Biomedical Data Science for Precision Medicine vanathi@pitt.edu Room 530, BAUM Biomedical Data Science for Precision Medicine Vanathi Gopalakrishnan, PhD Associate Professor of Biomedical Informatics Associate Professor of Intelligent Systems Associate Professor of Computational & Systems Biology Director, PRoBE Laboratory for Pattern Recognition from Biomedical Evidence Co-director, Bioengineering, Biotechnology and Innovation Area of Concentration School of Medicine University of Pittsburgh

PRoBE Lab: Research Interests To Accelerate Biomedical Knowledge Discovery by Developing and Applying novel hybrid Artificial Intelligence Methods Obtained for example from: Biomedical Assays, Clinical Assessments, Imaging tests Obtained for example from: Literature, Ontologies, Past analyses PRIOR KNOWLEDGE DATA Integrative Modeling RULE LEARNING METHODOLOGIES Example: Transfer Rule Learning project funded by NIGMS R01GM100387 to VG PREDICTIVE MODELS FOR BIOMEDICINE Enable predictions about for example: mechanism of action, target selection, useful biomarkers for early detection or monitoring of disease, toxicity risk, and clinical trial outcomes.

SOME TYPES OF DATA WE WORK WITH From biomarker* discovery studies for early detection of: Amyotrophic Lateral Sclerosis Alzheimer’s Lung Cancer Breast Cancer Esophageal Cancer Inflammatory Bowel Diseases Ulcerative Colitis, Crohn’s Coronary Artery Disease Pediatric Cardiomyopathy Proteomic Mass Spectral (large/wide/deep) Immunoassay Genomic GWAS/SNPs(very large) DNA Methylation Gene Expression microRNA Images Cardiac MRI Brain fMRI *The term “biomarker” or “biological marker”, refers to “a broad subcategory of medical signs – that is, objective indications of medical state observed from outside the patient – which can be measured accurately and reproducibly. Medical signs stand in contrast to medical symptoms, which are limited to those indications of health or illness perceived by patients themselves.” - Kyle Strimbu, Jorge A. Tavel. What are Biomarkers? Curr Opin HIV AIDS. 2010 Nov 5(6): 463–466.

Example: Lung Cancer Early Detection RISK PREDICTION MODEL Clinical History ? Accuracy of early detection # Unnecessary biopsies CT screen Results Blood Test Multiplexed serologic Quantitative immunoassays Luminex xMAP® technology Bigbee, W. L*., Gopalakrishnan, V.*, Weissfeld J, L., Wilson, D. O., Dacic, S. Lokshin, A. E., Siegfried, J. M. A Multiplexed Serum Biomarker Immunoassay Panel Discriminates Clinical Lung Cancer Patients from High-Risk Individuals Found to be Cancer-Free by CT Screening. J Thorac Oncol. 2012 Apr;7(4):698-708. (*These authors contributed equally to the study).

Some Current Projects Transfer Learning of Classification Rules applied to microbiome data modeling – NIGMS funded R01 – collaboration with Washington University. Biomarker/Pathway discovery from metabolomic data for personalizing heart disease treatment – collaboration with Industry (Metabolon, Inc.) and Dr. Steve Reis (CTSI director @ Pitt). Computer Aided Imaging Diagnostics to support Children’s Hospital of UPMC Radiologists – collaboration with Dr. Ashok Panigrahy and group.