Abstract: A fast bi-clustering algorithm, Bi-EB, is developed to detect the local pattern of integrated data, typically multi-omics biclusters. Bi-EB adopts a data-driven statistics strategy by using Expected-Maximum algorithm to extract the foreground bicluster pattern from its background noise data in an iterative search. Bi-EB has a higher recovery and relevance than all other seven biclustering algorithms in searching scale-shifted row and column biclusters, and outperforms algorithms CC, spectral and xMotif in constant bicluster pattern.

Learning Objective 1: Learn about a new bicluster algorithm which can incorporate multi-omics data and apply this algorithm in their research.


Aida Yazdanparast (Presenter)
Indiana University

Lang Li, Indiana University
Lijun Cheng, Indiana University

Presentation Materials: