PROJECT SUMMARY
Alzheimer’s disease (AD) and related dementias (ADRD) are highly heritable, severe, and complex brain
disorders. Molecular profiling efforts of the brain have enabled us to understand more about how common
variants may be associated with disease risk via regulating gene expression. However, the status quo as it
pertains to the use of multiomics data for gene-trait association, is limited to independent transcriptome- and
proteome-wide association studies. Here, we propose to: (1) additionally leverage epigenetic datasets and (2)
apply a data-driven correlation-aware meta-analytic approach to integrate a wide range of brain cell-type and
area specific imputed transcriptomes, proteomes and epigenomes. Our preliminary data suggest that this will
greatly increase the power to identify genetically driven gene dysregulation associated with these disorders while
controlling for the correlation between different genomic features.
First, we will impute brain cell-type (based on FANS/FACS-sorted and single-nucleus data) and brain area
(PFC, ACC, EC, STG, IFG) specific transcriptomes, proteomes and epigenomes for AD, frontotemporal lobar
degeneration, and dementia with Lewy bodies. Secondly, for each disease, we will perform a data-driven
correlation-aware meta-analysis of these imputed genomic features to identify key genes. Lastly, we will develop
an open resource that provides for the rapid dissemination and open access to analyses and outcomes. To our
knowledge, the proposed approach will result in the largest multiscale modeling of genetically driven gene
dysregulation in the brain, will identify novel genes associated with dementias and will enable similar studies in
other brain disorders as well.
