def read_simdata_file(simdata_file, one_hot_encode=False, ids_to_load=None):
ids = []
sequences = []
embeddings = []
labels = []
if (ids_to_load is not None):
ids_to_load = set(ids_to_load)
def action(inp, line_number):
if (line_number > 1):
if (ids_to_load is None or (inp[0] in ids_to_load)):
ids.append(inp[0])
sequences.append(inp[1])
embeddings.append(getEmbeddingsFromString(inp[2]))
labels.append([int(x) for x in inp[3:]])
util.perform_action_on_each_line_of_file(
file_handle=util.get_file_handle(simdata_file),
action=action,
transformation=util.default_tab_seppd)
return util.enum(
ids=ids,
sequences=sequences,
embeddings=embeddings,
labels=np.array(labels))
#!/usr/bin/env python
import os
import sys
import simdna
import simdna.util as util
import simdna.synthetic as synthetic
import simdna.pwm as pwm
generationSettings = util.enum(
allBackground="allBackground",
singleMotif1="singleMotif1" #embeds first motif
,
singleMotif2="singleMotif2" #embeds second motif
,
twoMotifs="twoMotifs" #embeds one of both motifs
,
twoMotifsFixedSpacing=
"twoMotifsFixedSpacing" #embeds both motifs with a fixed spacing
,
twoMotifsVariableSpacing=
"twoMotifsVariableSpacing" #embeds both motifs with a variable spacing
)
def motifGrammarSimulation(options):
pc = 0.001
bestHit = options.bestHit
pathToMotifs = options.pathToMotifs
loadedMotifs = synthetic.LoadedEncodeMotifs(pathToMotifs,
pseudocountProb=pc)
motifName1 = options.motifName1