Métodos para a avaliação da integração entre caracteres filogenéticos discretos

Abstract

Phylogenetics is the field that aims to understand the relationships between different organisms in terms of their development and evolution. A key question in this area is how to analyze the integration and modularity of different characteristics of individuals. Integration refers to the association between characteristics, while modularity focuses on the investigation of groups of characters that have greater dependence on some than others. Despite the abundance of papers in this field that use continuous data, there are fewer papers that focus on the discrete case. In this paper, we present an approach for evaluating the integration between discrete phylogenetic characters, for this the methodology consisting of two steps. The first step is to calculate the similarity between characters using simple correlations (Pearson and Spearman) and by utilizing topology (Threshold Model and Phylogenetic Logistic Regression- PLR). In using PLR, we consider the absolute values of the coefficients and the p-value as measures of association. The second step involves using the information obtained in step one to build a hierarchical Cluster, in order to visualize modules. We use simulated datasets from Markov and Threshold models. To compare the results of each technique, we employ three metrics: Rand Index (RI), Normalized Mutual nformation (NMI) e o Fowlkes Mallows Index (FMI). This allows us to assess how incorporating phylogenetic information impacts the analyses through data simulation.