In this week’s biology seminar, the invited speaker Prof. Hidetoshi Shimodaira introduced his recent work about selective inference. Selective inference is an important statistical problem described also by another word “the file-drawer effect”. For example, Journals are much more likely to publish studies with low P values, and the readers never hear about the great number of studies that showed no effect and were filed way. This makes it difficult to assess the strength of a reported P value. The challenge of correcting for the effects of selection is a complex one. Prof. Hidetoshi Shimodaira explained a method to improve the previously proposed approximately unbiased test by adjusting the selection bias. This method is applied to predict trees and edges in phylogeny. In the seminar, he started by an interesting introduction of Darwin's theory of evolution and the tree of life. Based on this theory, modern phylogenetic inference is developed by analyzing DNA sequences of species as system relationships like trees. He also showed us, different statistical tests can give different P values of trees and edges in the trees. Therefore, we should be careful doing the tests and aware of the bias types in the problem.
We thank Prof. Hidetoshi Shimodaira for his great talk and precious time with us!

Reported by Yingying Xu