plot_binary_RF: Plot binary rule-based heatmaps

In NourMarzouka/multiclassPairs: Build MultiClass Pair-Based Classifiers using TSPs or RF

Plot binary rule-based heatmaps

Description

plot_binary_RF Plot binary heatmaps for datasets based on rule-based random forest classifier

Usage

  plot_binary_RF(Data,
                  classifier,
                  ref              = NULL,
                  prediction       = NULL,
                  as_training      = FALSE,
                  platform         = NULL,
                  classes          = NULL,
                  platforms_ord    = NULL,
                  top_anno         = c("ref", "prediction", "platform")[1],
                  title            = "",
                  binary_col       = c("white", "black", "gray"),
                  ref_col          = NULL,
                  pred_col         = NULL,
                  platform_col     = NULL,
                  show_ref         = TRUE,
                  show_predictions = TRUE,
                  show_platform    = TRUE,
                  show_scores      = TRUE,
                  show_rule_name   = TRUE,
                  legend           = TRUE,
                  cluster_cols     = TRUE,
                  cluster_rows     = TRUE,
                  anno_height      = 0.03,
                  score_height     = 0.03,
                  margin           = c(0, 5, 0, 5))

Arguments

Data	a matrix or a dataframe, samples as columns and row as features/genes. Can also take ExpressionSet, or data_object generated by ReadData function.
classifier	Classifier as a rule_based_RandomForest object, generated by train_RF function
ref	Optional vector with the reference labels. Ref labels in data_object will be used if not ref input provided. For ExpressionSet, the name of the ref variable in the pheno data can be used.
prediction	Optional. "ranger.prediction" object for the class scores generated by predict_RF function.
as_training	Logical indicates if the plot is for the training data. It means the predictions will be extracted from the classifier itself and any prediction object will be ignored. If TRUE, then the training data/object should be used for Data argument.
platform	Optional vector with the platform/study labels or any additional annotation. Platform labels in data_object will be used if no platform input is provided. For ExpressionSet, the name of the variable in the pheno data can be used.
classes	Optional vector with class names. This will determine which classes will be plotted and in which order. It is not recommended to use both "classes" and "platforms_ord" arguments together to avoid sample order conflict and may result in an improper plotting for samples.
platforms_ord	Optional vector with the platform/study names. This will determine which platform/study will be plotted and in which order. This will be used when top_anno="platform". It is not recommended to use both "classes" and "platforms_ord" arguments together.
top_anno	Determine the top annotation level. Samples will be grouped based on the top_anno. Input can be one of three options: "ref", "prediction", "platform". Default is "ref".
title	Character input as a title for the whole heatmap. Default is "".
binary_col	Vector determines the colors of the binary heatmap. Default is c("white", "black", "gray"). First color for the color when the rule is false in the sample. Second color for the color when the rule is true. Third color is for NAs.
ref_col	Optional named vector determines the colors of classes for the reference labels. Default is NULL. Vector names should match with the ref labels.
pred_col	Optional named vector determines the colors of classes for the prediction labels. Default is NULL. Vector names should match with the prediction labels in the prediction labels.
platform_col	Optional named vector determines the colors of platforms/study labels. Default is NULL. Vector names should match with the platforms/study labels.
show_ref	Logical. Determines if the ref labels will be plotted or not. If the top_anno argument is "ref" then show_ref will be ignored and ref labels will be plotted.
show_predictions	Logical. Determines if the prediction labels will be plotted or not. If the top_anno argument is "prediction" then show_predictions will be ignored and predictions will be plotted.
show_platform	Logical. Determines if the platform/study labels will be plotted or not. If the top_anno argument is "platform" then show_platform will be ignored and platforms will be plotted.
show_scores	Logical. Determines if the prediction scores will be plotted or not. To visualize scores, the classifier should be trained with probability=TRUE otherwise show_scores will be turned FALSE automatically.
show_rule_name	Logical. Determines if the rule names will be plotted on the left side of the heatmapp or not.
legend	Logical. Determines if a legend will be plotted under the heatmap.
cluster_cols	Logical. Clustering the samples in each class (i.e. not all samples in the cohort) based on the binary rules for that class. If top_anno is "platform" then the rules from all classes are used to cluster the samples in each platform.
cluster_rows	Logical. Clustering the rules in each class.
anno_height	Determines the height of the annotations. It is recommended not to go out of this range 0.01<height<0.1. Default is 0.03.
score_height	Determines the height of the score bars. It is recommended not to go out of this range 0.01<height<0.1. Default is 0.03.
margin	Determines the margins of the heatmap. Default is c(0, 5, 0, 5).

Value

returns a heatmap plot for the binary rule

Author(s)

Nour-al-dain Marzouka <nour-al-dain.marzouka at med.lu.se>

Examples

# generate random data
Data <- matrix(runif(8000), nrow=100, ncol=80,
               dimnames = list(paste0("G",1:100), paste0("S",1:80)))

# generate random labels
L <- sample(x = c("A","B","C","D"), size = 80, replace = TRUE)

# generate random platform labels
P <- sample(c("P1","P2","P3"), size = 80, replace = TRUE)

# create data object
object <- ReadData(Data = Data,
                   Labels = L,
                   Platform = P,
                   verbose = FALSE)

# sort genes
genes_RF <- sort_genes_RF(data_object = object,
                          seed=123456, verbose = FALSE)

# to get an idea of how many genes we will use
# and how many rules will be generated
# summary_genes_RF(sorted_genes_RF = genes_RF,
#                  genes_altogether = c(10,20,50,100,150,200),
#                  genes_one_vs_rest = c(10,20,50,100,150,200))

# creat and sort rules
# rules_RF <- sort_rules_RF(data_object = object,
#                           sorted_genes_RF = genes_RF,
#                           genes_altogether = 100,
#                           genes_one_vs_rest = 100,
#                           seed=123456,
#                           verbose = FALSE)

# parameters <- data.frame(
#   gene_repetition=c(3,2,1),
#   rules_one_vs_rest=0,
#   rules_altogether=c(2,3,10),
#   run_boruta=c(FALSE,"produce_error",FALSE),
#   plot_boruta = FALSE,
#   num.trees=c(100,200,300),
#   stringsAsFactors = FALSE)
# parameters

# Or you can use expand.grid to generate dataframe with all parameter combinations
# parameters <- expand.grid(
#   gene_repetition=c(3,2,1),
#   rules_one_vs_rest=0,
#   rules_altogether=c(2,3,10),
#   num.trees=c(100,500,1000),
#   stringsAsFactors = FALSE)
# parameters


# test <- optimize_RF(data_object = object,
#                     sorted_rules_RF = rules_RF,
#                     test_object = NULL,
#                     overall = c("Accuracy"),
#                     byclass = NULL, verbose = FALSE,
#                     parameters = parameters)
# test
# test$summary[which.max(test$summary$Accuracy),]
#
# # train the final model
# # it is preferred to increase the number of trees and rules in case you have
# # large number of samples and features
# # for quick example, we have small number of trees and rules here
# # based on the optimize_RF results we will select the parameters
# RF_classifier <- train_RF(data_object = object,
#                           gene_repetition = 1,
#                           rules_altogether = 0,
#                           rules_one_vs_rest = 10,
#                           run_boruta = FALSE,
#                           plot_boruta = FALSE,
#                           probability = TRUE,
#                           num.trees = 300,
#                           sorted_rules_RF = rules_RF,
#                           boruta_args = list(),
#                           verbose = TRUE)
#
# # training accuracy
# # get the prediction labels
# # if the classifier trained using probability	= FALSE
# training_pred <- RF_classifier$RF_scheme$RF_classifier$predictions
# if (is.factor(training_pred)) {
#   x <- as.character(training_pred)
# }
#
# # if the classifier trained using probability	= TRUE
# if (is.matrix(training_pred)) {
#   x <- colnames(training_pred)[max.col(training_pred)]
# }
#
# # training accuracy
# caret::confusionMatrix(data =factor(x),
#                 reference = factor(object$data$Labels),
#                 mode = "everything")

# not to run
# visualize the binary rules in training dataset
# plot_binary_RF(Data = object,
#                classifier = RF_classifier,
#                prediction = NULL, as_training = TRUE,
#                show_scores = TRUE,
#                top_anno = "ref",
#                show_predictions = TRUE,
#                title = "Training data")

# not to run
# Extract and plot the proximity matrix from the classifier for the training data
# it takes long time for large data
# proximity_mat <- proximity_matrix_RF(object = object,
#                       classifier = RF_classifier,
#                       plot=TRUE,
#                       return_matrix=TRUE,
#                       title = "Test",
#                       cluster_cols = TRUE)

# not to run
# predict
# test_object # any test data
# results <- predict_RF(classifier = RF_classifier, impute = TRUE,
#                       Data = test_object)
#
# # visualize the binary rules in training dataset
# plot_binary_RF(Data = test_object,
#                classifier = RF_classifier,
#                prediction = results, as_training = FALSE,
#                show_scores = TRUE,
#                top_anno = "ref",
#                show_predictions = TRUE,
#                title = "Test data")

NourMarzouka/multiclassPairs documentation built on May 3, 2023, 7:20 p.m.