ontophylo: Reconstructing the evolutionary dynamics of phenomes using new ontology-informed phylogenetic methods

Abstract

1. Reconstructing ancestral states for discrete characters is essential for understanding trait evolution in organisms. However, most existing methods are limited to individual characters and often overlook the hierarchical and interactive nature of traits. Recent advances in phylogenetics now offer the possibility of integrating knowledge from anatomy ontologies to reconstruct multiple discrete character histories. However, practical applications that fully harness the potential of these new approaches are still lacking.2. This paper introducesontophylo, an R package that extends the PARAMO pipeline to address these limitations.Ontophyloenables the reconstruction of phenotypic entities composed of amalgamated characters, such as entire phenomes or anatomical regions. It offers three new applications: (1) reconstructing evolutionary rates of amalgamated characters using phylogenetic non-homogeneous Poisson process (pNHPP), allowing for rate variation and shifts over time and phylogeny; (2) explicit reconstruction of morphospace dynamics; and (3) direct visualization of evolutionary rates and statistics on vector images of organisms.Ontophyloincorporates ontological knowledge to facilitate these applications.3. Benchmarking confirms the accuracy of pNHPP in estimating character rates under different evolutionary scenarios, and example applications demonstrate the utility ofontophyloin studying morphological evolution in Hymenoptera using simulated data.4.Ontophylois easily integrated with other ontologyoriented and general-purpose R packages and offers new opportunities to examine morphological evolution on a phenomic level using new and legacy data.