E. Keith Davies & W. Graham Richards
Drug Discovery Today Volume 7, Issue 11, 2002. Pages: S99-S103
Drug discovery calculations such as protein folding or predicting small molecule protein inhibitors on a sufficiently large scale or with adequate precision have frustrated computational chemists because available supercomputers are not sufficiently powerful. The increasing power of PCs offers an alternative: harness the idle time available on corporate and home computers connected to an intranet or the Internet. However, while this has the potential of offering hundreds or thousands of years of computer time per elapsed day, the architecture constraints require computational chemists to choose their methods and applications problems with care. Some algorithms, such as those for molecular simulations, are generally not appropriate whereas virtual screening of molecules for protein inhibition works well.