A novel biclustering of gene expression data based on hybrid BAFS-BSA algorithm

As one of usual concepts, co-expressed genes can represent co-regulated genes in gene expression data. This strategy can be refined further because co-expression of the genome may be the result of independent activation under same experimental samples, rather than the same regulatory regime. Therefore, traditional clustering techniques are proposed to find significant clusters, especially, the biclustering technology. By combining Binary Artificial Fish Swarm (BAFS) with Binary Simulated Annealing (BSA) algorithms, the hybrid algorithm named BAFS-BSA-BIC was proposed in this paper. When this method of biclustering was applied to several datasets, lots of biological significant bifclusters were searched, and the results demonstrate the promising clustering performance of our method. The proposed technology was also compared to classical biclustering technologies-CC, QUBIC, FLOC and original BAFS algorithm, and its robustness and quality are better than these algorithms in searching optimal biclusters of co-expressed genes.