def binary_classification_dataset_setup(iterable_seq=None, negative_shuffle_ratio=None, shuffle_order=None):
iter1, iter2 = tee(iterable_seq)
iterable_graph = rnafold_to_eden(iter1)
iter3 = seq_to_seq(iter2, modifier=shuffle_modifier,
times=negative_shuffle_ratio, order=shuffle_order)
iterable_graph_neg = rnafold_to_eden(iter3)
return iterable_graph, iterable_graph_neg
def transform(self, seqs=None, mfe=False):
if mfe is False:
graphs = rnashapes_to_eden(seqs,
shape_type=self.shape_type,
energy_range=self.energy_range,
max_num=self.max_num,
split_components=self.split_components)
else:
graphs = rnafold_to_eden(seqs)
return graphs
def transform(self, seqs=None, mfe=False):
if mfe is False:
graphs = rnashapes_to_eden(seqs,
shape_type=self.shape_type,
energy_range=self.energy_range,
max_num=self.max_num,
split_components=self.split_components)
else:
graphs = rnafold_to_eden(seqs)
return graphs
def rna_refold(self, digraph=None, seq=None, vectorizer=None):
"""
:param digraph:
:param seq:
:return: will extract a sequence, RNAfold it and create a abstract graph
"""
# get a sequence no matter what :)
if not seq:
seq = get_sequence(digraph)
# print 'seq:',seq
graph = rnafold_to_eden([("emptyheader", seq)], shape_type=5, energy_range=30, max_num=3).next()
expanded_graph = self.vectorizer._edge_to_vertex_transform(graph)
ex_di_graph = graphlearn.minor.rnaabstract.expanded_rna_graph_to_digraph(expanded_graph)
ex_di_graph.graph["sequence"] = seq
# abstract_graph = directedgraphtools.direct_abstraction_wrapper(graph,0)
return ex_di_graph
def get_graphs(rfam_id = '../example/RF00005',size=9999):
seqs = fasta_to_sequence(rfam_uri(rfam_id))
graphs = islice( clean(rnafold_to_eden(seqs, shape_type=5, energy_range=30, max_num=3)), size)
return graphs
def get_graphs(rfam_id='../example/RF00005', size=9999):
seqs = fasta_to_sequence(rfam_uri(rfam_id))
graphs = islice(
clean(rnafold_to_eden(seqs, shape_type=5, energy_range=30, max_num=3)),
size)
return graphs
def pre_processor_rnafold(seqs):
graphs = rnafold_to_eden(seqs)
#from eden.modifier.graph import structure
#graphs = structure.basepair_to_nesting(graphs)
return graphs
