Elizabeth Powell, Ph.D.

September 07, 2023

Purpose

The goal is to promote data science concepts and tools in alcohol research, integrating data across disciplines and clinical and basic sciences realms.

Background

Data science has been a major focus of NIH, including the establishment of the Office for Data Science Strategy. Data science approaches have been used to make key findings in other research areas such as cancer and Parkinson’s disease research. The flood of data generated by NIAAA supported studies in genomics, imaging, electrophysiology and optogenetics, electronic health records, and personal wearable devices presents new challenges in analyses and interpretations and opportunities for discovery. Since 2019, NIAAA has required that human research data be stored in the NIAAA Data Archive (NOT-AA-23-002 for most recent notice).

Statement of Work/Project Objectives

The large databases of biological and behavioral and imaging studies supported by NIAAA provide ample information for data science approaches. However, the investigators lack the tools to participate in the data ecosystem and take advantage of current statistical and computational approaches. The state of the data science field in alcohol research has advanced only slightly since this concept was introduced in 2018. While the scope of the data is broad, many of the tools needed to answer questions in alcohol research require specific applications, algorithms or toolkits that are not currently available. This initiative is expected to:

• Generate intellectual property, analytical tools and methods for alcohol research that interface within modern data ecosystems for use by entire scientific community.
• Promote harmonization of data sets within specific disciplines of alcohol research to improve scientific reproducibility and increase sharing of data across multiple scientific teams.
• Transform fragmented sets of individual data components into a coordinated ecosystem.
• Enable multiscale analysis of clinical and basic science datasets, employ modern data science techniques of artificial intelligence, machine learning and deep learning.
• Promote interdisciplinary collaborations between neuroscientists and data scientists.
• Adapt NIH data science tools and tactics for use in alcohol research.

Justification

The volumes of data produced by NIAAA-supported research, along with publicly available databases and future results, can be analyzed using data science approaches to find new therapeutic targets and approaches for diagnosis and treatment of alcohol use disorder. Data science includes and extends beyond bioinformatics and computational neuroscience to discover new relationships and pathways for complex systems of normal human function and during adaptations due to disorders or disease. Data science is not widespread alcohol research, and thus the field is missing opportunities for discovery and treatment.

The Final NIH Policy for Data Management and Sharing (NOT-OD-21-013) requires data sharing, yet there are limited tools and resources for combining and analyzing data from alcohol research. Since the concept was introduced in 2018, NIAAA has funded two SBIR projects for new algorithms and automated data harmonization and imputation tools. These projects are currently in Phase II. Additional tools and strategies are needed to analyze data from NIAAA research, and tools are needed to make best use of the investment in the NIAAA Data Archive.