Analytics and Data Science

The HEAL Stewards’ mission is to work with NIH HEAL Investigators to help them fulfill HEAL data sharing expectations and requirements. The Stewards provide resources, guidance and consulting to support diverse study teams and data types across the Initiative. The Stewards collaborate with NIH and the HEAL Platform team at the University of Chicago to build the HEAL Data Ecosystem to enable secondary data reuse. They lead HEAL Data Ecosystem outreach and education efforts, producing the Fresh FAIR webinar and tutorial series. The HEAL Stewards apply the HeLx platform and Dug semantic search engine to HEAL Semantic Search, an advanced search tool released in September 2023 and under ongoing development.

The Imageomics Institute is an NSF Harnessing the Data Revolution Institute established in September 2021 that will create machine learning tools using publicly funded collections of digital image data from national centers, field stations, museums and individual laboratories that will enable scientists to study how function follows form in all areas of biology and expand public understanding of the rules of Life on Earth and how it evolves.

Clinical, biomedical, and translational science has reached an inflection point in the breadth and diversity of available data and the potential impact of such data to improve human health and well-being. However, the data are often siloed, disorganized, and not broadly accessible due to discipline-specific differences in terminology and representation. To address these challenges, NCATS ahs formed the Biomedical Data Translator Consortium, a diverse community of nearly 200 basic and clinical scientists, informaticians, ontologists, software developers, and practicing clinicians distributed over 11 teams and 28 institutions. The consortium has developed and tested a pilot knowledge graph-based “Translator” system capable of integrating existing biomedical data sets and “translating” those data into insights intended to augment human reasoning and accelerate translational science. Having demonstrated feasibility of the Translator system, the Translator program has since moved into development, and the Translator Consortium has made significant progress in the research, design, and implementation of an operational system.

The North Carolina Translational and Clinical Sciences (NC TraCS) Institute is the NCATS-funded CTSA at UNC. NC TraCS supports, funds, and connects the translational research community to improve the way biomedical research is conducted in North Carolina and across the country. RENCI leadership has served the NC TraCS Institute since 2011 as director of the Informatics and Data Science (IDSci) Component.

Phyloreferencing aims to provide a standard mechanism for defining biological clades with precise and fully machine-processable semantics. This project uses ontologies and OWL (a formal ontology language developed by the W3C) to enable users to create definitions with precise, unambiguous, and fully computable semantics for any organismal clade. Such definitions are called phyloreferences, in analogy to georeferences, which have enabled countless geo-enabled applications. The project employs machine reasoners to determine which elements in a phylogeny match a phyloreference. The team is developing the OWL ontologies, OWL data models, and tooling infrastructure needed to put their approach into practice, and to validate that it works correctly and is scaleable. The team has built JavaScript libraries and a web application that allow users to create, find, and reuse phyloreferences, and to apply them to phylogenies both small and Tree of Life-scale, for the purpose of integrating biodiversity data of interest.

ROBOKOP (Reasoning Over Biomedical Objects linked in Knowledge-Oriented Pathways) is a biomedical question-answering system based on a knowledge graph, meaning that it expresses data as a collection of nodes—such as genes and diseases—and edges that represent the relationships between the nodes. ROBOKOP uses multiple open biomedical databases to explore links between various biomedical data types and utilizes the BIOLINK model to express those data types. For example, it can be used to examine connections between a disease and a drug and then explore which genes might be involved in that relationship. ROBOKOP’s publicly accessible user interface (UI) allows users to address both relatively simple questions such as “what genes are associated with drug-induced liver injury?” and more complex ones such as “what drugs could be used to treat airborne pollutant–induced asthma exacerbations in patients who are non-responsive to traditional medications?”

The Gene Ontology (GO) is by far the largest knowledgebase of how genes function, and has become a critical component of the computational infrastructure enabling the genomic revolution. It has become nearly indispensable in the interpretation of large-scale molecular measurements in biological research. All knowledge in GO is represented using semantic web technologies and so is amenable to computational integration and consistency checking.

Data visualization is a critical and ubiquitous tool used to support data analysis tasks across a variety of domains. Visualizations are valued for their ability to “show the data” graphically, rather than using letters and numbers, in a way that enables users to assign meaning to what they see. This in turn helps users analyze complex data, discover new insights, make data-driven decisions, and communicate with other people about their findings. The correctness of these findings is therefore clearly contingent upon the correctness of the inferences that users make when viewing or interacting with a data visualization tool. However, recent studies have shown that people often interpret visualized patterns as indicators of causal relationships between variables in their data even when no causal relationships exist. The result is that visualizations can dramatically mislead users into drawing erroneous conclusions. This project develops a new approach to visualization, based on the concept of counterfactual reasoning, designed to help users draw more accurate and generalizable inferences when analyzing data using visualization tools. The project’s results, including open-source software, are intended to be broadly applicable across domains. In addition, the project will be evaluated with data and users in the population health domain with the potential to contribute to improvements to human health.

Diverse communities of microscopic organisms (microbiota) live in and on the human body. The skin microbiota aid in skin barrier function, help our immune system mature, and produce body odor by metabolizing sweat and other bodily secretions. Many factors impact which microbes live on the skin, and these microbes impact our health.

This project collects skin microbiota and host DNA samples from public participants to explore the connection between daily habits, host genetics, and skin microbiota composition. This participatory science project engages the public in science impacting their daily lives, providing valuable insight into human-microbe co-evolution, and human health.

Monkey Health Explorer is a participatory science project that is part of a larger collaborative project (with the NC Museum of Natural Sciences and colleagues from numerous universities from around the world) studying monkey health, social behavior, and aging. The larger multidisciplinary team includes geneticists, behavioral scientists, neuroscientists and others, studying rhesus macaque monkeys on Cayo Santiago island in Puerto Rico. Rhesus macaque monkey behavior has been studied for decades, and genetics and health assessments are now expanding. Blood smear slides help assess monkey health, and scanned images are run through an AI pipeline so individual white blood cell images can be uploaded to the people-powered science platform, Zooniverse.org, for classification help. Public participants trained in medical or veterinary fields greatly enhance the knowledge gained, by identifying monkeys with potential health conditions not originally investigated. This project engages K-12 classrooms across North Carolina and Puerto Rico to expand participant reach, while providing real-world research examples and early career training to students interested in health and data science careers.

Dug is a flexible, scalable, semantic search tool for biomedical datasets that leverages emerging knowledge graph-based methods to intelligently suggest relevant connections derived from peer-reviewed research. Developed through the National Heart, Lung, and Blood Institute’s (NHLBI) BioData Catalyst® (BDC) platform and the National Institutes of Health’s (NIH) HEAL Initiative, Dug can efficiently ingest, index, and search over nearly any type of study metadata and utilizes evidence-based relationships from curated knowledge graphs to explain why certain results are being returned. The Dug search engine is publicly available as an open source package to provide the biomedical community with a powerful, flexible tool for searching intuitively across an increasingly sprawling public data landscape.

MM-ARIEL is a cutting-edge research project aimed at revolutionizing the diagnosis and treatment of lupus and related autoimmune diseases through advanced multimodal machine learning approaches. This initiative focuses on integrating diverse patient data—including text, images, omics, activity monitor data, and electronic health records—into unified models that support early, equitable, and efficient diagnosis. The project prioritizes reducing health disparities, improving diagnostic accessibility, and addressing biases in data and models, particularly for racially and ethnically minoritized women who are disproportionately affected by lupus. MMR-Lupus is designed to enhance diagnostic capabilities for primary care physicians, empowering them to assess complex diseases without requiring immediate specialist intervention.