Grams Markus, Torres Ambrosio, Wirkner Christian S, Richter Stefan
Allgemeine & Spezielle Zoologie, Universität Rostock, Universitätsplatz 2, Rostock, 18055, Germany.
Centre for Integrative Biodiversity Discovery, Museum für Naturkunde - Leibniz-Institut für Evolutions- und Biodiversitätsforschung, Invalidenstr. 43, Berlin, 10115, Germany.
Cladistics. 2025 Jun;41(3):283-303. doi: 10.1111/cla.12611. Epub 2025 Apr 4.
Using a new character matrix composed of revised matrices of previous analyses and new morphological findings, the phylogeny of Malacostraca (Pancrustacea) is analysed anew with 207 characters for 35 terminal taxa across all recognized orders. Particular emphasis was placed on methodological versatility, including different degrees of implied weighting and one of the first applications of methods recently developed in TNT (with the xlinks-command) for considering character dependencies. With >67% of ontological dependencies our character matrix offers a perfect opportunity for putting this new methodology to the test. In particular, we can demonstrate the significant impact of character dependencies and conclusively argue the usefulness of "xlinks" (or the consideration of character dependencies in general). Furthermore, the multimethod framework also enables a comparative evaluation of established and new approaches, and the resulting cladograms thereof. Although our various results leave many questions about the phylogeny of Malacostraca unanswered, clear support is emerging for some monophyla, whereas some surprising findings give reason for methodological reflection. Also, the necessity for an increased attention in terms of taxon sampling and additional character examinations in certain groups becomes obvious. We herein provide (i) an R-function for automatically translating the character dependency syntax proposed by Grams and Richter (Cladistics, 2023, 39, 437) into xlinks-commands for TNT; and (ii) a TNT-script for analysing a character matrix successively under various k-values for implied weighting.
利用一个由先前分析的修正矩阵和新的形态学发现组成的新特征矩阵,对软甲纲(泛甲壳亚门)的系统发育进行了重新分析,涉及所有公认目下的35个终端分类单元,共207个特征。特别强调了方法的通用性,包括不同程度的隐含加权,以及首次应用TNT中最近开发的方法(使用xlinks命令)来考虑特征依赖性。我们的特征矩阵中有超过67%的本体依赖性,这为测试这种新方法提供了绝佳机会。特别是,我们可以证明特征依赖性的显著影响,并最终论证“xlinks”(或一般而言考虑特征依赖性)的有用性。此外,多方法框架还能够对既定方法和新方法及其生成的分支图进行比较评估。尽管我们的各种结果仍留下许多关于软甲纲系统发育的问题未得到解答,但对于一些单系群已有明确支持,而一些惊人的发现则引发了对方法学的反思。同样,在某些类群中增加分类单元采样和额外特征检验的必要性也变得明显。我们在此提供:(i)一个R函数,用于自动将Grams和Richter(《系统发育分析》,2023年,第39卷,第437页)提出的特征依赖性语法转换为TNT的xlinks命令;以及(ii)一个TNT脚本,用于在不同的k值下对特征矩阵进行连续分析以进行隐含加权。