Abstract: Apoptosis proteins have a central role in the development and homeostasis of an organism. These proteins are very important for understanding the mechanism of programmed cell death. A new encoding method based on tri-polypeptide composition is presented. By use of adjacent triune residues contents in the protein primary sequences, the increment of diversity is calculated to predict the subcellular location of apoptosis proteins. The content-balancing accuracy index presented by Zhang CT is extended to solve any classification problem. The experiment results show that for apoptosis protein subcellular location prediction, the method of tri-polypeptide diversity source can not only improve the overall prediction accuracy, but also solve the imbalance problem of samples. While the content-balancing accuracy index is much superior to the widely used overall prediction accuracy for evaluate prediction algorithms.