Yes, I was maybe too positive in my reply. Looking more closely, classical MD may not be appropriate here. The better way is QM/MM but that is not quick and dirty. It depends whether the main effect of the mutations is in the geometry of the protein or in the electronic structure of the metal binding site.
m On Thu, 2008-07-03 at 15:42 +0100, Andrew Purkiss-Trew wrote: > Just a word of warning regarding metals in macromolecular simulation. > There is no 'good' way of modeling protein-metal interactions in many > cases. Some metal ions are very polarizable (Ca 2+ being one example) > and the simple mechanics used in protein simulation don't allow for > this. Even using a polarizable forcefield is unlikely to give accurate > answers due to the complex nature of the atomic orbitals in many metal > ions when compared to protein atoms (mainly due to the addition of > d-orbitals to the mix). So, Na+, Mg2+ are generally OK, Ca2+, Fe, Mn, > etc. are much more tricky, especially in a quick and dirty simulation. > > On Thu, 2008-07-03 at 12:20 +0100, Martyn Winn wrote: > > Gromacs www.gromacs.org > > > > It is GPL. It can certainly handle metals, although as is always the > > case, you have to be careful in the choice of force field. It is fairly > > user friendly (by the standards of that field ;-). Not sure what the > > situation is for Macs but certainly runs well on Linux. > > > > Alternatives: charmm, amber, namd, etc etc > > > > Your suggested project is perfectly reasonable, but don't expect it to > > be a day's work .... > > > > HTH > > Martyn > > > > > > On Thu, 2008-07-03 at 11:30 +0100, David Briggs wrote: > > > Hi everyone. > > > > > > I have a slightly off topic question I hope someone can help with. > > > > > > I have a structure of a wild type domain, which binds metal ions. > > > Certain mutations in chelating residues cause changes in the apparent > > > affinity for said metal ions. > > > > > > As I have (so far) failed to crystallise the mutant, I would like to > > > do some molecular dynamics in a water box, (with afore mentioned > > > mutations) to get a handle on the origin of the change in metal ion > > > affinity. > > > > > > Does anyone know of a suitable package which: > > > > > > a) is open source, or basically free to academics > > > b) allows inclusion of metal ions > > > c) gives an understandable and interpretable output, in terms of > > > changes in energy, etc. > > > d) Preferably runs on mac, (although I do have access to linux as well.) > > > > > > Am I asking for the moon on a stick? > > > > > > Looking forward to your responses. > > > > > > Regards, > > > > > > Dave > -- *********************************************************************** * * * Dr. Martyn Winn * * * * STFC Daresbury Laboratory, Daresbury, Warrington, WA4 4AD, U.K. * * Tel: +44 1925 603455 E-mail: [EMAIL PROTECTED] * * Fax: +44 1925 603825 Skype name: martyn.winn * * URL: http://www.ccp4.ac.uk/martyn/ * ***********************************************************************
