Abstract for PRAGMA 13 Poster Session

From PRAGMAgridWIKI

Avian Flu Grid: International Collaborative Environment for Team Science on Avian Influenza

Rommie Amaro1, Irene Newhouse2, Dong Xu1, Youngjin Choi3, Lily Cheng1, Zhaohui Ding1,4, Ze Luo5, Wes Goodman1, Jung-Hsin Lin6, Habibah Wahab7, Osamu Tatebe8, Yusuke Tanimura9, Karpjoo Jeong3, Kai Nan5, Xiaohui Wei4, Maqsudul Alam2, Wilfred W. Li1, Peter W. Arzberger1

University of California, San Diego1, University of Hawaii at Manoa2, Konkuk University3, Jilin University4, Chinese Network Information Center5, National Taiwan University6, University of Science, Malaysia7, Tsukuba University8, National Institute of Advanced Industrious Science and Technology9

Abstract

The avian influenza virus type A, especially subtype H5N1, is becoming the world's largest pandemic threat due to its high virulence and lethality rate in birds, quickly expanding host reservoir, and high rate of mutations. The two surface glycoproteins, hemagglutinin (HA) and neuraminidase (NA) of influenza virus play important roles in the interactions with cellular receptors containing terminal N-acetylneuraminic acid (Neu5Ac, or NANA) moieties, aka, sialic acids. The approved anti-influenza drugs, oseltamivir and zanamivir, inhibit H5N1 infection by targeting the NA active site, thereby blocking the release of newly formed viral particles. However, research has shown that antigenic drift may give rise to new strains that are resistant to existing NA inhibitors and antigenic shift could give rise to new virulent subtypes of the flu virus. Thus, it is crucial to design novel HA- and NA-targeted inhibitors, which can be used in combination for optimal prophylaxis and treatment. The Relaxed Complex (RC) scheme and Molecular Dynamics (MD) simulations have been applied on the two target proteins to capture key protein dynamics information and accounting for receptor flexibility. Research is under way to take advantage of novel loop flexibilities and changing cavity shapes adjacent to NA active site to discover novel NA inhibitors that may work in a way similar to the HIV integrase inhibitor, raltegravir, inspired by the RC/MD simulation procedures. Further investigation involves statistical cluster analysis for rational selection of representative HA/NA protein structure snapshots, which are used in the virtual screening with synthetic and natural compound libraries. Finally, the binding energies of the high scoring hits are re-scored using Molecular Mechanics-Poisson Bolzmann Surface Area (MM-PBSA) method before experimental validation and subsequent lead optimization.

The RC/MD workflow is complex and computationally demanding, but with strong potentials for application to other infectious diseases. While the MD simulations are more scalable in a high performance computing cluster with low latency network, the virtual screening process requires little interprocessor communication and is highly suitable for a distributed grid environment. The Avian Flu Grid is an integrative effort based on the technology developed by several member institutes to support advanced scientific research for avian flu. The calculations based on these state-of-the-art computational approaches is managed by the CSF4 meta-scheduler (http://gcsf.sourceforge.net) through either a portal environment (https://portal.pragma-grid.net:9443) or Opal-based application specific web services (http://nbcr.net/serivces) which leverages CSF4 for job distribution. Computational data management and sharing across different sites is simplified using Gfarm (http://datafarm.apgrid.org). The simulation setup, and post-analysis are carried out using M*Grid and Glyco-M*Grid (http://www.mgrid.or.kr/), and virtual screening results are managed using the Scientific Data Grid (http://pragma.sdg.ac.cn/). In addition, Glyco-M*Grid, a grid portal-based integrated environment for e-Glycomics, provides a powerful tool for tackling the glycobiology in the avian flu systems. Preliminary equilibration and minimization steps of the MD simulations as well as virtual screening experiments are carried using PRAGMA grid. Natural compound library is screened using the NaPIMM portal (http://www.usm.my). Long running MD simulations with a large number of processors are done using TeraGrid (http://www.teragrid.org) resources, and HPC resources available at the National Biomedical Computation Resource (http://nbcr.net), and Maui High Performance Computing Center (http://www.mhpcc.edu). The complex work flows are captured using the Vision workflow management tools (http://mgltools.scripps.edu).

Underlining the significance of international collaboration in the fight against the pandemic threat of avian flu and other emerging infectious diseases, the integration of scientific research to PRAGMA infrastructure, is expected to greatly facilitate the drug discovery process, improve the accuracy in the search for novel inhibitors, and subsequently generate drug candidates with potentials for further validation and development in biological assays and experiments. For more information, please visit the project website at http://avianflugrid.pragma-grid.net.