Re: Running multiple-replicas metadynamics

From: Giacomo Fiorin (giacomo.fiorin_at_gmail.com)
Date: Tue Nov 14 2017 - 09:07:43 CST

The message indicates that the fitting procedure is not continuous over
time, most likely because of an internal conformational change. For
example, a peptide helix unfolds into a random coil. This is also visible
in VMD, but it is not a problem there because you are not doing dynamics.
Try changing the definition of the group so that it preserve its basic
structure.

For completeness, in more recent versions of the code the full message is
now:

colvars: Too many iterations in routine jacobi.
This is usually the result of an ill-defined set of atoms for rotational
alignment (RMSD, rotateReference, etc).

On Mon, Nov 13, 2017 at 11:58 PM, Prapasiri Pongprayoon <fsciprpo_at_ku.ac.th>
wrote:

> Hi Giacomo,
>
> Finally, All previous problems were sorted. 5 replicas can communicate to
> each others properly. However, I got another problem (pls see below). All
> replicas died with the error below.
>
> *4 replicas got an error below: *
> LDB: ============== END OF LOAD BALANCING =============== 1602.49
> Info: build node-aware spanning tree with send: 1, recv: 0 with branch
> factor 4
> Info: useSync: 0 useProxySync: 0
> LDB: =============== DONE WITH MIGRATION ================ 1602.49
> colvars: Metadynamics bias "meta_3": received a hill from replica "5" at
> step 138000.
> colvars: Metadynamics bias "meta_3": received a hill from replica "7" at
> step 133000.
> colvars: Error: in parsing "weight".
> FATAL ERROR: Error in the collective variables module: exiting.
>
> *1 replica got a msg below:*
> colvars: Metadynamics bias "meta_3": received a hill from replica "3" at
> step 129000.
> colvars: Metadynamics bias "meta_3": received a hill from replica "5" at
> step 137000.
> colvars: Metadynamics bias "meta_3": received a hill from replica "7" at
> step 131000.
> colvars: Metadynamics bias "meta_3": received a hill from replica "3" at
> step 130000.
> colvars: Metadynamics bias "meta_3": received a hill from replica "5" at
> step 138000.
> colvars: Metadynamics bias "meta_3": received a hill from replica "7" at
> step 132000.
> colvars: Too many iterations in routine jacobi.
> FATAL ERROR: Error in the collective variables module: exiting.
>
> Do you have any idea what went wrong?
> I have seen some previous threads with similar error msg which was caused
> by inappropriate setting of colvars. I’m not sure my case is the same so a
> part of my colvars file is shown below.
>
> colvar {
> name z_pos
> lowerBoundary -57.0
> lowerWallconstant 10.0
> upperBoundary 9.50
> upperWallconstant 10.0
> width 0.5
> distanceZ {
> main {
> atomNumbers 1 5 6 10 11 12 13 17 18 19 20 23 25 26
> 27 28 29 30 31 32 37 39 40 41 42 44 45 50
> }
> ref {
> atomNumbers 418 1043 2341 3541 5563
> }
> axis (0.0, 0.0, 1.0)
> }
> }
> colvar {
> name angle
> lowerBoundary -1
> upperBoundary 1
> width 0.5
> tilt {
> atoms {
> atomNumbers 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
> 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39
> 40 41 42 43 44 45 46 47 48 49 50 51
> }
> refPositionsFile {./ref.pdb} # all atoms in a
> drug have B=1
> refPositionsCol {B}
> refPositionsColValue {1.00}
>
> axis (0.0, 0.0, 1.0)
> }
> }
>
> colvar {
> name r_pos
> lowerBoundary 0.0
> lowerWallconstant 10.0
> upperBoundary 9.0
> upperWallconstant 10.0
> width 0.5
> distanceXY {
> main {
> atomNumbers 1 5 6 10 11 12 13 17 18 19 20 23 25 26
> 27 28 29 30 31 32 37 39 40 41 42 44 45 50
> }
> ref {
> atomNumbers 418 1043 2341 3541 5563
> }
> axis (0.0, 0.0, 1.0)
> }
> }
>
> Thanks for your help in advance.
>
> Regards,
> Prapasiri
>
>
> On Nov 10, 2560 BE, at 6:45 AM, Giacomo Fiorin <giacomo.fiorin_at_gmail.com>
> wrote:
>
> Well, I think there is one thing you should try, don't you agree?
>
> Giacomo
>
> On Thu, Nov 9, 2017 at 6:19 PM, Prapasiri Pongprayoon <fsciprpo_at_ku.ac.th>
> wrote:
>
>> Sorry I forgot to tell that I used NAMD 2.10.
>>
>> On Nov 10, 2560 BE, at 4:38 AM, Giacomo Fiorin <giacomo.fiorin_at_gmail.com>
>> wrote:
>>
>> I suggest you do a "run 0" on each replica at a time, not simultaneously--089e082209f829e466055df2c28f--

This archive was generated by hypermail 2.1.6 : Sun Dec 31 2017 - 23:21:46 CST