RB2015: Molecular Networks that Lead to Cognitive Decline and Alzheimer’s Disease: Philip De Jager

Neurologist Philip De Jager took us through the challenge of identifying novel targets in Alzheimer’s disease (AD) and age-related cognitive decline, understanding the structure of the AD population and understanding each individual’s location along the trajectory leading to AD. Using datasets from Catholic priests, nuns and older Chicago-area individuals who were free of dementia at enrollment, the study on 3000 participants looked at a range of risk factors, checking their cognitive function and PBMC with imaging. They obtained 1200 autopsies, froze each half of brain and have generated different types of data over the last seven years.

Looking at the overlap between genes and various traits, Not surprisingly, they found that the strongest association with the transcriptome was with cognition. Using data reduction, they discovered 47 modules, ranging in size from 20-600 genes. They then related them to trait correlations such as immune response, metabolism of proteins, transcription, translation, etc. Then they looked at specific components such as astrocytes, microglia and neurons as they correlated with traits of interest like cognitive decline. These 47 modules presented in a bayesian network offer a model with a directionality – however not with causality. Their goal is to test robustness of network and prioritize network elements and key gens for further work including a design of a small molecule screen.

Within their in-vitro module, they similarly identified where genes are expressed to manipulate them with shRNA, ORF, CRISPR and 400 targets. Further, they can assess altered target RNA expression and therefore an altered network. It’s currently just the beginning validation. By knocking down each of these genes, like KIF5B, they can test the methods, and see changes.

In summary, their group aims to map the causal chain leading from health to neurodegenerative disease in order to develop a primary prevention strategy and identify the right population at which to intervene. In developing these network diagrams relating risk factors and outcome, they can help guide target gene identification and compound discovery.