Comparison of BiClusO with Five Different Biclustering Algorithms Using Biological and Synthetic Data

摘要

Over the past decade, different biclustering techniques have been widely used in analyzing bipartite relationship dataset in biology. According to different comparison studies, the performance of these algorithms vary upon dataset size, pattern, and property which makes it difficult for a researcher to take the right decision for selecting a good biclustering algorithm. In this work, we compare our previously developed biclustering algorithm BiClusO with five different algorithms using biological and synthetic data and evaluate the performances. We use data folding mechanism to convert the biclustering problem to a simple graph clustering problem where polynomial heuristic algorithm DPClusO is used. Using two different scoring methods, the performance of our algorithm is evaluated. Our algorithm shows the best performance over the selected five biclustering algorithms.