In this project, we implemented a software tool, namely PGRANK, which employs the random walk with restart algorithm in a heterogeneous network of genes and diseases. It is developed to overcome the limitations of existing disease gene prediction tools. Beside the capability of prioritization of candidate genes, PGRANK can also rank candidate diseases. Therefore, it can discover not only novel gene-disease associations but also new disease-disease associations. In addition, it can identify novel genes associated with diseases without known molecular basis. Moreover, it is also convenient for users with freedom input of network of genes/proteins. Furthermore, novel promising gene-disease and disease-disease associations can be supported with network- and rank-based visualization as well as evidences and annotations collected from biomedical data. A case study on prediction of novel breast cancer-associated genes and diseases was performed to show the abilities of PGRANK. In addition, we also showed that PGRANK is much better than other tools (i.e., GPEC and PRINCIPLE) in prioritizing candidate disease genes.