Hi,
The molsurf code is very sensitive to certain molecular geometries,
particularly when only doing parts of molecules (which it appears you are
doing). The molsurf code in cpptraj is basically the original code from
Paul Beroza's molsurf program, which hard-coded some memory limits. I've
wanted to change this for a while but its low on my to-do list.
You can try a few things. Adjusting the default probe radius by a bit (say
0.05 Ang) may help. Or you could try setting the probe radius to 0.0 and
the offset to 1.4 (identifying which frame(s) cause issues and testing on
them will probably be helpful in testing these options). Or you could try
the more approximate but less temperamental 'surf' (LCPO surface area)
command. Others may have better suggestions.
Hope this helps,
-Dan
On Wednesday, January 27, 2016, anu chandra <anu80125.gmail.com> wrote:
> Hello Amber users,
>
> The molsurf calculation with simulation trajectories failed with the
> following error messages.
>
>
> make_cones() MAXTOR_PROBE exceeded
> make_cones() MAXTOR_PROBE exceeded
> terminate called after throwing an instance of 'std::length_error'
> what(): vector::_M_fill_insert
> Aborted (core dumped)
>
>
> I have used the following input parameters for the calculation. I am using
> AmberTools 15.
>
>
> trajin ../20md.nc
>
> autoimage
>
>
> molsurf s1 :7-450 out s1_SASA.dat probe 1.4
>
> molsurf s2 :7-213 out s2_SASA.dat probe 1.4
>
> molsurf s3 :214-450 out s3_SASA.dat probe 1.4
>
>
> Any suggestions would be very much appreciated
>
>
> Anu
> _______________________________________________
> 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 Wed Jan 27 2016 - 10:30:03 PST
