Biological pathway‐based Alzheimer’s Disease polygenic risk score and proteinopathy
Yuriko Katsumata, Xian Wu, Khine Zin Aung, Inori Tsuchiya, Lincoln MP Shade, Shama D Karanth, Kathryn Gauthreaux, Charles Mock, Walter A. Kukull, Erin L. Abner, Peter T Nelson, David W. FardoAbstract
Background
We recently reported genetic associations with dementia‐related proteinopathies. Using multidimensional generalized partial credit modeling, we constructed three continuous latent variables, corresponding to TDP‐43, Aβ/Tau, and a‐synuclein related neuropathology endophenotype scores.
Method
Participant data were drawn from the National Alzheimer’s Coordinating Center (NACC) neuropathology (NP) data (from the September 2023 data freeze) linked to Alzheimer’s Disease Genetics Consortium (ADGC) genotype data. ADGC genotype data were imputed by TOPMed Imputation Server (
Result
The total 19,596,813 single nucleotide variants (SNVs) or short indels were overlapped in ADGC imputed genotype and the GWAS summary statistics. We extracted 1,850 biological process, 542 molecular function, and 405 cellular component GO terms that were constructed of 5 or more SNVs. The AD PRSs in “membrane” cellular component and “protein binding” molecular function were strongly associated with the Ab/Tau pathology related latent score (Bonferroni corrected p‐value = 1.7 × 10‐6 and 0.00011, respectively). There was no biological pathway‐based AD PRS associated with TDP‐43 and α‐synuclein pathology‐related latent scores.
Conclusion
The biological pathway‐based AD PRSs were associated with only Aβ/Tau pathway related latent proteinopathy score. We will explore PRSs for other diseases that are potentially related to AD etiology including Parkinson’s disease linked to α‐synuclein pathology, amyotrophic lateral sclerosis (ASL) with TDP‐43 aggregation, cardiovascular diseases, cardiovascular risk factors, and psychiatric diseases.