bp_kfold_VIP_analysis: K-fold bootstrap and permutation over PLS-VIP

In AlpsNMR: Automated spectraL Processing System for NMR

Description

Bootstrap and permutation over PLS-VIP on AlpsNMR can be performed on both nmr_dataset_1D full spectra as well as nmr_dataset_peak_table peak tables.

Usage

bp_kfold_VIP_analysis(dataset, y_column, k = 4, ncomp = 3, nbootstrap = 300)

Arguments

dataset	An nmr_dataset_family object
y_column	A string with the name of the y column (present in the metadata of the dataset)
k	Number of folds, recomended between 4 to 10
ncomp	number of components for the bootstrap models
nbootstrap	number of bootstrap dataset

Details

Use of the bootstrap and permutation methods for a more robust variable importance in the projection metric for partial least squares regression, in a k-fold cross validation

Value

A list with the following elements:

• important_vips: A list with the important vips selected

• relevant_vips: List of vips with some relevance

• wilcoxon_vips: List of vips that pass a wilcoxon test

• vip_means: Means of the vips scores

• vip_score_plot: plot of the vips scores

• kfold_resuls: results of the k bp_VIP_analysis

• kfold_index: list of index of partitions of the folds

Examples

# Data analysis for a table of integrated peaks

## Generate an artificial nmr_dataset_peak_table:
### Generate artificial metadata:
num_samples <- 64 # use an even number in this example
num_peaks <- 20
metadata <- data.frame(
    NMRExperiment = as.character(1:num_samples),
    Condition = rep(c("A", "B"), times = num_samples/2),
    stringsAsFactors = FALSE
)

### The matrix with peaks
peak_means <- runif(n = num_peaks, min = 300, max = 600)
peak_sd <- runif(n = num_peaks, min = 30, max = 60)
peak_matrix <- mapply(function(mu, sd) rnorm(num_samples, mu, sd),
                                            mu = peak_means, sd = peak_sd)
colnames(peak_matrix) <- paste0("Peak", 1:num_peaks)
rownames(peak_matrix) <- paste0("Sample", 1:num_samples)

## Artificial differences depending on the condition:
peak_matrix[metadata$Condition == "A", "Peak2"] <- 
    peak_matrix[metadata$Condition == "A", "Peak2"] + 70

peak_matrix[metadata$Condition == "A", "Peak6"] <- 
    peak_matrix[metadata$Condition == "A", "Peak6"] - 60
    
### The nmr_dataset_peak_table
peak_table <- new_nmr_dataset_peak_table(
    peak_table = peak_matrix,
    metadata = list(external = metadata)
)

## We will use bootstrap and permutation method for VIPs selection 
## in a a k-fold cross validation 
bp_results <- bp_kfold_VIP_analysis(peak_table, # Data to be analized
                           y_column = "Condition", # Label
                           k = 3,
                           nbootstrap = 10)

message("Selected VIPs are: ", bp_results$importarn_vips)

AlpsNMR documentation built on April 1, 2021, 6:02 p.m.