Related Experiment Videos
The potential of Internet computing for drug discovery
E Keith Davies1, W Graham Richards
1Dept of Chemistry, Oxford University, South Parks Road, Oxford, UK OX13QT. keith.davies@chemistry.oxford.ac.uk
Drug Discovery Today
|June 6, 2002
Summary
Harnessing idle personal computer (PC) power offers a novel approach to large-scale drug discovery calculations. This method, while powerful, requires careful selection of computational algorithms for optimal performance in areas like virtual screening.
Area of Science:
- Computational chemistry
- Drug discovery
- Bioinformatics
Background:
- Supercomputers lack the power for high-precision, large-scale drug discovery calculations.
- Personal computers (PCs) offer a distributed computing alternative by utilizing idle processing time.
Purpose of the Study:
- To explore the feasibility of using networked PCs for complex computational chemistry tasks.
- To identify suitable algorithms for drug discovery on distributed PC architectures.
Main Methods:
- Leveraging idle time from internet-connected corporate and home PCs.
- Analyzing architectural constraints of distributed computing for computational chemistry.
Main Results:
- Distributed PC networks can provide substantial computational resources (hundreds or thousands of years per day).
- Certain algorithms, like virtual screening for protein inhibitors, are well-suited for this architecture.
- Molecular simulation algorithms are generally not appropriate due to architectural limitations.
Conclusions:
- Networked PCs present a viable, powerful alternative for specific large-scale drug discovery computations.
- Careful algorithm selection is crucial for successful implementation in distributed computing environments.
- Virtual screening is a promising application for harnessing collective PC power in drug discovery.