[Paper Review] Fit of Fossils and Mammalian Molecular Trees: Dating Inconsistencies Revisited

Peter J. Waddell|ArXiv.org|Dec 30, 2008

Evolution and Paleontology Studies20 references8 citations

TL;DR

This study re-evaluates divergence time estimates in placental mammals by reconciling fossil calibrations with molecular phylogenies, testing nine evolutionary rate models against revised continuous probability distributions for key fossil nodes. It finds a significant age discrepancy between Supraprimates (Euarchontoglires) and Laurasiatheria calibrations, with combined calibrations causing a drastic increase in penalty function regardless of model choice, indicating persistent dating inconsistencies.

ABSTRACT

Divergence time estimation requires the reconciliation of two major sources of data. These are fossil and/or biogeographic evidence that give estimates of the absolute age of nodes (ancestors) and molecular estimates that give us estimates of the relative ages of nodes in a molecular evolutionary tree. Both forms of data are often best characterized as yielding continuous probability distributions on nodes. Here, the distributions modeling older fossil calibrations within the tree of placental (eutherian) mammals are reconsidered. In particular the Horse/Rhino, Human/Tarsier, Whale/ Hippo, Rabbit/Pika and Rodentia calibrations are reexamined and adjusted. Inferring the relative ages of nodes in a phylogeny also requires the assumption of a model of evolutionary rate change across the tree. Here nine models of evolutionary rate change, are combined with various continuous distributions modeling fossil calibrations. Fit of model is measured both relative to a normalized fit, which assumes that all models fit well in the absence of multiple fossil calibrations, and also by the linearity of their residuals. The normalized fit used attempts to track twice the log likelihood difference from the best expected model. The results suggest there is a very large difference in the age of the root proposed by calibrations in Supraprimates (informally Euarchontoglires) versus Laurasiatheria. Combining both sets of calibrations results in the penalty function vastly increasing in all cases. These issues remain irrespective of the model used or whether the newer calibrations are used.

Motivation & Objective

• To reassess fossil calibration distributions for key placental mammal clades, including Horse/Rhino, Human/Tarsier, Whale/Hippo, Rabbit/Pika, and Rodentia.
• To evaluate the fit of nine evolutionary rate change models against these recalibrated fossil distributions.
• To investigate whether combining calibrations from different mammalian clades (Supraprimates vs. Laurasiatheria) improves or worsens divergence time estimation.
• To quantify the impact of fossil calibration uncertainty on molecular divergence time estimation using a normalized fit measure based on twice the log-likelihood difference from the best model.
• To determine whether newer fossil calibrations resolve long-standing inconsistencies in molecular dating of placental mammals.

Proposed method

• Re-evaluated fossil calibration points using continuous probability distributions to represent uncertainty in absolute node ages.
• Applied nine distinct models of evolutionary rate change across the phylogenetic tree to estimate relative divergence times.
• Used a normalized fit metric based on twice the log-likelihood difference from the best-fitting model to assess model performance.
• Assessed model fit through residual linearity to detect systematic deviations from expected patterns.
• Combined fossil calibrations from Supraprimates and Laurasiatheria clades to test for cumulative incompatibility.
• Evaluated the robustness of results across different models and calibration sets, including newer fossil data.

Experimental results

Research questions

• RQ1 Do revised fossil calibration distributions for key placental mammal clades improve the fit of molecular divergence time estimates?
• RQ2 Is there a significant discrepancy in the estimated root age of placental mammals when calibrating with Supraprimates versus Laurasiatheria fossils?
• RQ3 Does combining fossil calibrations from both Supraprimates and Laurasiatheria clades lead to a measurable increase in model penalty, indicating incompatibility?
• RQ4 How do different evolutionary rate models affect the fit of molecular trees to fossil calibration data?
• RQ5 Are the inconsistencies in divergence time estimation robust to the inclusion of newer fossil calibrations?

Key findings

• A substantial age discrepancy exists between fossil calibrations in Supraprimates (Euarchontoglires) and those in Laurasiatheria, with the former suggesting a much older root than the latter.
• Combining calibrations from both clades results in a dramatic increase in the penalty function across all tested evolutionary rate models, indicating strong incompatibility.
• The inconsistency persists regardless of the choice of evolutionary rate model or whether newer fossil calibrations are used.
• The normalized fit metric, based on twice the log-likelihood difference from the best model, shows that no model fits well when both calibration sets are combined.
• Residuals from the fitted models show non-linearity, suggesting systematic deviations from expected evolutionary patterns when multiple fossil calibration sets are used simultaneously.