Department of Biological Sciences, Louisiana State University, 202 Life Sciences Building, Baton Rouge, LA 70803, USA.
Mol Ecol Resour. 2011 May;11(3):473-80. doi: 10.1111/j.1755-0998.2010.02947.x. Epub 2010 Dec 2.
We describe a software package (SpedeSTEM) that allows researchers to conduct a species delimitation analysis using intraspecific genetic data. Our method operates under the assumption that a priori information regarding group membership is available, for example that samples are drawn from some number of described subspecies, races or distinct morphotypes. SpedeSTEM proceeds by calculating the maximum likelihood species tree from all hierarchical arrangements of the sampled alleles and uses information theory to quantify the model probability of each permutation. SpedeSTEM is tested here against empirical and simulated data; results indicate that evolutionary lineages that diverged as few as 0.5N generations in the past can be validated as distinct using sequence data from little as five loci. This work enables speciation investigations to identify lineages that are evolutionarily distinct and thus have the potential to form new species before these lineages acquire secondary characteristics such as reproductive isolation or morphological differentiation that are commonly used to define species.
我们描述了一个软件包（SpedeSTEM），它允许研究人员使用种内遗传数据进行物种划分分析。我们的方法假设存在关于群体归属的先验信息，例如样本取自一定数量的已描述亚种、变种或不同形态型。SpedeSTEM 通过计算所有抽样等位基因的分层排列的最大似然种系发生树，并使用信息理论来量化每种排列的模型概率。SpedeSTEM 在这里针对经验数据和模拟数据进行了测试；结果表明，过去分化时间仅为 0.5N 代的进化谱系可以通过来自五个基因座的序列数据来验证为不同的物种。这项工作使物种形成研究能够识别出具有进化意义的谱系，这些谱系在获得生殖隔离或形态分化等通常用于定义物种的次要特征之前，有可能形成新的物种。
