Dear AMBER community,
I'm trying to perform some RAMD simulations with amber22, and I have one
question and two remarks:
Question:
I struggled to find proper information about RAMD in AMBER, but as far as I
understand, its behavior differs from the NAMD implementation.
In NAMD the force direction is maintained at the next RAND cycle (after
randint steps in AMBER) if the ligand has moved less than a certain amount,
defined by the user.
As far as I understand this behavior is not reproducible in AMBER, as at
every "randint" a new random force is applied, with a new random direction.
The simulation stops when the ligand has moved away a "randmaxdist"
distance, and that's it.
Differently from NAMD, in AMBER the user can change how often
(ramdboostfreq) and how much (ramdboostrate) the boost acceleration is
modified.
I would like to know if it would be possible to have an implementation that
better reproduces the NAMD behavior. I think that would be able to more
efficiently "scan" for unbinding pathways, as with the current
implementation we are adding continually-changing random forces, which can
easily "counteract" their effect, slowing the process.
Maybe I'm just not understanding properly how RAMD works or how it's
implemented in AMBER, so any help would be greatly appreciated!
remark1
I just wanted to point out that the manual refers to the protein mask as
"randprotmask", while both pmemd and pmemd.cuda need the keyword
"randprotEINmask". I wanted to point that out because when running the
simulations with the cuda version of pmemd the error is not reported, and
this hopefully can help someone else spare some time (and it's gently
reminder to correct the manual :-) )
remark2
ramdboostfreq and ramdboostrate should be set by the user to a value
different than 0, otherwise, a general floating point error is raised
(invalid operation or something like that, I can't recall the error right
now). This I think should be stated in the manual, as it can be confusing
for a new user, and the error is not that informative.
Thank you all for your help, and for developing this amazing software!
Best regards,
Domenico
*Domenico Marson*, Ph.D.
Ricercatore T.D. (A) | Research Fellow
Università degli Studi di Trieste | University of Trieste
Department of Engineering and Architecture (DEA)
Molecular Biology and Nanotechnology Laboratory @ UniTS (MolBNL.UniTs)
Via Valerio, 10 - 34127 Trieste (Italy)
domenico.marson.dia.units.it
+39 040 5583437
Skype: domenicomars
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Jan 11 2024 - 03:00:02 PST
