L. A. Sugden

On convolutional neural networks for selection inference: revealing the lurking role of preprocessing, and the surprising effectiveness of summary statistics

This paper evaluates convolutional neural networks for detecting genomic selection footprints, finding that preprocessing significantly influences model behavior and that human decisions impact model interpretability.

Localization of adaptive variants in human genomes using averaged one-dependence estimation

The paper introduces SWIF(r), a probabilistic method to detect selective sweeps and localize adaptive mutations in human genomes using population-genetic data.