CS seminar series presents
Semantic biclustering: a new way to analyze and interpret gene expression data.
by Jiří Kléma & František Malinka
Thursday, April 7 at 14:00 in 205
The objective of biclustering (or block-clustering, co-clustering) is to find submatrices (biclusters) of a data matrix such that these submatrices exhibit an interesting pattern in their contained values. In our presentation we motivate and define the task of semantic biclustering. In an input binary matrix, in our case a gene expression matrix, the task is to discover homogeneous biclusters allowing joint characterization of the contained elements in terms of knowledge pertaining to both the rows (e.g. genes) and the columns (e.g. biological situations). We propose two approaches to solve the task, based on adaptations of current biclustering, enrichment, and rule and tree learning methods. We compare the approaches in experiments with Drosophila ovary gene expression data under a newly proposed validation protocol stemming from the predictive classification.