12  Community composition

12.1 Composition barplot

A typical way to visualize microbiome composition is by using a composition barplot which shows the relative abundances of selected taxonomic features. In the following code chunk, relative abundance is calculated, and top taxonomic features are retrieved for the phylum rank. Then, the barplot is visualized ordering rank by abundance values and samples by “Bacteroidetes”:

library(miaViz)
data("GlobalPatterns")
tse <- GlobalPatterns

# Computing relative abundance
tse <- transformAssay(tse, assay.type = "counts", method = "relabundance")

# Getting top taxa on a Phylum level
tse <- agglomerateByRank(tse, rank = "Phylum")
top_taxa <- getTop(tse, top = 10, assay.type = "relabundance")

# Renaming the "Phylum" rank to keep only top taxa and assign the rest to "Other"
phylum_renamed <- lapply(rowData(tse)$Phylum, function(x) {
    if (x %in% top_taxa) {
        x
    } else {
        "Other"
    }
})
rowData(tse)$Phylum_sub <- as.character(phylum_renamed)

# Agglomerate the data based on specified taxa
tse_sub <- agglomerateByVariable(tse, by = "rows", f = "Phylum_sub")

# Visualizing the composition barplot, with samples ordered by "Bacteroidetes"
plotAbundance(
    tse_sub,
    assay.type = "relabundance",
    order.row.by = "abund", order.col.by = "Bacteroidetes"
)

12.2 Composition heatmap

Community composition can be visualized with a heatmap, where the horizontal axis represents samples and the vertical axis the taxonomic features. The color of each intersection point represents abundance of a taxon in a specific sample.

Here, abundances are first transformed with CLR (centered log-ratio) to remove compositionality bias. Then standardization transformation is applied to the CLR-transformed data. This shifts all taxonomic features to zero mean and unit variance, allowing visual comparison between taxonomic features that have different absolute abundance levels. For more information on transformation methods, refer to Chapter 11.

tse <- GlobalPatterns

# Agglomerate to phylum level
tse <- agglomerateByPrevalence(tse, rank = "Phylum")

# Add clr-transformation on samples
tse <- transformAssay(
    tse,
    assay.type = "counts", method = "relabundance", pseudocount = 1
)
tse <- transformAssay(tse, assay.type = "relabundance", method = "clr")

# Add scale features (taxa)
tse <- transformAssay(
    tse,
    assay.type = "clr", MARGIN = "rows", method = "standardize",
    name = "clr_z"
)

We can visualize the heatmap with the sechm package. It is a wrapper for the ComplexHeatmap package (Gu 2022).

library(sechm)

# Plot heatMap with sechm
heatmap <- sechm(
    tse,
    assayName = "clr_z", features = rownames(tse),
    show_rownames = TRUE, show_colnames = TRUE,
    row_names_gp = gpar(fontsize = 6), column_names_gp = gpar(fontsize = 8),
    breaks = 1, hmcols = c("blue", "white", "red")
)

heatmap

Another method to visualize community composition is by plotting a NeatMap, which employs radial theta sorting when plotting the heatmap (Rajaram and Oono 2010). The getNeatOrder() function in the miaViz package allows us to achieve this. This method sorts data points based on their angular position in a 2D space, typically after an ordination technique such as PCA or NMDS has been applied. The getNeatOrder() method calculates the angle (theta) for each point relative to the centroid and sorts data points based on these theta values in ascending order. This approach preserves the relationships between data points according to the ordination method’s spatial configuration, rather than relying on hierarchical clustering.

Rajaram, Satwik, and Yoshi Oono. 2010. “NeatMap - Non-Clustering Heat Map Alternatives in R.” BMC Bioinformatics 11: 45. https://doi.org/10.1186/1471-2105-11-45.

Now, we’ll perform the aforementioned steps to create a NeatMap using the sechm package and the getNeatOrder() function.

library(scater)
library(sechm)

# Perform PCA on the dataset
tse <- runPCA(tse, ncomponents = 10, assay.type = "clr_z")

# Sort by radial theta using the first two principal components
sorted_order <- getNeatOrder(
    reducedDim(tse, "PCA")[, c(1, 2)],
    centering = "mean"
)
tse <- tse[, sorted_order]

# Plot NeatMap with sechm
neatmap <- sechm(
    tse,
    assayName = "clr_z", features = rownames(tse),
    show_rownames = TRUE, show_colnames = TRUE,
    do.scale = FALSE, cluster_rows = FALSE, sortRowsOn = NULL,
    row_names_gp = gpar(fontsize = 6), column_names_gp = gpar(fontsize = 8),
    breaks = 1, hmcols = c("blue", "white", "red")
)

neatmap

TipAdditional heatmap visualization

In addition, there are also other packages that provide functions for more complex heatmaps, such as those provided by iheatmapr and ComplexHeatmap (Gu 2022). The utilization of ComplexHeatmap for clustered heatmaps is explained in Section 15.3.

Gu, Zuguang. 2022. “Complex Heatmap Visualization.” iMeta 1 (3): e43. https://doi.org/https://doi.org/10.1002/imt2.43.
TipExercises

Goal: After this exercise, you are able to visualize a summary of microbial composition.

Exercise 1: Community composition

  1. Load any of the example datasets mentioned in Section 4.2.

  2. Agglomerate data to phylum level.

  3. Transform data to relative abundances.

  4. Visualize community composition with a relative abundance bar plot. Which taxonomic features are most prevalent?

  5. Explore colData. Select the sample metadata variable and visualize the composition with a bar plot in respect to sample grouping. Does the community composition associate with the grouping?

  6. Normalize abundances to a standardized scale, i.e., calculate Z-scores.

  7. Plot normalized abundances with a heatmap.

Useful functions:

utils::data(), mia::agglomerateByRank(), mia::transformAssay(), miaViz::plotAbundance(), SummarizedExperiment::colData(), sechm::sechm()

Back to top