def test_tuple_proportion(self):
#TATTAGGCTGTGATGTGCTT
#01234567890123456789
GI = reader.gff3_interval(self.gff3_1, fasta=self.fa)
self.assertEqual(GI._tuple_proportion('Chr2', (0, 10, 0, 0)),
(0.2, 0.4, 0.3, 0.1))
self.assertEqual(GI._tuple_proportion('Chr2', (5, 15, 1, 1)),
(0.1, 0.3, 0.5, 0.1))
def test_gff3_12_fa_region_2_gene(self):
GI = reader.gff3_interval(self.gff3_1, fasta=self.fa)
GI.add_gff3(self.gff3_2, 'treat')
ret = GI.calc_intersect_2('Chr1',
'control',
'treat',
GI.element_dict['gene'],
1,
p=99)
self.assertEquals(ret, (1, 1, 1))
def test_gff3_1_fa(self):
GI = reader.gff3_interval(self.gff3_1, fasta=self.fa)
self._test_elem_sets(GI)
self.assertEqual(GI.gff3_names, ['control'])
for chrom in ('Chr1', 'Chr2'):
self.assertEqual(GI._get_max(chrom), self.fa_max[chrom])
# Check sums
for elem in GI.element_dict.keys():
fa, ra, ba = self._gen_arrays(GI, 'Chr1', elem, 1)
for i, a in enumerate((fa, ra, ba)):
self.assertEqual(a.shape, (1, self.fa_max['Chr1']))
self.assertEqual(np.sum(a), self.elems_sum1[elem][i])
def main():
fCheck = fileCheck() #class for checking parameters
parser = argparse.ArgumentParser(description="A tool for comparing GFF3 annotations")
parser.add_argument('-C', '--control', metavar='GFF3', help='Control GFF3. All comparisons are relative to this annotation.', required=True, type=fCheck.gff3)
parser.add_argument('-R', '--reference', metavar='FASTA', \
help='Control reference (required for base pair metrics)', type=fCheck.fasta)
parser.add_argument('--cname', metavar='STR', help='Name of control GFF3', default='control', type=str)
parser.add_argument('-T', '--treat', metavar='GFF3', \
help='Space separated list of GFF3 files for comparison against the control', \
type=fCheck.gff3, required=True, nargs='+')
parser.add_argument('-N', '--names', metavar='STR', \
help='Space separated list of names for treatment GFF3 files (name order must match file order)', \
type=str, required=True, nargs='+')
parser.add_argument('-p', '--percent', metavar='INT', \
help='Reciprocal percent overlap threshold [%(default)s]', type=int, default=90)
parser.add_argument('--plot', action="store_true", help="Plot venn diagrams of results")
parser.add_argument('-e', '--ext', metavar='EXT', \
help='Figure extension [%(default)s]', default='png', \
type=argChecker(('pdf','png','eps'),'figure extension').check)
parser.add_argument('-v', '--verbose', action="store_true", help='Enable verbose logging')
parser.add_argument('--temd', action="store_true", help='Analyze TE metadata')
args = parser.parse_args()
################################
# Configure logging
################################
if args.verbose:
logger.setLevel(logging.DEBUG)
logger.debug("DEBUG logging enabled")
else:
logger.setLevel(logging.INFO)
################################
# Check arguments
################################
if len(args.names) != len(args.treat):
logger.error("treat(%i) != names(%i)"%(len(args.treat), len(args.names)))
raise ValueError
################################
# Create GFF3 intervals
################################
GI = reader.gff3_interval(args.control, name=args.cname, fasta=args.reference)
for f, n in zip(args.treat, args.names):
GI.add_gff3(f, n)
################################
# Generate results
################################
fig_ext = args.ext if args.plot else False
if args.reference:
logger.info("Basepair resolution results")
summaries.tabular(GI, fig_ext=fig_ext, temd=args.temd)
logger.info("Interval results")
summaries.tabular_region(GI, p=args.percent, fig_ext=fig_ext, temd=args.temd)
logger.info("Done")
def test_gff3_12_nofa(self):
GI = reader.gff3_interval(self.gff3_1)
GI.add_gff3(self.gff3_2, 'treat')
self._test_elem_sets(GI)
self.assertEqual(GI.gff3_names, ['control', 'treat'])
for chrom in ('Chr1', 'Chr2'):
self.assertEqual(GI._get_max(chrom), self.gff3_max[chrom])
# Check sums
for elem in GI.element_dict.keys():
fa, ra, ba = self._gen_arrays(GI, 'Chr1', elem, 1)
for i, a in enumerate((fa, ra, ba)):
self.assertEqual(a.shape, (2, self.gff3_max['Chr1']))
self.assertEqual(np.sum(a[0, :]), self.elems_sum1[elem][i])
self.assertEqual(np.sum(a[1, :]), self.elems_sum2[elem][i])
def test_get_proportion_arrays(self):
#TATTAGGCTGTGATGTGCTT
#01234567890123456789
# ----- ------
GI = reader.gff3_interval(self.gff3_1, fasta=self.fa)
prop_array = GI.get_proportion_arrays('Chr2',
'control',
'exon',
1,
strand=False)
self.assertEqual(sorted(prop_array[0]),
sorted([float(1) / 5, float(1) / 6])) #A
self.assertEqual(sorted(prop_array[1]),
sorted([float(2) / 5, float(3) / 6])) #T
self.assertEqual(sorted(prop_array[2]),
sorted([float(2) / 5, float(2) / 6])) #G
self.assertEqual(sorted(prop_array[3]),
sorted([float(0) / 5, float(0) / 6])) #C
