William J. Godinez

This paper introduces a diffusion model that predicts high-content images based on compound bioactivity profiles, showing improved performance for chemically dissimilar compounds.

The paper introduces TAT models to predict compound activity using dose-dependent transcriptomic profiles, achieving 51% assay prediction success and validating malaria inhibitor identification.

This paper develops an image-based assay with machine learning to profile beta-lactam antibiotic activity in E. coli by analyzing morphological changes linked to PBP specificity.

This paper develops an unsupervised deep learning method for analyzing cellular images to identify phenotypes and compound effects without segmentation or manual annotations.