In a gene expression microarray data set, there could be tens or hundreds of dimensions, each of which corresponds to an experimental condition. But handling large value data set is a time consuming task, so we have introduced the fuzzy concept in order to discretize the dataset within the range 0 to 1. For our experiment, the gene expression data has been considered because of the common challenges. The first one is the curse of dimensionality. With increasing dimensionality, this data set becomes computationally intractable and therefore inapplicable in many real applications. Secondly, the specificity of similarities between points in a high dimensional data diminishes. It was proven that: for any point in a high dimensional data, the expected gap between the Euclidean distance to the closest neighbour and that to the farthest point shrinks as the dimensionality grows. In this paper, a key step in the analysis of gene expression data is the clustering using A-means and fuzzy A-means which group the genes that manifest similar expression patterns. From the experimental evaluation: it has been analyzed that fuzzy Jfc-mcans shows better performance. The optimization techniques like Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) for clustering on gene expression data has been used because: many mathematical features that make them attractive for gene expression analysis including their flexibility in choosing a similarity function, when dealing with large data set. As compared to all the techniques it has been observed that the number of clusters is more and runtime of fuzzy k- means is less and it provides crisp boundary among the clusters; and the application of soft computing techniques shows better result than the traditional algorithms. © 2012 Published by Elsevier Ltd.