Episode 11: Reproducible Voice Data as a Biomarker for AI Models

Description

In this episode, Allissa Dillman chats with David Dorr and Andrea Krussel about their roles in the Bridge2AI (https://bridge2ai.org/) program, which aims to enhance AI readiness by improving data standards and developing machine learning practices. Co-PIs in B2AI Voice (https://b2ai-voice.org/voice-ai-summer-school/), Dorr and Krussel discussed their work on creating reproducible voice data as a biomarker for AI models, including training programs for researchers, successful examples of career development, publication of their competency framework (https://www.frontiersin.org/journals/digital-health/articles/10.3389/fdgth.2025.1610253/abstract), and upcoming symposium. You can also find the B2AI Voice Data online (https://docs.b2ai-voice.org/) and an AI curricula (https://instructure.okta.com/app/).

Guest Bios

David Dorr, MD, MS earned his bachelor's degree in economics (with minors in mathematics and psychology) and his MD from Washington University in St. Louis. He then completed his internal medicine residency at Oregon Health & Science University and earned a master's degree in medical informatics and health services administration from the University of Utah. Broadly, Dr. Dorr's interests lie in complex care management, especially for older adults and other at-risk populations, coordination of care, collaborative care, chronic disease management, quality and the requirements of clinical information systems to support these areas. From these interests, he has broadened into clinical information needs, Electronic Health Record (EHR) deployment and Health Information Exchange, as a way to expand systems-based approaches to all of health care. Finally, Dr. Dorr performs evaluations of care management and informatics initiatives using a variety of methodologies. Learn more on Dr. Dorr's OHSU profile (https://www.ohsu.edu/people/david-a-dorr-md-ms).

Andrea Krussel is the Director of Education and Workforce Development in the Office of Health Information and Data Science at WashU Medicine. In this capacity, she oversees the operations, development, and implementation of biostatistics, data science, and biomedical informatics academic programs. Andrea holds a BA in Theatre and an MA in Communication Arts and utilizes her teaching, acting, and science backgrounds to provide science communication courses and workshops for the WashU community. She also creates, leads, and evaluates competency-based workforce development initiatives in AI, informatics, and data science at WashU, nationally, and abroad. Andrea is a PhD student in Education Policy, Organization, and Leadership at the University of Illinois, Urbana-Champaign. Her research focuses on integrating artificial intelligence into medical education to enhance clinical decision-making skills in medical students and trainees while easing the administrative burden of clinical instructors.

Learn more about the CFDE podcast here (https://orau.org/cfde-trainingcenter/podcast.html).

Episode 10: Overview of the Common Fund Data Ecosystem and how the Training Center supports the CFDE

Description

The National Institutes of Health Common Fund Data Ecosystem (CFDE) aims to enable broad use of Common Fund data to accelerate discovery. NIH Common Fund programs generate a wide range of diverse and valuable data sets and knowledge designed to be used by the research community. The CFDE aims to facilitate improved discovery, reuse, integration, and analyses of these datasets to form novel hypothesis for accelerating discoveries in biomedical research. ORAU received a contract from the NIH to develop a state-of-the-art training center for users of the CFDE. In this cross-over episode with Further Together: The ORAU Podcast, Allissa Dillman, Ph.D., co-principal investigator and owner of BioData Sage LLC, and Jennifer Burnette, ORAU project director, talk to host Michael Holtz about the importance of the CFDE Training Center and how it can help researchers mine data for omics research of various kinds.

Guest Bios

Allissa Dillman, PhD, co-PI and Training and Engagement Director for the CFDE Training Center, is the founder and CEO of BioData Sage LLC, a company focused on providing a holistic approach to data science integration in the biomedical and biological science fields. She works with clients in industry, academia, government, and nonprofit sector to create and support training programs on bioinformatics, cloud computing, and the tools and standards for reproducible data science practices for scientific and lay communities. She also creates community events, such as hackathons, where broad communities work towards solving real biomedical data challenges.

Jennifer Burnette, MPH, CFDE Training Center Director and project manager and director for ORAU expertly steers complex, interdisciplinary programs that bridge local, state, and federal stakeholders. Her career is a testament to her ability to navigate complex networks, having successfully coordinated projects for over 20 federal agency customers including the Centers for Disease Control and Prevention (CDC), Department of Energy (DOE), Department of Homeland Security (DHS), Federal Emergency Management Agency (FEMA), National Institutes of Health (NIH), National Science Foundation (NSF), and 13 DOE and other federal national laboratories.

Learn more about the CFDE podcast here (https://orau.org/cfde-trainingcenter/podcast.html).

Episode 09: MoTrPAC and the Science of Fitness

Description

In this episode, Allissa Dillman chats with researchers Malene Lindholm and Dan Katz about their work on MoTrPAC. Malene has a 20-year background in molecular exercise physiology and Dan started his medical training in cardiology and now uses high-dimensional data to understand heart failure. MoTrPAC (Molecular Transducers of Physical Activity Consortium) is a NIH-funded initiative studying exercise effects across multiple sites in the US using both human and animal studies. Dan and Malene explain what the dataset is all about, including what data is collected, from where, and the kinds of raw and pre-processed data available. They also describe how it can be accessed through various user-friendly tools for scientists of all technical abilities. Find out more at https://motrpac-data.org/ and contact them with questions at motrpac-helpdesk@lists.stanford.edu.

Guest Bios

Daniel Katz, MD, is an Assistant Professor of Medicine in the Stanford Center for Biomedical Informatics Research (BMIR) and the Cardiovascular Medicine Divisions. He practices as an Advanced Heart Failure and Transplant Cardiologist. He completed internal medicine residency at Massachusetts General Hospital, general cardiology training at Beth Israel Deaconess Medical Center, and joined Stanford in 2021 for his advanced heart failure training. His research focuses on identifying the various pathophysiological patterns and mechanisms that lead to the heterogeneous syndrome of heart failure. His efforts leverage high dimensional data in many forms including clinical phenotypes, plasma proteomics, metabolomics, and genetics. He is presently engaged in analysis of multi-omics data from MoTrPAC, and the NHLBI Trans-Omics for Precision Medicine (TOPMed) Program.

Malene Lindholm, PhD, is a senior research engineer and Director of the Human Molecular Athlete Moonshot for the Wu Tsai Human Performance Alliance at Stanford University. After earning her PhD in Medicine at the Karolinska Institute in Sweden, she performed her postdoctoral studies at the Cardiovascular Institute at Stanford University. Her current research focuses on unraveling the multi-omics adaptation mechanisms to exercise across tissues and the genetic basis of extreme human performance. The ultimate objective is to transform the findings into tangible application in the field of precision exercise health and medicine.

Learn more about the CFDE podcast here (https://orau.org/cfde-trainingcenter/podcast.html).

Episode 08: An Early Career Researcher’s Experience with the Common Fund Data Ecosystem

Description

In this episode, Allissa Dillman interviews Seth Berke about his educational journey from pre-med to genomics, and the importance of public datasets and cloud computing in scientific research. Seth highlights the value of, and opportunities provided by the Common Fund Data Ecosystem for early career researchers, shares insights on integrating multiple types of biomedical data, and offers advice for students interested in the field of genomics. Seth’s recent publication referenced in the podcast can be found here: Fundamentals of FAIR biomedical data analyses in the cloud using custom pipelines.

Guest Bio

Seth Berke is a graduate of Johns Hopkins University with a B.S. in Molecular and Cellular Biology and an incoming Ph.D. student in Quantitative and Computational Biology at Princeton. His undergraduate research focused on leveraging biomedical cloud platforms to analyze large-scale pediatric genomics data, work that led to presentations at several international conferences and recent manuscript acceptances in leading journals. He currently works as a Research Assistant at the Johns Hopkins Bloomberg School of Public Health, contributing to the NIH Acute to Chronic Pain Signatures Consortium by integrating and analyzing multimodal data. Seth aims to advance precision health by uniting data across biopsychosocial domains to improve the prediction and treatment of complex disease phenotypes.

Learn more about the CFDE podcast here (https://orau.org/cfde-trainingcenter/podcast.html).

Episode 07: Insights into Training and Engagement with CFDE Data

Description

In this episode, Allissa Dillman chats with Kelli Bursey about the current state of training and mentoring within the Common Fund Data Ecosystem (CFDE), and Diane Krause about academic engagement with CFDE datasets. This information was gathered by the Training Center as part of a mixed-methods landscape analysis used to identify necessary CFDE data science skills for CFDE learners, describe the existing CFDE training landscape, and identify needed training activities and resources for CFDE learners and stakeholders.

Guest Bios

Kelli Bursey has over 15 years of experience at ORAU as a mixed-method researcher, evaluator, and project manager. She actively contributes to a diverse range of evaluation, health communication, and formative research projects for government agencies and private sector clients, providing strategic insights and expertise to drive project success. Ms. Bursey has served over 12 years on the Institutional Review Board (IRB) and currently serves as the Chair for Oak Ridge Sitewide Institutional Review Board, and is well-versed in federal regulations for the protection of human subjects. She holds a certificate of proficiency in qualitative research and brings strong skills in both qualitative and quantitative analysis.

Diane Krause, MS-MPH, CPH, RDN, has over 15 years of experience in health communication at ORAU, specializing in media monitoring, social media listening, and information analysis. She contributed to a landscape analysis for the CFDE Training Center, where she analyzed scholarly output related to CFDE datasets and online discussions of related technical topics.

Learn more about the CFDE podcast here (https://orau.org/cfde-trainingcenter/podcast.html).

Episode 06: Integrating Biomedical Knowledge at Scale with Petagraph

Description

In this episode, Allissa Dillman and Ben Stear discuss knowledge graphs and why they are useful. Tune in as they explore how these tools integrate vast and diverse datasets into unified, queryable networks that support discovery and predictive modeling. Ben shares how knowledge graphs like Petagraph are advancing FAIR data principles, enabling new research insights, and laying the foundation for future integration with large language models.

More information on Petagraph can be found on the project GitHub or in the recently published article: Petagraph: a large-scale unifying knowledge graph framework for integrating biomolecular and biomedical data.

Guest Bio

Ben Stear is a bioinformatics scientist at Children’s Hospital of Philadelphia. His recent work has focused on large data integration and analysis using graphs. He is also working on a PhD at Drexel University focusing on how to detect pleiotropy using graph machine learning and graph analysis.

Learn more about the CFDE podcast here (https://orau.org/cfde-trainingcenter/podcast.html).

Episode 05: Bio-IT World Hackathon Final Presentations

Description

The CFDE Training Center was a sponsor of the Bio-IT World Hackathon, held April 1 – 2, 2025. The event was facilitated by Dr. Allissa Dillman from CFDE Training Center. The event focused on Open Source and FAIR Data (Findable, Accessible, Interoperable, Reusable) principles, and an emphasis was placed on projects leveraging omics data and integrating CFDE tools, with a goal of improving interoperability across datasets to accelerate discoveries.

Data scientists, developers, and life science professionals collaborated on five projects to tackle real-world challenges using data from GlyGen, IDG, HuBMAP, MoTrPAC, and tools from CFDE.

This episode will consist of clips from various presentations from the hackathon.

Learn more about the CFDE podcast here (https://orau.org/cfde-trainingcenter/podcast.html).

Episode 4: How the Data Resource Center Uses the FAIR Principles to Increase Accessibility and Interoperability

Description

This month, join Dr. Allissa Dillman and guest Dr. Shankar Subramaniam, a PI of the Common Fund Data Ecosystem (CFDE) Data Resource Center (DRC), as they discuss the DRC role in integrating and making Common Fund data more accessible. They will highlight the CFDE Workbench, a robust information and data portal designed to help researchers access, analyze, and understand large datasets more effectively. They will also discuss how the DRC supports the FAIR data principles and shar tips for researchers using its tools.

Guest Bio

Shankar Subramaniam is a Distinguished Professor of Bioengineering, Computer Science and Engineering, Cellular and Molecular Medicine, Computer Science and Engineering and Nano Engineering. He is a pioneer in bioinformatics with his development of the Biology Workbench, the first of its kind in web-based infrastructures. He has fostered training and research in systems biology and bioinformatics at the national level, serving on the NIH Director’s Advisory Committee on Bioinformatics and played a key role in the formulation of the NIH Director’s Roadmap which places a major emphasis on the use of quantitative approaches of engineering to biomedical research in health and disease. He has been instrumental in raising national awareness of the roles of engineering approaches to biomedical research. In addition to inventing new methods for analysis of complex systems, he pioneered a novel technology for RNA sequencing with the smallest quantities of RNA leading to our ability to analyze human tissues at the microscale. His contributions to models of human disease are wide and profound and have strong implications for precision and personalized systems medicine.

Learn more about the CFDE podcast here (https://orau.org/cfde-trainingcenter/podcast.html).