Rob B. Scharpf, PhD

Primary Academic Title

Associate Professor of Oncology

Research Interests

Biostatistics, Computational Biology, Machine learning applications in cancer genomics, Non-invasive molecular-based biomarkers for early detection of cancer and monitoring late stage disease, Bayesian modeling, Experimental design and batch effects, Reproducibility

Lab Website

Cancer Genomics Lab

Research Summary

My research includes the use of artificial intelligence (AI) for predictive models of cancer detection and progression from minimally invasive blood-based molecular biomarkers, statistical methods and software for DNA copy number estimation, experimental design and modeling of batch effects, methods for pre-processing and summarizing high-throughput sequence data, and the development of code and data resources for reproducibility of scientific papers. The scope and impact of this work is best exemplified by the methods and applications we have developed for genome-wide analysis of cell-free DNA (cfDNA) fragmentation patterns, leading to the development of an approach called DELFI (DNA evaluation of fragments for early interception). Extensions and applications of DELFI include the development of classifiers for early detection of lung, ovarian, and liver cancer, methods for the identification of tissue of origin of tumor-derived cfDNA fragments, detection of minimal residual disease after surgical resection, and models to estimate and monitor tumor burden dynamics for patients with late stage disease undergoing treatment.

Google Scholar

https://scholar.google.com/citations?user=Zh6R2YYAAAAJ

PubMed

https://www.ncbi.nlm.nih.gov/myncbi/robert.scharpf.1/bibliography/public/

Graduate Program Affiliations

• Biostatistics, Johns Hopkins Bloomberg School of Public Health
• Human Genetics and Genomics, Johns Hopkins School of Medicine
• Cellular and Molecular Medicine, Johns Hopkins School of Medicine