Share |

Grid portal goes (anti) viral

Image of interactive 3D model of a virus protein in GVSS.

The newly launched web portal of GVSS, a grid-based platform that predicts how specific virus proteins interact with drug components, features more than 70.000 virus proteins. The results can be displayed as scalable, interactive 3D models. Image courtesy Hsi-Kai Wang, Academia Sinica Grid Computing Center.

Every year, viruses trigger new outbreaks of illnesses, often affecting millions of people around the world. As globalization has resulted in the rapid spread of these infectious diseases, a host of new challenges to public health needs to be addressed on a global and regional level. A fast, accurate diagnosis and the subsequent treatment, for instance, largely rely on the timely development of diagnostic tests, vaccines, and drugs.

However, the road to finding a suitable antidote is a long and rocky one; drug discovery usually requires a substantial amount of information and resources in addition to the time needed for development and testing. Each preparatory stage of the process, such as isolating the virus or identifying an effective counteragent, can take up to five years.

During the International Symposium on Grids and Clouds 2012 in Taipei last month, the Academia Sinica Grid Computing Center (ASGC), in close collaboration with Ying-Ta Wu of Academia Sinica's Genomics Research Center, launched the new web portal of the Grid Virtual Screening Service (GVSS), a grid-based platform that predicts how specific virus proteins will interact with drug components.

Scientists can save valuable testing time and resources, as it simplifies the large-scale screenings in the initial stages of the drug discovery process by harnessing the processing power of grids to deliver high-performance docking simulations that can be easily adjusted to accommodate projects of any size.

In 2009, GVSS played a central role in an initiative called Dengue Fever Drug Discovery, which aimed at finding a suitable drug for the infectious tropical disease by making use of grid resources to perform docking simulations. Over a period of several months, Wu and his colleagues used GVSS to carry out simulations for over 300,000 compounds, which would have taken more than 12 years on a single desktop computer.

As an alternative to the existing command-line interface, the new GVSS web portal is fully compatible with any standard browser and does not require prior installation. Users currently have access to EUAsia service grid resources to perform their simulations.The integration of desktop grids resources into the newly launched web portal is also currently under development, and will be released in the near future.Users are only required to obtain a valid user certificate and become a member of the EUAsia Virtual Organization.

Image of chemical compound drawing interface in GVSS.

The intuitive 2D compound search engine allows users to "draw" compounds and filter them by specifying chemical properties. Image courtesy Academia Sinica Grid Computing Center.

Facilitating drug discovery

"GVSS was created to help run molecular docking simulations using grid resources. This allows us to perform a large number of docking jobs within a reasonable time frame. GVSS also offers an intuitive user interface, so users don't need to be technical experts to create and submit grid jobs," Wu said. "The portal makes running docking simulations on a grid platform a very simple task, while maintaining a high degree of flexibility for users to experiment with the parameters."

The portal includes a range of new innovative analysis and visualization tools. Users can choose from more than 70.000 proteins included in the Protein Data Bank, which can be displayed as scalable, interactive 3D models. Using this visualization tool, researchers can define their own docking site on the selected protein. Real-time job monitoring of submitted docking jobs is also featured.

In addition to the protein data, GVSS currently provides more than 10 million drug compounds from the ZINC database, with plans to further increase the number of included databases in the future. The intuitive 2D compound search engine allows users to "draw" the compounds they are looking for and filter them by specifying chemical properties. Compounds that are not included in the database can be directly uploaded to the server.

"The results can be displayed as interactive 2D and 3D models or downloaded for further analysis," Wu said. "A simple clustering function allows users to rank the most frequent hits and their structure families."

Your rating: None Average: 4.7 (9 votes)