Thank you for your kindly reply,
However, I've used a truncated octahedron box to test the simulation box vector transformation from amber to plumed by using plumed "cell : CELL \ PRINT ARG=cell.*" command,
I'm using plumed-2.9.0 with amber22.
When I ran plumed command with pmemd.cuda, the cell vectors were printed as:
#! FIELDS time cell.ax cell.ay cell.az cell.bx cell.by cell.bz cell.cx cell.cy cell.cz
0.000000 5.360858 5.360858 5.360858 -5.360858 -5.360858 5.360858 5.360858 -5.360858 -5.360858
When I ran plumed command with pmemd.MPI, the cell vectors were printed the same as I ran pmemd.cuda:
#! FIELDS time cell.ax cell.ay cell.az cell.bx cell.by cell.bz cell.cx cell.cy cell.cz
0.000000 5.360858 5.360858 5.360858 -5.360858 -5.360858 5.360858 5.360858 -5.360858 -5.360858
However, when I ran plumed command with sander.MPI, the cell vectors printed were different with the above two:
#! FIELDS time cell.ax cell.ay cell.az cell.bx cell.by cell.bz cell.cx cell.cy cell.cz
0.000000 9.285279 0.000000 0.000000 -3.095093 8.754245 0.000000 -3.095093 -4.377123 7.581399
It seems that sander are using a different box vector tranformation method comparing to pmemd,
I'm wondering that whether this difference can cause problems,
many thanks to your precious time!
Regards,
XIA
________________________________
发件人: David A Case <david.case.rutgers.edu>
发送时间: 2023年6月28日 20:22
收件人: Songyan Xia (LHS, 118010341) <songyanxia.link.cuhk.edu.cn>; AMBER Mailing List <amber.ambermd.org>
主题: Re: [AMBER] Using plumed with amber
On Wed, Jun 28, 2023, Songyan Xia (LHS, 118010341) via AMBER wrote:
> I want to use plumed with amber to do some CV
>calculations, I noticed that when compiling amber, amber describes plumed
>as "used as an alternate MD backend for Sander", is this means that plumed
>can only used with sander but not pmemd to do the MD simulation?
Plumed is available in pmemd as well. But it runs on the CPU (I believe)
and can therefore slow down GPU-based (pmemd.cuda) simulations.
....dac
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Jun 28 2023 - 19:30:02 PDT
