George Sandler

Interpretable neural networks reveal global epistasis in yeast

A benchmark for the next generation of genotype–phenotype mapping

Equivalent linear mappings of deep networks are a promising path for biology

A quantitative-genetic decomposition of a neural network

From black box to glass box\: Making UMAP interpretable with exact feature contributions

Epistasis and deep learning in quantitative genetics

Cross-trait learning with a canonical transformer tops custom attention in genotype–phenotype mapping

Predicting antimicrobial resistance phenotypes across 7,000 *E. coli* genomes