Re: [AMBER] Error: an illegal memory access was encountered launching kernel kNLSkinTest

From: Christian Seitz via AMBER <amber.ambermd.org>
Date: Thu, 1 Sep 2022 08:59:14 -0700

 Hi David,

I don't see any bad bonds using the default parameters of check:

bic = loadpdb bicarbonate_amber.pdb
check bic
prot = loadpdb protein.pdb
check prot

Checking parameters for unit 'bic'.
Checking for bond parameters.
Checking for angle parameters.
check: Warnings: 1
Unit is OK.
Checking parameters for unit 'prot'.
Checking for bond parameters.
Checking for angle parameters.
check: Warnings: 1
Unit is OK.

When I tried to use checkstructure, I got syntax errors. If checkstructure
is still supported, it does not contain the same syntax as check does, even
though the documentation would indicate that they are identical:
https://amber-md.github.io/cpptraj/CPPTRAJ.xhtml#index-check
The warnings for my protein and bicarbonate ion are just that they are not
at a charge of exactly 0.

Best,
Christian

On Thu, Sep 1, 2022 at 4:43 AM David A Case <david.case.rutgers.edu> wrote:

> On Wed, Aug 31, 2022, Christian Seitz via AMBER wrote:
>
> >2. I can make it through two short rounds of minimization, using a couple
> >hundred cycles total. The final energy is pretty low:
> >
> > Maximum number of minimization cycles reached.
> >
> >
> > FINAL RESULTS
> >
> >
> >
> > NSTEP ENERGY RMS GMAX NAME NUMBER
> > 80 -1.0728E+06 2.1341E+01 1.0576E+04 O2 6984
> >
> > BOND = 589516.2494 ANGLE = 4608.9671 DIHED =
> 3317.8961
> > VDWAALS = 174751.9838 EEL = -1839252.0573 HBOND =
> 0.0000
> > 1-4 VDW = 1490.7925 1-4 EEL = -7562.9974 RESTRAINT =
> 0.0000
> > CMAP = 350.6766
>
> Your bond energy is very high, so you probably have a bad bond somewhere,
> which doesn't go away on minimizaiton. Use the checkstructure action in
> cpptraj to look at bond lengths.
>
> ....dac
>
>

-- 
*Christian Seitz*
PhD Candidate, Biochemistry & Biophysics | UC-San Diego
NSF GRFP Fellow, Amgen Scholar
McCammon lab <https://mccammon.ucsd.edu/> and Amaro lab
<https://amarolab.ucsd.edu/>
cseitz.ucsd.edu <cseitz.elon.edu>
[image: www.linkedin.com/in/christianseitz21]
<http://www.linkedin.com/in/christianseitz21>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Sep 01 2022 - 09:02:33 PDT
Custom Search