Ancestral Inference from Functional Data: Statistical Methods and Numerical Examples

TL;DR

This paper develops statistical methods using phylogenetic Gaussian process regression and IPCA to infer ancestral function-valued traits from tip data in evolutionary studies.

Contribution

It introduces a novel approach combining Gaussian process regression and IPCA for ancestral inference of function-valued traits in phylogenetics.

Findings

Effective modeling of evolution of function-valued traits

Accurate estimation of ancestral traits from tip data

Parameter estimation of evolutionary dynamics

Abstract

Many biological characteristics of evolutionary interest are not scalar variables but continuous functions. Here we use phylogenetic Gaussian process regression to model the evolution of simulated function-valued traits. Given function-valued data only from the tips of an evolutionary tree and utilising independent principal component analysis (IPCA) as a method for dimension reduction, we construct distributional estimates of ancestral function-valued traits, and estimate parameters describing their evolutionary dynamics.