BMI Department Seminars

To subscribe to the BMI Seminar mailing list email join-biostat-seminar [at] lists.wisc.edu (Subject: Subscribe%20to%20BMI%20Seminar%20mailing%20list) .

Upcoming Seminars:

Upcoming BMI Seminar Events
Title Date Location Presenter Abstract
Lessons (Not) to be Learned from the Debate about Genetic Privacy Feb 20 2018 - 9:30am Orchard View Room, Discovery Building

Ellen Wright Clayton, MD, JD

Vanderbilt University
...

Predictive gene regulatory models for precision medicine Feb 20 2018 -
10:45am to 11:45am
Biotechnology Center Auditorium

Hatice Ulku Osmanbeyoglu

Postdoctoral Research fellow
Memorial Sloan-Kettering Cancer...

PDF icon OSMANBEYOLU Poster.pdf
Doctor AI - Interpretable Deep Learning Methods for modeling Electronic Health Records Feb 22 2018 -
10:45am to 11:45am
Biotechnology Center Auditorium

Jimeng Sun, PhD

Associate Professor,
College of Computing
Georgia Institute of...

PDF icon SUN Poster.pdf

BMI Seminar List:

Title Presenter(s) Date Location Abstract
Lessons (Not) to be Learned from the Debate about Genetic Privacy

Ellen Wright Clayton, MD, JD

Vanderbilt University


Abstract:

Genetic data are frequently characterized as special. Arguments in support of this assertion are that the data are unique, immutable, identifying, and that they reveal information not only about the individual but also about other people, including family members and larger groups. People fear use by employers, insurers, and the government. Proposals to amend the regulations for human research to deem all biospecimens identifiable per se and to impose additional consent requirements were seriously considered and ultimately rejected, allegedly because the public feared that these requirements would limit research. Demonstrations that some people can be identified from research DNA sequence data have been highly publicized. Laws have been enacted that purport to prohibit “genetic discrimination.” I will demonstrate that these perceptions of uniqueness and risk are out of proportion to reality and will argue that examining these data in context can provide valuable insights into developing more effective and appropriate protections for data more generally.

Speaker:
Ellen Wright Clayton, MD, JD, is the Craig-Weaver Professor of Pediatrics and Professor of Health Policy and member of the Center for Biomedical Ethics and Society at Vanderbilt University Medical Center and Professor of Law at Vanderbilt University. She has been studying ethical and legal issues in genetics research and its translation to the clinic for her entire career. She has served on the Advisory Council for National Human Genome Research and co-Chair of the ELSI Working Group of the International HapMap Project. She is currently Co-Chair of the Center for Excellence on ELSI Research entitled Genetic Privacy and Identity Settings, Co-Chair of LawSeqSM, and a co-Investigator in the eMERGE (electronic records and genomics) project at Vanderbilt, which is currently studying the impact of returning research results to participants. She is a member of the National Academy of Medicine, having served on 11 committees, chairing five as well as the Executive Committee of its Advisory Council. She is currently Co-Chair of the Report Review Committee of the National Academies Societies of Science, Engineering, and Medicine. Most important, she did her residency in pediatrics at the University of Wisconsin.

Tuesday, February 20, 2018 - 9:30am Orchard View Room, Discovery Building
Predictive gene regulatory models for precision medicine

Hatice Ulku Osmanbeyoglu

Postdoctoral Research fellow
Memorial Sloan-Kettering Cancer Center


Abstract:
The process of matching pathway-targeted drugs to tumor mutational profile regardless of cancer type is critical in the development of targeted therapies. However, ‘actionable mutations’ interact with distinct gene regulatory programs and signaling networks, and can occur against different tumor-specific genetic backgrounds. To better model the context-dependent role of somatic alterations, we developed a novel computational strategy to integrate parallel phosphoproteomic and mRNA sequencing data across the TCGA, linking dysregulation of upstream signaling pathways with altered transcriptional response. We then developed a statistical approach to interpret the impact of somatic alterations in terms of functional outcomes, such as altered signaling and transcription factor activity. Our analysis predicted distinct dysregulated transcriptional regulators downstream of similar somatic alterations in different cancers. These results have implications for the pancancer use of targeted drugs and potentially for the design of combination therapies.

Tuesday, February 20, 2018 -
10:45am to 11:45am
Biotechnology Center Auditorium PDF icon OSMANBEYOLU Poster.pdf
Doctor AI - Interpretable Deep Learning Methods for modeling Electronic Health Records

Jimeng Sun, PhD

Associate Professor,
College of Computing
Georgia Institute of Technology


Abstract:
Abstract: Deep neural networks provide great potential to create better models for longitudinal electronic health records (EHRs). In this talk, we will present a series of case studies of deep learning for modeling EHR.
1) We illustrate how recurrent neural networks (RNN) can be used to model temporal relations among
events in electronic health records (EHRs) to predict heart failures.
2) We introduce an interpretable predictive model RETAIN which achieves high accuracy while remaining
clinically interpretable and is based on a two-level neural attention model that detects influential past
visits and significant clinical variables within those visits (e.g. key diagnoses).
3) Finally, we present a new approach, medical Generative Adversarial Network (medGAN), to generate realistic synthetic patient records.


Bio: Jimeng Sun is an Associate Professor of College of Computing at Georgia Tech. Prior to Georgia Tech, he was a researcher at IBM TJ Watson Research Center. His research focuses on health analytics and machine learning, especially in designing tensor factorizations, deep learning methods, and largescale predictive modeling systems. He published over 120 papers and filed over 20 patents (5 granted). He has received SDM/IBM early career research award 2017, ICDM best research paper award in 2008, SDM best research paper award in 2007, and KDD Dissertation runner-up award in 2008. Dr. Sun received B.S. and M.Phil. in Computer Science from Hong Kong University of Science and Technology in 2002 and 2003, M.Sc and PhD in Computer Science from Carnegie Mellon University in 2006 and 2007.

Thursday, February 22, 2018 -
10:45am to 11:45am
Biotechnology Center Auditorium PDF icon SUN Poster.pdf