Abstract: In-frame InDel is a common type of insertion and deletion mutations in coding regions, which are closely associated with the occurrence and development of cancer. However, there is currently a lack of clear consensus on the efficacy of computation methods for predicting cancer driver in-frame InDels. In this paper, eight computational methods are comprehensively and systematically compared and evaluated, confirming their applicability and reliability of these methods in identifying cancer driver in-frame InDels. Then, four computational methods with outstanding performance are selected to mine potential driver in-frame InDels in the cancer genome and explore the rationality of these mutations as cancer driver InDels. Finally, a user-friendly online database dbCCID that integrates multiple prediction methods and annotation information is constructed to create convenience for researchers. It is expected this work will provide a theoretical support for the selection and development of in-frame InDel prediction methods for cancer.