The abstract is read at the publisher; the summary is JClub's.
ShiVa detects concurrent evolutionary shifts in both OU optima (θ) and diffusion variance (σ²), reducing false positives for θ shifts when σ² shifts occur along phylogeny.
- Why it matters: Existing OU shift-detection methods focus mainly on changes in optimal trait values and often ignore simultaneous variance shifts, risking misleading inference when variance changes are present.
- What they did: The authors model trait evolution with a multi-optima and multi-variance OU framework and cast joint shift detection as variable selection using an ℓ1-penalized loss in the R package ShiVa.
- The result: In simulations, ShiVa yields lower mean integrated squared error for diffusion variance and significantly reduces false positives for detecting θ shifts under variance shifts, while matching existing performance when only θ shifts occur.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.