def map_kmers_to_read(self, sequence, read, candidate_gene_names):
self.logger.info("Map k-mers to read")
seq_length = len(sequence)
end = seq_length - self.k
kmers_obj = Kmers(sequence, self.k, self.homopolyer_compression)
read_kmers = kmers_obj.get_all_kmers(
max_kmer_count=self.max_kmer_count)
is_read_matching = False
sequence_hits, hit_counter, read_kmer_hits = self.put_kmers_in_read_bins(
seq_length, end, self.fasta_kmers, read_kmers)
blocks_obj = Blocks(self.k, self.min_block_size, self.max_gap,
self.margin)
block_start, block_end = blocks_obj.find_largest_block(sequence_hits)
block_start = blocks_obj.adjust_block_start(block_start)
block_end = blocks_obj.adjust_block_end(block_end, seq_length)
block_kmers = self.create_kmers_for_block(block_start, block_end,
read_kmer_hits)
is_read_matching = self.apply_kmers_to_genes(self.fasta_obj,
block_kmers,
candidate_gene_names)
if self.filtered_reads_file:
self.append_read_to_fastq_file(read, block_start, block_end)
return is_read_matching
def does_read_contain_quick_pass_kmers(self, sequence):
self.logger.info("Perform quick pass k-mer check on read")
seq_length = len(sequence)
if seq_length < self.min_block_size:
self.logger.info("Read below minimum size")
return {}
kmers_obj = Kmers(sequence, self.k, self.homopolyer_compression)
read_onex_kmers = kmers_obj.get_one_x_coverage_of_kmers()
intersect_read_fasta_kmers = self.fasta_obj.kmer_keys_set & set(
read_onex_kmers)
if len(intersect_read_fasta_kmers) > self.min_kmers_for_onex_pass:
gene_names = self.genes_containing_first_pass_kmers(
self.fasta_obj, intersect_read_fasta_kmers)
candidate_gene_names = {
k: v
for k, v in gene_names.items()
if v > self.min_kmers_for_onex_pass
}
return candidate_gene_names
return {}
def test_four_kmers_compression(self):
k = Kmers('GCAAAAATTTTTGC', 4, True)
self.assertEqual(k.get_all_kmers_counter(), {
'GCAT': 0,
'CATG': 0,
'ATGC': 0
})
def test_four_kmers(self):
k = Kmers('AAAAATTTTT', 4, False)
self.assertEqual(k.get_all_kmers_counter(), {
'AAAA': 0,
'AAAT': 0,
'AATT': 0,
'ATTT': 0,
'TTTT': 0
})
def sequence_kmers(self):
seq_counter = 0
kmer_to_sequences = {}
for record in SeqIO.parse(self.filename, "fasta"):
kmers = Kmers(str(record.seq), self.k, self.homopolyer_compression)
# We assume here that the sequence name is unique in the FASTA file
kmer_to_sequences[record.id] = kmers.get_all_kmers_counter(
max_kmer_count=self.max_kmer_count)
seq_counter += 1
return kmer_to_sequences
def test_short_sequence(self):
k = Kmers('A', 10, False)
self.assertEqual(k.get_all_kmers_counter(), {})
def test_four_kmers_compression_all_repeats(self):
k = Kmers('AAAAATTTTTGGGGGCCCCC', 4, True)
self.assertEqual(k.get_all_kmers_counter(), {'ATGC': 0})