A Unified Method for Detecting Phylogenetic Signals in Continuous, Discrete, and Multiple Trait Combinations

ABSTRACT Phylogenetic signals are widely used in ecological and evolutionary studies. Trait data used to detect phylogenetic signals can be continuous or discrete, but existing indices are designed for either type, not both. Moreover, most existing methods can only perform phylogenetic detection of...

Full description

Saved in:
Bibliographic Details
Main Authors: Liang Yao, Ye Yuan
Format: Article
Language:English
Published: Wiley 2025-03-01
Series:Ecology and Evolution
Subjects:
Online Access:https://doi.org/10.1002/ece3.71106
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:ABSTRACT Phylogenetic signals are widely used in ecological and evolutionary studies. Trait data used to detect phylogenetic signals can be continuous or discrete, but existing indices are designed for either type, not both. Moreover, most existing methods can only perform phylogenetic detection of individual traits, despite the fact that biological functions are often the result of interactions among multiple traits. Some attempts to detect phylogenetic signals across multiple trait combinations have employed alternative indicators, which may not align perfectly with the rigorous criteria for defining phylogenetic signals. In this study, we developed a new index (the M statistic) to detect phylogenetic signals for continuous traits, discrete traits, and multiple trait combinations. This capability is inherited from Gower's distance, which is used in the calculation of the M statistic to convert various types of traits into distances. The M statistic strictly adheres to the definition of phylogenetic signals and detects them by comparing these distances from phylogenies and traits. Using simulated data, we compared the performance of our new approach with that of existing commonly used indices. The results show that our method is not inferior to the existing methods. It performs well in handling continuous variables, discrete variables, and multiple trait combinations. We used trait data of turtles (Testudines) to demonstrate the utility of our new method. We suggest this new index as an original method for the detection of phylogenetic signals across various variable types. We provide an R package called “phylosignalDB” to facilitate all calculations.
ISSN:2045-7758