AI reveals insights into link between CD33 and cognitive impairment in Alzheimer’s Disease

• DOI: 10.1371/journal.pcbi.1009894
• Publication Date: 2023-02-13T20:30:37Z
• Authors: Tamara Raschka; Meemansa Sood; Bruce Schultz; Aybuge Altay; Christian Ebeling; Holger Fröhlich
• Keywords: Autoencoder

Modeling biological mechanisms is a key for disease understanding and drug-target identification. However, formulating quantitative models in the field of Alzheimer’s Disease challenged by lack detailed knowledge relevant biochemical processes. Additionally, fitting differential equation systems usually requires time resolved data possibility to perform intervention experiments, which difficult neurological disorders. This work addresses these challenges employing recently published Variational Autoencoder Modular Bayesian Networks (VAMBN) method, we here trained on combined clinical patient level gene expression while incorporating focused graph. Our approach, called iVAMBN, resulted model that allowed us simulate down-expression putative drug target CD33, including potential impact cognitive impairment brain pathophysiology. Experimental validation demonstrated high overlap molecular mechanism predicted be altered CD33 perturbation with cell line data. Altogether, our modeling approach may help select promising targets.