On September 24, 2026, teams led by Daifeng Wang at the University of Wisconsin and Panos Roussos at Mount Sinai Hospital published a study in Nature Medicine. Using single-nucleus RNA-seq data from 1,494 donors and over 6 million brain cells from the PsychAD consortium, they developed a deep learning framework called PASCode, identified about 1.5 million phenotype-associated cells, and revealed molecular mechanisms underlying cognitive resilience and AD-related depression. The related tools have been open-sourced.