Advances in next generation sequencing technologies have detected numerous synonymous mutations in the human genome that do not alter amino acids. In recent years, much work have pointed out the important role of synonymous mutations in many human diseases, including psychiatric disease, congenital heart disease, and cancer. However, it is difficult to distinguish disease-associated from benign synonymous mutations. Experimental characterization of all identified synonymous mutations is unpractical and usually time-consuming, costly and labor intensive. Consequently, the scientific community needs computational methods to highlight the most likely deleterious synonymous mutations.

In this study, we develop an ensemble predictor PrDSM (v1.0) by conducting a comprehensive performance evaluation of existing tools for predicting deleterious synonymous mutations. Independent tests demonstrate that the ensemble predictor outperform the current tools.

PrDSM, the pre-computed score is available on here.