-------------------------------------------------------------------- COLLOQUIUM OF THE COMPUTATIONAL MATERIALS SCIENCE CENTER AND THE SCHOOL OF PHYSICS, ASTRONOMY AND COMPUTATIONAL SCIENCES (CSI 898-Sec 001) -------------------------------------------------------------------- Exploring Novel Approaches in Rational Drug Design Iwona E. Weidlich Chemistry and Biochemistry Department, University of Maryland, Baltimore County, and Chemical Biology Laboratory,Center for Cancer Research, National Cancer Institute, NIH, DHHS, Frederick, MD Rational drug design requires novel computational approaches and research which merges chemistry with biology and focuses on the design and development of new chemicals and biological compounds. Identification of major opportunities and gaps in biomedical research as well as applying new promising technologies becomes a priority. Novel approaches contribute to the development of selective lifesaving drugs needed by patients. Large comprehensive data sets have already been compiled in wide scope of databases. We need to develop novel computational tools, align and integrate new data to profile and analyze them in a systematic way. It is imperative to implement several complementary strategies and expertise to reshape clinical research and provide the international scientific community with new insights into drug design. Molecular docking and virtual screening are important tools in drug discovery that are used to significantly reduce the number of possible chemical compounds to be investigated. I discuss how to improve the interpretation of virtual screening results, docking and handling large databases of small molecules (NCI Database, ChemNavigator iResearch Library[1] PubChem[2]) as well as large-scale high-throughput screening (HTS). A concrete example of applying virtual screening methods for inhibitors of human Tyrosyl-DNA phosphodiesterase will be presented. Additionally I focus on the recent development of services available from NCI CADD group web server[3], chemical structure identifiers, and the chemical databases in the open chemistry field. I will also address the challenges we face in academic and pharmaceutical drug discovery. One of the supplemental goals is to investigate how to improve the prediction of drug-like compounds by applying machine learning classifiers for HTS Data Analysis and Screening. As an example I present robust QSAR model for Hepatitis C Virus RNA Polymerase. This model was built with Random Forest algorithm using Morgan Fingerprints (based onRDkit)[4]. I propose using effective and efficient computational methods to infer functional linkage from the data accumulated either directly or derived from the high throughput technologies in large scale experiments. 1. http://www.chemnavigator.com 2. http://pubchem.ncbi.nlm.nih.gov 3. http://cactus.nci.nih.gov 4. http://en.wikipedia.org/wiki/Random forest; http://rdkit.org Monday, October 31, 2011 4:30 pm Room 301, Research I, Fairfax Campus Refreshments will be served at 4:15 PM. ---------------------------------------------------------------------- Find the schedule at http://cmasc.gmu.edu/seminar/schedule.html