Finding a drug for a specific purpose is without a doubt a very difficult task, and today it is largely a trial and error experimental process. A first screening that could simplify the process would be to automatically detect which molecules will stick to certain proteins. This effect is called protein docking. If a molecule sticks to a protein, it is certainly more likely to affect it. In theory one can numerically compute the electric potential around a protein to see if it matches the corresponding potential generated by the molecule. If done naively, the predicted results do not coincide with the experiments. The underlying reason seems to be that the large molecules are surrounded by liquid, mostly water. Their potential interacts with the ions in the water. The ions change the orientation, which affects the potential effectively transforming it from the classical coulomb potential (i.e. the fundamental solution of the Laplacian) to the potential of an integral operator (the fractional Laplacian in the simplest case). Experimentally, this has shown to provide a more accurate model to predict protein docking.
There is a group in the center for Bioinformatics in Saarland University doing research in this field actively. They have a webside describing the project http://bioinf-www.bioinf.uni-sb.de/projects/solvation.html
- ↑ Ishizuka, R; Chong, S-H; Hirata, F (2008), "An integral equation theory for inhomogeneous molecular fluids: the reference interaction site model approach.", The Journal of Chemical Physics (AIP) 128 (3): 034504, http://www.ncbi.nlm.nih.gov/pubmed/18205507
- ↑ Hildebrandt, A.; Blossey, R.; Rjasanow, S.; Kohlbacher, O.; Lenhof, H.P. (2007), "Electrostatic potentials of proteins in water: a structured continuum approach", Bioinformatics (Oxford Univ Press) 23 (2): e99
- ↑ Scott, R.; Boland, M.; Rogale, K.; Fernández, A. (2004), Continuum equations for dielectric response to macro-molecular assemblies at the nano scale, IOP Publishing