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

Area of Science:

• Ecology
• Evolutionary Biology
• Phylogenetics

Background:

• Phylogenetic signals are crucial for understanding ecological and evolutionary processes.
• Existing indices for detecting phylogenetic signals are limited to either continuous or discrete traits, not both.
• Current methods often analyze traits individually, overlooking the interactive nature of biological functions.

Purpose of the Study:

• To introduce a novel index, the M statistic, for detecting phylogenetic signals.
• To enable the detection of phylogenetic signals across continuous traits, discrete traits, and multiple trait combinations.
• To provide a unified and rigorous method for phylogenetic signal detection.

Main Methods:

• Developed the M statistic, leveraging Gower's distance to convert diverse trait types into a unified distance measure.
• Calculated the M statistic by comparing phylogenetic distances with trait distances.
• Validated the M statistic's performance against existing indices using simulated data.

Main Results:

• The M statistic effectively detects phylogenetic signals in continuous, discrete, and combined traits.
• Performance of the M statistic is comparable to or better than existing methods.
• Demonstrated the practical application of the M statistic using turtle trait data.

Conclusions:

• The M statistic offers a versatile and accurate approach for phylogenetic signal detection across various data types.
• This new index provides a rigorous method for analyzing evolutionary relationships.
• An R package, "phylosignalDB," is available to facilitate the use of the M statistic.