miferreiro/D2-MCS

Drugs Discovery Multi-Clustering System

D2-MCS toolkit: Drugs Discover Multi-Clustering System

R framework to automatically predict the biological activity of chemical compounds regarding to theirs chemical substructures

Operation mode

Required libraries

On linux (Ubuntu)

• r-base
• r-base-core

On Windows

Download from CRAN Project: https://cran.r-project.org/bin/windows/

R required packages

		R Libraries
R6	car	plyr	dplyr	caret
mltools	here	parallel	RWeka	devtools
ModelMetrics	tools	graphics	gridExtra	grid
lattice	rJava	FSelector	BBmisc	ggalt
tictoc	ggrepel	ggplot2	reshape2

Execution example (get classification performance)

source("sources.R")
data <- Dataset$new( filepath ="<path_to_dataset_file>", header=TRUE, 
                     sep="\t",skip = 1, normalize.names=TRUE, classIndex = 1 )
data$executePartition(4)
  
trFunction <- TwoClass$new( method = "cv", number = 10, savePredictions = "final", 
                            classProbs = TRUE, allowParallel = TRUE, verboseIter = FALSE)
  
fisherC <- BinaryFisherClustering$new(dataset = data$getSubset(1,2), maxClusters = 50)
fisherC$execute(positive.class = "<value_positive_class>")
train.subset <- fisherC$createSubset( subset = data$getSubset(c(2,3)) )
test.subset <- data$getSubset(4)
  
Benchmarking <- D2MCS$new( path = "models/BinaryFisherCluster", trainFunction = trFunction )
Benchmarking$train( train.set = train.subset, metric = "MCC" )
classify <- Benchmarking$classify( test.set = test.subset), voting.scheme = ClassWeightedVoting$new(), 
                                   positive.class = "<value_positive_class>" )
classify$computePerformance( ob = test.subset$getClass(), list(MCC$new(), PPV$new(), Accuracy$new()) )
  

Output visualization example

	Measure	Value
1	MCC	0.7858148
2	PPV	0.8864542
3	Accuracy	0.8928571

Execution example (execute NSGAII and SMSEMOA optimization algorithms)

optimization <- Benchmarking$optimize( opt.set = test.subset, 
                                       voting.scheme = ClassWeightedVoting$new(), 
                                       opt.algorithm = list(NSGAII$new( min.function  = FPFN$new(), 
                                                                        n.generations = 25000,
                                                                        n.iteractions = 1 ),
                                                            SMSEMOA$new( min.function = FPFN$new(), 
                                                                         n.generations = 25000,
                                                                         n.iteractions = 1 ) ), 
                                       positive.class = "<value_positive_class>" )

Developers

David Ruano Ordás

Citation

To cite D2-MCS please use:

Ruano-Ordás, D; Yevseyeva, I, Basto-Fernandes,V; Méndez, José R; Emmerichc, Michael T.M. (2018). Improving the drug discovery process by using multiple classifier systems Expert Systems with Applications. Volume 121, pp. 292-303. Elsevier. https://doi.org/10.1016/j.eswa.2018.12.032