Ronquist Fredrik, Larget Bret, Huelsenbeck John P, Kadane Joseph B, Simon Donald, van der Mark Paul
School of Computational Science, Florida State University, Tallahassee, FL 32306-4120, USA.
Science. 2006 Apr 21;312(5772):367; author reply 367. doi: 10.1126/science.1123622.
Mossel and Vigoda (Reports, 30 September 2005, p. 2207) show that nearest neighbor interchange transitions, commonly used in phylogenetic Markov chain Monte Carlo (MCMC) algorithms, perform poorly on mixtures of dissimilar trees. However, the conditions leading to their results are artificial. Standard MCMC convergence diagnostics would detect the problem in real data, and correction of the model misspecification would solve it.
莫塞尔和维戈达（《报告》，2005年9月30日，第2207页）表明，系统发育马尔可夫链蒙特卡罗（MCMC）算法中常用的最近邻交换转换在不同树的混合情况下表现不佳。然而，导致他们得出这些结果的条件是人为设定的。标准的MCMC收敛诊断方法会在实际数据中检测到问题，而对模型错误设定的修正将解决该问题。
