partVar_plot: Partitioned Variance Plot

In EvaYiwenWang/PLSDAbatch: PLSDA-batch

Partitioned Variance Plot

Description

This function draws a partitioned variance plot explained by different sources.

Usage

partVar_plot(
    prop.df,
    text.cex = 3,
    x.angle = 60,
    x.hjust = 1,
    title = NULL,
    color.set = NULL
)

Arguments

prop.df	A data frame that contains the proportion of variance explained by different sources.
text.cex	Numeric, the size of text on the plot.
x.angle	Numeric, angle of x axis, in the range of 0 to 360.
x.hjust	Numeric, horizontal justification of x axis, in the range of 0 to 1.
title	Character, the plot title.
color.set	A vector of characters, indicating the set of colors to use. The colors are represented by hexadecimal color code.

Value

None.

Author(s)

Yiwen Wang, Kim-Anh Lê Cao

See Also

Scatter_Density, box_plot, density_plot and alignment_score as the other methods for batch effect detection and batch effect removal assessment.

Examples

## First example
library(vegan) # for function varpart()
library(TreeSummarizedExperiment) # for functions assays(),rowData()
data('AD_data')
# centered log ratio transformed data
ad.clr <- assays(AD_data$EgData)$Clr_value
ad.batch <- rowData(AD_data$EgData)$Y.bat # batch information
ad.trt <- rowData(AD_data$EgData)$Y.trt # treatment information
names(ad.batch) <- names(ad.trt) <- rownames(AD_data$EgData)

ad.factors.df <- data.frame(trt = ad.trt, batch = ad.batch)
rda.res <- varpart(ad.clr, ~ trt, ~ batch,
                    data = ad.factors.df, scale = TRUE)

ad.prop.df <- data.frame(Treatment = NA, Batch = NA,
                            Intersection = NA,
                            Residuals = NA)
ad.prop.df[1,] <- rda.res$part$indfract$Adj.R.squared

ad.prop.df <- ad.prop.df[, c(1,3,2,4)]

ad.prop.df[ad.prop.df < 0] <- 0
ad.prop.df <- as.data.frame(t(apply(ad.prop.df, 1, function(x){x/sum(x)})))

partVar_plot(prop.df = ad.prop.df)

## Second example
# a list of data corrected from different methods
ad.corrected.list <- assays(AD_data$CorrectData)
ad.prop.df <- data.frame(Treatment = NA, Batch = NA,
                            Intersection = NA,
                            Residuals = NA)
for(i in seq_len(length(ad.corrected.list))){
    rda.res <- varpart(ad.corrected.list[[i]], ~ trt, ~ batch,
                    data = ad.factors.df, scale = TRUE)
    ad.prop.df[i, ] <- rda.res$part$indfract$Adj.R.squared}

rownames(ad.prop.df) <- names(ad.corrected.list)

ad.prop.df <- ad.prop.df[, c(1,3,2,4)]

ad.prop.df[ad.prop.df < 0] <- 0
ad.prop.df <- as.data.frame(t(apply(ad.prop.df, 1,
                                    function(x){x/sum(x)})))

partVar_plot(prop.df = ad.prop.df)

EvaYiwenWang/PLSDAbatch documentation built on Sept. 25, 2024, 8:54 p.m.