The aim of this project is to automatically identify the relevant cluster for micro array gene expression data sets to keep high qualitative system update of newly recorded data. However, the selection of the most representative feature of the strongly related target clusters is needed for the appropriate selection of subset of features. Features are divided into clusters by using graph-theoretic clustering methods such as the minimum cost spanning tree clustering method, which ensures the identification of the exact cause of immune deficiency disorder in PIDs. A modified version of vector quantization is used in Flexible Fuzzy Inference System (FLEXFIS) specifically for Takagi Sugeno fuzzy model to identify the appropriate cluster. A fast clustering-based feature selection algorithm for subset selection is proposed. This strategy has a high probability of producing a subset of useful and independent features specific to the PIDs. The system adopts the minimum-spanning tree (MST) clustering method to ensure efficiency. It builds classifiers on the basis of the decision rules arising from these genes or gene pairs. This gained high-quality solutions to classification problems in the analysis of high-dimensional gene expression. A method is needed to select clusters genes and conditions simultaneously, finding distinctive clusters with less number of rules generated. An evaluation is done on the microarray gene expression data for the purpose of identification of the subset selection. Feature Selection is the process of selecting a subset of relevant features (variables, predictors) for use in model construction. Here fast Algorithm is used for selection of features. In FAST algorithm clustering based strategy is used where the data are clustered according to the irrelevant features. 35 publicly available real-world high-dimensional image, microarray, and text data are used for clustering.