Quantitative simultaneous monitoring of the expression levels of thousands of genes under various experimental conditions is now possible using microarrays. However, there are still gaps toward wholegenome functional annotation of genes using gene expression data. In this paper, we propose a novel technique called Fuzzy Nearest Clusters for functional annotation of unclassified genes. The technique consists of two steps: a hierarchical clustering step to detect homogeneous co-expressed gene clusters in each possibly heterogeneous functional class; followed by a classification step to predict the functional roles of unclassified genes based on their similarities to these clusters. Our experimental results with yeast gene expression data showed that the proposed method can accurately predict the genes' functions, even those with multiple functional roles, and the performance is most independent of the heterogeneity of the complex functional classes, as compared to other approaches.