Exploratory data analysis (EDA) relies on visualization to reveal patterns, structures, and relationships before formal modeling. As modern datasets increase in size, dimension, and heterogeneity, exposing the association structure becomes challenging. Matrix visualization (MV) methods (heatmaps) address this by arranging samples and variables in a matrix layout, with values encoded as colors to highlight structural patterns. We propose the GAPR package that implements generalized association plots (GAP), a framework for rearranging heatmap layouts through seriation and flipping mechanisms to visualize association structures in data matrices. Written in R with optimized C++ backends, GAPR provides efficient MV for EDA for statisticians and data scientists. Its flexibility and efficiency make it suitable for diverse applications, and it is available on CRAN and GitHub for reproducible research.
Lin et al. (Thu,) studied this question.