linda.plot: Plot linda results

In zhouhj1994/LinDA: Linear model for differential abundance analysis

Plot linda results

Description

The function plots the effect size plot and volcano plot based on the output from linda.

Usage

linda.plot(
  linda.obj,
  variables.plot,
  titles = NULL,
  alpha = 0.05,
  lfc.cut = 1,
  legend = FALSE,
  directory = NULL,
  width = 11,
  height = 8
)

Arguments

linda.obj	return from function linda.
variables.plot	vector; variables whose results are to be plotted. For example, suppose the return value variables is equal to ('x1', 'x2', 'x3b', 'x3c', 'x1:x2'), then one could set variables.plot = c('x3b', 'x1:x2').
titles	vector; titles of the effect size plot and volcano plot for each variable in variables.plot. The default is NULL. If NULL, the titles will be set as variables.plot.
alpha	a real value between 0 and 1; cutoff for padj.
lfc.cut	a positive value; cutoff for log2FoldChange.
legend	TRUE or FALSE; whether to show the legends of the effect size plot and volcano plot.
directory	character; the directory to save the figures, e.g., getwd(). The default is NULL. If NULL, figures will not be saved.
width	the width of the graphics region in inches. See R function pdf.
height	the height of the graphics region in inches. See R function pdf.

Value

A list of ggplot2 objects.

plot.lfc	a list of effect size plots. Each plot corresponds to one variable in variables.plot.
plot.volcano	a list of volcano plots. Each plot corresponds to one variable in variables.plot.

Author(s)

Huijuan Zhou [e-mail] Jun Chen [e-mail] Xianyang Zhang [e-mail]

References

Huijuan Zhou, Kejun He, Jun Chen, and Xianyang Zhang. LinDA: Linear Models for Differential Abundance Analysis of Microbiome Compositional Data.

Examples

#install package "phyloseq" for importing "smokers" dataset
ind <- smokers$meta$AIRWAYSITE == 'Throat'
otu.tab <- as.data.frame(smokers$otu[, ind])
meta <- cbind.data.frame(Smoke = factor(smokers$meta$SMOKER[ind]),
                         Sex = factor(smokers$meta$SEX[ind]),
                         Site = factor(smokers$meta$SIDEOFBODY[ind]),
                         SubjectID = factor(smokers$meta$HOST_SUBJECT_ID[ind]))
ind1 <- which(meta$Site == 'Left')
res.left <- linda(otu.tab[, ind1], meta[ind1, ], formula = '~Smoke+Sex', alpha = 0.1,
                  prev.cut = 0.1, lib.cut = 1000, winsor.quan = 0.97)
ind2 <- which(meta$Site == 'Right')
res.right <- linda(otu.tab[, ind2], meta[ind2, ], formula = '~Smoke+Sex', alpha = 0.1,
                  prev.cut = 0.1, lib.cut = 1000, winsor.quan = 0.97)
rownames(res.left$output[[1]])[which(res.left$output[[1]]$reject)]
rownames(res.right$output[[1]])[which(res.right$output[[1]]$reject)]

linda.obj <- linda(otu.tab, meta, formula = '~Smoke+Sex+(1|SubjectID)', alpha = 0.1,
                   prev.cut = 0.1, lib.cut = 1000, winsor.quan = 0.97)
linda.plot(linda.obj, c('Smokey', 'Sexmale'),
           titles = c('Smoke: n v.s. y', 'Sex: female v.s. male'), alpha = 0.1, lfc.cut = 1,
           legend = TRUE, directory = NULL, width = 11, height = 8)

zhouhj1994/LinDA documentation built on Jan. 30, 2024, 5:54 p.m.