Application of Fuzzy c-Means and Self-organizing maps for genes clustering in mouse brain microarray data analysis

Brain is an important organ because it controls organism´s emotions, thinking and living. Many scientists are interested in brain research. Therefore, brain science research has become more and more important. Microarray is the most mature technology for studying the whole genome from specific organs, and hoping to discover new marker genes from them. However, the brain science research still does not have a standard platform to analyze microarray data. In this paper, we proposed a hybrid system by using Fuzzy C-Means (supervised) and Self-organizing map (unsupervised) to analyze microarray data from mice´s with different calcium channel phenotypes. Based on these experimental results, 14 target genes are found and these genes were also confirmed by biological references. The 14 target genes are all related to memory function. In addition, the cluster results from Fuzzy C-Means are 100%, but only 50% for SOM. In the future this method can be applied to more biological datasets for genetic pathways, which can be used in the understanding of directions and strengths of gene regulatory circuit. Consequently, the relationship between calcium channels and the memory and learning function of brains can be further studied.