AMBER Archive (2009)

Subject: RE: [AMBER] Regd : gaff force field

From: Ross Walker (ross_at_rosswalker.co.uk)
Date: Fri Aug 07 2009 - 10:52:07 CDT

> You can simply remove the warnings and charge errors.
> AMBER insert a space before the name of each heavy atom in the output
> file. and most heavy atom names begin in column 14. and generally 2 two
> heavy atoms FE(Iron) or CA (calcium) will not be correct. So you
> should check the output file to edit it manually.
> But regarding the warnings, you do one thing run the simulation with
> cut=999, this will remove all such warnings. Do a minimization with
> cut=999 initially and do your simulation from the resulting pdb file.

> > WARNING: There is a bond of 0.045292 angstroms between:
> > ------- .R<S1 10>.A<C8 14> and .R<S1 10>.A<O2 17>
> > WARNING: There is a bond of 49.381530 angstroms between:
> > ------- .R<Q2 2>.A<O1 12> and .R<R1 5>.A<C7 10>
> > WARNING: There is a bond of 36.657305 angstroms between:
> > ------- .R<Q2 2>.A<O3 14> and .R<R1 6>.A<C7 10>

Ashish, with all due respect, these warnings are important and should NOT just be ignored. They are here for a reason and one should pay attention to them. Do you really somehow think that a bond of 0.04 angstroms and another of 49.38 is in some way reasonable? This structure is completely messed up. Either the initial structure coordinates are simply wrong or the pdb is corrupted in some way and needs to be fixed. Messing with things like the cutoff to remove warnings without actually thinking about what the cutoff actually means is a VERY VERY bad idea.

If this is going to be an explicit solvent simulation with PME then the cutoff should be set between 8 and 9 angstroms. If it is an implicit solvent simulation with GB then if it can be afforded setting cut=999.0 is a good idea. If it is going to be a vacuum simulation then set it to whatever you want since the results will likely be meaningless regardless of what it is set to.

A few things to note. 1 minimization will NOT fix the issues with the huge bonds, this is a fundamental problem with the initial structure that needs to be fixed by hand. Be very wary of running GB with iron sulphur clusters.

Additionally I reiterate my earlier message! Read the LITERATURE!!! There is no excuse for NOT doing this and there are no shortcuts that can be taken here. Using GAFF for iron sulphur clusters is a BAD idea. Go and see what others in the literature have done. If you ultimately want to publish this work you will have to be able to defend your choice of parameters etc so taking shortcuts now will hurt you later.

Also be aware that these types of systems represent VERY advanced simulations so make sure you understand what you are doing both if you are actually running the simulations AND/OR if you are offering advice to others.

All the best
Ross

/\
\/
|\oss Walker

| Assistant Research Professor |
| San Diego Supercomputer Center |
| Tel: +1 858 822 0854 | EMail:- ross_at_rosswalker.co.uk |
| http://www.rosswalker.co.uk | PGP Key available on request |

Note: Electronic Mail is not secure, has no guarantee of delivery, may not be read every day, and should not be used for urgent or sensitive issues.

_______________________________________________
AMBER mailing list
AMBER_at_ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber