RE: Distorted peptide bond with new residue

From: Vermaas, Joshua (Joshua.Vermaas_at_nrel.gov)
Date: Thu Jul 05 2018 - 17:05:54 CDT

Ahh, that's a different problem. Did you regenerate angles and dihedrals after a patch? A mistake I've made more often than I'd care to admit is forgetting this after a patch, and as a result not having *any* bonded terms at the join point to keep the geometries sane.

-Josh

On 2018-07-05 16:00:39-06:00 Nick Palmer wrote:

I will try the extra bonds term, however the bond is not just cis, the n-co-ca bond angle is 180 degrees by the end of the minimization.

On Thu, Jul 5, 2018, 5:34 PM Vermaas, Joshua <Joshua.Vermaas_at_nrel.gov<mailto:Joshua.Vermaas_at_nrel.gov>> wrote:
Hi Nick,

The initial geometry probably has more to do with it. The dihdedral that describes a peptide bond has the functional form cos(2*phi-180), which has minima at 0 and 180. Normally, the initial geometry is such that you get the correct trans peptide bond upon minimization, but in your case the minimizer just found the cis-peptide potential energy minimum, which you'll need to fix. VMD has a snazzy cis-peptide plugin (http://www.ks.uiuc.edu/Research/vmd/plugins/cispeptide/d7e28cd4f5484427885aee7c080%7C0%7C0%7C636664248394423262&sdata=qCZhFr2%2FYRVEFvufgRzcNnrhUu%2BKCbQZyke1BFYk%2ByM%3D&reserved=0>) that will let you fix this for normal amino acids, although I don't know if the selections it uses under the hood will recognize your modified residue. Another possibility is to add a ridiculous extra-bonds dihedral term in addition to the regular bond to make the cis-well disappear (this is what the plugin sets up for you). See http://www.ks.uiuc.edu/Research/namd/2.12/ug/node27.html#SECTION00086400000000000000ermaas%40nrel.gov%7C43be3d95ce30446d6a4d08d5e2c2bda8%7Ca0f29d7e28cd4f5484427885aee7c080%7C0%7C0%7C636664248394433267&sdata=8CvZWgHNPZIbSMReh0p20wOlsbtoK3Auay0gQjBsOBs%3D&reserved=0>

-Josh

On 2018-07-05 15:22:30-06:00 owner-namd-l_at_ks.uiuc.edu<mailto:owner-namd-l_at_ks.uiuc.edu> wrote:

Hello everyone,
I have been trying to create a new residue topology and add parameters to it. The topology and parameters work, however when minimizing the structure the peptide bond connected to the c terminal of the new residue is distorted so that the peptide bond has a 0 dihedral angle rather than the 180 it is supposed to have. When creating the psf for this molecule, I had to manually link this residue to the rest of the chain with a LINK patch. Could the link patch change the type of the bond somehow?
Thank you in advance

--
Nicholas J. Palmer

This archive was generated by hypermail 2.1.6 : Mon Dec 31 2018 - 23:21:15 CST