Center for Alzheimer's and Related Dementias (CARD): Harmonized Data-Derived Resources for the Alzheimer's Disease and Related Dementias Community

National Institute On Aging

Investigators

Linked publications & trials

Abstract

Current data harmonization work has been underway to maximize the utility of both datasets and analytics workflows across ADRD domains. We have also attempted to improve and standardize compute infrastructure to accomplish these tasks. All publicly available genomics data in the ADRD space is currently being aggregated and standardized as per Nalls et al 2019. Clinical data from longitudinal ADRD studies has been accessed and is being harmonized to mirror work represented in Iwaki et al 2019. Additionally, we have been initiating testing of hybrid cloud infrastructure to support this work, maximizing the power of Google Cloud, Microsoft Azure and the NIH's Biowulf compute resource to ensure efficient analytics. Additionally we have been shepherding ADRD related datasets for public access on open science platforms as well as participating in the development and improvement of these platforms to help build a robust open science community around CARD. Our team is also supporting data science and bioinformatic efforts for marquee CARD projects like iNDI. A major focus of the data science team at CARD is inclusivity and democratization of complex workflows. In addition to supporting open code and open data efforts, we have been building streamlined tools for use on cloud data platforms to allow everyone to participate in global open science. Two focal points of the CARD data science team's inclusivity efforts are first growing collaborations with Mexican and Latin American dementia researchers and secondly maximizing the contributions of diverse populations in genetics, such as current trans-ethnic fine-mapping studies underway that seek to use diversity to uncover causal risk variants for Alzheimer's disease including collaborating investigators from the previously described dementia research centers. Using these approaches, we have begun to develop several web-based applications relevant to ADRD research. In particular, we are creating a platform for the curation, sharing, and integration of multi-modal -omics data to gain mechanistic insights for early-stage ADRD drug development. We are also developing a web-based repository for automated machine learning workflows and democratized genomics. Finally, we have created an open-science data commons for functional genomics screens in human cell types, in collaboration with investigators at University of California-San Francisco.