應用Adjusted K-means方法選取適當的演化樹

Abstract

演化樹的建立是演化研究裡重要又有趣的問題之一。在文獻中已經有許多有關演化樹建的方法而每個方法都有自己的準則以及演化模型。然而，在建立演化樹過程中，這些準則與演化模型有可能會導致演化樹在拓樸上的誤差。因為已經有許多不同的方法建立演化樹，所以我們所感興趣的是選取可靠的演化樹。在這篇論文裡，我們提出adjusted k-means方法與misclassification error score準則來解決問題。這個方法是利用統計的觀點看資料的特質來評估演化樹。我們應用這個方法在Owlet-Nightjars的實際資料上，顯示出Dumbacher et al.(2003)與其他方法所建構的演化樹，可以達到最小的misclassification error score。

The reconstruction of phylogenetic trees is one of the most important and interesting problems of the evolutionary study. There are many methods proposed in the literature for constructing phylogenetic trees. Each method has its own criterion and bases on a selected evolutionary model. However, the topologies for the trees constructed from different methods may be quite different. The topology error may due to the unsuitable criterion or evolutionary model. Since there are many trees built from different methods, we are interested in selecting a valid tree. In this study, we propose an adjusted k-means approach and a misclassification error score criterion to solve the problem. This approach evaluates the trees by looking at the feature of the data from a statistical point view. It can provide an object criterion to select a valid tree from the statistics perspective. We apply the approach to the real data of phylogeny of the owlet-nightjars. It shows that the phylogeny tree constructed by Dumbacher et al. (2003) can reach minimum misclassification error score compared with the other several methods.