def initialize_distance(sequences):
n_seqs = len(sequences)
D_dtw = np.zeros((n_seqs, n_seqs))
D_sw = np.zeros((n_seqs, n_seqs))
scrappie_dfs = [None for _ in range(n_seqs)]
squiggles = [None for _ in range(n_seqs)]
for j in range(n_seqs):
if scrappie_dfs[j] is None:
scrappie_dfs[j] = simulate_squiggle(sequences[j])
squiggles[j] = list(scrappie_dfs[j]["current"])
for i in range(n_seqs):
if scrappie_dfs[i] is None:
scrappie_dfs[i] = simulate_squiggle(sequences[i])
squiggles[i] = list(scrappie_dfs[i]["current"])
if i > j:
continue
else:
dtw_dist = calc_dtw(squiggles[i], squiggles[j])
D_dtw[i, j] = dtw_dist
D_dtw[j, i] = dtw_dist
_, sw_dist, _, _, _ = s_w(sequences[j],
sequences[i],
cost_fn={
"match": 1,
"mismatch": -1,
"gap": -8
})
D_sw[i, j] = sw_dist
D_sw[j, i] = sw_dist
return D_dtw, D_sw, scrappie_dfs
def calculate_sw(sequences):
D_sw = np.zeros((96, 96))
for j in range(96):
for i in range(96):
if i >= j:
continue
else:
_, sw, a, _, _ = s_w(sequences[j], sequences[i], cost_fn={"match": 1, "mismatch": -1, "gap": -8})
D_sw[i, j] = sw
D_sw[j, i] = sw
return D_sw
def helper_run_sw(read, molbit_seqs=None, search_len=100,
cost_fn={"match": 2, "mismatch": -2, "gap": -1}):
read_id, seq = read
sw_search_end = min(len(seq), search_len)
search_seq = seq[:sw_search_end]
results = [read_id]
for header_i, header_seq in enumerate(molbit_seqs):
_, sw_score, _, sw_loc, _ = s_w.s_w(search_seq, header_seq,
cost_fn=cost_fn)
results.append(sw_score)
return results
