|
Folding@Home
with AMBER FAQ
 The
Folding@Home developers (primarily Jim Caldwell and Young Min
Rhee) have been working with the AMBER
development team (of which Jim Caldwell is a long time member)
to incorporate AMBER into Folding@Home. This FAQ follows the form
of our Gromacs FAQ.
-
As a participant of Folding@Home, is there anything I need
to do? No. For the participants of Folding@Home, nothing
new need be done, the update is automatic.
-
Why include AMBER?
What do you gain? For the scientists
using Folding@Home, the results will be dramatic. For GB/SA
calculations (see the AMBER
GB page for details), AMBER
affords benefits over the current GB/SA code in Folding@Home
(the
current
code
is
based on
Jay
Ponder's
Tinker), especially
a great speed increase (perhaps up to 5x). This will
substantially
increase the power of Folding@Home for many calculations. Also,
AMBER adds not just speed, but new functionality in its
ability to perform polarizable force field calculations
(which cannot be performed with Gromacs or our current
Tinker 3.8 implementation).
-
If it's faster, will the participants get more points? Sorry,
the points in the stats will still be benchmarked by the CPU
time, so your stats won't be affected either.
-
How can AMBER
be that much
faster? AMBER's GB/SA code is more recently written
and more heavily optimized than Tinker. Gromacs does not
have GB/SA functionality.
-
What about SSE/3D-Now/Altivec
support?
Currently, SSE and SSE2 are not supported. AMBER is run in
double precision so SSE would not help. SSE2 could help, but
would not provide a significant speed increase. Our first goal
was to get AMBER running stably and we will then try to further
tweak optimizations with SSE2. However, we expect to gain only
10% speed increase.
- What about Mac
OS X support? We will be adding OS X support once the Win and
Lin ports settle down. We are currently looking into what would
be the best FORTRAN 90 compiler for OS X.
- Why didn't you
use AMBER earlier? AMBER has just been recently rewritten
in FORTRAN90 which makes it more friendly for Windows compilation.
Also, Jim Caldwell (long time AMBER developer) has just
recently joined the FAH team.
-
Has the current
scientific code been slowing you down unnecessarily?
No: since AMBER's fast GB/SA code is itself pretty new
to AMBER, there hasn't been a significant speed advantage,
until recently.
-
What about the
current scientific codes? What will happen to them? We
will continue to perform most calculations with Gromacs.
However, we expect that AMBER will take over for many of
the calculations which we would start on Tinker.
-
Is there a charge
for using AMBER? FAH has permission (from Dave Case
at Scripps) to run AMBER on the FAH supercomputer and no
FAH donators will have
to pay any sort of AMBER fee. However,
if someone wants to use
AMBER for their own research (i.e. not for running FAH,
but for some other purpose), then there would be a fee.
The AMBER core in FAH has been designed to run only FAH calculations and the
FAH AMBER core cannot be used for any other purpose.
-
What modifications
will you be making to AMBER? Will
your modifications to AMBER be available?When will we
see those patches released? Essentially all of our AMBER
modifications are involved with connecting AMBER to FAH. We will
make these available to AMBER developers, but likely they will
not include them in the AMBER distribution.
- I'd like to learn
more about AMBER. Where should I look? First,
check out the AMBER
web page. Also, a good
general overview of the Amber codes can be found in: D.A.
Pearlman, D.A. Case, J.W. Caldwell, W.R. Ross, T.E. Cheatham,
III, S. DeBolt, D. Ferguson, G. Seibel and P. Kollman. AMBER,
a computer program for applying molecular mechanics, normal
mode analysis, molecular dynamics and free energy calculations
to elucidate the structures and energies of molecules. Comp.
Phys. Commun. 91, 1-41 (1995). An overview of the
Amber protein force fields, and how they were developed,
can be found in: J.W. Ponder and D.A.
Case. Force fields for protein simulations. Adv. Prot.
Chem. 66, 27-85 (2003). Similar information for nucleic
acids is given by T.E. Cheatham, III and M.A. Young. Molecular
dynamics simulation of nucleic acids: Successes, limitations
and promise. Biopolymers 56, 232-256 (2001).
updated on
October 28, 2004
|
