Cancer Identification and Gene Classification using DNA Microarray Gene Expression Patterns

DNA microarray gene expression patterns of several model organisms provide a fascinating opportunity to explore important abnormal biological phenomena. The development of cancer is a multi-step process in which several genes and other environmental and hormonal factors play an important role. In this paper, a new algorithm is proposed to analysis DNA microarray gene expression patterns efficiently for huge amount of DNA microarray data. For better visibility and understanding, experimental results of DNA microarray gene pattern analysis are represented graphically. The shape of each graph corresponding to a DNA microarray gene expression pattern is determined by using an eight-directional chain code sequence, which is invariant to translation, scaling, and rotation. The cancer development is identified based on the variations of DNA microarray gene expression patterns of the same organism by simultaneously monitoring the expression of thousand of genes. At the end, classification of cancer genes is also focused based on the distribution probability of codes of the eight-directional chain code sequences representing DNA microarray gene expression patterns and the experimental result is provided.