Abstract
Full-sibling (FS) identification is a crucial aspect of kinship analysis. However, current FS identification methods need to be more accurate and effective for routine FS identification. In this study, the optimal cut-off values for FS identification by incorporating both the identical by state (IBS) and likelihood ratio (LR) methods under four different levels of error rates were assessed. The corresponding sensitivity, specificity, accuracy, and effectiveness were calculated for different numbers of short tandem repeats (STRs), ranging from 19 to 55. The identified cut-off values were validated using authentic pedigrees from the Han population in East China. It was determined that the cut-off values derived from the IBS method and LR method with an error rate below 0.01% were dependable and practical. The study also explored the impact of half-sibling relationships and the inclusion of reference relatives in FS identification. Currently, an updated technical specification for FS identification in China was released, serving as a valuable resource for FS testing in forensic applications.