ISCB Overton Prize Keynote

Jian Peng

Jian Peng

College of Medicine (by courtesy), Institute of Genomic Biology (affiliate)
Cancer Center at Illinois (affiliate), National Center of Supercomputing and Applications (affiliate), University of Illinois at Urbana-Champaign
United States

Introduced by: Shoshana Wodak, Conference Co-chair
Time: Wednesday, July 15, 9:30 am - 10:30 am (Eastern Daylight Time)
Presentation Title: Machine learning for structural and functional genomics



The Overton Prize recognizes the research, education, and service accomplishments of early to mid-career scientists who are emerging leaders in computational biology and bioinformatics. The Overton Prize was instituted in 2001 to honor the untimely loss of G. Christian Overton, a leading bioinformatics researcher and a founding member of the ISCB Board of Directors. Jian Peng is being recognized as the 2020 winner of the Overton Prize.

Abstract:

Recent advances in network and functional genomics have enabled large-scale measurements of molecular interactions, functional activities, and the impact of genetic perturbations. Identifying connections, patterns, and deeper functional annotations from heterogeneous measurements will enhance our capability to predict protein function, discover their roles in biological processes underlying diseases, and develop novel therapeutics. In this talk, I will cover a few machine algorithms that interrogate molecular interactions, perturbation screens, structural data, and evolutionary information to understand protein functions. First, I will present a network-based representation learning algorithm that integrates multiple molecular interactomes into compact topological embeddings for protein function inference. I will describe its application to discovering new disease factors and subnetworks from genetic perturbations and variations. Finally, I will present a few deep learning algorithms for protein structure prediction and sequence-to-function mapping. Applications to protein engineering and disease-associated mutation prediction will be introduced.

Biography:

Jian Peng is an Assistant Professor, Department of Computer Science at the University of Illinois at Urbana-Champaign. Since his undergraduate degree in China, Jian Peng has pioneered the application of deep learning techniques to computational biology, including protein structure prediction, biological network analysis, and drug discovery, by developing highly innovative methods to successfully address non-trivial challenges.

Jian Peng has spearheaded some of the most impressive contributions to our field such as the application of deep learning techniques to protein contact map prediction, his invention of a novel computational framework for simultaneous dimensionality reduction of multiple heterogeneous biological networks, enabling state-of-the-art function prediction and drug discovery and the development of TransposeNet, an approach that translates discoveries from model organisms to human, for which adequate approaches did not previously exist.

Dr. Peng has established himself as internationally-renowned researcher in structure-based, genome scale prediction.