Multilayer Cluster Heat Map Visualizing Biological Tensor Data

Abstract

AbstractRecent advances in biological high-throughput technology are generating a broad range of omics data. Facing a torrent of massive biological data, visual data mining can be considered an intuitive and powerful approach for hypothesis generation. The cluster heat map approach has been popularly used to visualize the matrix types of biological data. In this study, we extended the use of the cluster heat map to reveal informative patterns hidden in third-order tensor-type biological data. By applying the extended method, a multilayer cluster heat map, to trans-omics and network tensor data, we successfully demonstrated the proof-of-concept of our approach. Our new visual data mining method will be a useful tool for increasing the amount of biological tensor data. Our implementation and the tensor data studied are available from http://www.hgc.jp/~niiyan/MCHM.KeywordsH3k4 MethylationAdjacency MatriceInformative PatternNetwork TensorTensor DataThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.