Abstracts (first author)
Statistical test to reliably detect recent positive selection by utilizing tree shape and branch lengths
Many methods have been proposed in last two decades to search for traces of recent positive selection and the “maximum frequency of derived mutations” (MFDM) test, which is only based on the unbalanced tree topology, is theoretically insensitive to demography, including bottlenecks and expansions. In this study, we analyzed the statistical features of unbalanced tree in the neutral model and further proposed a new neutrality test that integrates the information of tree topology and the branch lengths of unbalanced tree. Simulations based on wide-range parameters demonstrated that the new test is robust to different demographic models and has a high statistical power to detect recent positive selection. Especially, this test does not need the information of outgroup, and thus will not be affected by the misinference of derived and ancestral variants of segregating sites due to multiple hits. This work provides a reliable method that can distinguish selection from demography when single-locus or genome-wide DNA polymorphism data is available.