David F. Stein

LoGoFunc predicts pathogenic gain- and loss-of-function variants using ensemble learning with diverse biological features, outperforming existing tools.

This study evaluates AI-generated kinase structures to explore underrepresented conformational states, showing that reducing MSA depth in AlphaFold2 can predict novel conformations for 398 kinases.

The paper compiles 820 host genetic variants associated with COVID-19 susceptibility, develops a machine learning classifier to predict severity variants, and creates a web-based resource for exploring these variants.

This paper shows that the m.3243A>G mtDNA mutation levels in blood change in a bi-phasic way with age, requiring an age correction dependent on mutation level, and finds evidence for positive germline selection of this mutation.

The paper identifies gene- and protein-level features distinguishing gain-of-function and loss-of-function mutations using NLP and machine learning on a new database of pathogenic mutations.

singlecellVR provides an accessible VR tool for visualizing single-cell data, supporting various modalities and offering a companion package for data conversion and pre-analyzed datasets.