MSiP (version 1.3.7)

rfTrain: rfTrain

Description

The labeled feature matrix can be used as input for Random Forest (RF) classifier. The classifier then assigns each bait-prey pair a confidence score, indicating the level of support for that pair of proteins to interact. Hyperparameter optimization can also be performed to select a set of parameters that maximizes the model's performance. This function also computes the areas under the precision-recall (PR) and ROC curve to evaluate the performance of the classifier.

Usage

rfTrain(
  dtInput,
  impute = TRUE,
  p = 0.3,
  parameterTuning = TRUE,
  mtry = seq(from = 1, to = 10, by = 2),
  min_node_size = seq(from = 1, to = 9, by = 2),
  splitrule = c("gini"),
  metric = "Accuracy",
  resampling.method = "repeatedcv",
  iter = 5,
  repeats = 5,
  pr.plot = TRUE,
  roc.plot = TRUE
)

Arguments

dtInput

Data frame containing instances with class labels

impute

Logical value, indicating whether to impute missing values

p

The percentage of data that goes to training; defaults to 0.3

parameterTuning

Logical value; indicating whether to tune rf hyper parameters

mtry

Number of variables to possibly split at in each node and it is bound by the number of variables in your model

min_node_size

Minimal node size

splitrule

Splitrule rule for classification: 'gini', 'extratrees' or 'hellinger' with default 'gini'

metric

A string that specifies what summary metric will be used to select the optimal model; default to Accuracy

resampling.method

The resampling method:'boot', 'boot632', 'optimism_boot', 'boot_all', 'cv', 'repeatedcv', 'LOOCV', 'LGOCV'; defaults to repeatedcv

iter

Number of resampling iterations; defaults to 5

repeats

for repeated k-fold cross validation only; defaults to 5

pr.plot

Logical value, indicating whether to plot precision-recall (PR) curve

roc.plot

Logical value, indicating whether to plot ROC curve

Value

Data frame containing a classification results for all instances in the data set, where positive confidence score corresponds to the level of support for the pair of proteins to be true positive, whereas negative score corresponds to the level of support for the pair of proteins to be true negative.

Examples

# NOT RUN {
data(testdfClassifier)
# }
# NOT RUN {
predidcted_RF <-
rfTrain(testdfClassifier,impute = FALSE, p = 0.3, parameterTuning = FALSE,
mtry  = seq(from = 1, to = 5, by = 1),
min_node_size = seq(from = 1, to = 5, by = 1),
splitrule =c("gini"),metric = "Accuracy",
resampling.method = "cv",iter = 2,repeats = 2,
pr.plot = TRUE, roc.plot = FALSE)
head(predidcted_RF)
# }