Marinela Capanu, PhD

Associate Attending Biostatistician
Epidemiology & Biostatistics
Directory
Epidemiology
Biostatistics
Health Outcomes
MSK Computational Oncology Department & Staff
Administrative Resources
Career Opportunities
Educational Opportunities
Departmental Newsletter

Marinela Capanu, PhD

Associate Attending Biostatistician
Share
Share
Marinela Capanu, Associate Attending Biostatistician

Email

[email protected]

Education

University of Florida

Current Research Interests

Dr. Capanu’s current methodological research interests are in the development of optimized variable selection procedures for regression models. She has recently developed novel variable selection techniques and showed that they have superior performance compared to competitor methods for low to moderate scale linear and generalized linear models. She is currently extending these techniques to the setting of high dimensional generalized linear models. As part of her collaborative work, Dr. Capanu has been involved in clinical collaborations with the Gastrointestinal Oncology Service since joining Memorial Sloan Kettering Cancer Center in 2005. She has assisted the oncologists in this group at every stage in their clinical trials from the design to the statistical analysis and publication of their studies. Dr. Capanu is also assisting investigators in the Department of Radiology with the design and analysis of various projects investigating different imaging modalities and potential biomarkers.

Publications

Selected peer-reviewed publications:

  1. Booth J, Capanu M, Heigenhauser L (2005) Exact conditional p-value calculation for the quasi-symmetry model, Journal of Computational and Graphical Statistics, 14(3):716-725.
  2. Capanu M, Orlow I, Berwick M, Hummer AJ, Thomas DC, Begg CB (2008) The Use of Hierarchical Models for Estimating Relative Risks of Individual Genetic Variants: An Application to a Study of Melanoma, Statistics in Medicine, 27(11):1973-1992
  3. Capanu M, Begg CB (2011) Hierarchical Modeling for Estimating Relative Risks of Rare Genetic Variants: Properties of the Pseudo-Likelihood Method, Biometrics, 61:371-380
  4. Capanu M, Giurcanu M, Begg CB, Gönen M. Optimized variable selection via repeated data splitting. Stat Med. 2020 Jul 20;39(16):2167-2184. doi: 10.1002/sim.8538. Epub 2020 Apr 13. PubMed PMID: 32282097; PubMed Central PMCID: PMC8547352.
  5. Capanu M, Giurcanu M, Begg CB, Gönen M. Subsampling based variable selection for generalized linear models. Comput Stat Data Anal. 2023 Aug;184:107740. doi: 10.1016/j.csda.2023.107740. Epub 2023 Mar 11. PMID: 37090139; PMCID: PMC10118238.

 

Visit PubMed for a full listing of Marinela Capanu’s journal articles

Pubmed is an online index of biomedical articles maintained by the U.S. National Library of Medicine and the National Institutes of Health.

My Downloads

Disclosures

Doctors and faculty members often work with pharmaceutical, device, biotechnology, and life sciences companies, and other organizations outside of MSK, to find safe and effective cancer treatments, to improve patient care, and to educate the health care community.

MSK requires doctors and faculty members to report (“disclose”) the relationships and financial interests they have with external entities. As a commitment to transparency with our community, we make that information available to the public.

Marinela Capanu discloses the following relationships and financial interests:

No disclosures meeting criteria for time period


The information published here is for a specific annual disclosure period. There may be differences between information on this and other public sites as a result of different reporting periods and/or the various ways relationships and financial interests are categorized by organizations that publish such data.

This page and data include information for a specific MSK annual disclosure period (January 1, 2022 through disclosure submission in spring 2023). This data reflects interests that may or may not still exist. This data is updated annually.

Learn more about MSK’s COI policies here. For questions regarding MSK’s COI-related policies and procedures, email MSK’s Compliance Office at [email protected].


View all disclosures