def sequence_variant(self, sequence, mutation, strand="+"):
"""
Return the possible mutation(s) in the sequence due to
the sequence. For now, return the original sequence.
# TODO: probably not return the original sequence
"""
# first step: get positive string, standardizing so it's
# easier for me to think about
if sequence.strand == "-":
seq = sequence.reverse_complement()
else:
seq = sequence
if mutation["observed"] == "lengthTooLong":
# raise UnknownMutationError, "Could not read the variants for %s SNP %s at chr%s:%s-%s" % (mutation['class'], mutation['name'], mutation['chrom'], mutation['chromStart'], mutation['chromEnd'])
sequences = []
dup = self.__class__.duplicate(sequence)
dup2 = self.__class__.duplicate(sequence, name=mutation["name"])
xform = TransformedGenomeSequence.replace(
dup2,
mutation["chromStart"],
mutation["chromEnd"],
"?" * (mutation["chromEnd"] - mutation["chromStart"] + 1),
)
sequences.append(dup)
sequences.append(xform)
else:
klass = mutation["class"].lower().replace("-", "")
sequences = getattr(self.__class__, "_%s_sequence_variant" % klass, self.__class__._unknown_variant)(
seq, mutation
)
if strand == "-":
return [reverse_complement(s) for s in sequences]
else:
return sequences
:param transformer: The MutationTransformer that takes in the original sequence
and applies the SNP mutations. Mutation structures will vary
with each transformer.
"""
xf = transformer
e = None
try:
sequences = xf.sequence_variants(original, mutations, combination_width, strand)
except ReturnWithCaveats, e:
rc = e
sequences = e.return_value
sequences = [seq for seq in sequences if seq.positive_strand_sequence != original.positive_strand_sequence]
if original.strand == strand:
sequences.insert(0, TransformedGenomeSequence.from_sequence(original))
else:
sequences.insert(0, TransformedGenomeSequence.from_sequence(original.reverse_complement()))
if e:
raise ReturnWithCaveats(e.explanations, sequences)
else:
return sequences
def find_in_sequence(fragment, sequence):
regex = base_regexp_expand(fragment, overlap=True)
return __find_in_sequence_regex(regex, sequence, overlapped=True)
def __find_in_sequence_regex(regex, sequence, overlapped=True):
seq_matches = []
def duplicate(cls, sequence, name=None):
return TransformedGenomeSequence.from_sequence(sequence, name=name)
