High-performance computing (HPC) has become a state strategic technology in a number of countries. One hypothesis is that HPC can accelerate biopharmaceutical innovation. Our experimental data demonstrate that HPC can significantly accelerate biopharmaceutical innovation by employing molecular dynamics-based virtual screening (MDVS). Without using HPC, MDVS for a 10K compound library with tens of nanoseconds of MD simulations requires years of computer time. In contrast, a state of the art HPC can be 600 times faster than an eight-core PC server is in screening a typical drug target (which contains about 40K atoms). Also, careful design of the GPU/CPU architecture can reduce the HPC costs. However, the communication cost of parallel computing is a bottleneck that acts as the main limit of further virtual screening improvements for drug innovations.
Journal article
2013-10-28T00:00:00+00:00
53
2757 - 2764
7
Algorithms, Artificial Intelligence, Binding Sites, Databases, Chemical, Databases, Protein, Drug Discovery, HIV Protease, HIV Protease Inhibitors, High-Throughput Screening Assays, Humans, Ligands, Molecular Docking Simulation, Molecular Dynamics Simulation, Neuraminidase, PPAR alpha, Protein Binding, Small Molecule Libraries, Structure-Activity Relationship, Thermodynamics, User-Computer Interface