A new metric to cut through the noise in evolutionary trees
A persistent challenge in phylogenetic analysis is “entropic site saturation,” where repeated, independent mutations at the same amino acid site obscure the true evolutionary signal, leading to inaccurate tree reconstructions. Researchers have introduced the Dayhoff Exchange Score (DE-score), a novel metric designed to quantify this specific form of saturation within and between amino acid datasets. The tool provides both an overall dataset score and taxon-specific values, pinpointing which lineages contribute most to the noise. Rigorously tested on over 20,000 simulated datasets, the DE-score outperforms existing methods like the Slope R² score in detecting problematic saturation, even when confounded by factors like missing data or increased taxon sampling. This advancement offers a critical pre-analysis checkpoint for evolutionary biologists aiming to build more reliable phylogenies.
Why it might matter to you: For professionals focused on phylogenetics and molecular evolution, this tool directly addresses a core methodological hurdle that can undermine the accuracy of ancestral reconstructions and divergence time estimates. Implementing the DE-score as a standard diagnostic step could enhance the robustness of your phylogenetic inferences, leading to more confident conclusions about speciation events, evolutionary rates, and common ancestry. It represents a practical advance in the analytical toolkit for tackling complex evolutionary questions with genomic data.
Source →Stay curious. Stay informed — with Science Briefing.
Always double check the original article for accuracy.