Clustering of samples and variables with mixed-type data

Abstract

Analysis of data measured on different scales is a relevant challenge. Biomedical studies often focus on high-throughput datasets of, e.g., quantitative measurements. However, the need for integration of other features possibly measured on different scales, e.g. clinical or cytogenetic factors, becomes increasingly important. The analysis results (e.g. a selection of relevant genes) are then visualized, while adding further information, like clinical factors, on top. However, a more integrative approach is desirable, where all available data are analyzed jointly, and where also in the visualization different data sources are combined in a more natural way. Here we specifically target integrative visualization and present a heatmap-style graphic display. To this end, we develop and explore methods for clustering mixed-type data, with special focus on clustering variables. Clustering of variables does not receive as much attention in the literature as does clustering of samples. We extend the variables clustering methodology by two new approaches, one based on the combination of different association measures and the other on distance correlation. With simulation studies we evaluate and compare different clustering strategies. Applying specific methods for mixed-type data proves to be comparable and in many cases beneficial as compared to standard approaches applied to corresponding quantitative or binarized data. Our two novel approaches for mixed-type variables show similar or better performance than the existing methods ClustOfVar and bias-corrected mutual information. Further, in contrast to ClustOfVar, our methods provide dissimilarity matrices, which is an advantage, especially for the purpose of visualization. Real data examples aim to give an impression of various kinds of potential applications for the integrative heatmap and other graphical displays based on dissimilarity matrices. We demonstrate that the presented integrative heatmap provides more information than common data displays about the relationship among variables and samples. The described clustering and visualization methods are implemented in our R package CluMix available from https://cran.r-project.org/web/packages/CluMix.

Identification of an early diagnostic biomarker of lung adenocarcinoma based on co-expression similarity and construction of a diagnostic model

“…The transformation of this expression pattern also indicates that the progression of the disease affects the function of gene regulation, which leads to some function level changes. The clustering results are visualized using heatmap thermal maps [ 15 ].…”

A Machine Learning Methodology for Diagnosing Chronic Kidney Disease

“…From Tables 10 and 11, it can be seen that the proposed methodology improves the performance of the otherwise independent models and achieves comparable or better performance compared to the models proposed in previous studies. In addition, the CKD data set is composed of mixed variables (numeric and category), so the similarity evaluation methods based on mixed data could be used to calculate the similarity between samples, such as general similarity coefficient [37]. In this study, we used euclidean distance to evaluate the similarity between samples, and KNN could obtain a good result based on euclidean distance with the highest accuracy of 99.25%.…”

Socially engaged calves are more likely to be colonised by VTEC O157:H7 than individuals showing signs of poor welfare

“…Thus, of the 318 observed calves, 191 were included in the analysis. To visualize the data, a hierarchical cluster analysis of variables was performed using the package CluMix 57 . Calves were clustered by similarity using Gowers distance and variables clustered using a combination of association measures according to the CluMix-ama approach 58 .…”

“…To visualize the data, a hierarchical cluster analysis of variables was performed using the package CluMix 57 . Calves were clustered by similarity using Gowers distance and variables clustered using a combination of association measures according to the CluMix-ama approach 58 . The combined effects of each individual risk factors and their association with colonisation of VTEC O157:H7 were analysed using Elastic net regression in the package glmnet (alpha = 0.5) 59 .…”