Comparative Advantages of Novel Algorithms Using MSR Threshold and MSR Difference Threshold for Biclustering Gene Expression Data

摘要: The goal of biclustering in gene expression data matrix is to find a submatrix such that the genes show highly correlated activities across all conditions submatrix. A measure called mean squared residue (MSR) used simultaneously evaluate coherence rows and columns within MSR difference incremental increase when or condition added bicluster. In this chapter, three algorithms using threshold (MSRT) (MSRDT) are experimented compared. All these methods use seeds generated from K-Means clustering algorithm. Then enlarged by adding more conditions. first algorithm makes MSRT alone. Both second third make newly introduced concept MSRDT. Highly coherent biclusters obtained concept. algorithm, different method calculate results on bench mark datasets prove better than many metaheuristic algorithms.