def check_get_transcript_from_args_add_three_tabix(fmt):
# check endpoints of transcript and CDS
# GTF2 and BED format
bed_ref = REF_FILES["100transcripts_bed"]
cds = {}
tx = {}
for line in open(bed_ref):
items = line.strip("\n").split("\t")
name = items[3]
tx_start = int(items[1])
tx_end = int(items[2])
tx[name] = (tx_start,tx_end)
cds_start = int(items[6])
cds_end = int(items[7])
if cds_start < cds_end:
cds[name] = (cds_start,cds_end)
else:
cds[name] = (None,None)
ref_files = {
"BED" : REF_FILES["100transcripts_bed_tabix"],
"GTF2" : REF_FILES["100transcripts_gtf_tabix"]
}
parser = get_annotation_file_parser()
argstr = "--annotation_format %s --annotation_files %s --tabix" % (fmt,ref_files[fmt])
args = parser.parse_args(shlex.split(argstr))
transcripts = list(get_transcripts_from_args(args))
for my_tx in transcripts:
name = my_tx.get_name()
expected_cds_start, expected_cds_end = cds[name]
found_cds_start = my_tx.cds_genome_start
found_cds_end = my_tx.cds_genome_end
found_tx_start = my_tx.spanning_segment.start
found_tx_end = my_tx.spanning_segment.end
expected_tx_start, expected_tx_end = tx[name]
if found_cds_start is None:
assert_true(found_cds_end is None)
assert_true(expected_cds_start is None,"Unequal CDS start (expected %s, got %s) on transcript %s" % (expected_cds_start,found_cds_start,name))
assert_true(expected_cds_end is None,"Unequal CDS stop (expected %s, got %s) on transcript %s" % (expected_cds_end,found_cds_end,name))
else:
assert_equal(found_cds_start,expected_cds_start,"Unequal CDS start (expected %s, got %s) on transcript %s" % (expected_cds_start,found_cds_start,name))
assert_equal(found_cds_end,expected_cds_end,"Unequal CDS stop (expected %s, got %s) on transcript %s" % (expected_cds_end,found_cds_end,name))
assert_equal(found_tx_start,expected_tx_start,"Unequal transcript start (expected %s, got %s) on transcript %s" % (expected_tx_start,found_tx_start,name))
assert_equal(found_tx_end,expected_tx_end,"Unequal transcript end (expected %s, got %s) on transcript %s" % (expected_tx_end,found_tx_end,name))
assert_equal(len(transcripts),len(cds),"Not all transcripts found in input. Expected %s. Got %s." % (len(cds),len(transcripts)))
def check_get_transcript_from_args_add_three(parser,fmt,add_three,n=1):
"""Make sure stop codons are moved by three if ``add_three`` is specified
in argument string and not moved by three if not specified.
These are checked against expected values found in a BED file.
Parameters
----------
parser : :py:class:`argparse.ArgumentParser`
Argument parser for annotation files
fmt : str
Format of input file ("BED", "BigBed", "GTF2", "GFF3", or "PSL")
add_three : bool
Whether or not to move stop codons by three nucleotides
n : int
Number of times input file was passed to parser (Default : 1)
"""
# GTF file does not include stop codon; others do.
# So, GTF file will be off by three relative to others
# unless it includes explicit stop codon features
#
# we correct for that here
gtf_offset = 3 if fmt == "GTF2" else 0
bed_ref = MINI["bed_file"]
dtmp = {}
for line in open(bed_ref):
items = line.strip("\n").split("\t")
name = items[3]
strand = items[5]
cds_start = int(items[6])
cds_end = int(items[7])
if cds_start < cds_end:
if strand == "+":
cds_end -= gtf_offset
else:
cds_start += gtf_offset
if add_three == True:
if strand == "+":
cds_end += 3
else:
cds_start -= 3
dtmp[name] = (cds_start,cds_end)
else:
dtmp[name] = (None,None)
files = " ".join([MINI["%s_file" % fmt.lower()]] * n)
argstr = "--annotation_format %s --annotation_files %s " % (fmt,files)
if add_three == True:
argstr += " --add_three"
args = parser.parse_args(shlex.split(argstr))
transcripts = list(get_transcripts_from_args(args))
for tx in transcripts:
name = tx.get_name()
expected_start = dtmp[name][0]
expected_end = dtmp[name][1]
found_start = tx.cds_genome_start
found_end = tx.cds_genome_end
if found_start is None:
assert_true(found_end is None)
assert_true(expected_start is None,"Unequal CDS start (expected %s, got %s) on transcript %s" % (expected_start,found_start,name))
assert_true(expected_end is None,"Unequal CDS stop (expected %s, got %s) on transcript %s" % (expected_end,found_end,name))
else:
assert_equal(found_start,expected_start,"Unequal CDS start (expected %s, got %s) on transcript %s" % (expected_start,found_start,name))
assert_equal(found_end,expected_end,"Unequal CDS stop (expected %s, got %s) on transcript %s" % (expected_end,found_end,name))
assert_equal(len(transcripts),n*len(dtmp),"Not all transcripts found in input. Expected %s. Got %s." % (n*len(dtmp),len(transcripts)))
