def test_kmer_generator(self):
self.assertEqual(kmer_generator("", 0), {})
self.assertEqual(kmer_generator("A", 1), {0: ['A', 'T']})
self.assertEqual(
kmer_generator(["ATTCG", "GGGG"], 3), {
0: [
'ATT', 'TTC', 'TCG', 'CGA', 'GAT', 'ATC', 'TCG', 'CGA',
'GAA', 'AAT'
],
1: ['GGG', 'GGG', 'GGG', 'GGG', 'CCC', 'CCC', 'CCC', 'CCC']
})
self.assertEqual(
kmer_generator(["ATTCG", "GGGG"], 3, extension=False), {
0: ['ATT', 'TTC', 'TCG', 'CGA', 'GAA', 'AAT'],
1: ['GGG', 'GGG', 'CCC', 'CCC']
})
def compute_match_score(query, template_obj, kmerLength, primers):
'''
Inputs:
- (str) data_prefix: the prefix of the NGS dataset paths
'''
record1, record2, gzip1, gzip2, filename = get_reads_records(query)
#Run reads through Coarse Filtering to drastically reduce computation for Fine Filtering
reads = combine_records(record1,record2)
flanking_sequences = set(kmer for kmers in kmer_generator([sequence for primer_set in primers.values() for sequence in primer_set], kmerLength, extension=False).values() for kmer in kmers)
nucleotides_seen,recruitedReads = coarse_filtering(reads, kmerLength, template_obj["Kmers"], flanking_sequences)
# Close all the things
record1.close()
if record2 is not None:
record2.close()
if gzip1 is not None:
gzip1.close()
if gzip2 is not None:
gzip2.close()
coverage = nucleotides_seen/1000000.0 #assuming all genomes are roughly the same length - nucleotides seen should be proportional to coverage
#Run reads through Fine Filtering to get score for each template
matchScore, flanking_coverage = fine_filtering(template_obj, recruitedReads, kmerLength, primers)
print(matchScore)
print(flanking_coverage)
matchScore = [score/float(1+flanking_coverage[i]) for i,score in enumerate(matchScore)]
print(matchScore)
print("\n")
return matchScore, filename
def setUp(self):
self.PATH_PREFIX = PATH_PREFIX
templates = list(SeqIO.parse(resource_filename('prince.resources', "templates.fasta"), "fasta"))
templateNames = [t.id for t in templates]
templates = [str(t.seq) for t in templates]
#Generate k-mers
templateKmers = kmer_generator(templates, 25)
self.template_obj = {"Names":templateNames, "Sequences":templates, "Kmers":templateKmers}
with open(resource_filename('prince.resources', "TB_primers_extended.json")) as primers:
self.primers=json.load(primers)