基于IBS算法预测亲缘关系准确性研究

Accuracy Research on The Kinship Relationship Prediction by IBS Algorithm

目的评估基于状态一致性(identity-by-state,IBS)算法预测个体间亲缘关系的准确性。方法采用Illumina GSA芯片对253份样本进行全基因组检测,基于高密度单核苷酸多态性(single nucleotide polymorphism,SNP)数据计算两两个体间IBS共享统计量预测亲缘关系。通过不同条件参数筛选SNP,评估位点数对算法预测准确性的影响。结果1~4级亲缘关系预测准确率高达99%,预测误差在1级以内且无假阳性。SNP数量减少对预测准确率无显著影响,即使在较低密度的SNP标记中,该算法也能获得较高的准确率。结论IBS算法是法医系谱推断的有效方法,且对于微量降解的法医现场检材具有很好的应用价值。

Objective To evaluate the accuracy of predicting the kinship relationship between individuals based on the identity-by-state(IBS)algorithm.Methods The Illumina GSA chip was used to perform whole-genome detection on 253 samples.Based on high-density single nucleotide polymorphism(SNP)data,the IBS sharing statistics between two individuals was calculated to predict the kinship relationship.Filtering SNP by different conditional parameters to evaluate the influence of the number of sites on the accuracy of the algorithm’s prediction.Results The prediction accuracy rate of 1st-4th degree of relatives proved to be as high as 99%,with a paired difference of 1st degree and no false positive.Decrease in the number of SNPs has no significant impact on the accuracy of prediction,and the algorithm still achieves a higher accuracy rate even in the lower density of SNP markers.Conclusion The IBS algorithm provides an effective method for forensic genealogy inference,which has good application value for forensic on-site inspection materials with trace degradation.