Advanced Heat Map and Clustering Analysis Using Heatmap3

TLDR

Heat maps and clustering are widely used in expression analysis for visualization and quality control, but the base R ‘heatmap’ function lacks advanced customizability and dendrogram features. To address these limitations, the authors developed the R package ‘heatmap3’, which extends the original ‘heatmap’ function with powerful, convenient features. Built on the original ‘heatmap’ function, ‘heatmap3’ enables highly customizable heat maps and dendrograms, offering adjustable legends, side annotations, a broader color palette, multi‑layer phenotype labeling, automatic association tests, and alternative agglomeration methods for clustering.

Abstract

Heat maps and clustering are used frequently in expression analysis studies for data visualization and quality control. Simple clustering and heat maps can be produced from the “heatmap” function in R. However, the “heatmap” function lacks certain functionalities and customizability, preventing it from generating advanced heat maps and dendrograms. To tackle the limitations of the “heatmap” function, we have developed an R package “heatmap3” which significantly improves the original “heatmap” function by adding several more powerful and convenient features. The “heatmap3” package allows users to produce highly customizable state of the art heat maps and dendrograms. The “heatmap3” package is developed based on the “heatmap” function in R, and it is completely compatible with it. The new features of “heatmap3” include highly customizable legends and side annotation, a wider range of color selections, new labeling features which allow users to define multiple layers of phenotype variables, and automatically conducted association tests based on the phenotypes provided. Additional features such as different agglomeration methods for estimating distance between two samples are also added for clustering.