Premium
P1‐163: MOLECULAR MECHANISMS OF THE ALZHEIMER'S RISK GENE UNC5C IN NEURONAL DEATH
Author(s) -
Karunakaran Devi Krishna Priya,
Sadleir Katherine R.,
Kemal Shahrnaz,
Cuddy Leah K.,
Watts Ryan,
Atwal Jasvinder,
Vassar Robert J.
Publication year - 2018
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2018.06.167
Subject(s) - programmed cell death , biology , neun , senile plaques , microbiology and biotechnology , tunel assay , microglia , apoptosis , neuron , alzheimer's disease , neuroscience , pathology , genetics , immunology , immunohistochemistry , medicine , inflammation , disease
nucleotide polymorphisms (SNPs) associated with AD. Even though genetic factors account for up to 80% of the liability for AD, each AD associated SNPs from the previous GWAS do not able to address the onset of AD. Methods: In order to make an AD prediction model, we used GWAS data for AD cases (n1⁄4986) and cognitively normal (n1⁄41155) from Korean elderly. Using a combination of the top-ranked SNPs, we generated AD risk prediction models. In the training data set, 10-fold cross validation was used to evaluate the prediction of the polygenic risk models, which is generated from hundreds of independent SNPs. Furthermore, the prediction models are assessed whether these models are able to predict AD in an independent sample, Japanese AD cohort. Results: Among the AD prediction model, prediction accuracy from age stratification more effectively predicts the onset of AD than no considering age effect. These results suggest that PRS models, considering age effect, are able to predict the onset of AD with relatively small sample sizes. Conclusions: The AD prediction models based on the polygenic risk score implies that many common variants with the small effects are associated with AD. Also, the results suggest that our model considering aging has the potential to develop an accurate predictive model with hundreds of SNPs for practical clinical use.