Re: AW: Various questions

From: Douglas Houston (DouglasR.Houston_at_ed.ac.uk)
Date: Wed Jan 22 2014 - 04:30:17 CST

Hi Norman,

Many thanks for your prompt reply.

1. I guess you've listed all the reasons why I only read about
periodic boundary conditions being used, so what might you simulate in
a sphere using non-periodic boundary conditions for?

2. Indeed it's true that a program could not guess the box size of a
system it doesn't know anything about. No doubt that is why in the
tutorial you load the protein molecule into VMD before using the
'measure minmax' and 'measure center' functions. By loading the
molecule, the program knows something about the system. Then you must
calculate by hand the dimensions (there does not appear to be a VMD
function that does this, correct me if I'm wrong) and, again by hand,
type these into NADM's .conf file. I guess I can script it though.

3. To clarify, I'm not looking for a MD production run protocol,
simply for an example of a setup protocol. The following is an example
relaxation protocol from a Gromacs "Simulation Setup" tutorial, for
which all the keywords are provided (as the NAMD tutorial says "you
can perform several different types of simulation using this structure
as a starting point"):
Add hydrogens
Steepest descent minimise in vacuum, no restraints
Add waters & ions
Steepest descent minimise in vacuum, no restraints, periodic boundary
conditions
Short MD with solute atoms fixed
Short unrestrained MD with temperature coupling

Here is an example relaxation protocol from the "Preparing a Protein
for MD Simulations" Desmond tutorial, again for which all the keywords
are provided:
Add hydrogens, water and ions
Steepest descent minimise with restraints on solute (a force constant,
not fixed)
Steepest descent minimisation without any restraints
Short MD using NVT, force restraints on solute (heavy atoms only)
Short MD using NPT, force restraints on solute (heavy atoms only)
Short MD using NPT, no restraints

Note the subtle differences between this and the Gromacs example,
although they both achieve the same ends.

I bet that just about everyone who uses NAMD has a .conf file they use
to achieve the above (i.e. set up and relax their system, before any
production run). I would really appreciate it if someone sent me this
file.

Many thanks again,

Doug

Quoting Norman Geist <norman.geist_at_uni-greifswald.de> on Wed, 22 Jan
2014 09:42:07 +0100:

> Douglas,
>
>> -----Ursprüngliche Nachricht-----
>> Von: owner-namd-l_at_ks.uiuc.edu [mailto:owner-namd-l_at_ks.uiuc.edu] Im
>> Auftrag von Douglas Houston
>> Gesendet: Dienstag, 21. Januar 2014 16:42
>> An: namd-l_at_ks.uiuc.edu
>> Betreff: namd-l: Various questions
>>
>> Dear all,
>>
>> I am a beginner when it comes to NAMD but have been working through
>> the helpful tutorials. However, I have a few questions that I have not
>> been able to find the answers to:
>>
>> 1. The NAMD tutorial includes an example of a simulation in a sphere
>> with non-periodic boundary conditions. Why do I only ever read about
>> simulating in a box with periodic boundary conditions in the
>> literature? Maybe I'm just not reading enough ;-) but is there a
>> discussion of the pros and cons of each anywhere?
>
> There are several reasons why one uses periodic boundary conditions:
>
> 1. Easier to parallelize
> 2. Fast PME long range electrostatics
> 3. More realistic environment to system
> 4. NVE difficult for reflective boundary
> 5. Energy conservation difficult for reflective boundary
> 6. Analysation of diffusion difficult with reflective boundary
>
> [...]
>
>>
>> 2. I have used mainly Gromacs in the past - Gromacs automatically
>> detects and applies its equivalents of NAMD's Cellorigin and
>> BasisVector 1, 2 and 3 keywords. Is there a way to get NAMD to do this?
>
> To me this doesn't make much sense since this values must be set really
> carefully and no program can't ever
> guess the real box size of a system it doesn't know anything about. Example:
> periodic crystal.
>
>>
>> 3. From the NAMD tutorial: "One typically minimizes the system and
>> then equilibrates with the atoms in the protein fixed in space." But
>> there is no mention of this in the example .conf file that is
>> provided. What are the keywords relevant for fixing solute? Does
>> anyone have an example .conf file of a "standard" protein relaxation
>> protocol? I have seen example protocols for Gromacs and Desmond but is
>> there a "default" one for NAMD?
>
> Check the namd manual for "fixedatoms". The simulation protocol depends on
> your needs. This has nothing to do with namd anyway but with md itself and
> do always depend on what you are doing.
>
>>
>> 4. I have been working on generating a .psf for an olefin stapled
>> peptide. I have added an entry to the CHARMM parameter file and got
>> psfgen to output a .psf that looks OK in VMD (by OK I mean that the
>> right bonds are visible and hydrogen addition looks sensible).
>> However, I'm worried there may be some BOND, DIHE, IMPR or other
>> inappropriate descriptions I've missed (for example, I've used patches
>> to connect the linker to the backbone, which necessitated deleting the
>> existing backbone peptide bonds - but have I deleted everything I
>> should have?). I understand a validation run might identify incorrect
>> atomic motions but before that is there a way to visualise all entries
>> in the .psf file using VMD (or should I direct this to the VMD mailing
>> list)?
>>
>> 5. Related to 4., are entries like IMPR disregarded if there is not a
>> corresponding BOND? If so that would make me less worried about
>> catching every single one.
>>
>> 6. Why are the hexene methanediyl hydrogens in the
>> top_all27_prot_lipid.inp file a different atom type (HAL2) than e.g.
>> lysine side chain methanediyl hydrogens (i.e. HA)? I note the charges
>> are the same. The olefin staple is essentially a monounsaturated lipid
>> and I've built it using -CH2 parameters taken from amino acid side
>> chains but the hexene parameters seem equally correct to me. What's
>> the difference between these atom types?
>>
>> 7. I think that's enough questions for now!
>>
>> Many thanks for any advice.
>>
>> cheers,
>>
>> Doug
>>
>>
>> _____________________________________________________
>> Dr. Douglas R. Houston
>> Lecturer
>> Room 3.23
>> Institute of Structural and Molecular Biology
>> Michael Swann Building
>> King's Buildings
>> University of Edinburgh
>> Edinburgh, EH9 3JR, UK
>> Tel. 0131 650 7358
>>
>> --
>> The University of Edinburgh is a charitable body, registered in
>> Scotland, with registration number SC005336.
>
>
>
> ---
> Diese E-Mail ist frei von Viren und Malware, denn der avast!
> Antivirus Schutz ist aktiv.
> http://www.avast.com
>
>
>

_____________________________________________________
Dr. Douglas R. Houston
Lecturer
Room 3.23
Institute of Structural and Molecular Biology
Michael Swann Building
King's Buildings
University of Edinburgh
Edinburgh, EH9 3JR, UK
Tel. 0131 650 7358

-- 
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.

This archive was generated by hypermail 2.1.6 : Thu Dec 31 2015 - 23:20:25 CST