Christopher Ré

A 40B-parameter DNA model trained on 9.3 trillion tokens scores coding and noncoding variants zero-shot, exposes interpretable sequence features, and writes chromosome-scale sequences.

7B-parameter convolution-attention DNA model with 131 kb context, trained on prokaryotic genomes, performs zero-shot fitness and essentiality prediction and generates Cas loci, transposons, and 650 kb sequences.

AMELIE 2 improves Mendelian disorder diagnosis by matching patient phenotypes and genotypes to primary literature, achieving high accuracy and efficiency.

This paper introduces IVY, a method that improves Mendelian Randomization by combining genetic variants using probabilistic models, showing better results than allele scoring at moderate sample sizes.

This paper develops a weakly supervised deep learning model to classify aortic valve malformations using unlabeled cardiac MRI sequences, showing improved performance over supervised models.

AMELIE automates Mendelian disorder diagnosis by analyzing literature and prioritizing candidate genes, achieving high diagnostic accuracy for singleton patients.