Skip to main content

IMAP: Integrated Microbiome Analysis Pipelines: End-to-End Practical User Guides Using Integrated Approaches

3 Exploration of Visualization Types

3.1 Import data 

# Load required packages
library(phyloseq)
library(tidyverse)

cat("\nSaved RData objects\n\n")

Saved RData objects
load("data/external_ps_objects.rda", verbose = T)
Loading objects:
  df_GlobalPatterns
  df_dietswap
  df_caporaso
  df_kostic_crc
  ps_GlobalPatterns
  ps_dietswap
  ps_caporaso
  ps_kostic_crc
load("data/ps_transformed.rda", verbose = T)
Loading objects:
  ps_asin
  ps_identity
  ps_compositional
  ps_z_otu
  ps_z_sample
  ps_log10
  ps_log10p
  ps_clr
  ps_shift
  ps_scale
load("data/bray_distances.rda", verbose = T)
Loading objects:
  ps_asin_bray_dist
  ps_compositional_bray_dist
  ps_z_otu_bray_dist
  ps_z_sample_bray_dist
  ps_log10_bray_dist
  ps_log10p_bray_dist
  ps_clr_bray_dist
  ps_shift_bray_dist
  ps_scale_bray_dist
load("data/psextra_distances.rda", verbose = T)
Loading objects:
  psextra_clr_asin_bray_dist
  psextra_id_asin_bray_dist
  psextra_log10p_asin_bray_dist
load("data/reduced_dimension.rda", verbose = T)
Loading objects:
  pca_asin_bray_metrics
  mds_asin_bray_metrics
  pcoa_asin_bray_metrics
  tsne_asin_bray_metrics
load("data/phyloseq_extra_objects.rda", verbose = T)
Loading objects:
  psextra_clr_dietswap
  psextra_id_dietswap
  psextra_log10p_dietswap
load("data/phyloseq_raw_rel_psextra_df_objects.rda", verbose = T)
Loading objects:
  ps_raw
  ps_rel
  psextra_raw
  psextra_rel
  ps_df

3.2 Major visualization R colors

In R, there are several built-in palettes that we can use for color schemes in plots. Some commonly used palettes include:

[1] "#66C2A5" "#FC8D62" "#8DA0CB" "#E78AC3" "#A6D854" "#FFD92F" "#E5C494"
[8] "#B3B3B3"
[1] "#B3B3B3" "#E5C494" "#FFD92F" "#A6D854" "#E78AC3" "#8DA0CB" "#FC8D62"
[8] "#66C2A5"
[1] "#440154FF" "#46337EFF" "#365C8DFF" "#277F8EFF" "#1FA187FF" "#4AC16DFF"
[7] "#9FDA3AFF" "#FDE725FF"
[1] "#FDE725FF" "#9FDA3AFF" "#4AC16DFF" "#1FA187FF" "#277F8EFF" "#365C8DFF"
[7] "#46337EFF" "#440154FF"
[1] "#FF0000" "#FFBF00" "#80FF00" "#00FF40" "#00FFFF" "#0040FF" "#8000FF"
[8] "#FF00BF"
[1] "#FF00BF" "#8000FF" "#0040FF" "#00FFFF" "#00FF40" "#80FF00" "#FFBF00"
[8] "#FF0000"

  • viridis: A perceptually uniform and colorblind-friendly palette.

  • magma: A palette with a dark-to-light color scheme.

  • inferno: A palette with a light-to-dark color scheme.

  • plasma: A palette with a dark-to-light color scheme.

  • cool: A palette with cool colors.

  • hot: A palette with hot colors.

  • terrain.colors: A palette with colors resembling a terrain map.

  • rainbow: A palette with colors of the rainbow.

  • heat.colors: A palette with colors ranging from dark red to yellow.

  • In ggpubr: “npg”, “aaas”, “lancet”, “jco”, “ucscgb”, “uchicago”, “simpsons” and “rickandmorty”.

3.3 Major visualization techniques

Below are some of the key visualization techniques used in microbiome research, along with their descriptions and the corresponding tools in R.

Visual Type Description
Barplots Display the relative abundances of different taxa across groups.
Heatmaps Represent the abundance or presence/absence of taxa across samples.
Scatter plots Useful for visualizing relationships between numerical variables
Box plots Summarize the distribution of a variable
PCA plots Dimensionality reduction technique for visualizing similarities or dissimilarities between samples based on their microbial composition
Alpha diversity plots Measure the diversity within a sample, e.g. rarefaction plot
Beta diversity plots Measure the dissimilarity between samples, e.g. PCoA ordination
Line plot Visualize changes in the abundance of specific taxa over time or across different conditions.
Network plots Depict interactions or associations between taxa
Volcano plots Identify statistically significant differences in abundance between groups
Correlation plots Visualize correlations between taxa or between taxa and metadata variables
UpSet plots Display intersections of sets and their size in a matrix layout
Venn diagrams Show overlap between taxa or groups
Differential abundance plots Visualize differences in abundance between groups while controlling for confounding factors
Indicator species analysis plots Identify taxa associated with specific groups or conditions