def do(options):
outputFileName_core = util.addArguments("EmptyBackground", [
util.ArgumentToAdd(options.seqLength, "seqLength"),
util.ArgumentToAdd(options.numSeqs, "numSeqs")
])
embedInBackground = sn.EmbedInABackground(
backgroundGenerator=sn.ZeroOrderBackgroundGenerator(
seqLength=options.seqLength),
embedders=[])
sequenceSet = sn.GenerateSequenceNTimes(embedInBackground, options.numSeqs)
sn.printSequences(outputFileName_core + ".simdata",
sequenceSet,
includeFasta=True,
includeEmbeddings=True)
def do(options):
if (options.seed is not None):
import numpy as np
np.random.seed(options.seed)
import random
random.seed(options.seed)
outputFileName_core = util.addArguments("DensityEmbedding", [
util.ArgumentToAdd(options.prefix, "prefix"),
util.BooleanArgument(options.bestHit, "bestHit"),
util.ArrArgument(options.motifNames, "motifs"),
util.ArgumentToAdd(options.min_motifs, "min"),
util.ArgumentToAdd(options.max_motifs, "max"),
util.ArgumentToAdd(options.mean_motifs, "mean"),
util.FloatArgument(options.zero_prob, "zeroProb"),
util.ArgumentToAdd(options.seqLength, "seqLength"),
util.ArgumentToAdd(options.numSeqs, "numSeqs")
])
loadedMotifs = synthetic.LoadedEncodeMotifs(options.pathToMotifs,
pseudocountProb=0.001)
Constructor = synthetic.BestHitPwmFromLoadedMotifs if options.bestHit else synthetic.PwmSamplerFromLoadedMotifs
embedInBackground = synthetic.EmbedInABackground(
backgroundGenerator=synthetic.ZeroOrderBackgroundGenerator(
seqLength=options.seqLength),
embedders=[
synthetic.RepeatedEmbedder(
synthetic.SubstringEmbedder(
synthetic.ReverseComplementWrapper(
substringGenerator=Constructor(
loadedMotifs=loadedMotifs, motifName=motifName),
reverseComplementProb=options.rc_prob),
positionGenerator=synthetic.UniformPositionGenerator()),
quantityGenerator=synthetic.ZeroInflater(
synthetic.MinMaxWrapper(synthetic.PoissonQuantityGenerator(
options.mean_motifs),
theMax=options.max_motifs,
theMin=options.min_motifs),
zeroProb=options.zero_prob))
for motifName in options.motifNames
])
sequenceSet = synthetic.GenerateSequenceNTimes(embedInBackground,
options.numSeqs)
synthetic.printSequences(outputFileName_core + ".simdata",
sequenceSet,
includeFasta=True,
includeEmbeddings=True,
prefix=options.prefix)