| title | author | date | output |
|---|---|---|---|
Large-scale structure-activity relationship of host defense peptides |
Saw Simeon, Hao Li, Theeraphon Piacham, Virapong Prachayasittikul and Chanin Nantasanamat |
October 13, 2015 |
html_document |
###HDPs having their respective activity were merge with negative data set to build seperate models
## $cancer
## Training set Internal validation External set
## ACC_Mean 99.2613 97.5583 97.6046
## ACC_SD 0.1433 0.3102 0.6682
## Sens_Mean 99.3600 98.3807 98.3492
## Sens_SD 0.1644 0.2511 0.6887
## Spec_Mean 98.8989 94.5317 94.9041
## Spec_SD 0.5444 0.9232 2.1142
## MCC_Mean 0.9780 0.9274 0.9287
## MCC_SD 0.0043 0.0092 0.0201
##
## $fungus
## Training set Internal validation External set
## ACC_Mean 99.1317 96.9126 97.1210
## ACC_SD 0.1578 0.2887 0.5901
## Sens_Mean 99.3237 96.8298 96.9850
## Sens_SD 0.1842 0.3971 0.9161
## Spec_Mean 98.9841 96.9786 97.2420
## Spec_SD 0.2723 0.3417 0.8210
## MCC_Mean 0.9824 0.9373 0.9416
## MCC_SD 0.0032 0.0059 0.0120
##
## $bacteria
## Training set Internal validation External set
## ACC_Mean 99.1115 96.9422 97.0605
## ACC_SD 0.1624 0.2981 0.6091
## Sens_Mean 99.3654 96.9297 97.1218
## Sens_SD 0.1644 0.4589 1.0143
## Spec_Mean 98.9161 96.9535 97.0302
## Spec_SD 0.2763 0.2843 0.8232
## MCC_Mean 0.9820 0.9379 0.9404
## MCC_SD 0.0033 0.0061 0.0124
##
## $virus
## Training set Internal validation External set
## ACC_Mean 99.1646 96.0039 95.9189
## ACC_SD 0.1988 0.4324 0.8257
## Sens_Mean 98.4089 91.5197 91.3986
## Sens_SD 0.4843 1.0566 2.3605
## Spec_Mean 99.3926 97.3520 97.2890
## Spec_SD 0.2455 0.3853 0.9133
## MCC_Mean 0.9765 0.8876 0.8848
## MCC_SD 0.0056 0.0123 0.0238
###HDPs having anti--bacteria, anti--cancer, anti--fungus and virus-- were combined into one to represent HDPs while the negative set was used as non--HDPs.###
## Training set Internal validation External set
## ACC_Mean 99.3873 98.1257 98.1922
## ACC_SD 0.0763 0.1237 0.2867
## Sens_Mean 99.4387 98.5458 98.5944
## Sens_SD 0.0878 0.1001 0.2640
## Spec_Mean 99.1291 95.9758 96.1536
## Spec_SD 0.1609 0.5045 1.1747
## MCC_Mean 0.9781 0.9325 0.9350
## MCC_SD 0.0027 0.0045 0.0104
###HDPs having their respective activity were combined together to build a union predictive model###
Results for training
## Bacteria Cancer Fungus Virus Overall
## ACC_Mean 96.0307 96.0505 96.0500 96.0401 96.042825
## ACC_SD 0.2174 0.1845 0.2185 0.2011 0.205375
## Sens_Mean 96.1914 96.2048 96.1826 96.1909 96.192425
## Sens_SD 0.2364 0.2113 0.2100 0.2265 0.221050
## Spec_Mean 93.3186 93.4566 93.7839 93.5015 93.515150
## Spec_SD 1.5234 1.5337 1.5985 1.5726 1.557050
## MCC_Mean 0.7225 0.7241 0.7239 0.7234 0.723475
## MCC_SD 0.0169 0.0145 0.0170 0.0158 0.016050
Results for CV
## Bacteria Cancer Fungus Virus Overall
## ACC_Mean 92.9778 93.0262 93.0751 93.0796 93.039675
## ACC_SD 0.2731 0.2657 0.2426 0.2499 0.257825
## Sens_Mean 97.9807 98.0357 98.0257 98.0608 98.025725
## Sens_SD 0.2337 0.2109 0.2006 0.2191 0.216075
## Spec_Mean 42.9199 42.8544 43.5194 43.2379 43.132900
## Spec_SD 1.8937 2.0311 1.8305 1.9767 1.933000
## MCC_Mean 0.5057 0.5078 0.5130 0.5124 0.509725
## MCC_SD 0.0201 0.0202 0.0181 0.0189 0.019325
Results for Testing
## Bacteria Cancer Fungus Virus Overall
## ACC_Mean 93.3841 93.3183 93.3104 93.2103 93.305775
## ACC_SD 0.5439 0.5368 0.5752 0.5747 0.557650
## Sens_Mean 94.6636 94.7143 94.5986 94.5303 94.626700
## Sens_SD 0.4644 0.3942 0.3781 0.4320 0.417175
## Spec_Mean 71.5641 70.7074 71.3498 70.4909 71.028050
## Spec_SD 5.1636 6.2726 6.3225 6.0771 5.958950
## MCC_Mean 0.5263 0.5260 0.5206 0.5114 0.521075
## MCC_SD 0.0452 0.0397 0.0418 0.0449 0.042900
###Function and Scriptrs for Analysis### #####HDPs having their respective activity were merge with negative data set to build seperate models#####
### read FASTA
library(protr)
cancer <- readFASTA("cancer.fasta")
fungus <- readFASTA("fungus.fasta")
bacteria <- readFASTA("bacteria.fasta")
virus <- readFASTA("virus.fasta")
negative <- readFASTA("negative_Protein.fasta")
### removed wired protein
cancer <- cancer[(sapply(cancer, protcheck))]
fungus <- fungus[(sapply(fungus, protcheck))]
bacteria <- bacteria[(sapply(bacteria, protcheck))]
virus <- virus[(sapply(virus, protcheck))]
negative <- negative[(sapply(negative, protcheck))]
### Descriptors generation
cancer_des <- t(sapply(cancer, extractAAC))
fungus_des <- t(sapply(fungus, extractAAC))
bacteria_des <- t(sapply(bacteria, extractAAC))
virus_des <- t(sapply(virus, extractAAC))
negative_des <- t(sapply(negative, extractAAC))
#### label the labels
cancer_des <- as.data.frame(cancer_des)
cancer_des$Label <- "Cancer"
fungus_des <- as.data.frame(fungus_des)
fungus_des$Label <- "Fungus"
bacteria_des <- as.data.frame(bacteria_des)
bacteria_des$Label <- "Bacteria"
virus_des <- as.data.frame(virus_des)
virus_des$Label <- "Virus"
negative_des <- as.data.frame(negative_des)
negative_des$Label <- "Negative"
cancer <- rbind(cancer_des, negative_des)
cancer$Label <- as.factor(cancer$Label)
fungus <- rbind(fungus_des, negative_des)
fungus$Label <- as.factor(fungus$Label)
bacteria <- rbind(fungus_des, negative_des)
bacteria$Label <- as.factor(bacteria$Label)
virus <- rbind(virus_des, negative_des)
virus$Label <- as.factor(virus$Label)
input <- list(cancer = cancer, fungus = fungus, bacteria = bacteria,
virus = virus)
#### training results using J48
J48_training <- function(x) {
results <- list(100)
for (i in 1:100) {
in_train <- caret::createDataPartition(x$Label, p = 0.80, list = FALSE)
train <- x[in_train, ]
test <- x[-in_train, ]
model_train <- RWeka::J48(Label~., data = train)
summary <- summary(model_train)
confusionmatrix <- summary$confusionMatrix
results[[i]] <- as.numeric(confusionmatrix)
}
return(results)
}
mean_and_sd <- function(x) {
c(round(mean(x, na.rm = TRUE), digits = 4),
round(sd(x, na.rm = TRUE), digits = 4))
}
J48_train <- function(x) {
ok <- J48_training(x)
results <- data.frame(ok)
data <- data.frame(results)
m = ncol(data)
ACC <- matrix(nrow = m, ncol = 1)
SENS <- matrix(nrow = m, ncol = 1)
SPEC <-matrix(nrow = m, ncol = 1)
MCC <- matrix(nrow = m, ncol = 1)
for(i in 1:m){
ACC[i,1] = (data[1,i]+data[4,i])/(data[1,i]+data[2,i]+data[3,i]+data[4,i])*100
SENS[i,1] = (data[4,i])/(data[3,i]+data[4,i])*100
SPEC[i,1] = (data[1,i]/(data[1,i]+data[2,i]))*100
MCC1 = (data[1,i]*data[4,i]) - (data[2,i]*data[3,i])
MCC2 = (data[4,i]+data[2,i])*(data[4,i]+data[3,i])
MCC3 = (data[1,i]+data[2,i])*(data[1,i]+data[3,i])
MCC4 = sqrt(MCC2)*sqrt(MCC3)
MCC[i,1] = MCC1/MCC4
}
results_ACC <- mean_and_sd(ACC)
results_SENS <- mean_and_sd(SENS)
results_SPEC <- mean_and_sd(SPEC)
results_MCC <- mean_and_sd(MCC)
results_all <- (data.frame(c(results_ACC, results_SENS, results_SPEC, results_MCC)))
rownames(results_all) <- c("ACC_Mean", "ACC_SD", "Sens_Mean", "Sens_SD", "Spec_Mean", "Spec_SD",
"MCC_Mean", "MCC_SD")
return(results_all)
}
#### 10-fold results using J48
J48_10fold <- function(x) {
results <- list(100)
for (i in 1:100) {
in_train <- caret::createDataPartition(x$Label, p = 0.80, list = FALSE)
train <- x[in_train, ]
test <- x[-in_train, ]
model_train <- RWeka::J48(Label~., data = train)
eval_j48 <- RWeka::evaluate_Weka_classifier(model_train, numFolds = 10, complexity = FALSE, seed = 1, class = TRUE)
confusionmatrix <- eval_j48$confusionMatrix
results[[i]] <- as.numeric(confusionmatrix)
}
return(results)
}
J48_cross_validation <- function(x) {
ok <- J48_10fold(x)
results <- data.frame(ok)
data <- data.frame(results)
m = ncol(data)
ACC <- matrix(nrow = m, ncol = 1)
SENS <- matrix(nrow = m, ncol = 1)
SPEC <-matrix(nrow = m, ncol = 1)
MCC <- matrix(nrow = m, ncol = 1)
for(i in 1:m){
ACC[i,1] = (data[1,i]+data[4,i])/(data[1,i]+data[2,i]+data[3,i]+data[4,i])*100
SENS[i,1] = (data[4,i])/(data[3,i]+data[4,i])*100
SPEC[i,1] = (data[1,i]/(data[1,i]+data[2,i]))*100
MCC1 = (data[1,i]*data[4,i]) - (data[2,i]*data[3,i])
MCC2 = (data[4,i]+data[2,i])*(data[4,i]+data[3,i])
MCC3 = (data[1,i]+data[2,i])*(data[1,i]+data[3,i])
MCC4 = sqrt(MCC2)*sqrt(MCC3)
MCC[i,1] = MCC1/MCC4
}
results_ACC <- mean_and_sd(ACC)
results_SENS <- mean_and_sd(SENS)
results_SPEC <- mean_and_sd(SPEC)
results_MCC <- mean_and_sd(MCC)
results_all <- (data.frame(c(results_ACC, results_SENS, results_SPEC, results_MCC)))
rownames(results_all) <- c("ACC_Mean", "ACC_SD", "Sens_Mean", "Sens_SD", "Spec_Mean", "Spec_SD",
"MCC_Mean", "MCC_SD")
return(results_all)
}
### testing results using J48
J48_testing <- function(x) {
results <- list(100)
for (i in 1:100) {
in_train <- caret::createDataPartition(x$Label, p = 0.80, list = FALSE)
train <- x[in_train, ]
test <- x[-in_train, ]
model_train <- RWeka::J48(Label~., data = train)
eval_external <- RWeka::evaluate_Weka_classifier(model_train, newdata = test, numFolds = 0, complexity = FALSE, seed = 1, class = TRUE)
confusionmatrix <- eval_external$confusionMatrix
results[[i]] <- as.numeric(confusionmatrix)
}
return(results)
}
J48_external <- function(x) {
ok <- J48_testing(x)
results <- data.frame(ok)
data <- data.frame(results)
m = ncol(data)
ACC <- matrix(nrow = m, ncol = 1)
SENS <- matrix(nrow = m, ncol = 1)
SPEC <-matrix(nrow = m, ncol = 1)
MCC <- matrix(nrow = m, ncol = 1)
for(i in 1:m){
ACC[i,1] = (data[1,i]+data[4,i])/(data[1,i]+data[2,i]+data[3,i]+data[4,i])*100
SENS[i,1] = (data[4,i])/(data[3,i]+data[4,i])*100
SPEC[i,1] = (data[1,i]/(data[1,i]+data[2,i]))*100
MCC1 = (data[1,i]*data[4,i]) - (data[2,i]*data[3,i])
MCC2 = (data[4,i]+data[2,i])*(data[4,i]+data[3,i])
MCC3 = (data[1,i]+data[2,i])*(data[1,i]+data[3,i])
MCC4 = sqrt(MCC2)*sqrt(MCC3)
MCC[i,1] = MCC1/MCC4
}
results_ACC <- mean_and_sd(ACC)
results_SENS <- mean_and_sd(SENS)
results_SPEC <- mean_and_sd(SPEC)
results_MCC <- mean_and_sd(MCC)
results_all <- (data.frame(c(results_ACC, results_SENS, results_SPEC, results_MCC)))
rownames(results_all) <- c("ACC_Mean", "ACC_SD", "Sens_Mean", "Sens_SD", "Spec_Mean", "Spec_SD",
"MCC_Mean", "MCC_SD")
return(results_all)
}
results_J48 <- function(x) {
training <- J48_train(x)
cross_validation <- J48_cross_validation(x)
testing <- J48_external(x)
results <- data.frame(Training = training, Cross_Validation = cross_validation, Testing = testing)
colnames(results) <- c("Training set", "Internal validation", "External set")
return(results)
}
suppressPackageStartupMessages(library(parallel))
suppressPackageStartupMessages(library(doSNOW))
cl <- makeCluster(8)
registerDoSNOW(cl)
clusterExport(cl = cl, ls())
result_J48_performance <- parLapply(cl = cl, input, function(x) {
models <- suppressWarnings(results_J48(x))
return(models)
})
print(result_J48_performance)
stopCluster(cl)
#####HDPs having anti--bacteria, anti--cancer, anti--fungus and virus-- were combined into one to represent HDPs while the negative set was used as non--HDPs#####
### read FASTA
library(protr)
cancer <- readFASTA("cancer.fasta")
fungus <- readFASTA("fungus.fasta")
bacteria <- readFASTA("bacteria.fasta")
virus <- readFASTA("virus.fasta")
negative <- protr::readFASTA("negative_Protein.fasta")
### removed wired protein
cancer <- cancer[(sapply(cancer, protcheck))]
fungus <- fungus[(sapply(fungus, protcheck))]
bacteria <- bacteria[(sapply(bacteria, protcheck))]
virus <- virus[(sapply(virus, protcheck))]
negative <- negative[(sapply(negative, protcheck))]
### Descriptors generation
cancer_des <- t(sapply(cancer, extractAAC))
fungus_des <- t(sapply(fungus, extractAAC))
bacteria_des <- t(sapply(bacteria, extractAAC))
virus_des <- t(sapply(virus, extractAAC))
negative_des <- t(sapply(negative, extractAAC))
#### label the labels
cancer_des <- as.data.frame(cancer_des)
fungus_des <- as.data.frame(fungus_des)
bacteria_des <- as.data.frame(bacteria_des)
virus_des <- as.data.frame(virus_des)
hdp <- rbind(cancer_des, fungus_des, bacteria_des, virus_des)
hdp$Label <- "Positive"
hdp <- as.data.frame(hdp)
negative_des <- as.data.frame(negative_des)
negative_des$Label <- "Negative"
data <- rbind(hdp, negative_des)
data$Label <- as.factor(data$Label)
results_J48 <- function(x) {
training <- J48_train(x)
cross_validation <- J48_cross_validation(x)
testing <- J48_external(x)
results <- data.frame(Training = training, Cross_Validation = cross_validation, Testing = testing)
colnames(results) <- c("Training set", "Internal validation", "External set")
return(results)
}
result_J48_performance <- suppressMessages(results_J48(data))
print(result_J48_performance)
#####HDPs having their respective activity were combined together to build a union predictive model#####
library(protr)
cancer <- readFASTA("cancer.fasta")
fungus <- readFASTA("fungus.fasta")
bacteria <- readFASTA("bacteria.fasta")
virus <- readFASTA("virus.fasta")
negative <- protr::readFASTA("negative_Protein.fasta")
### removed wired protein
cancer <- cancer[(sapply(cancer, protcheck))]
fungus <- fungus[(sapply(fungus, protcheck))]
bacteria <- bacteria[(sapply(bacteria, protcheck))]
virus <- virus[(sapply(virus, protcheck))]
negative <- negative[(sapply(negative, protcheck))]
### function for amino acid composition
composition <- function(x) {
library(protr)
c(extractAAC(x))
}
### generation the Descriptors
cancer_des <- t(sapply(cancer, extractAAC))
fungus_des <- t(sapply(fungus, extractAAC))
bacteria_des <- t(sapply(bacteria, extractAAC))
virus_des <- t(sapply(virus, extractAAC))
negative_des <- t(sapply(negative, extractAAC))
#### label the labels
cancer_des <- as.data.frame(cancer_des)
cancer_des$Label <- "Cancer"
fungus_des <- as.data.frame(fungus_des)
fungus_des$Label <- "Fungus"
bacteria_des <- as.data.frame(bacteria_des)
bacteria_des$Label <- "Bacteria"
virus_des <- as.data.frame(virus_des)
virus_des$Label <- "Virus"
combine_data <- rbind(cancer_des, fungus_des,
bacteria_des, virus_des)
combine_data$Label <- as.factor(combine_data$Label)
#### training results using J48
J48_training <- function(x, Label){
if (Label == "Bacteria") {
suppressPackageStartupMessages(library(parallel))
suppressPackageStartupMessages(library(doSNOW))
cl <- makeCluster(8)
registerDoSNOW(cl)
ok <- list(100)
ok <- foreach(i = 1:100) %dopar% {
in_train <- caret::createDataPartition(x$Label, p = 0.80, list = FALSE)
train <- x[in_train, ]
test <- x[-in_train, ]
rm(in_train)
rm(test)
model_train <- RWeka::J48(Label~., data = train)
actual <- train$Label
prediction <- predict(model_train, train)
rm(train)
rm(model_train)
results <- caret::confusionMatrix(prediction, actual)
results <- results$table
results <- table(prediction, actual)
rm(prediction)
rm(actual)
results <- as.numeric(results)
ok[[i]] <- cbind(results[[1]], (results[[5]] + results[[9]] + results[[13]]),
(results[[2]] + results[[3]] + results[[4]]), (results[[6]] + results[[11]] + results[[16]]))
Cancer <- cbind(results[[6]], (results[[2]] + results[[10]] + results[[14]]),
(results[[5]] + results[[7]] + results[[8]]), (results[[1]] + results[[11]] + results[[16]]))
Fungus <- cbind(results[[11]], (results[[3]] + results[[7]] + results[[15]]),
(results[[9]] + results[[10]] + results[[12]]), (results[[1]] + results[[6]] + results[[16]]))
Virus <- cbind(results[[16]], (results[[4]] + results[[8]] + results[[12]]),
(results[[13]] + results[[14]] + results[[15]]), (results[[1]] + results[[6]] + results[[11]]))
}
} else if (Label == "Cancer") {
cl <- makeCluster(8)
registerDoSNOW(cl)
ok <- list(100)
ok <- foreach(i = 1:100) %dopar% {
in_train <- caret::createDataPartition(x$Label, p = 0.80, list = FALSE)
train <- x[in_train, ]
test <- x[-in_train, ]
rm(in_train)
rm(test)
model_train <- RWeka::J48(Label~., data = train)
actual <- train$Label
prediction <- predict(model_train, train)
rm(train)
rm(model_train)
results <- caret::confusionMatrix(prediction, actual)
results <- results$table
results <- table(prediction, actual)
results <- as.numeric(results)
rm(prediction)
rm(actual)
Bacteria <- cbind(results[[1]], (results[[5]] + results[[9]] + results[[13]]),
(results[[2]] + results[[3]] + results[[4]]), (results[[6]] + results[[11]] + results[[16]]))
ok[[i]] <- cbind(results[[6]], (results[[2]] + results[[10]] + results[[14]]),
(results[[5]] + results[[7]] + results[[8]]), (results[[1]] + results[[11]] + results[[16]]))
Fungus <- cbind(results[[11]], (results[[3]] + results[[7]] + results[[15]]),
(results[[9]] + results[[10]] + results[[12]]), (results[[1]] + results[[6]] + results[[16]]))
Virus <- cbind(results[[16]], (results[[4]] + results[[8]] + results[[12]]),
(results[[13]] + results[[14]] + results[[15]]), (results[[1]] + results[[6]] + results[[11]]))
}
} else if (Label == "Fungus") {
cl <- makeCluster(8)
registerDoSNOW(cl)
ok <- list(100)
ok <- foreach(i = 1:100) %dopar% {
in_train <- caret::createDataPartition(x$Label, p = 0.80, list = FALSE)
train <- x[in_train, ]
test <- x[-in_train, ]
rm(test)
rm(in_train)
model_train <- RWeka::J48(Label~., data = train)
actual <- train$Label
prediction <- predict(model_train, train)
rm(model_train)
rm(train)
results <- caret::confusionMatrix(prediction, actual)
results <- results$table
results <- table(prediction, actual)
results <- as.numeric(results)
rm(actual)
rm(prediction)
Bacteria <- cbind(results[[1]], (results[[5]] + results[[9]] + results[[13]]),
(results[[2]] + results[[3]] + results[[4]]), (results[[6]] + results[[11]] + results[[16]]))
Cancer <- cbind(results[[6]], (results[[2]] + results[[10]] + results[[14]]),
(results[[5]] + results[[7]] + results[[8]]), (results[[1]] + results[[11]] + results[[16]]))
ok[[i]] <- cbind(results[[11]], (results[[3]] + results[[7]] + results[[15]]),
(results[[9]] + results[[10]] + results[[12]]), (results[[1]] + results[[6]] + results[[16]]))
Virus <- cbind(results[[16]], (results[[4]] + results[[8]] + results[[12]]),
(results[[13]] + results[[14]] + results[[15]]), (results[[1]] + results[[6]] + results[[11]]))
}
} else if (Label == "Virus") {
cl <- makeCluster(8)
registerDoSNOW(cl)
ok <- list(100)
ok <- foreach(i = 1:100) %dopar% {
in_train <- caret::createDataPartition(x$Label, p = 0.80, list = FALSE)
train <- x[in_train, ]
test <- x[-in_train, ]
rm(in_train)
rm(test)
model_train <- RWeka::J48(Label~., data = train)
actual <- train$Label
prediction <- predict(model_train, train)
rm(model_train)
rm(train)
results <- caret::confusionMatrix(prediction, actual)
results <- results$table
results <- table(prediction, actual)
results <- as.numeric(results)
rm(prediction)
rm(actual)
Bacteria <- cbind(results[[1]], (results[[5]] + results[[9]] + results[[13]]),
(results[[2]] + results[[3]] + results[[4]]), (results[[6]] + results[[11]] + results[[16]]))
Cancer <- cbind(results[[6]], (results[[2]] + results[[10]] + results[[14]]),
(results[[5]] + results[[7]] + results[[8]]), (results[[1]] + results[[11]] + results[[16]]))
Fungus <- cbind(results[[11]], (results[[3]] + results[[7]] + results[[15]]),
(results[[9]] + results[[10]] + results[[12]]), (results[[1]] + results[[6]] + results[[16]]))
ok[[i]] <- cbind(results[[16]], (results[[4]] + results[[8]] + results[[12]]),
(results[[13]] + results[[14]] + results[[15]]), (results[[1]] + results[[6]] + results[[11]]))
}
return(ok)
stopCluster(cl)
} }
mean_and_sd <- function(x) {
c(round(mean(x, na.rm = TRUE), digits = 4),
round(sd(x, na.rm = TRUE), digits = 4))
}
results_training_Bacteria <- function(x) {
yes <- J48_training(x, Label = "Bacteria")
great <- data.frame(yes)
rm(yes)
TP <- seq(from = 1, to = 400, by = 4)
FN <- seq(from = 2, to = 400, by = 4)
FP <- seq(from = 3, to = 400, by = 4)
TN <- seq(from = 4, to = 400, by = 4)
results <- mapply(c, great[TP], great[FN], great[FP], great[TN])
rm(great)
data <- data.frame(results)
rm(results)
m = ncol(data)
ACC <- matrix(nrow = m, ncol = 1)
SENS <- matrix(nrow = m, ncol = 1)
SPEC <-matrix(nrow = m, ncol = 1)
MCC <- matrix(nrow = m, ncol = 1)
for(i in 1:m){
ACC[i,1] = (data[1,i]+data[4,i])/(data[1,i]+data[2,i]+data[3,i]+data[4,i])*100
SENS[i,1] = (data[4,i])/(data[3,i]+data[4,i])*100
SPEC[i,1] = (data[1,i]/(data[1,i]+data[2,i]))*100
MCC1 = (data[1,i]*data[4,i]) - (data[2,i]*data[3,i])
MCC2 = (data[4,i]+data[2,i])*(data[4,i]+data[3,i])
MCC3 = (data[1,i]+data[2,i])*(data[1,i]+data[3,i])
MCC4 = sqrt(MCC2)*sqrt(MCC3)
MCC[i,1] = MCC1/MCC4
}
rm(TP)
rm(FP)
rm(TN)
rm(FN)
results_ACC <- mean_and_sd(ACC)
results_SENS <- mean_and_sd(SENS)
results_SPEC <- mean_and_sd(SPEC)
results_MCC <- mean_and_sd(MCC)
rm(ACC)
rm(SENS)
rm(SPEC)
rm(MCC)
results_all <- (data.frame(c(results_ACC, results_SENS, results_SPEC, results_MCC)))
rownames(results_all) <- c("ACC_Mean", "ACC_SD", "Sens_Mean", "Sens_SD", "Spec_Mean", "Spec_SD",
"MCC_Mean", "MCC_SD")
rm(results_ACC)
rm(results_SENS)
rm(results_SPEC)
rm(results_MCC)
return(results_all)
}
results_training_Cancer <- function(x) {
yes <- J48_training(x, Label = "Cancer")
great <- data.frame(yes)
rm(yes)
TP <- seq(from = 1, to = 400, by = 4)
FN <- seq(from = 2, to = 400, by = 4)
FP <- seq(from = 3, to = 400, by = 4)
TN <- seq(from = 4, to = 400, by = 4)
results <- mapply(c, great[TP], great[FN], great[FP], great[TN])
rm(great)
data <- data.frame(results)
rm(results)
m = ncol(data)
ACC <- matrix(nrow = m, ncol = 1)
SENS <- matrix(nrow = m, ncol = 1)
SPEC <-matrix(nrow = m, ncol = 1)
MCC <- matrix(nrow = m, ncol = 1)
for(i in 1:m){
ACC[i,1] = (data[1,i]+data[4,i])/(data[1,i]+data[2,i]+data[3,i]+data[4,i])*100
SENS[i,1] = (data[4,i])/(data[3,i]+data[4,i])*100
SPEC[i,1] = (data[1,i]/(data[1,i]+data[2,i]))*100
MCC1 = (data[1,i]*data[4,i]) - (data[2,i]*data[3,i])
MCC2 = (data[4,i]+data[2,i])*(data[4,i]+data[3,i])
MCC3 = (data[1,i]+data[2,i])*(data[1,i]+data[3,i])
MCC4 = sqrt(MCC2)*sqrt(MCC3)
MCC[i,1] = MCC1/MCC4
}
rm(TP)
rm(FP)
rm(TN)
rm(FN)
results_ACC <- mean_and_sd(ACC)
results_SENS <- mean_and_sd(SENS)
results_SPEC <- mean_and_sd(SPEC)
results_MCC <- mean_and_sd(MCC)
rm(ACC)
rm(SENS)
rm(SPEC)
rm(MCC)
results_all <- (data.frame(c(results_ACC, results_SENS, results_SPEC, results_MCC)))
rownames(results_all) <- c("ACC_Mean", "ACC_SD", "Sens_Mean", "Sens_SD", "Spec_Mean", "Spec_SD",
"MCC_Mean", "MCC_SD")
rm(results_ACC)
rm(results_SENS)
rm(results_SPEC)
rm(results_MCC)
return(results_all)
}
results_training_Fungus <- function(x) {
yes <- J48_training(x, Label = "Fungus")
great <- data.frame(yes)
rm(yes)
TP <- seq(from = 1, to = 400, by = 4)
FN <- seq(from = 2, to = 400, by = 4)
FP <- seq(from = 3, to = 400, by = 4)
TN <- seq(from = 4, to = 400, by = 4)
results <- mapply(c, great[TP], great[FN], great[FP], great[TN])
rm(great)
data <- data.frame(results)
rm(results)
m = ncol(data)
ACC <- matrix(nrow = m, ncol = 1)
SENS <- matrix(nrow = m, ncol = 1)
SPEC <-matrix(nrow = m, ncol = 1)
MCC <- matrix(nrow = m, ncol = 1)
for(i in 1:m){
ACC[i,1] = (data[1,i]+data[4,i])/(data[1,i]+data[2,i]+data[3,i]+data[4,i])*100
SENS[i,1] = (data[4,i])/(data[3,i]+data[4,i])*100
SPEC[i,1] = (data[1,i]/(data[1,i]+data[2,i]))*100
MCC1 = (data[1,i]*data[4,i]) - (data[2,i]*data[3,i])
MCC2 = (data[4,i]+data[2,i])*(data[4,i]+data[3,i])
MCC3 = (data[1,i]+data[2,i])*(data[1,i]+data[3,i])
MCC4 = sqrt(MCC2)*sqrt(MCC3)
MCC[i,1] = MCC1/MCC4
}
rm(TP)
rm(FP)
rm(TN)
rm(FN)
results_ACC <- mean_and_sd(ACC)
results_SENS <- mean_and_sd(SENS)
results_SPEC <- mean_and_sd(SPEC)
results_MCC <- mean_and_sd(MCC)
rm(ACC)
rm(SENS)
rm(SPEC)
rm(MCC)
results_all <- (data.frame(c(results_ACC, results_SENS, results_SPEC, results_MCC)))
rownames(results_all) <- c("ACC_Mean", "ACC_SD", "Sens_Mean", "Sens_SD", "Spec_Mean", "Spec_SD",
"MCC_Mean", "MCC_SD")
rm(results_ACC)
rm(results_SENS)
rm(results_SPEC)
rm(results_MCC)
return(results_all)
}
results_training_Virus <- function(x) {
yes <- J48_training(x, Label = "Virus")
great <- data.frame(yes)
rm(yes)
TP <- seq(from = 1, to = 400, by = 4)
FN <- seq(from = 2, to = 400, by = 4)
FP <- seq(from = 3, to = 400, by = 4)
TN <- seq(from = 4, to = 400, by = 4)
results <- mapply(c, great[TP], great[FN], great[FP], great[TN])
rm(great)
data <- data.frame(results)
rm(results)
m = ncol(data)
ACC <- matrix(nrow = m, ncol = 1)
SENS <- matrix(nrow = m, ncol = 1)
SPEC <-matrix(nrow = m, ncol = 1)
MCC <- matrix(nrow = m, ncol = 1)
for(i in 1:m){
ACC[i,1] = (data[1,i]+data[4,i])/(data[1,i]+data[2,i]+data[3,i]+data[4,i])*100
SENS[i,1] = (data[4,i])/(data[3,i]+data[4,i])*100
SPEC[i,1] = (data[1,i]/(data[1,i]+data[2,i]))*100
MCC1 = (data[1,i]*data[4,i]) - (data[2,i]*data[3,i])
MCC2 = (data[4,i]+data[2,i])*(data[4,i]+data[3,i])
MCC3 = (data[1,i]+data[2,i])*(data[1,i]+data[3,i])
MCC4 = sqrt(MCC2)*sqrt(MCC3)
MCC[i,1] = MCC1/MCC4
}
rm(TP)
rm(FP)
rm(TN)
rm(FN)
results_ACC <- mean_and_sd(ACC)
results_SENS <- mean_and_sd(SENS)
results_SPEC <- mean_and_sd(SPEC)
results_MCC <- mean_and_sd(MCC)
rm(ACC)
rm(SENS)
rm(SPEC)
rm(MCC)
results_all <- (data.frame(c(results_ACC, results_SENS, results_SPEC, results_MCC)))
rownames(results_all) <- c("ACC_Mean", "ACC_SD", "Sens_Mean", "Sens_SD", "Spec_Mean", "Spec_SD",
"MCC_Mean", "MCC_SD")
rm(results_ACC)
rm(results_SENS)
rm(results_SPEC)
rm(results_MCC)
return(results_all)
}
J48_training_all <- function(x) {
bacteria <- results_training_Bacteria(x)
cancer <- results_training_Cancer(x)
fungus <- results_training_Fungus(x)
virus <- results_training_Virus(x)
results_all <- cbind(bacteria, cancer, fungus, virus)
rm(bacteria)
rm(cancer)
rm(fungus)
rm(virus)
total <- apply(results_all, 1, mean)
results_all_mean <- cbind(results_all, total)
rm(results_all)
rm(total)
colnames(results_all_mean) <- c("Bacteria", "Cancer", "Fungus", "Virus", "Overall")
return(results_all_mean)
}
### 10 fold Cross Validations
J48_10_CV <- function(x, Label){
if (Label == "Bacteria") {
suppressPackageStartupMessages(library(parallel))
suppressPackageStartupMessages(library(doSNOW))
cl <- makeCluster(8)
registerDoSNOW(cl)
ok <- list(100)
ok <- foreach(i = 1:100) %dopar% {
in_train <- caret::createDataPartition(x$Label, p = 0.80, list = FALSE)
train <- x[in_train, ]
test <- x[-in_train, ]
rm(in_train)
rm(test)
model_train <- RWeka::J48(Label~., data = train)
results <- RWeka::evaluate_Weka_classifier(model_train, newata = NULL, numFolds = 10, complexity = FALSE)
rm(train)
confusionMatrix <- results$confusionMatrix
rm(model_train)
results <- as.numeric(confusionMatrix)
rm(confusionMatrix)
ok[[i]] <- cbind(results[[1]], (results[[5]] + results[[9]] + results[[13]]),
(results[[2]] + results[[3]] + results[[4]]), (results[[6]] + results[[11]] + results[[16]]))
Cancer <- cbind(results[[6]], (results[[2]] + results[[10]] + results[[14]]),
(results[[5]] + results[[7]] + results[[8]]), (results[[1]] + results[[11]] + results[[16]]))
Fungus <- cbind(results[[11]], (results[[3]] + results[[7]] + results[[15]]),
(results[[9]] + results[[10]] + results[[12]]), (results[[1]] + results[[6]] + results[[16]]))
Virus <- cbind(results[[16]], (results[[4]] + results[[8]] + results[[12]]),
(results[[13]] + results[[14]] + results[[15]]), (results[[1]] + results[[6]] + results[[11]]))
}
} else if (Label == "Cancer") {
cl <- makeCluster(8)
registerDoSNOW(cl)
ok <- list(100)
ok <- foreach(i = 1:100) %dopar% {
in_train <- caret::createDataPartition(x$Label, p = 0.80, list = FALSE)
train <- x[in_train, ]
test <- x[-in_train, ]
rm(in_train)
rm(test)
model_train <- RWeka::J48(Label~., data = train)
results <- RWeka::evaluate_Weka_classifier(model_train, newata = NULL, numFolds = 10, complexity = FALSE)
rm(train)
confusionMatrix <- results$confusionMatrix
rm(model_train)
results <- as.numeric(confusionMatrix)
rm(confusionMatrix)
Bacteria <- cbind(results[[1]], (results[[5]] + results[[9]] + results[[13]]),
(results[[2]] + results[[3]] + results[[4]]), (results[[6]] + results[[11]] + results[[16]]))
ok[[i]] <- cbind(results[[6]], (results[[2]] + results[[10]] + results[[14]]),
(results[[5]] + results[[7]] + results[[8]]), (results[[1]] + results[[11]] + results[[16]]))
Fungus <- cbind(results[[11]], (results[[3]] + results[[7]] + results[[15]]),
(results[[9]] + results[[10]] + results[[12]]), (results[[1]] + results[[6]] + results[[16]]))
Virus <- cbind(results[[16]], (results[[4]] + results[[8]] + results[[12]]),
(results[[13]] + results[[14]] + results[[15]]), (results[[1]] + results[[6]] + results[[11]]))
}
} else if (Label == "Fungus") {
cl <- makeCluster(8)
registerDoSNOW(cl)
ok <- list(100)
ok <- foreach(i = 1:100) %dopar% {
in_train <- caret::createDataPartition(x$Label, p = 0.80, list = FALSE)
train <- x[in_train, ]
test <- x[-in_train, ]
rm(in_train)
rm(test)
model_train <- RWeka::J48(Label~., data = train)
results <- RWeka::evaluate_Weka_classifier(model_train, newata = NULL, numFolds = 10, complexity = FALSE)
rm(train)
rm(model_train)
confusionMatrix <- results$confusionMatrix
rm(results)
results <- as.numeric(confusionMatrix)
rm(confusionMatrix)
Bacteria <- cbind(results[[1]], (results[[5]] + results[[9]] + results[[13]]),
(results[[2]] + results[[3]] + results[[4]]), (results[[6]] + results[[11]] + results[[16]]))
Cancer <- cbind(results[[6]], (results[[2]] + results[[10]] + results[[14]]),
(results[[5]] + results[[7]] + results[[8]]), (results[[1]] + results[[11]] + results[[16]]))
ok[[i]] <- cbind(results[[11]], (results[[3]] + results[[7]] + results[[15]]),
(results[[9]] + results[[10]] + results[[12]]), (results[[1]] + results[[6]] + results[[16]]))
Virus <- cbind(results[[16]], (results[[4]] + results[[8]] + results[[12]]),
(results[[13]] + results[[14]] + results[[15]]), (results[[1]] + results[[6]] + results[[11]]))
}
} else if (Label == "Virus") {
cl <- makeCluster(8)
registerDoSNOW(cl)
ok <- list(100)
ok <- foreach(i = 1:100) %dopar% {
in_train <- caret::createDataPartition(x$Label, p = 0.80, list = FALSE)
train <- x[in_train, ]
test <- x[-in_train, ]
rm(in_train)
rm(test)
model_train <- RWeka::J48(Label~., data = train)
results <- RWeka::evaluate_Weka_classifier(model_train, newata = NULL, numFolds = 10, complexity = FALSE)
rm(model_train)
rm(train)
confusionMatrix <- results$confusionMatrix
results <- as.numeric(confusionMatrix)
rm(confusionMatrix)
Bacteria <- cbind(results[[1]], (results[[5]] + results[[9]] + results[[13]]),
(results[[2]] + results[[3]] + results[[4]]), (results[[6]] + results[[11]] + results[[16]]))
Cancer <- cbind(results[[6]], (results[[2]] + results[[10]] + results[[14]]),
(results[[5]] + results[[7]] + results[[8]]), (results[[1]] + results[[11]] + results[[16]]))
Fungus <- cbind(results[[11]], (results[[3]] + results[[7]] + results[[15]]),
(results[[9]] + results[[10]] + results[[12]]), (results[[1]] + results[[6]] + results[[16]]))
ok[[i]] <- cbind(results[[16]], (results[[4]] + results[[8]] + results[[12]]),
(results[[13]] + results[[14]] + results[[15]]), (results[[1]] + results[[6]] + results[[11]]))
}
return(ok)
stopCluster(cl)
} }
mean_and_sd <- function(x) {
c(round(mean(x, na.rm = TRUE), digits = 4),
round(sd(x, na.rm = TRUE), digits = 4))
}
results_CV_Bacteria <- function(x) {
yes <- J48_10_CV(x, Label = "Bacteria")
great <- data.frame(yes)
rm(yes)
TP <- seq(from = 1, to = 400, by = 4)
FN <- seq(from = 2, to = 400, by = 4)
FP <- seq(from = 3, to = 400, by = 4)
TN <- seq(from = 4, to = 400, by = 4)
results <- mapply(c, great[TP], great[FN], great[FP], great[TN])
data <- data.frame(results)
rm(great)
rm(results)
m = ncol(data)
ACC <- matrix(nrow = m, ncol = 1)
SENS <- matrix(nrow = m, ncol = 1)
SPEC <-matrix(nrow = m, ncol = 1)
MCC <- matrix(nrow = m, ncol = 1)
for(i in 1:m){
ACC[i,1] = (data[1,i]+data[4,i])/(data[1,i]+data[2,i]+data[3,i]+data[4,i])*100
SENS[i,1] = (data[4,i])/(data[3,i]+data[4,i])*100
SPEC[i,1] = (data[1,i]/(data[1,i]+data[2,i]))*100
MCC1 = (data[1,i]*data[4,i]) - (data[2,i]*data[3,i])
MCC2 = (data[4,i]+data[2,i])*(data[4,i]+data[3,i])
MCC3 = (data[1,i]+data[2,i])*(data[1,i]+data[3,i])
MCC4 = sqrt(MCC2)*sqrt(MCC3)
MCC[i,1] = MCC1/MCC4
}
rm(TP)
rm(FP)
rm(TN)
rm(FN)
results_ACC <- mean_and_sd(ACC)
results_SENS <- mean_and_sd(SENS)
results_SPEC <- mean_and_sd(SPEC)
results_MCC <- mean_and_sd(MCC)
rm(data)
rm(ACC)
rm(SENS)
rm(SPEC)
rm(MCC)
results_all <- (data.frame(c(results_ACC, results_SENS, results_SPEC, results_MCC)))
rownames(results_all) <- c("ACC_Mean", "ACC_SD", "Sens_Mean", "Sens_SD", "Spec_Mean", "Spec_SD",
"MCC_Mean", "MCC_SD")
rm(results_ACC)
rm(results_SENS)
rm(results_SPEC)
rm(results_MCC)
return(results_all)
}
results_CV_Cancer <- function(x) {
yes <- J48_10_CV(x, Label = "Cancer")
great <- data.frame(yes)
rm(yes)
TP <- seq(from = 1, to = 400, by = 4)
FN <- seq(from = 2, to = 400, by = 4)
FP <- seq(from = 3, to = 400, by = 4)
TN <- seq(from = 4, to = 400, by = 4)
results <- mapply(c, great[TP], great[FN], great[FP], great[TN])
data <- data.frame(results)
rm(great)
rm(results)
m = ncol(data)
ACC <- matrix(nrow = m, ncol = 1)
SENS <- matrix(nrow = m, ncol = 1)
SPEC <-matrix(nrow = m, ncol = 1)
MCC <- matrix(nrow = m, ncol = 1)
for(i in 1:m){
ACC[i,1] = (data[1,i]+data[4,i])/(data[1,i]+data[2,i]+data[3,i]+data[4,i])*100
SENS[i,1] = (data[4,i])/(data[3,i]+data[4,i])*100
SPEC[i,1] = (data[1,i]/(data[1,i]+data[2,i]))*100
MCC1 = (data[1,i]*data[4,i]) - (data[2,i]*data[3,i])
MCC2 = (data[4,i]+data[2,i])*(data[4,i]+data[3,i])
MCC3 = (data[1,i]+data[2,i])*(data[1,i]+data[3,i])
MCC4 = sqrt(MCC2)*sqrt(MCC3)
MCC[i,1] = MCC1/MCC4
}
rm(TP)
rm(FP)
rm(TN)
rm(FN)
results_ACC <- mean_and_sd(ACC)
results_SENS <- mean_and_sd(SENS)
results_SPEC <- mean_and_sd(SPEC)
results_MCC <- mean_and_sd(MCC)
rm(data)
rm(ACC)
rm(SENS)
rm(SPEC)
rm(MCC)
results_all <- (data.frame(c(results_ACC, results_SENS, results_SPEC, results_MCC)))
rownames(results_all) <- c("ACC_Mean", "ACC_SD", "Sens_Mean", "Sens_SD", "Spec_Mean", "Spec_SD",
"MCC_Mean", "MCC_SD")
rm(results_ACC)
rm(results_SENS)
rm(results_SPEC)
rm(results_MCC)
return(results_all)
}
results_CV_Fungus <- function(x) {
yes <- J48_10_CV(x, Label = "Fungus")
great <- data.frame(yes)
rm(yes)
TP <- seq(from = 1, to = 400, by = 4)
FN <- seq(from = 2, to = 400, by = 4)
FP <- seq(from = 3, to = 400, by = 4)
TN <- seq(from = 4, to = 400, by = 4)
results <- mapply(c, great[TP], great[FN], great[FP], great[TN])
data <- data.frame(results)
rm(great)
rm(results)
m = ncol(data)
ACC <- matrix(nrow = m, ncol = 1)
SENS <- matrix(nrow = m, ncol = 1)
SPEC <-matrix(nrow = m, ncol = 1)
MCC <- matrix(nrow = m, ncol = 1)
for(i in 1:m){
ACC[i,1] = (data[1,i]+data[4,i])/(data[1,i]+data[2,i]+data[3,i]+data[4,i])*100
SENS[i,1] = (data[4,i])/(data[3,i]+data[4,i])*100
SPEC[i,1] = (data[1,i]/(data[1,i]+data[2,i]))*100
MCC1 = (data[1,i]*data[4,i]) - (data[2,i]*data[3,i])
MCC2 = (data[4,i]+data[2,i])*(data[4,i]+data[3,i])
MCC3 = (data[1,i]+data[2,i])*(data[1,i]+data[3,i])
MCC4 = sqrt(MCC2)*sqrt(MCC3)
MCC[i,1] = MCC1/MCC4
}
rm(TP)
rm(FP)
rm(TN)
rm(FN)
results_ACC <- mean_and_sd(ACC)
results_SENS <- mean_and_sd(SENS)
results_SPEC <- mean_and_sd(SPEC)
results_MCC <- mean_and_sd(MCC)
rm(data)
rm(ACC)
rm(SENS)
rm(SPEC)
rm(MCC)
results_all <- (data.frame(c(results_ACC, results_SENS, results_SPEC, results_MCC)))
rownames(results_all) <- c("ACC_Mean", "ACC_SD", "Sens_Mean", "Sens_SD", "Spec_Mean", "Spec_SD",
"MCC_Mean", "MCC_SD")
rm(results_ACC)
rm(results_SENS)
rm(results_SPEC)
rm(results_MCC)
return(results_all)
}
results_CV_Virus <- function(x) {
yes <- J48_10_CV(x, Label = "Virus")
great <- data.frame(yes)
rm(yes)
TP <- seq(from = 1, to = 400, by = 4)
FN <- seq(from = 2, to = 400, by = 4)
FP <- seq(from = 3, to = 400, by = 4)
TN <- seq(from = 4, to = 400, by = 4)
results <- mapply(c, great[TP], great[FN], great[FP], great[TN])
data <- data.frame(results)
rm(great)
rm(results)
m = ncol(data)
ACC <- matrix(nrow = m, ncol = 1)
SENS <- matrix(nrow = m, ncol = 1)
SPEC <-matrix(nrow = m, ncol = 1)
MCC <- matrix(nrow = m, ncol = 1)
for(i in 1:m){
ACC[i,1] = (data[1,i]+data[4,i])/(data[1,i]+data[2,i]+data[3,i]+data[4,i])*100
SENS[i,1] = (data[4,i])/(data[3,i]+data[4,i])*100
SPEC[i,1] = (data[1,i]/(data[1,i]+data[2,i]))*100
MCC1 = (data[1,i]*data[4,i]) - (data[2,i]*data[3,i])
MCC2 = (data[4,i]+data[2,i])*(data[4,i]+data[3,i])
MCC3 = (data[1,i]+data[2,i])*(data[1,i]+data[3,i])
MCC4 = sqrt(MCC2)*sqrt(MCC3)
MCC[i,1] = MCC1/MCC4
}
rm(TP)
rm(FP)
rm(TN)
rm(FN)
results_ACC <- mean_and_sd(ACC)
results_SENS <- mean_and_sd(SENS)
results_SPEC <- mean_and_sd(SPEC)
results_MCC <- mean_and_sd(MCC)
rm(data)
rm(ACC)
rm(SENS)
rm(SPEC)
rm(MCC)
results_all <- (data.frame(c(results_ACC, results_SENS, results_SPEC, results_MCC)))
rownames(results_all) <- c("ACC_Mean", "ACC_SD", "Sens_Mean", "Sens_SD", "Spec_Mean", "Spec_SD",
"MCC_Mean", "MCC_SD")
rm(results_ACC)
rm(results_SENS)
rm(results_SPEC)
rm(results_MCC)
return(results_all)
}
J48_CV_all <- function(x) {
bacteria <- results_CV_Bacteria(x)
cancer <- results_CV_Cancer(x)
fungus <- results_CV_Fungus(x)
virus <- results_CV_Virus(x)
results_all <- cbind(bacteria, cancer, fungus, virus)
rm(bacteria)
rm(cancer)
rm(fungus)
rm(virus)
total <- apply(results_all, 1, mean)
results_all_mean <- cbind(results_all, total)
rm(results_all)
rm(total)
colnames(results_all_mean) <- c("Bacteria", "Cancer", "Fungus", "Virus", "Overall")
return(results_all_mean)
}
#### function for testing
J48_testing <- function(x, Label){
if (Label == "Bacteria") {
suppressPackageStartupMessages(library(parallel))
suppressPackageStartupMessages(library(doSNOW))
cl <- makeCluster(8)
registerDoSNOW(cl)
ok <- list(100)
ok <- foreach(i = 1:100) %dopar% {
in_train <- caret::createDataPartition(x$Label, p = 0.80, list = FALSE)
train <- x[in_train, ]
test <- x[-in_train, ]
rm(in_train)
model_train <- RWeka::J48(Label~., data = train)
rm(train)
actual <- test$Label
prediction <- predict(model_train, test)
rm(model_train)
rm(test)
results <- caret::confusionMatrix(prediction, actual)
results <- results$table
results <- table(prediction, actual)
rm(prediction)
rm(actual)
results <- as.numeric(results)
ok[[i]] <- cbind(results[[1]], (results[[5]] + results[[9]] + results[[13]]),
(results[[2]] + results[[3]] + results[[4]]), (results[[6]] + results[[11]] + results[[16]]))
Cancer <- cbind(results[[6]], (results[[2]] + results[[10]] + results[[14]]),
(results[[5]] + results[[7]] + results[[8]]), (results[[1]] + results[[11]] + results[[16]]))
Fungus <- cbind(results[[11]], (results[[3]] + results[[7]] + results[[15]]),
(results[[9]] + results[[10]] + results[[12]]), (results[[1]] + results[[6]] + results[[16]]))
Virus <- cbind(results[[16]], (results[[4]] + results[[8]] + results[[12]]),
(results[[13]] + results[[14]] + results[[15]]), (results[[1]] + results[[6]] + results[[11]]))
}
} else if (Label == "Cancer") {
cl <- makeCluster(8)
registerDoSNOW(cl)
ok <- list(100)
ok <- foreach(i = 1:100) %dopar% {
in_train <- caret::createDataPartition(x$Label, p = 0.80, list = FALSE)
train <- x[in_train, ]
test <- x[-in_train, ]
rm(in_train)
model_train <- RWeka::J48(Label~., data = train)
rm(train)
actual <- test$Label
prediction <- predict(model_train, test)
rm(model_train)
rm(test)
results <- caret::confusionMatrix(prediction, actual)
results <- results$table
results <- table(prediction, actual)
results <- as.numeric(results)
rm(prediction)
rm(actual)
Bacteria <- cbind(results[[1]], (results[[5]] + results[[9]] + results[[13]]),
(results[[2]] + results[[3]] + results[[4]]), (results[[6]] + results[[11]] + results[[16]]))
ok[[i]] <- cbind(results[[6]], (results[[2]] + results[[10]] + results[[14]]),
(results[[5]] + results[[7]] + results[[8]]), (results[[1]] + results[[11]] + results[[16]]))
Fungus <- cbind(results[[11]], (results[[3]] + results[[7]] + results[[15]]),
(results[[9]] + results[[10]] + results[[12]]), (results[[1]] + results[[6]] + results[[16]]))
Virus <- cbind(results[[16]], (results[[4]] + results[[8]] + results[[12]]),
(results[[13]] + results[[14]] + results[[15]]), (results[[1]] + results[[6]] + results[[11]]))
}
} else if (Label == "Fungus") {
cl <- makeCluster(8)
registerDoSNOW(cl)
ok <- list(100)
ok <- foreach(i = 1:100) %dopar% {
in_train <- caret::createDataPartition(x$Label, p = 0.80, list = FALSE)
train <- x[in_train, ]
test <- x[-in_train, ]
rm(in_train)
model_train <- RWeka::J48(Label~., data = train)
rm(train)
actual <- test$Label
prediction <- predict(model_train, test)
rm(model_train)
rm(test)
results <- caret::confusionMatrix(prediction, actual)
results <- results$table
results <- table(prediction, actual)
results <- as.numeric(results)
rm(prediction)
rm(actual)
Bacteria <- cbind(results[[1]], (results[[5]] + results[[9]] + results[[13]]),
(results[[2]] + results[[3]] + results[[4]]), (results[[6]] + results[[11]] + results[[16]]))
Cancer <- cbind(results[[6]], (results[[2]] + results[[10]] + results[[14]]),
(results[[5]] + results[[7]] + results[[8]]), (results[[1]] + results[[11]] + results[[16]]))
ok[[i]] <- cbind(results[[11]], (results[[3]] + results[[7]] + results[[15]]),
(results[[9]] + results[[10]] + results[[12]]), (results[[1]] + results[[6]] + results[[16]]))
Virus <- cbind(results[[16]], (results[[4]] + results[[8]] + results[[12]]),
(results[[13]] + results[[14]] + results[[15]]), (results[[1]] + results[[6]] + results[[11]]))
}
} else if (Label == "Virus") {
cl <- makeCluster(8)
registerDoSNOW(cl)
ok <- list(100)
ok <- foreach(i = 1:100) %dopar% {
in_train <- caret::createDataPartition(x$Label, p = 0.80, list = FALSE)
train <- x[in_train, ]
test <- x[-in_train, ]
rm(in_train)
model_train <- RWeka::J48(Label~., data = train)
rm(train)
actual <- test$Label
prediction <- predict(model_train, test)
rm(model_train)
rm(test)
results <- caret::confusionMatrix(prediction, actual)
results <- results$table
results <- table(prediction, actual)
results <- as.numeric(results)
rm(prediction)
rm(actual)
Bacteria <- cbind(results[[1]], (results[[5]] + results[[9]] + results[[13]]),
(results[[2]] + results[[3]] + results[[4]]), (results[[6]] + results[[11]] + results[[16]]))
Cancer <- cbind(results[[6]], (results[[2]] + results[[10]] + results[[14]]),
(results[[5]] + results[[7]] + results[[8]]), (results[[1]] + results[[11]] + results[[16]]))
Fungus <- cbind(results[[11]], (results[[3]] + results[[7]] + results[[15]]),
(results[[9]] + results[[10]] + results[[12]]), (results[[1]] + results[[6]] + results[[16]]))
ok[[i]] <- cbind(results[[16]], (results[[4]] + results[[8]] + results[[12]]),
(results[[13]] + results[[14]] + results[[15]]), (results[[1]] + results[[6]] + results[[11]]))
}
return(ok)
stopCluster(cl)
} }
mean_and_sd <- function(x) {
c(round(mean(x, na.rm = TRUE), digits = 4),
round(sd(x, na.rm = TRUE), digits = 4))
}
results_testing_Bacteria <- function(x) {
yes <- J48_testing(x, Label = "Bacteria")
great <- data.frame(yes)
rm(yes)
TP <- seq(from = 1, to = 400, by = 4)
FN <- seq(from = 2, to = 400, by = 4)
FP <- seq(from = 3, to = 400, by = 4)
TN <- seq(from = 4, to = 400, by = 4)
results <- mapply(c, great[TP], great[FN], great[FP], great[TN])
data <- data.frame(results)
rm(great)
rm(results)
m = ncol(data)
ACC <- matrix(nrow = m, ncol = 1)
SENS <- matrix(nrow = m, ncol = 1)
SPEC <-matrix(nrow = m, ncol = 1)
MCC <- matrix(nrow = m, ncol = 1)
for(i in 1:m){
ACC[i,1] = (data[1,i]+data[4,i])/(data[1,i]+data[2,i]+data[3,i]+data[4,i])*100
SENS[i,1] = (data[4,i])/(data[3,i]+data[4,i])*100
SPEC[i,1] = (data[1,i]/(data[1,i]+data[2,i]))*100
MCC1 = (data[1,i]*data[4,i]) - (data[2,i]*data[3,i])
MCC2 = (data[4,i]+data[2,i])*(data[4,i]+data[3,i])
MCC3 = (data[1,i]+data[2,i])*(data[1,i]+data[3,i])
MCC4 = sqrt(MCC2)*sqrt(MCC3)
MCC[i,1] = MCC1/MCC4
}
rm(TP)
rm(FP)
rm(TN)
rm(FN)
results_ACC <- mean_and_sd(ACC)
results_SENS <- mean_and_sd(SENS)
results_SPEC <- mean_and_sd(SPEC)
results_MCC <- mean_and_sd(MCC)
rm(ACC)
rm(SENS)
rm(SPEC)
rm(MCC)
rm(data)
results_all <- (data.frame(c(results_ACC, results_SENS, results_SPEC, results_MCC)))
rownames(results_all) <- c("ACC_Mean", "ACC_SD", "Sens_Mean", "Sens_SD", "Spec_Mean", "Spec_SD",
"MCC_Mean", "MCC_SD")
rm(results_ACC)
rm(results_SENS)
rm(results_SPEC)
rm(results_MCC)
return(results_all)
}
results_testing_Cancer <- function(x) {
yes <- J48_testing(x, Label = "Cancer")
great <- data.frame(yes)
rm(yes)
TP <- seq(from = 1, to = 400, by = 4)
FN <- seq(from = 2, to = 400, by = 4)
FP <- seq(from = 3, to = 400, by = 4)
TN <- seq(from = 4, to = 400, by = 4)
results <- mapply(c, great[TP], great[FN], great[FP], great[TN])
data <- data.frame(results)
rm(great)
rm(results)
m = ncol(data)
ACC <- matrix(nrow = m, ncol = 1)
SENS <- matrix(nrow = m, ncol = 1)
SPEC <-matrix(nrow = m, ncol = 1)
MCC <- matrix(nrow = m, ncol = 1)
for(i in 1:m){
ACC[i,1] = (data[1,i]+data[4,i])/(data[1,i]+data[2,i]+data[3,i]+data[4,i])*100
SENS[i,1] = (data[4,i])/(data[3,i]+data[4,i])*100
SPEC[i,1] = (data[1,i]/(data[1,i]+data[2,i]))*100
MCC1 = (data[1,i]*data[4,i]) - (data[2,i]*data[3,i])
MCC2 = (data[4,i]+data[2,i])*(data[4,i]+data[3,i])
MCC3 = (data[1,i]+data[2,i])*(data[1,i]+data[3,i])
MCC4 = sqrt(MCC2)*sqrt(MCC3)
MCC[i,1] = MCC1/MCC4
}
rm(TP)
rm(FP)
rm(TN)
rm(FN)
results_ACC <- mean_and_sd(ACC)
results_SENS <- mean_and_sd(SENS)
results_SPEC <- mean_and_sd(SPEC)
results_MCC <- mean_and_sd(MCC)
rm(ACC)
rm(SENS)
rm(SPEC)
rm(MCC)
rm(data)
results_all <- (data.frame(c(results_ACC, results_SENS, results_SPEC, results_MCC)))
rownames(results_all) <- c("ACC_Mean", "ACC_SD", "Sens_Mean", "Sens_SD", "Spec_Mean", "Spec_SD",
"MCC_Mean", "MCC_SD")
rm(results_ACC)
rm(results_SENS)
rm(results_SPEC)
rm(results_MCC)
return(results_all)
}
results_testing_Fungus <- function(x) {
yes <- J48_testing(x, Label = "Fungus")
great <- data.frame(yes)
rm(yes)
TP <- seq(from = 1, to = 400, by = 4)
FN <- seq(from = 2, to = 400, by = 4)
FP <- seq(from = 3, to = 400, by = 4)
TN <- seq(from = 4, to = 400, by = 4)
results <- mapply(c, great[TP], great[FN], great[FP], great[TN])
data <- data.frame(results)
rm(great)
rm(results)
m = ncol(data)
ACC <- matrix(nrow = m, ncol = 1)
SENS <- matrix(nrow = m, ncol = 1)
SPEC <-matrix(nrow = m, ncol = 1)
MCC <- matrix(nrow = m, ncol = 1)
for(i in 1:m){
ACC[i,1] = (data[1,i]+data[4,i])/(data[1,i]+data[2,i]+data[3,i]+data[4,i])*100
SENS[i,1] = (data[4,i])/(data[3,i]+data[4,i])*100
SPEC[i,1] = (data[1,i]/(data[1,i]+data[2,i]))*100
MCC1 = (data[1,i]*data[4,i]) - (data[2,i]*data[3,i])
MCC2 = (data[4,i]+data[2,i])*(data[4,i]+data[3,i])
MCC3 = (data[1,i]+data[2,i])*(data[1,i]+data[3,i])
MCC4 = sqrt(MCC2)*sqrt(MCC3)
MCC[i,1] = MCC1/MCC4
}
rm(TP)
rm(FP)
rm(TN)
rm(FN)
results_ACC <- mean_and_sd(ACC)
results_SENS <- mean_and_sd(SENS)
results_SPEC <- mean_and_sd(SPEC)
results_MCC <- mean_and_sd(MCC)
rm(ACC)
rm(SENS)
rm(SPEC)
rm(MCC)
rm(data)
results_all <- (data.frame(c(results_ACC, results_SENS, results_SPEC, results_MCC)))
rownames(results_all) <- c("ACC_Mean", "ACC_SD", "Sens_Mean", "Sens_SD", "Spec_Mean", "Spec_SD",
"MCC_Mean", "MCC_SD")
rm(results_ACC)
rm(results_SENS)
rm(results_SPEC)
rm(results_MCC)
return(results_all)
}
results_testing_Virus <- function(x) {
yes <- J48_testing(x, Label = "Virus")
great <- data.frame(yes)
rm(yes)
TP <- seq(from = 1, to = 400, by = 4)
FN <- seq(from = 2, to = 400, by = 4)
FP <- seq(from = 3, to = 400, by = 4)
TN <- seq(from = 4, to = 400, by = 4)
results <- mapply(c, great[TP], great[FN], great[FP], great[TN])
data <- data.frame(results)
rm(great)
rm(results)
m = ncol(data)
ACC <- matrix(nrow = m, ncol = 1)
SENS <- matrix(nrow = m, ncol = 1)
SPEC <-matrix(nrow = m, ncol = 1)
MCC <- matrix(nrow = m, ncol = 1)
for(i in 1:m){
ACC[i,1] = (data[1,i]+data[4,i])/(data[1,i]+data[2,i]+data[3,i]+data[4,i])*100
SENS[i,1] = (data[4,i])/(data[3,i]+data[4,i])*100
SPEC[i,1] = (data[1,i]/(data[1,i]+data[2,i]))*100
MCC1 = (data[1,i]*data[4,i]) - (data[2,i]*data[3,i])
MCC2 = (data[4,i]+data[2,i])*(data[4,i]+data[3,i])
MCC3 = (data[1,i]+data[2,i])*(data[1,i]+data[3,i])
MCC4 = sqrt(MCC2)*sqrt(MCC3)
MCC[i,1] = MCC1/MCC4
}
rm(TP)
rm(FP)
rm(TN)
rm(FN)
results_ACC <- mean_and_sd(ACC)
results_SENS <- mean_and_sd(SENS)
results_SPEC <- mean_and_sd(SPEC)
results_MCC <- mean_and_sd(MCC)
rm(ACC)
rm(SENS)
rm(SPEC)
rm(MCC)
rm(data)
results_all <- (data.frame(c(results_ACC, results_SENS, results_SPEC, results_MCC)))
rownames(results_all) <- c("ACC_Mean", "ACC_SD", "Sens_Mean", "Sens_SD", "Spec_Mean", "Spec_SD",
"MCC_Mean", "MCC_SD")
rm(results_ACC)
rm(results_SENS)
rm(results_SPEC)
rm(results_MCC)
return(results_all)
}
J48_testing_all <- function(x) {
bacteria <- results_testing_Bacteria(x)
cancer <- results_testing_Cancer(x)
fungus <- results_testing_Fungus(x)
virus <- results_testing_Virus(x)
results_all <- cbind(bacteria, cancer, fungus, virus)
rm(bacteria)
rm(cancer)
rm(fungus)
rm(virus)
total <- apply(results_all, 1, mean)
results_all_mean <- cbind(results_all, total)
rm(results_all)
rm(total)
colnames(results_all_mean) <- c("Bacteria", "Cancer", "Fungus", "Virus", "Overall")
return(results_all_mean)
}
###Results for training
training_results <- suppressWarnings(J48_training_all(combine_data))
print(training_results)
####Results for CV
CV_results <- suppressWarnings(J48_CV_all(combine_data))
print(CV_results)
####Results for Testing
{r, echo=FALSE, cache = TRUE}
testing_results <- suppressWarnings(J48_testing_all(combine_data))
print(testing_results)