def build_initial_best_motifs():
length = len(fst(dna_strings))
def random_mapper(dna_string):
i = randrange(length - k + 1)
return dna_string[i: i + k]
return list(map(random_mapper, dna_strings))
def detect(filename):
langs, kws = cfg.read_kws(relative_path(KEYWORDS_DIR, __file__))
inputs, _ = inputs_dict(filename, kws, langs)
model_coeffs = cfg.read_config(relative_path(MODEL_CONFIG, __file__))
responses_dict = compute_neurons(inputs, model_coeffs)
print(responses_dict)
min_response = max(responses_dict.iteritems(), key=snd)
return fst(min_response).capitalize()
def matcher(token_tuple):
token_type = fst(token_tuple)
token_value = snd(token_tuple)
should_be_discarded = not token_value or \
token_type == NUMBER or \
token_type == STRING or \
token_type == N_TOKENS or \
any(token_value.startswith(quot) and token_value.endswith(quot) for quot in QUOTES)
return should_be_discarded
def clean_tokens(readline):
def matcher(token_tuple):
token_type = fst(token_tuple)
token_value = snd(token_tuple)
should_be_discarded = not token_value or \
token_type == NUMBER or \
token_type == STRING or \
token_type == N_TOKENS or \
any(token_value.startswith(quot) and token_value.endswith(quot) for quot in QUOTES)
return should_be_discarded
return map(lambda t: (fst(t), snd(t)), ifilterfalse(matcher, generate_tokens(readline)))
def greedy_motif_search(dna_strings, k, profile_builder=build_profile_columns,
initial_motifs_builder=first_kmer_builder):
best_motifs = initial_motifs_builder(dna_strings, k)
for kmer in all_possible_kmers_from_text(fst(dna_strings), k):
current_score = matrix_score(best_motifs)
motifs = [kmer]
for dna_string in tail(dna_strings):
profile = profile_builder(motifs)
additional_motif = most_probable_kmer(dna_string, k, profile)
motifs.append(additional_motif)
if current_score > matrix_score(motifs):
best_motifs = motifs
return best_motifs
def column_consensus(column):
return fst(top_by_snd_and_ignore(((nucleotide, column.count(nucleotide))
for nucleotide in column), descending=True))
def mapper(read):
return '({0}|{1})'.format(fst(read), snd(read))
def most_probable_kmer(dna_string, k, probabilities):
all_kmers = all_possible_kmers_from_text(dna_string, k)
return fst(top_by_snd_and_ignore(list((kmer, probability_score(kmer, probabilities))
for kmer in all_kmers), descending=True))
def space_paired_read(read, d):
return fst(read) + ' ' * d + snd(read)
def space_paired_read(read, d):
return fst(read) + ' ' * d + snd(read)