GPU Accelerated Chemical Similarity Calculation for Compound Library Comparison
Department of Computational and Systems Biology, Joint Pitt/CMU Computational Biology Program, University of Pittsburgh, Pittsburgh, Pennsylvania 15260, United States
Journal of Chemical Information and Modeling, 2011
@article{doi:10.1021/ci1004948,
author={Ma, Chao and Wang, Lirong and Xie, Xiang-Qun},
title={GPU Accelerated Chemical Similarity Calculation for Compound Library Comparison},
journal={Journal of Chemical Information and Modeling},
year={2011},
doi={10.1021/ci1004948},
URL={http://pubs.acs.org/doi/abs/10.1021/ci1004948},
eprint={http://pubs.acs.org/doi/pdf/10.1021/ci1004948},
publisher={ACS Publications}
}
Chemical similarity calculation plays an important role in compound library design, virtual screening, and "lead" optimization. In this manuscript, we present a novel GPU-accelerated algorithm for all-vs-all Tanimoto matrix calculation and nearest neighbor search. By taking advantage of multicore GPU architecture and CUDA parallel programming technology, the algorithm is up to 39 times superior to the existing commercial software that runs on CPUs. Because of the utilization of intrinsic GPU instructions, this approach is nearly 10 times faster than existing GPU-accelerated sparse vector algorithm, when Unity fingerprints are used for Tanimoto calculation. The GPU program that implements this new method takes about 20 min to complete the calculation of Tanimoto coefficients between 32 M PubChem compounds and 10K Active Probes compounds, i.e., 324G Tanimoto coefficients, on a 128-CUDA-core GPU.
July 6, 2011 by hgpu
