Re: [AMBER] Ligand clustering input .pdb file.

From: Daniel Roe <>
Date: Tue, 20 Jul 2021 12:06:44 -0400


On Mon, Jul 19, 2021 at 5:14 AM Samuele Di Cristofano
<> wrote:
> The problem is that only few of my output files ( .pdb of representative
> clusters members) exhibits wrong connectivity when visualized.

The PDB format doesn't store topology information, so most programs
(including cpptraj, VMD, etc) have to try and guess at how things are
bonded. Depending on the structure, those guesses are not always
correct. You would be better off using a format that has actual
topology information (i.e. mol2, amber topology, psf, etc) when
clustering/visualizing the output. In fact, I recommend that you write
out mol2 files instead of PDBs from your clustering so you can see
exactly what cpptraj thinks the PDB topology is.


> This doesn't happen when I use only 20 models (i.e. output from a single
> docking run) as input.
> Could someone help me to figure out where the problem is?
> The 'trajin' command needs a particular .pdb format ?
> Thanks in advance.
> Samuele
> _______________________________________________
> AMBER mailing list

AMBER mailing list
Received on Tue Jul 20 2021 - 09:30:02 PDT
Custom Search