From: Jérôme Hénin (jerome.henin_at_ibpc.fr)
Date: Thu Dec 13 2018 - 06:56:02 CST
If you are using an orthorhombic cell, you can do that rapidly using qwrap (
https://github.com/jhenin/qwrap) (easier to use under a Unix system, as far
as I know).
Load a trajectory that starts with the protein in one piece (as is usually
the case); then:
qunwrap sel protein compound fragment ;# to get the protein back into one
piece throughout the trajectory
qwrap sel "not protein" center protein ;# to wrap everything else in the
box around it
On Thu, 13 Dec 2018 at 01:14, Vermaas, Joshua <Joshua.Vermaas_at_nrel.gov>
> Hi Jacob,
> #3 looks close to what I would do. Compounding on fragment, not residue,
> would at least prevent a connected protein subunit from itself containing
> long bonds when it shouldn't, which is what I'm assuming is happening.
> On 2018-12-12 16:58:18-07:00 owner-namd-l_at_ks.uiuc.edu wrote:
> Hello all,
> I have a rather long simulation of a protein with many subunits in
> solvent. I wrapped everything during simulation, and now I wish to change
> things such that the protein subunits do not jump around across PBC
> boundaries and such that bonds do not stretch. How do I achieve this goal?
> I have tried a few different methods and have made some progress, but
> still have issues. Here is what I've done so far:
> 1. I did basic wrapping and unwrapping with pbctools, and the system
> almost always exploded into a ball of broken bonds.
> 2. I referred to the PCBTools manual page
> and started using the -compound command. I've tinkered with *compound *
> parameter with pbc wrap. This helped avoid the broken bond explosion, but
> the different subunits still jump across boundaries.
> 3. Finally, I searched old listhost responses and found a few similar
> circumstances. I found this response
> which suggested sequential wrap commands, one to center the protein such
> that all chains avoid the PBC boundary, and a second to wrap it nicely.
> This response results in this sort of command structure:
> pbc wrap -center com -centersel "protein and chain A" -compound residue
> pbc wrap -center com -centersel "protein" -compound residue -all
> I found it quite difficult to avoid the PBC boundaries, so a small number
> of frames always contained PBC-related jumps. Further, this method
> introduced a few cases of bond stretching within residues.
> - - - - -
> Do you have any additional suggestions to try? Are there questions I
> should be asking to understand why I am having these issues?
> Thank you very much for your time and guidance,
> Jacob Usadi
> Jacob Usadi
> University of Chicago '19
> Mathematics / Computational and Applied Mathematics BS
This archive was generated by hypermail 2.1.6 : Sun Nov 17 2019 - 23:20:26 CST