Biclustering in gene-expression data is a subset of the genes demonstrating consistent patterns over a subset of the conditions. Recently, the most of research in biclustering involving statistical and graph-theoretic approaches by adding or deleting rows and/or columns in the data matrix based on some constraints. This is an exhaustive search of the space, and hence the solutions may not be feasible. The proposed work finds the significant biclusters in large expression data using shuffled cuckoo search with Nelder–Mead (SCS-NM). The diversification and intensification of the search space are obtained through shuffling and simplex NM, respectively. The proposed work is tested on four benchmark datasets, and the results are compared with the swarm intelligence techniques and the various biclustering algorithms. The results show that there is significant improvement in the fitness value of proposed work SCS-NM. In addition, the work determines the biological relevance of the biclusters with Gene Ontology in terms of function, process and component. © 2018, © 2018 Taylor & Francis.