A simulative deep learning model of SNP interactions on chromosome 19 for predicting Alzheimer's disease risk and rates of disease progression.

BACKGROUND: Identifying genetic patterns that contribute to Alzheimer's disease (AD) is important not only for pre-symptomatic risk assessment but also for building personalized therapeutic strategies. METHODS: We implemented a novel simulative deep learning model to chromosome 19 genetic data from the Alzheimer's Disease Neuroimaging Initiative and the Imaging and Genetic Biomarkers of Alzheimer's Disease datasets. The model quantified the contribution of each single nucleotide polymorphism (SNP) and their epistatic impact on the likelihood of AD using the occlusion method. The top 35 AD-risk SNPs in chromosome 19 were identified, and their ability to predict the rate of AD progression was analyzed. RESULTS: Rs561311966 (APOC1) and rs2229918 (ERCC1/CD3EAP) were recognized as the most powerful factors influencing AD risk. The top 35 chromosome 19 AD-risk SNPs were significant predictors of AD progression. DISCUSSION: The model successfully estimated the contribution of AD-risk SNPs that account for AD progression at the individual level. This can help in building preventive precision medicine.

CADRE: A Collaborative, Cloud-Based Solution for Big Bibliographic Data Research in Academic Libraries.

Big bibliographic datasets hold promise for revolutionizing the scientific enterprise when combined with state-of-the-science computational capabilities. Yet, hosting proprietary and open big bibliographic datasets poses significant difficulties for libraries, both large and small. Libraries face significant barriers to hosting such assets, including cost and expertise, which has limited their ability to provide stewardship for big datasets, and thus has hampered researchers' access to them. What is needed is a solution to address the libraries' and researchers' joint needs. This article outlines the theoretical framework that underpins the Collaborative Archive and Data Research Environment project. We recommend a shared cloud-based infrastructure to address this need built on five pillars: 1) Community-a community of libraries and industry partners who support and maintain the platform and a community of researchers who use it; 2) Access-the sharing platform should be accessible and affordable to both proprietary data customers and the general public; 3) Data-Centric-the platform is optimized for efficient and high-quality bibliographic data services, satisfying diverse data needs; 4) Reproducibility-the platform should be designed to foster and encourage reproducible research; 5) Empowerment-the platform should empower researchers to perform big data analytics on the hosted datasets. In this article, we describe the many facets of the problem faced by American academic libraries and researchers wanting to work with big datasets. We propose a practical solution based on the five pillars: The Collaborative Archive and Data Research Environment. Finally, we address potential barriers to implementing this solution and strategies for overcoming them.