def test_min_scores_filter(self):
'We can keep the hits scores above the given one'
blast_file = open(os.path.join(TEST_DATA_DIR, 'blast.xml'))
#with evalue
filters = [{'kind': 'score_threshold',
'score_key': 'expect',
'max_score': 1e-34,
}]
expected = {'cCL1Contig2': 2, 'cCL1Contig3': 0,
'cCL1Contig4': 2, 'cCL1Contig5': 2}
blasts = BlastParser(fhand=blast_file)
filtered_blasts = filter_alignments(blasts, config=filters)
match_summary = _summarize_matches(filtered_blasts)
_check_match_summary(match_summary, expected)
#with similartiry
filters = [{'kind': 'score_threshold',
'score_key': 'similarity',
'min_score': 92,
}]
expected = {'cCL1Contig2': 0, 'cCL1Contig3': 0,
'cCL1Contig4': 1, 'cCL1Contig5': 2}
blasts = BlastParser(fhand=blast_file)
filtered_blasts = filter_alignments(blasts, config=filters)
match_summary = _summarize_matches(filtered_blasts)
_check_match_summary(match_summary, expected)
def _do_blast_2(db_fpath,
queries,
program,
dbtype=None,
blast_format=None,
params=None,
remote=False):
'''It returns an alignment result with the blast.
It is an alternative interface to the one based on fpaths.
db_fpath should be a plain sequence file.
queries should be a SeqRecord list.
If an alternative blast output format is given it should be tabular, so
blast_format is a list of fields.
'''
query_fhand = write_seqs(queries, file_format='fasta')
query_fhand.flush()
if remote:
blastdb = db_fpath
fmt = 'XML' if blast_format is None else blast_format.upper()
else:
blastdb = get_or_create_blastdb(db_fpath, dbtype=dbtype)
if blast_format is None:
blast_format = [
'query',
'subject',
'query_length',
'subject_length',
'query_start',
'query_end',
'subject_start',
'subject_end',
'expect',
'identity',
]
fmt = generate_tabblast_format(blast_format)
if params is None:
params = {}
params['outfmt'] = fmt
blast_fhand = tempfile.NamedTemporaryFile(suffix='.blast')
do_blast(query_fhand.name,
blastdb,
program,
blast_fhand.name,
params,
remote=remote)
if remote:
blasts = BlastParser(blast_fhand)
else:
blasts = TabularBlastParser(blast_fhand, blast_format)
return blasts, blast_fhand
def test_no_filter(self):
'It test the blast parser'
blast_file = open(os.path.join(TEST_DATA_DIR, 'blast.xml'))
parser = BlastParser(fhand=blast_file)
match_summary = _summarize_matches(parser)
#lcl|2_0 cCL1Contig2
#lcl|3_0 cCL1Contig3
#lcl|4_0 cCL1Contig4
#lcl|5_0 cCL1Contig5
expected = {'cCL1Contig2': 3, 'cCL1Contig3': 1,
'cCL1Contig4': 5, 'cCL1Contig5': 8}
_check_match_summary(match_summary, expected)
def test_min_length_filter(self):
'We can keep the hits length above the given one'
blast_file = open(os.path.join(TEST_DATA_DIR, 'blast.xml'))
#with the min length given in base pairs
filters = [{'kind': 'min_length',
'min_num_residues': 500,
'length_in_query': True
}]
expected = {'cCL1Contig2': 3, 'cCL1Contig3': 0,
'cCL1Contig4': 1, 'cCL1Contig5': 1}
blasts = BlastParser(fhand=blast_file)
filtered_blasts = filter_alignments(blasts, config=filters)
match_summary = _summarize_matches(filtered_blasts)
_check_match_summary(match_summary, expected)
#with the min length given in query
filters = [{'kind': 'min_length',
'min_percentage': 70,
'length_in_query': True
}]
expected = {'cCL1Contig2': 0, 'cCL1Contig3': 0,
'cCL1Contig4': 2, 'cCL1Contig5': 0}
blasts = BlastParser(fhand=blast_file)
filtered_blasts = filter_alignments(blasts, config=filters)
match_summary = _summarize_matches(filtered_blasts)
#print match_summary
_check_match_summary(match_summary, expected)
#with the min length given in subject %
filters = [{'kind': 'min_length',
'min_percentage': 0.002,
'length_in_query': False
}]
expected = {'cCL1Contig2': 3, 'cCL1Contig3': 0,
'cCL1Contig4': 1, 'cCL1Contig5': 2}
blasts = BlastParser(fhand=blast_file)
filtered_blasts = filter_alignments(blasts, config=filters)
match_summary = _summarize_matches(filtered_blasts)
_check_match_summary(match_summary, expected)
def test_best_scores_filter(self):
'We can keep the hits with the bests expects'
blast_file = open(os.path.join(TEST_DATA_DIR, 'blast.xml'))
filters = [{'kind': 'best_scores',
'score_key': 'expect',
'max_score': 1e-4,
'score_tolerance': 10
}]
expected = {'cCL1Contig2': 2, 'cCL1Contig3': 1,
'cCL1Contig4': 1, 'cCL1Contig5': 2}
blasts = BlastParser(fhand=blast_file)
filtered_blasts = filter_alignments(blasts, config=filters)
match_summary = _summarize_matches(filtered_blasts)
_check_match_summary(match_summary, expected)
def test_blast_parser(self):
'It test the blast parser'
blast_file = open(os.path.join(TEST_DATA_DIR, 'blast.xml'))
parser = BlastParser(fhand=blast_file)
expected_results = [
{'query':{'name':'cCL1Contig2',
'description':"<unknown description>",
'length':1924},
'matches':[
{'subject':{'name':'chr18',
'description':'No definition line found',
'length':19691255},
'scores':{'expect':4.60533e-35},
'match_parts':[{'query_start':276, 'query_end':484,
'query_strand':-1,
'subject_start':477142,
'subject_end':477350,
'subject_strand':1,
'scores':{'expect': 4.60533e-35,
'similarity':84.2,
'identity': 84.2}
}],
}
]
},
{'query':{'name':'cCL1Contig3',
'description':"<unknown description>",
'length':629},
},
{}, {}
]
n_blasts = 0
for index, blast in enumerate(parser):
_check_blast(blast, expected_results[index])
n_blasts += 1
assert n_blasts == 4
#with the subject id given in the xml blast
expected_results = [
{'query':{'name':'cCL1Contig2',
'description':'<unknown description>',
'length':1924}}, {}, {}, {}]
parser = BlastParser(fhand=blast_file)
for index, blast in enumerate(parser):
_check_blast(blast, expected_results[index])
# Check using def as acceion in all the blasts
#It changes depending on the blast output format. depends on version
blast_file = open(os.path.join(TEST_DATA_DIR, 'melon_tair.xml'))
parser = BlastParser(fhand=blast_file)
assert parser.next()['matches'][0]['subject']['name'] == 'tair1'
def test_blast_no_result(self):
'It test that the xml output can be and empty string'
blast_file = NamedTemporaryFile()
blasts = BlastParser(fhand=blast_file)
filters = [{'kind': 'best_scores',
'score_key': 'expect',
'max_score': 1e-4,
'score_tolerance': 10
}]
filt_b = filter_alignments(blasts, config=filters,)
try:
filt_b.next()
self.fail()
except StopIteration:
pass