This article presents an efficient two-stage clustering method for clustering microarray gene expression time series data. The algorithm is based on the identification of genes having significant membership to multiple classes. A recently proposed variable string length genetic scheme and an iterated version of well known fuzzy C-means algorithm are utilized as the underlying clustering techniques. The performance of the two-stage clustering technique has been compared with the hierarchical clustering algorithms, those are widely used for clustering gene expression data, to prove its effectiveness on some publicly available gene expression data.