Hello!
There is something wrong in the 16np_test.nc. I tried to run the
minimization with AmberMdPrep.sh and indeed it crashed. I extracted
the last frame of the 16np_test.nc with the name lastframe.pdb,
deleted the waters and re-generated the topology with tleap using:
----------
source leaprc.protein.ff19SB
source leaprc.water.opc
x = loadpdb lastframe.pdb
solvateoct x OPCBOX 6
saveamberparm x x.topo x.coords
----------
Remember the calcium atoms need a TER card:
ATOM 1659 CA CA 110 48.780 43.538 48.529 1.00 0.00 CA
TER 1660 CA 110
ATOM 1660 CA CA 111 54.307 42.016 37.879 1.00 0.00 CA
TER 1661 CA 111
Then used:
$ AmberMdPrep.sh -p x.topo -c x.coords --temp 310
And it runs the 9 steps and the final minimization with no problem
-
#Frame EQ[A0] EQ[A1] EQ[A2] EQ[OneA1]
EQ[corr] EQ[vala] EQ[chisq] EQ[pltime] EQ[fslope]
EQ[name] EQ[result]
1 1.0647 0.008527 1.0710 117.2747
0.2862 0.0001909 0.02071 500.0000 1.093e-08 MD_Density
yes
-
Perhaps re-generating the topology would help.
Rodrigo.
On Mon, Aug 21, 2023 at 11:57 AM Loftus, Alec via AMBER
<amber.ambermd.org> wrote:
>
> Hi AMBER community,
>
>
>
> I am using amber18 to run simulations on a protein and when I attempt to use pmemd.cuda for equilibration I run into the following error:
>
>
>
>
>
> Program received signal SIGSEGV: Segmentation fault - invalid memory reference.
>
>
>
> Backtrace for this error:
>
> #0 0x2AAAB75FF697
>
> #1 0x2AAAB75FFCDE
>
> #2 0x2AAAB841C3AF
>
> #3 0x56F960 in gti_lj1264_nb_setup_
>
> #4 0x47F419 in __extra_pnts_nb14_mod_MOD_nb14_setup
>
> #5 0x4E2286 in __pme_alltasks_setup_mod_MOD_pme_alltasks_setup
>
> #6 0x4C2CE4 in MAIN__ at pmemd.F90:?
>
>
>
>
>
> I have successfully run equilibration using both sander and sander.MPI, so I’m unsure why pmemd.cuda is having problems.
>
>
>
> Here is a link to a folder that contains the input files for the command (in their normal forms and then also just as plain text files): https://drive.google.com/drive/folders/127hqpyQ5U0FhV6RdlZsn9HlKVQzxP1pj?usp=share_link
>
>
>
> Any information about how to make it run properly or what might be causing this error is greatly appreciated!!
>
>
>
> Thanks,
>
>
>
> Alec Loftus
>
> Graduate student
>
> Loyola University Chicago
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Aug 21 2023 - 13:00:02 PDT
