Short Course 2 

PubChem Mining - From Small Molecule to Structures and Bioactivity

(June 7, 2012 6:30pm – 8:30pm)

A number of universities recently started drug discovery centers. Among the list include Duke, University of North Carolina, Emory, Vanderbilt, Broad Institute, John Hopkins and many others. NIH has maintained its intramural drug discovery programs for a number of years. Recently NIH has started ambitious extramural funding programs to support academic-based drug discovery programs (with an estimated funding of ~$600M). With these new initiatives, huge volumes of data have been collected in an open and collaborative environment and such data are shared with the public free of charge.

In this short course, the presenter will review the knowledge discovery and management needs in the drug discovery process. One part of the talk will focus on an introduction to PubChem, a public repository for small molecule structures and bioactivity data, where several components facilitating data mining of biological assays test results, including data organization, search of chemicals and analysis for assay development will be covered. On the second part, latest computing and modeling methodology development, primarily those from data mining and machine learning will be overviewed.

The primary targeted audience of the tutorial is cheminformatics researchers and practitioners who are interested in developing or applying advanced computing techniques to support knowledge discovery in drug discovery and drug development programs. It may also help drug development and drug post-market safety monitoring. General knowledge of cheminformatics and statistics is assumed.

Agenda:

• 6:30- 6:55 Welcome and dinner
• 6:55-7:05 Drug Discovery Pipelines and Where Computing May Help
  
• 7:05-7:25 Overview of PubChem
  
• 7:25- 7:50 Kernel Methods in QSAR (quantitative structure activity relationship modeling)
  
• 7:50-8:25 Advanced Topics of Data Analysis in Drug Discovery
  
• 8:25-8:30 Q&A

Instructor:

Jun (Luke) Huan, Ph.D., Associate Professor, Electrical Engineering and Computer Science Department, University of Kansas

Dr. Jun (Luke) Huan is an associate professor in the Electrical Engineering and Computer Science department at the University of Kansas. He is an affiliated member of the Information and Telecommunication Technology Center (ITTC), Bioinformatics Center, Bioengineering Program, and the Center for Biostatistics and Advanced Informatics—all KU research organizations. Dr. Huan received his Ph.D. in Computer Science from the University of North Carolina at Chapel Hill in 2006 (with Wei Wang, Jan Prins, and Alexander Tropsha). Before joining KU in 2006, he worked at the Argonne National Laboratory (with Ross Overbeek) and the GlaxoSmithKline plc. (with Nicolas Guex). Dr. Huan was a recipient of the National Science Foundation Faculty Early Career Development Award in 2009. He has published more than 60 peer reviewed papers in leading conference and journals. His research team won the best student paper award in ICDM'11 and the best paper award (runner-up) in CIKM'09. He serves on the program committees of leading international conferences including ACM SIGKDD, IEEE ICDE, ACM CIKM, IEEE ICDM.