Biology subjects

Seselj, M.

Publications and source records attributed to Seselj, M..

3 recordsLinked to original sources

bioRxiv2026

A hierarchical Bayesian framework accommodates intraspecific and interspecific variation in multivariate traits

Phylogenetic comparative methods are a critical tool in biology, providing the framework to test evolutionary hypotheses of phenotypic diversification. Accommodating intraspecific variation in multivariate analyses is critical for accurate evolutionary inference, but current methods that incorporate intraspecific variation either 1) assume that traits evolve independently or 2) that all taxa share the same intraspecific covariance structure. Violations of these assumptions can produce biased estimates of evolutionary parameters. Here, we introduce a hierarchical Bayesian framework for multivariate traits that jointly estimates taxon-specific intraspecific covariance structures alongside the underlying evolutionary process. This framework propagates uncertainty from sample size discrepancies and missing data, enabling the incorporation of highly variable morphological traits into phylogenetic analyses. Analysis of simulated data confirms that the model and implementation are well calibrated under the assumed generative model, including challenging datasets with more traits than individuals and substantial missing observations. Applied to perikymata spacing across the great ape clade, including modern humans and Neandertals, the framework recovers intraspecific covariance structures that differ among taxa and yields evolutionary rate estimates markedly more uniform across the tooth crown than those obtained when taxon means are fixed. Our method, which is applicable to other multivariate traits, provide a flexible, tractable approach to joint estimation of intraspecific variation and evolutionary process in multivariate traits.

evolutionary biology↗

bioRxiv2025

Principal Components Analysis fails to recover phylogenetic structure in hominins

ObjectivesPaleoanthropologists often utilize geometric morphometrics and principal components analysis (PCA) to interpret shape variation within the hominin fossil record. It is common practice to interpret proximity in principal components (PC) space among taxa as indicative of not just morphological, but also phylogenetic affinity. This interpretation, however, has not been directly evaluated for hominins. Materials and MethodsFirst, we inferred the posterior distribution of hominin phylogenetic trees and subsampled trees from this distribution. On these phylogenies, we simulated 2D and 3D geometric morphometric datasets and traditional morphological datasets, containing traits analogous to measurements of size or length, with varying numbers of landmarks or traits and evolutionary rates. On each dataset, we conducted a PCA and used neighbor-joining to infer evolutionary relationships from the PC scores of each taxon. We measure the difference between the PCA tree and sampled tree with subtree pruning and regrafting distance and Robinson-Foulds distance. ResultsPCA trees inferred from traditional morphometric data were identical to the sampled tree in 0.11% of datasets when we only considered PC axes 1 and 2, and in 2.9% of datasets when we considered all axes. No PCA tree inferred from any of the 2,400,000 shape datasets was identical to the sampled tree, regardless of the number of axes. DiscussionPhylogenetic interpretations of the hominin fossil record based on proximity in PC space are inherently flawed and likely to be erroneous. Arguments in the hominin systematics literature based on PCA should therefore be reevaluated using phylogenetically-informed alternatives.

evolutionary biology↗

bioRxiv2024

The effects of trait redundancy and information content on hominin phylogenetic inference

Paleoanthropological phylogenetic inference is based on characters assumed to be phylogenetically informative and independent. Yet, our understanding of whether these criteria are met in published character supermatrices is limited. We assess the phylogenetic information content (PHIC) of 107 discrete craniodental traits from a widely-used hominin character matrix. We compare test topologies -- inferred by permuting single traits or removing single traits, anatomical units (AUs), and operational taxonomic units -- to the baseline topology, inferred from the unmodified matrix. In this dataset, only 31 traits have some degree of PHIC: 23 uniquely informative traits -- sufficient, as a set, to closely approach the baseline topology -- and eight redundant traits. No single AU, nor the combination of mandibular and dentition AUs, contains sufficient PHIC to approach the baseline topology, and only the maxilla contains more PHIC than expected. Therefore, phylogenetic placements of fossil hominins represented by isolated AUs should be regarded as putative until better-preserved specimens and more informative traits can be incorporated. Given the ubiquity of discrete morphological data in paleontology and that most of the history of life on Earth was only recorded through fossils, our methods should be broadly applicable to phylogenetic inference involving other paleontological clades.

evolutionary biology↗