def test_read_gff(self):
level_records, other_records = helper.read_gff(
os.path.join(data_dir, 'test_read_gff.gff'))
expected_level_records = [[
gff.GFF_record('\t'.join([
'seqname', 'SOURCE', 'gene', '42', '100', '.', '+', '.',
'ID=id'
]))
],
[
gff.GFF_record('\t'.join([
'seqname', 'SOURCE', 'mRNA', '42',
'100', '.', '+', '.',
'ID=id.1;Parent=id'
]))
],
[
gff.GFF_record('\t'.join([
'seqname', 'SOURCE', 'exon', '52',
'62', '.', '+', '.',
'ID=id.1:exon:1;Parent=id.1'
]))
]]
expected_other_records = {
'seqname': [
gff.GFF_record('\t'.join([
'seqname', 'SOURCE', 'eggs', '150', '200', '.', '+', '.',
'ID=spam'
]))
]
}
self.assertEqual(expected_level_records, level_records)
self.assertEqual(expected_other_records, other_records)
def test_interscts(self):
'''Test instersects'''
gff_1 = gff.GFF_record('\t'.join(['seq', 'SOURCE', 'gene', '42', '45', '.', '.', '.']))
intersects = [
['seq', 'SOURCE', 'gene', '43', '44', '.', '.', '.'],
['seq', 'SOURCE', 'gene', '42', '43', '.', '.', '.'],
['seq', 'SOURCE', 'gene', '43', '45', '.', '.', '.'],
['seq', 'SOURCE', 'gene', '41', '42', '.', '.', '.'],
['seq', 'SOURCE', 'gene', '41', '43', '.', '.', '.']
]
for l in intersects:
record = gff.GFF_record('\t'.join(l))
self.assertTrue(gff_1.intersects(record))
not_intersects = [
['seq', 'SOURCE', 'gene', '40', '41', '.', '.', '.'],
['seq', 'SOURCE', 'gene', '46', '50', '.', '.', '.'],
['seq', 'SOURCE', 'gene', '1', '10', '.', '.', '.'],
['seq', 'SOURCE', 'gene', '100', '200', '.', '.', '.'],
['different_seq', 'SOURCE', 'gene', '43', '45', '.', '.', '.'],
]
for l in not_intersects:
record = gff.GFF_record('\t'.join(l))
self.assertFalse(gff_1.intersects(record))
def test_load_cufflinks_gtf(self):
genes = helper.load_cufflinks_gtf(
os.path.join(data_dir, 'test_load_cufflinks_gtf.gtf'))
gene1_trans = gff.GFF_record('\t'.join([
'seq', 'Cufflinks', 'transcript', '1', '100', '1000', '+', '.',
'gene_id "CUFF.1"; transcript_id "CUFF.1.1";'
]))
gene1_exon = gff.GFF_record('\t'.join([
'seq', 'Cufflinks', 'exon', '1', '100', '1000', '+', '.',
'gene_id "CUFF.1"; transcript_id "CUFF.1.1";'
]))
gene2_trans = gff.GFF_record('\t'.join([
'seq2', 'Cufflinks', 'transcript', '1', '200', '2000', '+', '.',
'gene_id "CUFF.2"; transcript_id "CUFF.2.1";'
]))
gene2_exon1 = gff.GFF_record('\t'.join([
'seq2', 'Cufflinks', 'exon', '1', '100', '2000', '+', '.',
'gene_id "CUFF.2"; transcript_id "CUFF.2.1";'
]))
gene2_exon2 = gff.GFF_record('\t'.join([
'seq2', 'Cufflinks', 'exon', '150', '200', '2000', '+', '.',
'gene_id "CUFF.2"; transcript_id "CUFF.2.1";'
]))
gene1 = gene.Gene(gene1_trans)
gene1.add_gff_record(gene1_exon)
gene2 = gene.Gene(gene2_trans)
gene2.add_gff_record(gene2_exon1)
gene2.add_gff_record(gene2_exon2)
expected_genes = {'seq': [gene1], 'seq2': [gene2]}
self.assertEqual(expected_genes['seq2'], genes['seq2'])
with self.assertRaises(helper.Error):
helper.load_cufflinks_gtf(
os.path.join(data_dir,
'test_load_cufflinks_gtf.no_parent.gtf'))
def test_remove_all_but_longest_transcript(self):
'''Test remove_all_but_longest_transcript'''
self.gene.remove_all_but_longest_transcript()
self.assertEqual(['gene_id.1'], list(self.gene.transcripts.keys()))
gff_mRNA2 = gff.GFF_record('\t'.join([
'seqname', 'SOURCE', 'mRNA', '42', '100', '.', '+', '.',
'ID=gene_id.2;Parent=gene_id'
]))
gff_exon2 = gff.GFF_record('\t'.join([
'seqname', 'SOURCE', 'exon', '50', '65', '.', '+', '.',
'ID=gene_id.2:exon:1;Parent=gene_id.2'
]))
self.gene.add_gff_record(gff_mRNA2)
self.gene.add_gff_record(gff_exon2)
self.gene.remove_all_but_longest_transcript()
self.assertEqual(['gene_id.2'], list(self.gene.transcripts.keys()))
def test_longest_transcript_by_exon_length(self):
'''Test longest_transcript_by_exon_length'''
self.assertEqual('gene_id.1',
self.gene.longest_transcript_by_exon_length())
gff_mRNA2 = gff.GFF_record('\t'.join([
'seqname', 'SOURCE', 'mRNA', '42', '100', '.', '+', '.',
'ID=gene_id.2;Parent=gene_id'
]))
gff_exon2 = gff.GFF_record('\t'.join([
'seqname', 'SOURCE', 'exon', '50', '65', '.', '+', '.',
'ID=gene_id.2:exon:1;Parent=gene_id.2'
]))
self.gene.add_gff_record(gff_mRNA2)
self.gene.add_gff_record(gff_exon2)
self.assertEqual('gene_id.2',
self.gene.longest_transcript_by_exon_length())
def test_set_attribute(self):
'''Test get_attribute'''
gff_record = gff.GFF_record('\t'.join(['seq', 'SOURCE', 'feature', '42', '43', '.', '.', '.']))
gff_record.set_attribute('key1', '42')
self.assertTrue(gff_record.get_attribute('key1'), '42')
gff_record.set_attribute('key1', '43')
self.assertTrue(gff_record.get_attribute('key1'), '43')
def setUp(self):
self.gff_mRNA = gff.GFF_record('\t'.join([
'seqname', 'SOURCE', 'mRNA', '42', '100', '.', '+', '.',
'ID=gene_id.1;Parent=gene_id'
]))
self.gff_gene = gff.GFF_record('\t'.join([
'seqname', 'SOURCE', 'gene', '42', '100', '.', '+', '.',
'ID=gene_id'
]))
self.gff_exon1 = gff.GFF_record('\t'.join([
'seqname', 'SOURCE', 'exon', '50', '60', '.', '+', '.',
'ID=gene_id.1:exon:1;Parent=gene_id.1'
]))
self.gene = gene.Gene(self.gff_mRNA)
self.gene.add_gff_record(self.gff_gene)
self.gene.add_gff_record(self.gff_exon1)
def test_sort(self):
'''Test sort'''
unsorted_list = [
gff.GFF_record('\t'.join(['x', 'x', 'x', '42', '43', '.', '.', '.'])),
gff.GFF_record('\t'.join(['x', 'x', 'x', '12', '13', '.', '.', '.']))
]
sorted_list = list(unsorted_list)
sorted_list.sort()
for l in [self.trans.five_utr, self.trans.three_utr, self.trans.exons, self.trans.ncRNA ,self.trans.rRNA, self.trans.tRNA, self.trans.snRNA]:
l += unsorted_list
self.trans._sort()
for l in [self.trans.five_utr, self.trans.three_utr, self.trans.exons, self.trans.ncRNA ,self.trans.rRNA, self.trans.tRNA, self.trans.snRNA]:
self.assertEqual(sorted_list, l)
def test_update_other_records(self):
other_records = {}
gff1 = gff.GFF_record('\t'.join([
'seqname', 'SOURCE', 'eggs', '150', '200', '.', '+', '.', 'ID=spam'
]))
gff2 = gff.GFF_record('\t'.join([
'seqname', 'SOURCE', 'spam', '150', '200', '.', '+', '.', 'ID=eggs'
]))
expected = {}
helper.update_other_records(other_records, gff1)
expected['seqname'] = [copy.deepcopy(gff1)]
self.assertEqual(other_records, expected)
helper.update_other_records(other_records, gff2)
expected['seqname'].append(copy.deepcopy(gff2))
self.assertEqual(other_records, expected)
def test_can_extend(self):
new_left_exon = gff.GFF_record('\t'.join([
'seqname', 'SOURCE', 'exon', '40', '55', '.', '+', '.',
'ID=gene_id.1:exon:1;Parent=gene_id.1'
]))
new_right_exon = gff.GFF_record('\t'.join([
'seqname', 'SOURCE', 'exon', '55', '70', '.', '+', '.',
'ID=gene_id.1:exon:1;Parent=gene_id.1'
]))
gene2 = copy.deepcopy(self.gene)
self.assertFalse(self.gene.can_extend(gene2))
gene2.add_gff_record(new_left_exon)
self.assertTrue(self.gene.can_extend(gene2))
gene2.seqname = 'holyhandgrenade'
self.assertFalse(self.gene.can_extend(gene2))
gene2 = copy.deepcopy(self.gene)
gene2.add_gff_record(new_right_exon)
self.assertTrue(self.gene.can_extend(gene2))
def test_lenient_mode(self):
'''Test __init__ when being lenient'''
gff.lenient = True
test_input = [
['seq', 'Cufflinks', 'transcript', '42', '43', '.', '.', '.', 'key1 "val1"; key2 "val2";'],
['seq', 'SOURCE', 'gene', '42', '43', '.', '.', '.', 'attribute with no equals sign'],
]
for l in test_input:
record = gff.GFF_record('\t'.join(l))
self.assertEqual('\t'.join(l), str(record))
gff.lenient = False
def test_get_attribute(self):
'''Test get_attribute'''
gff.lenient = True
gff_record = gff.GFF_record('\t'.join(['seq', 'SOURCE', 'feature', '42', '43', '.', '.', '.', 'key1=val1;key2=val2;key3']))
self.assertEqual(gff_record.get_attribute('key1'), 'val1')
self.assertEqual(gff_record.get_attribute('key2'), 'val2')
self.assertEqual(gff_record.get_attribute('key3'), None)
with self.assertRaises(gff.Error):
gff_record.get_attribute('killer rabbit')
gff.lenient = False
def test_load_ref_gff(self):
#level_records, other_records = helper.read_gff(os.path.join(data_dir, 'test_get_genes_from_ref.gff'))
genes, other_records = helper.load_ref_gff(
os.path.join(data_dir, 'test_get_genes_from_ref.gff'))
gene1_gene = gff.GFF_record('\t'.join(
['seq', 'SOURCE', 'gene', '42', '100', '.', '+', '.', 'ID=gene']))
gene1_mRNA = gff.GFF_record('\t'.join([
'seq', 'SOURCE', 'mRNA', '42', '100', '.', '+', '.',
'ID=gene.1;Parent=gene'
]))
gene1_exon = gff.GFF_record('\t'.join([
'seq', 'SOURCE', 'exon', '42', '62', '.', '+', '.',
'ID=gene.1:exon:1;Parent=gene.1'
]))
gene1_exon2 = gff.GFF_record('\t'.join([
'seq', 'SOURCE', 'exon', '92', '100', '.', '+', '.',
'ID=gene.1:exon:2;Parent=gene.1'
]))
gene1 = gene.Gene(gene1_gene)
gene1.add_gff_record(gene1_mRNA)
gene1.add_gff_record(gene1_exon)
gene1.add_gff_record(gene1_exon2)
gene2_gene = gff.GFF_record('\t'.join(
['seq2', 'SOURCE', 'gene', '1', '10', '.', '+', '.', 'ID=gene2']))
gene2_mRNA = gff.GFF_record('\t'.join([
'seq2', 'SOURCE', 'mRNA', '1', '10', '.', '+', '.',
'ID=gene2.1;Parent=gene2'
]))
gene2_exon = gff.GFF_record('\t'.join([
'seq2', 'SOURCE', 'exon', '1', '10', '.', '+', '.',
'ID=gene2.1:exon:1;Parent=gene2.1'
]))
gene2 = gene.Gene(gene2_gene)
gene2.add_gff_record(gene2_mRNA)
gene2.add_gff_record(gene2_exon)
expected_genes = {'seq': [gene1], 'seq2': [gene2]}
expected_other = {
'seq': [
gff.GFF_record('\t'.join([
'seq', 'SOURCE', 'eggs', '150', '200', '.', '+', '.',
'ID=spam'
]))
]
}
self.assertEqual(expected_genes, genes)
self.assertEqual(expected_other, other_records)
def test_init_bad_input(self):
'''Test __init__ for bad input'''
test_input = [
['seq', 'SOURCE', 'gene', '42', '43', '.', '.'],
['seq', 'SOURCE', 'gene', '42', '43', '.', '.', '.', '.', 'too many'],
['seq', 'SOURCE', 'gene', 'not_int', '43', '.', '.', '.'],
['seq', 'SOURCE', 'gene', '42', 'not_int', '.', '.', '.'],
['seq', 'SOURCE', 'gene', '42', '43', 'not_float', '.', '.'],
['seq', 'SOURCE', 'gene', '42', '43', '.', 'bad_strand', '.'],
['seq', 'SOURCE', 'gene', '42', '43', '.', '.', '.', 'attribute with no equals sign'],
]
for l in test_input:
with self.assertRaises(gff.Error):
gff.GFF_record('\t'.join(l))
def test_init_good_input(self):
'''Test __init__ for good input'''
test_input = [
['seq', 'SOURCE', 'gene', '42', '43', '.', '.', '.'],
['seq', 'SOURCE', 'gene', '42', '43', '1.4', '-', '0'],
['seq', 'SOURCE', 'gene', '42', '43', '1.4', '+', '1'],
['seq', 'SOURCE', 'gene', '42', '43', '1.4', '+', '1', 'key=value'],
['seq', 'SOURCE', 'gene', '42', '43', '1.4', '+', '1', 'key=value;key2=value 2'],
['seq', 'Cufflinks', 'transcript', '42', '43', '.', '.', '.', 'key1 "val1";'],
['seq', 'Cufflinks', 'transcript', '42', '43', '.', '.', '.', 'key1 "val1";'],
['seq', 'Cufflinks', 'transcript', '42', '43', '.', '.', '.', 'key1 "val1"; key2 "val2";'],
]
for l in test_input:
record = gff.GFF_record('\t'.join(l))
self.assertEqual('\t'.join(l), str(record))
def test_less_than(self):
'''Test less than operator'''
gff_1 = gff.GFF_record('\t'.join(['seq', 'SOURCE', 'gene', '42', '44', '.', '.', '.']))
gff_2 = gff.GFF_record('\t'.join(['seq', 'SOURCE', 'gene', '42', '43', '.', '.', '.']))
gff_3 = gff.GFF_record('\t'.join(['seq', 'SOURCE', 'gene', '42', '45', '.', '.', '.']))
gff_4 = gff.GFF_record('\t'.join(['seq', 'SOURCE', 'gene', '41', '42', '.', '.', '.']))
gff_5 = gff.GFF_record('\t'.join(['seq', 'SOURCE', 'gene', '41', '45', '.', '.', '.']))
gff_6 = gff.GFF_record('\t'.join(['different_seq', 'SOURCE', 'gene', '42', '45', '.', '.', '.']))
self.assertFalse(gff_1 < gff_1)
self.assertFalse(gff_1 < gff_2)
self.assertTrue(gff_1 < gff_3)
self.assertFalse(gff_1 < gff_4)
self.assertFalse(gff_1 < gff_5)
self.assertFalse(gff_1 < gff_6)
def test_len(self):
'''Test __len__'''
g = gff.GFF_record('\t'.join(['seq', 'SOURCE', 'gene', '42', '44', '.', '.', '.']))
self.assertEqual(3, len(g))
def test_update_utrs(self):
'''Test update_utrs()'''
self.trans.add_gff_record(self.gff_exon)
self.trans.add_gff_record(self.gff_exon2)
self.trans.add_gff_record(self.gff_exon3)
new_left_exon = gff.GFF_record('\t'.join(['seqname', 'SOURCE', 'CDS', '30', '50', '.', '+', '.']))
new_left_exon2 = gff.GFF_record('\t'.join(['seqname', 'SOURCE', 'CDS', '10', '20', '.', '+', '.']))
new_right_exon = gff.GFF_record('\t'.join(['seqname', 'SOURCE', 'CDS', '90', '100', '.', '+', '.']))
new_right_exon2 = gff.GFF_record('\t'.join(['seqname', 'SOURCE', 'CDS', '110', '120', '.', '+', '.']))
new_five_utr = gff.GFF_record('\t'.join(['seqname', 'UTR_updater', 'five_prime_UTR', '30', '43', '.', '+', '.', 'ID=gene_id.1:5utr;Parent=gene_id.1']))
new_five_utr_2 = gff.GFF_record('\t'.join(['seqname', 'UTR_updater', 'five_prime_UTR', '30', '43', '.', '+', '.', 'ID=gene_id.1:5utr:2;Parent=gene_id.1']))
new_five_utr_1 = gff.GFF_record('\t'.join(['seqname', 'UTR_updater', 'five_prime_UTR', '10', '20', '.', '+', '.', 'ID=gene_id.1:5utr:1;Parent=gene_id.1']))
new_three_utr = gff.GFF_record('\t'.join(['seqname', 'UTR_updater', 'three_prime_UTR', '95', '100', '.', '+', '.', 'ID=gene_id.1:3utr;Parent=gene_id.1']))
new_three_utr_1 = gff.GFF_record('\t'.join(['seqname', 'UTR_updater', 'three_prime_UTR', '95', '100', '.', '+', '.', 'ID=gene_id.1:3utr:1;Parent=gene_id.1']))
new_three_utr_2 = gff.GFF_record('\t'.join(['seqname', 'UTR_updater', 'three_prime_UTR', '110', '120', '.', '+', '.', 'ID=gene_id.1:3utr:2;Parent=gene_id.1']))
before_adding = [copy.deepcopy(self.trans) for x in range(5)]
after_adding = [copy.deepcopy(self.trans) for x in range(5)]
to_add = [copy.deepcopy(self.trans) for x in range(5)]
# test extending the first exon
to_add[0].exons.pop(0)
to_add[0].add_gff_record(new_left_exon)
after_adding[0].exons[0].coords.start = 44
after_adding[0].add_gff_record(new_five_utr)
# test adding a spliced UTR at the start
to_add[1].exons.pop(0)
to_add[1].add_gff_record(new_left_exon)
to_add[1].add_gff_record(new_left_exon2)
after_adding[1].exons[0].coords.start = 44
after_adding[1].add_gff_record(new_five_utr_1)
after_adding[1].add_gff_record(new_five_utr_2)
# test extending the last exon
to_add[2].exons.pop()
to_add[2].add_gff_record(new_right_exon)
after_adding[2].exons[-1].coords.end = 94
after_adding[2].add_gff_record(new_three_utr)
# test adding a spliced UTR at the end
to_add[3].exons.pop()
to_add[3].add_gff_record(new_right_exon)
to_add[3].add_gff_record(new_right_exon2)
after_adding[3].exons[-1].coords.end = 94
after_adding[3].add_gff_record(new_three_utr_1)
after_adding[3].add_gff_record(new_three_utr_2)
# test trying to add too many new uts splices
to_add[4].add_gff_record(new_left_exon)
to_add[4].add_gff_record(new_left_exon2)
to_add[4].add_gff_record(new_left_exon2)
to_add[4].add_gff_record(new_left_exon2)
# test adding a 5'UTR that overlaps an existing gene. Should do nothing
t = copy.deepcopy(before_adding[0])
t.update_utrs(to_add[0], exclude_coords = [intervals.Interval(10,31)])
self.assertEqual(t, before_adding[0])
# test adding a 3'UTR that overlaps an existing gene. Should do nothing
t = copy.deepcopy(before_adding[2])
t.update_utrs(to_add[2], exclude_coords = [intervals.Interval(100,110)])
self.assertEqual(t, before_adding[2])
for i in range(len(before_adding)):
before_adding[i].update_utrs(to_add[i])
self.assertEqual(before_adding[i], after_adding[i])
def setUp(self):
self.gff_mRNA = gff.GFF_record('\t'.join(['seqname', 'SOURCE', 'mRNA', '42', '100', '.', '+', '.', 'ID=gene_id.1;Parent=gene_id']))
self.gff_five_utr = gff.GFF_record('\t'.join(['seqname', 'SOURCE', 'five_prime_UTR', '42', '43', '.', '+', '.', 'ID=gene_id.1:5utr;Parent=gene_id.1']))
self.gff_three_utr = gff.GFF_record('\t'.join(['seqname', 'SOURCE', 'three_prime_UTR', '95', '100', '.', '+', '.', 'ID=gene_id.1:3utr;Parent=gene_id.1']))
self.gff_exon = gff.GFF_record('\t'.join(['seqname', 'SOURCE', 'CDS', '44', '53', '.', '+', '.', 'ID=gene_id.1:exon:1;Parent=gene_id.1']))
self.gff_exon2 = gff.GFF_record('\t'.join(['seqname', 'SOURCE', 'exon', '60', '63', '.', '+', '.', 'ID=gene_id.1:exon:2;Parent=gene_id.1']))
self.gff_exon3 = gff.GFF_record('\t'.join(['seqname', 'SOURCE', 'exon', '90', '94', '.', '+', '.', 'ID=gene_id.1:exon:2;Parent=gene_id.1']))
self.gff_pseudogenic_exon = gff.GFF_record('\t'.join(['seqname', 'SOURCE', 'pseudogenic_exon', '60', '63', '.', '+', '.', 'ID=gene_id.1:exon:3;Parent=gene_id.1']))
self.gff_transcript = gff.GFF_record('\t'.join(['seqname', 'SOURCE', 'transcript', '42', '100', '.', '+', '.', 'ID=gene_id.1;Parent=gene_id']))
self.gff_polypeptide = gff.GFF_record('\t'.join(['seqname', 'SOURCE', 'polypeptide', '42', '100', '.', '+', '.', 'ID=gene_id.1:pep;Derives_from=gene_id.1;translation=abcdefghjiklmnopwhatisyourfavouritecolourqrstuvwxyz']))
self.gff_pseudogenic_transcript = gff.GFF_record('\t'.join(['seqname', 'SOURCE', 'pseudogenic_transcript', '42', '100', '.', '+', '.', 'ID=gene_id.1;Parent=gene_id']))
self.gff_ncRNA = gff.GFF_record('\t'.join(['seqname', 'SOURCE', 'ncRNA', '42', '43', '.', '+', '.', 'ID=gene_id.1:ncRNA;Parent=gene_id.1']))
self.gff_tRNA = gff.GFF_record('\t'.join(['seqname', 'SOURCE', 'tRNA', '42', '43', '.', '+', '.', 'ID=gene_id.1:tRNA;Parent=gene_id.1']))
self.gff_snRNA = gff.GFF_record('\t'.join(['seqname', 'SOURCE', 'snRNA', '42', '43', '.', '+', '.', 'ID=gene_id.1:snRNA;Parent=gene_id.1']))
self.gff_rRNA = gff.GFF_record('\t'.join(['seqname', 'SOURCE', 'rRNA', '42', '43', '.', '+', '.', 'ID=gene_id.1:rRNA;Parent=gene_id.1']))
self.gff_other = gff.GFF_record('\t'.join(['seqname', 'SOURCE', 'eggs', '42', '4242', '.', '+', '.']))
self.trans = transcript.Transcript(self.gff_mRNA)