# Create a distance matrix file
progress.StartTask("Computing dissimilarity matrix")
SVMlFile = "DissimilarityMatrix" + conf.SampleSize
# Calculate distance and identify redundant structures within the same condition
SF.DistanceStruct(os.path.join(OutputSamples, 'Samples.txt'), SVMlFile,
int(conf.SampleSize), ProbingConditions)
progress.EndTask()
################################# Calculate Conditional Boltzmann probabilities
# for each condition, calculate Z over all non redundant structures and return a conditional Boltzmann probability for all structures with null value for redundant ones.
progress.StartTask("Computing Boltzmann probabilities")
BoltzmannFactor = defaultdict(lambda: defaultdict())
ConditionalBoltzmannProbability = defaultdict(lambda: defaultdict())
Zprobabilities = defaultdict(lambda: defaultdict())
Redondantestructure = FF.UnpickleVariable(
"Redondantestructures_Id.pkl")
ConditionalBoltzmannProbability = SF.Boltzmann_Calc(
ProbingConditions, OutputSamples, int(conf.SampleSize),
RNASequence, Redondantestructure)
progress.EndTask()
################################# Clustering of structures based on their base pair distance
progress.StartTask("Iterative clustering")
# Load the pickled dissimilarity matrix
DM = FF.UnpickleVariable("dissmatrix.pkl")
# Get the list of redundant structures
Redundant = []
Redundant = FF.UnpickleVariable("Redondantestructures.pkl")
BoltzmannFactor = FF.UnpickleVariable("Boltzman.pkl")
method = "MiniBatchKMean"
Clusters, CentroidStructure = OC.DefineNumberCluster(
print("Distance Matrix generation for % d Structures started " %
(int(conf.numberofsruct) * len(listfiles) + MFESnbrstruct))
Redundant = []
Newnumberofsruct = defaultdict(lambda: defaultdict())
#calculate distance
SF.DistanceStruct(os.path.join(OutputSamples, 'Samples.txt'), SVMlFile,
int(conf.numberofsruct), MFESnbrstruct, listfiles)
endtime = time.time()
print(
"End of distance calculation between the structures in the sample %53f\t"
% (endtime - startime))
#!!!!!!!!!!!!!Clustering!!!!!!!!!!!
# load the pickled dissimilarity matrix
DM = FF.UnpickleVariable(os.path.join(conf.PickledData, "dissmatrix.pkl"))
Redundant = FF.UnpickleVariable(
os.path.join(conf.PickledData, "Redondantestructures.pkl"))
Redundant_Id = FF.UnpickleVariable(
os.path.join(conf.PickledData, "Redondantestructures_Id.pkl"))
Newnumberofsruct = FF.UnpickleVariable(
os.path.join(conf.PickledData, "Dicnumberofsruct.pkl"))
# clustering with DIANA Algorithm
if conf.Diana == "true":
startime = time.time()
Clusters = defaultdict(list)
structs = [i + 1 for i in range(len(DM))]
clusters = CL.DIANA.doClustering(DM, structs,
conf.maxDiameterThreshold,
conf.maxAverageDiameterThreshold)
#!!!!!!!!!!!!! Distance Matrix calculation !!!!!!!!!!!
#
startime = time.time()
print("Distance Matrix generation for % d Structures started "%(int(conf.numberofsruct)*len(conf.constraintes)+MFESnbrstruct))
Redundant=[]
Newnumberofsruct=defaultdict(lambda: defaultdict())
#calculate distance
SF.DistanceStruct(os.path.join(OutputSamples,'Samples.txt'),SVMlFile,int(conf.numberofsruct),MFESnbrstruct,listfiles)
endtime=time.time()
print("End of distance calculation between the structures in the sample %53f\t"%(endtime-startime))
#!!!!!!!!!!!!!Clustering!!!!!!!!!!!
# load the pickled dissimilarity matrix
DM=FF.UnpickleVariable(os.path.join(conf.PickledData,"dissmatrix.pkl"))
Redundant=FF.UnpickleVariable(os.path.join(conf.PickledData,"Redondantestructures.pkl"))
Redundant_Id=FF.UnpickleVariable(os.path.join(conf.PickledData,"Redondantestructures_Id.pkl"))
Newnumberofsruct=FF.UnpickleVariable(os.path.join(conf.PickledData,"Dicnumberofsruct.pkl"))
# clustering with DIANA Algorithm
if conf.Diana=="true":
startime = time.time()
Clusters = defaultdict(list)
structs = [i + 1 for i in range(len(DM))]
clusters = CL.DIANA.doClustering(DM, structs,conf.maxDiameterThreshold, conf.maxAverageDiameterThreshold)
for i in range(len(clusters)):
Clusters[i] = clusters[i]
endtime = time.time()
print ("Clusters using Diana algorithm: %s %53f\t"%(Clusters,endtime-startime))
