:0-417^655-697(-)']
}
unmatched_query_juncs = {
K: [SegmentChain.from_str(X) for X in V]
for K, V in unmatched_query_juncs.items()
}
"""Query junctions with no known matches"""
unmatched_noncan_query_juncs = [
"YNL130C:0-23^145-180(-)",
"YNL130C:0-53^165-180(-)",
"YNL130C:0-70^141-180(-)",
"YNL130C:0-49^121-180(-)",
]
unmatched_noncan_query_juncs = [
SegmentChain.from_str(X) for X in unmatched_noncan_query_juncs
]
"""Query junctions without canonical splice junctions in the match range"""
repetitive_regions = [
"YBR215W_mRNA_0:190-193(+)", # threeprime splice site plus
"YHL001W_mRNA_0:144-149(+)", # fiveprime splice site plus
"YIL133C_mRNA_0:935-940(-)", # threeprime splice site minus
"YMR194C_B_mRNA_0:325-328(-)", # fiveprime splice site minus
]
cross_hash = GenomeHash(
[SegmentChain(GenomicSegment.from_str(X)) for X in repetitive_regions])
cross_hash_seqs = {X.chrom for X in cross_hash.feature_dict.values()}
def test_exit_status():
# define columns
cols = {
"intA" : numpy.random.randint(0,high=2**16,size=size),
"intB" : numpy.random.randint(-10,high=20,size=size),
"idxA" : numpy.arange(size),
"chrA" : numpy.array([chr(65+(X%(91-65))) for X in range(size)]),
"strA" : numpy.array([str(GenomicSegment("chrA",X,X+500,"+")) for X in range(size)]),
"strB" : numpy.array([str(GenomicSegment("chrB",X/2,X/2+500,"-")) for X in range(size)]),
"floatA" : 10*numpy.random.randn(size) + 500,
"floatB" : (10**-5)*numpy.random.random(size),
"objA" : numpy.tile(None,5000),
"objB" : numpy.array([GenomicSegment("chrC",X,X+Y,"+") for X,Y in zip(range(size),numpy.random.randint(2,high=1000,size=size))]),
}
# allocate temp files we will use
headerfile = NamedTemporaryFile(delete=False,mode="w")
headerfile_extra_cols = NamedTemporaryFile(delete=False,mode="w")
headerfile_extra_cols_diff = NamedTemporaryFile(delete=False,mode="w")
headerfile_extra_cols_shuffled = NamedTemporaryFile(delete=False,mode="w")
headerfile_shuffled = NamedTemporaryFile(delete=False,mode="w")
headerfile_diff_vals = NamedTemporaryFile(delete=False,mode="w")
noheaderfile = NamedTemporaryFile(delete=False,mode="w")
noheaderfile_extra_cols = NamedTemporaryFile(delete=False,mode="w")
noheaderfile_extra_cols_diff = NamedTemporaryFile(delete=False,mode="w")
noheaderfile_extra_cols_shuffled = NamedTemporaryFile(delete=False,mode="w")
noheaderfile_shuffled = NamedTemporaryFile(delete=False,mode="w")
noheaderfile_diff_vals = NamedTemporaryFile(delete=False,mode="w")
# write values
keyorder = ["idxA"] + sorted(list(set(cols.keys()) - { "idxA" }))
table1 = pd.DataFrame(cols)
table1.to_csv(headerfile,index=False,header=True,sep="\t")
table1.to_csv(noheaderfile,index=False,header=False,sep="\t",
columns=keyorder)
headerfile.close()
noheaderfile.close()
table1["extra"] = 2**7 * numpy.random.random(size=size)
table1.to_csv(headerfile_extra_cols,index=False,header=True,sep="\t")
table1.to_csv(noheaderfile_extra_cols,index=False,header=False,sep="\t",
columns=["extra"]+keyorder)
headerfile_extra_cols.close()
noheaderfile_extra_cols.close()
table1["extra"] += 10**-4 * numpy.random.random(size=size)
table1.to_csv(headerfile_extra_cols_diff,index=False,header=True,sep="\t")
table1.to_csv(noheaderfile_extra_cols_diff,index=False,header=False,sep="\t",
columns=["extra"]+keyorder)
headerfile_extra_cols_diff.close()
noheaderfile_extra_cols_diff.close()
shufidx = numpy.arange(size)
shuffle(shufidx)
table2 = pd.DataFrame({ K : V[shufidx] for K,V in cols.items()})
table2.to_csv(headerfile_shuffled,index=False,header=True,sep="\t")
table2.to_csv(noheaderfile_shuffled,index=False,header=False,sep="\t",
columns=keyorder)
headerfile_shuffled.close()
noheaderfile_shuffled.close()
table2["extra"] = table1["extra"][shufidx]
table2.to_csv(headerfile_extra_cols_shuffled,index=False,header=True,sep="\t")
table2.to_csv(noheaderfile_extra_cols_shuffled,
index=False,header=False,sep="\t",
columns=["extra"]+keyorder)
headerfile_extra_cols_shuffled.close()
noheaderfile_extra_cols_shuffled.close()
# Define tests, as tuples of:
# -Test name/description
# -Command-line arguments to pass to :py:mod:`plastid.bin.test_table_equality`
# -Expected exit code/returns status for :py:func:`main`
tests = [
("same",
"%s %s" % (headerfile.name,headerfile.name),
0),
("diff_column_names",
"%s %s" % (headerfile.name,headerfile_extra_cols.name),
1),
("extra_column_names_ignored",
"%s %s --exclude extra" % (headerfile.name,headerfile_extra_cols.name),
0),
("shuffled_rows",
"%s %s" % (headerfile.name,headerfile_shuffled.name),
1),
("shuffled_rows_name_sort",
"%s %s --sort_keys idxA" % (headerfile.name,headerfile_shuffled.name),
0),
("shuffled_rows_multi_name_sort",
"%s %s --sort_keys strB chrA" % (headerfile.name,headerfile_shuffled.name),
0),
("same_column_names_diff_values",
"%s %s" % (headerfile_extra_cols.name,headerfile_extra_cols_diff.name),
1),
("same_column_names_diff_values_tol",
"%s %s --tol 0.01" % (headerfile_extra_cols.name,headerfile_extra_cols_diff.name),
0),
("same_column_names_diff_values_ignored",
"%s %s --exclude extra" % (headerfile_extra_cols.name,headerfile_extra_cols_diff.name),
0),
("shuffled_rows_extra_columns_ignored",
"%s %s --exclude extra" % (headerfile.name,headerfile_extra_cols_shuffled.name),
1),
("shuffled_rows_extra_columns_ignored_name_sort",
"%s %s --exclude extra --sort_keys idxA" % (headerfile.name,headerfile_extra_cols_shuffled.name),
0),
("noheader_same",
"%s %s --no_header" % (noheaderfile.name,noheaderfile.name),
0),
("noheader_extra_columns",
"%s %s --no_header" % (noheaderfile.name,noheaderfile_extra_cols.name),
1),
("noheader_shuffled_rows",
"%s %s --no_header" % (noheaderfile.name,noheaderfile_shuffled.name),
1),
("noheader_shuffled_rows_int_sort",
"%s %s --no_header --sort_keys 0" % (noheaderfile.name,noheaderfile_shuffled.name),
0),
("noheader_diff_values",
"%s %s --no_header" % (noheaderfile_extra_cols.name,noheaderfile_extra_cols_diff.name),
1),
("noheader_diff_values_tol",
"%s %s --no_header --tol 0.01" % (noheaderfile_extra_cols.name,noheaderfile_extra_cols_diff.name),
0),
("no_header_diff_values_ignored",
"%s %s --no_header --exclude 0" % (noheaderfile_extra_cols.name,noheaderfile_extra_cols_diff.name),
0),
("no_header_shuffled_rows_extra_columns_ignored_int_sort",
"%s %s --no_header --exclude 0 --sort_keys 1" % (noheaderfile_extra_cols.name,noheaderfile_extra_cols_shuffled.name),
0)
]
""" Tests to conduct, as tuples of:
- Test name/description
- Command-line arguments to pass to :py:mod:`plastid.bin.test_table_equality`
- Expected exit code/returns status for :py:func:`main`
"""
for test_name, argstr, expected_exit in tests:
yield check_exit_status, test_name, argstr, expected_exit
# clean up
os.unlink(headerfile.name )
os.unlink(headerfile_extra_cols.name )
os.unlink(headerfile_extra_cols_diff.name )
os.unlink(headerfile_extra_cols_shuffled.name )
os.unlink(headerfile_shuffled.name )
os.unlink(headerfile_diff_vals.name )
os.unlink(noheaderfile.name )
os.unlink(noheaderfile_extra_cols.name )
os.unlink(noheaderfile_extra_cols_diff.name )
os.unlink(noheaderfile_extra_cols_shuffled.name )
os.unlink(noheaderfile_shuffled.name )
os.unlink(noheaderfile_diff_vals.name )
cleanup_resources()
def setUpClass(cls):
cls.ivcs = {
"plus": [
SegmentChain(GenomicSegment("chrA", 0, 100, "+")),
SegmentChain(GenomicSegment("chrA", 50, 100, "+")),
SegmentChain(GenomicSegment("chrA", 50, 51, "+"))
],
"minus_k25_off0": [
SegmentChain(
GenomicSegment("chrA", 0 + 25 - 1, 100 + 25 - 1, "-")),
SegmentChain(
GenomicSegment("chrA", 50 + 25 - 1, 100 + 25 - 1, "-")),
SegmentChain(
GenomicSegment("chrA", 50 + 25 - 1, 51 + 25 - 1, "-"))
],
"minus_k50_off0": [
SegmentChain(
GenomicSegment("chrA", 0 + 50 - 1, 100 + 50 - 1, "-")),
SegmentChain(
GenomicSegment("chrA", 50 + 50 - 1, 100 + 50 - 1, "-")),
SegmentChain(
GenomicSegment("chrA", 50 + 50 - 1, 51 + 50 - 1, "-"))
],
"minus_k25_off10": [
SegmentChain(
GenomicSegment("chrA", 0 + 25 - 1 - 2 * 10,
100 + 25 - 1 - 2 * 10, "-")),
SegmentChain(
GenomicSegment("chrA", 50 + 25 - 1 - 2 * 10,
100 + 25 - 1 - 2 * 10, "-")),
SegmentChain(
GenomicSegment("chrA", 50 + 25 - 1 - 2 * 10,
51 + 25 - 1 - 2 * 10, "-"))
],
"minus_k50_off10": [
SegmentChain(
GenomicSegment("chrA", 0 + 50 - 1 - 2 * 10,
100 + 50 - 1 - 2 * 10, "-")),
SegmentChain(
GenomicSegment("chrA", 50 + 50 - 1 - 2 * 10,
100 + 50 - 1 - 2 * 10, "-")),
SegmentChain(
GenomicSegment("chrA", 50 + 50 - 1 - 2 * 10,
51 + 50 - 1 - 2 * 10, "-"))
],
}
def find_canonicals_in_range(query_junc,minus_range,plus_range,genome,canonicals):
"""Find any canonical splice junctions within in `minus_range...plus_range`
of `query_junc`
To be classified as within the range, the boundaries of the canonical
junction must be:
1. within `minus_range...plus_range` of the boundaries
of the the discovered junction.
2. separated by a nucleotide distance equal to the distance
separating the junction in `query_junc`.
3. On the same chromosome and strand.
Parameters
----------
query_junc : |SegmentChain|
A two-exon fragment representing a query splice junction
minus_range : int <= 0
Maximum number of nucleotides splice junction could be moved
to the left without reducing sequence support for the junction
see :py:func:`find_match_range`
plus_range : int >= 0
Maximum number of nucleotides splice junction could be moved
to the right without reducing sequence support for the junction
see :py:func:`find_match_range`
genome : dict
dict mapping chromosome names to :py:class:`Bio.SeqRecord.SeqRecord` s
canonicals : list
dinucleotide sequences to consider as canonical splice sites,
as a list of tuples. e.g. `[("GT","AG"), ("GC","AG")]`
Returns
-------
list
List of |SegmentChains| representing canonical splice junctions in
`minus_range...plus_range` of `query_junc`
"""
ltmp = []
chrom = query_junc.chrom
strand = query_junc.strand
iv1,iv2 = query_junc[0], query_junc[1]
iv1start, iv1end = iv1.start, iv1.end
iv2start, iv2end = iv2.start, iv2.end
for i in range(minus_range,plus_range+1):
for pair in canonicals:
if str(genome[chrom][iv1end + i:iv1end + i + 2].seq) == pair[0]\
and str(genome[chrom][iv2start - 2 + i:iv2start + i].seq) == pair[1]:
new_iv1 = GenomicSegment(chrom,
iv1start,
iv1end + i,
strand)
new_iv2 = GenomicSegment(chrom,
iv2start + i,
iv2end,
strand)
ltmp.append(SegmentChain(new_iv1,new_iv2))
return ltmp