Gene expression data analysis is used in several areas including drug discovery and clinical applications. Biclustering in gene expression data is a subset of the genes representing consistent patterns over a subset of the conditions. In this case the conditions can be related to the disease types, the biclustering method is much hopeful in this application field. The proposed work finds the significant biclusters in large expression data using modified the Nelder-Mead method. The Nelder-Mead minimises a function of n parameters by comparing the n + 1 vertices of a simplex, and updating the worst vertex by moving it around a centroid. This work considers the median instead of centroid and differential evolution that takes place in the simplex to get the true global minimum. It is tested on benchmark datasets and the results are compared with standard benchmark algorithms. The results indicate that there is a substantial betterment in the purported study. © 2016 Inderscience Enterprises Ltd.