Detecting Recent Positive Selection with High Accuracy and Reliability by Conditional Coalescent Tree

摘要: Studies of natural selection, followed by functional validation, are shedding light on understanding genetic mechanisms underlying human evolution and adaptation. Classic methods for detecting such as the integrated haplotype score (iHS) Fay Wu's H statistic, useful candidate gene searching positive selection. These methods, however, have limited capability to localize causal variants in selection target regions. In this study, we developed a novel method based conditional coalescent tree detect recent counting unbalanced mutations genealogies. Extensive simulation studies revealed that our is more robust than many other approaches against biases due various demographic effects, including population bottleneck, expansion, or stratification, while not sacrificing its power. Furthermore, demonstrated superiority localizing from massive linked variants. The rate successful localization was about 20-40% higher state-of-the-art simulated data sets. On empirical data, validated four well-known selected genes were all successfully localized method, ADH1B, MCM6, APOL1, HBB. Finally, computational efficiency new much iHS implementations, is, 24-66 times faster REHH package, 10,000 original implementation. magnitudes make suitable applying large sequencing Software can be downloaded https://github.com/wavefancy/scct.