Phenotyping & Precision Medicine Resource Core
Purpose

The Phenotyping and Precision Medicine Resource Core provides critical services to our research community that broadly include:
- Key scientific expertise and analytic resources needed for Prognostic and Prescriptive Phenotyping
- Consultative services for phenotyping and precision medicine analyses for our research community
- Providing educational opportunities relevant to data science approaches in rheumatic and musculoskeletal diseases (RMDs)
- Guidance and advice to investigators on phenotypic considerations
-
- Development of a high value real-world dataset for OA characterization and phenotyping in collaboration with other CCCRs and multiple entities at UNC
Resource Core Team
J.S. Marron (Steve Marron), PhD– Amos Hawley Professor of Statistics and Operations Research
Michael Kosorok, PhD – W.R. Kenan, Jr. Distinguished Professor of Biostatistics
Louise Thoma, PT, DPT, PhD – Assistant Professor of Physical Therapy
Astia Allenzara, MD – Assistant Professor of Medicine in the Division of Rheumatology, Allergy, and Immunology
Internal Advisory Board
Terry Magnuson, PhD – Chief Research and Strategy Officer for the UNC School of Data Science and Society (SDSS)
Jonathan Berg, MD, PhD – Director of the Program for Precision Medicine in Health Care (PPMH)
Nicholas Shaheen, MD, PhD – PI and Director of NC TraCS
Emily Pfaff, PhD, MS – Co-Director of Informatics and Data Science (IDSci) at NC TraCS
Kevin Anstrom, PhD, MS – Director of the Collaborative Studies Coordinating Center
Setting You Up For Success
Note: We invite you to view this recorded presentation on Precision Medicine, featuring Dr. Amanda Nelson: “Machine Learning for Phenotyping in OA.”
Resource Core Limitations and Proposed Solutions to Big Data Challenges
Limitations | Solutions |
Inter-institutional differences (EHR, analytics) | Obtain and characterize an integrated, multi-practice dataset |
Lack of open-source tools, reporting transparency | Utilize and develop all tools for open-access use |
Lack of entry points for Early Stage Investigators (ESI) | Encourage early and ongoing involvement by ESIs |
Insufficient data science workforce | Provide didactic and hands-on research training in data science |
Limited interpretability/generalizability of results | Utilize a real world dataset and ask clinically relevant questions |
Incomplete, missing, fractured data | Study missingness to understand potential bias, data fusion |
Bias and privacy concerns | Utilize a deidentified dataset from a consented population |
Key Publications
- PMID: 35609053 Biclustering reveals potential knee OA phenotypes in exploratory analyses: Data from the Osteoarthritis Initiative. Nelson AE, Keefe TH, Schwartz TA, Callahan LF, Loeser RF, Golightly YM, Arbeeva L, Marron JS. PLoS One. 2022 May 24;17(5):e0266964. doi: 10.1371/journal.pone.0266964. eCollection 2022.PMID: 35609053 Free PMC article.
- PMID: 37527856 Precision Medicine-Based Machine Learning Analyses to Explore Optimal Exercise Therapies for Individuals With Knee Osteoarthritis: Random Forest-Informed Tree-Based Learning. Kim S, Kosorok MR, Arbeeva L, Schwartz TA, Callahan LF, Golightly YM, Nelson AE, Allen KD. J Rheumatol. 2023 Oct;50(10):1341-1345. doi: 10.3899/jrheum.2022-1039. Epub 2023 Aug 1.PMID: 37527856 Free PMC article.
- PMID: 31002938 A machine learning approach to knee osteoarthritis phenotyping: data from the FNIH Biomarkers Consortium. Nelson AE, Fang F, Arbeeva L, Cleveland RJ, Schwartz TA, Callahan LF, Marron JS, Loeser RF. Osteoarthritis Cartilage. 2019 Jul;27(7):994-1001. doi: 10.1016/j.joca.2018.12.027. Epub 2019 Apr 16.PMID: 31002938 Free PMC article.
- PMID: 36817090 Patterns of variation among baseline femoral and tibial cartilage thickness and clinical features: Data from the osteoarthritis initiative. Keefe TH, Minnig MC, Arbeeva L, Niethammer M, Xu Z, Shen Z, Chen B, Nissman DB, Golightly YM, Marron JS, Nelson AE. Osteoarthr Cartil Open. 2023 Jan 24;5(1):100334. doi: 10.1016/j.ocarto.2023.100334. eCollection 2023 Mar.PMID: 36817090 Free PMC article.
- PMID: 32144896 Precision Medicine Approach to Develop and Internally Validate Optimal Exercise and Weight-Loss Treatments for Overweight and Obese Adults With Knee Osteoarthritis: Data From a Single-Center Randomized Trial. Jiang X, Nelson AE, Cleveland RJ, Beavers DP, Schwartz TA, Arbeeva L, Alvarez C, Callahan LF, Messier S, Loeser R, Kosorok MR. Arthritis Care Res (Hoboken). 2021 May;73(5):693-701. doi: 10.1002/acr.24179.PMID: 32144896 Free PMC article. Clinical Trial.
- PMID: 35840150 Narrative Review of Machine Learning in Rheumatic and Musculoskeletal Diseases for Clinicians and Researchers: Biases, Goals, and Future Directions. Nelson AE, Arbeeva L. J Rheumatol. 2022 Nov;49(11):1191-1200. doi: 10.3899/jrheum.220326. Epub 2022 Jul 15.PMID: 35840150 Free PMC article. Review.
For more information, or to speak with someone regarding our Phenotyping and Precision Medicine Core, please contact Amanda Nelson, MD: Amanda_Nelson@med.unc.edu.
To initiate the process of requesting information, data, and/or collaboration from UNC’s CCCR, please complete and submit this form.