There are a number of issues to consider. I'll do my best to explain.
Of course, ask if you have questions about it. You also sound as if
you might be a little new at this. If so, you will likely encounter
many unexpected requirements. Searching this archive will help a lot.
If that doesn't work, of course write back.
Force fields:
You need two force fields -- three, technically, if you count water
(ions, if present, being a fourth?). The rest of this focuses only on
the protein and carbohydrate. The best bet right now is ff99SB for
the protein and glycam06 for the carbohydrate. However, depending on
your system, you might need to make some modifications to the force
field information. At the simplest level, you can simply add them in
to your leap (tleap or xleap) input in the manner you would normally
add any other force field. Definitely check for newer force fields,
tho. I know some with improvements are in the works.
Carbohydrate:
Since you're using Amber 8, you will not be able to use mixed scaling
for the 1-4 interactions (electrostatic and van der Waals). If you
have no idea what that means, the most important issue is the
flexibility in your carbohydrate. If all parts of the carbohydrate
are reasonably tightly bound to the protein, then you should be ok
using the protein's scaling. If your carbohydrate is loosely bound,
or if parts of it are not closely associated with the protein, then it
might be advisable to consider moving to a more recent Amber. If you
are in the latter situation, let me know more about your system, and I
will try to help you decide.
Atom types:
If your system contains sialic acid (NeuNac, Neu5Ac), then there might
be a conflict of atom types. There is a fairly simple workaround to
this, and we will be implementing it soon. But, you do need to be
careful of it.
Coordinates:
At this time, you almost certainly need to have an initial set of
coordinates with the carbohydrate already bound. If you don't have a
crystal or nmr structure, you can try docking or some other method.
But, you need coordinates first.
top & crd:
You only need separate top & crd files if you plan to perform some
modeling or analysis with the complex separated. If you plan to do
that, again, be very careful. Make sure you understand very well what
you are doing. What might seem the most intuitive may not the the
best course. You might, however, build them separately first just to
minimize potential sources of errors. What I mean is that you will
likely encounter little things that need to be modified. If you work
all those out on the two parts separately first, then it will be less
headache than trying to figure out the whole thing at once. However,
you do not *need* to build separate top & crd for protein and
carbohydrate in order to build the complex. You only need them built
separately if you plan to do certain specialized types of simulation.
On Mon, Jan 16, 2012 at 2:42 AM, john Kerk <gor63.hotmail.de> wrote:
>
> Dear Dr. Lachele
> Thank you for reply. I simulate a protein-carbohydrate complex. But, as I think, I should at first create a .top and .crd files for protein and carbohydrate individually and then both files for the complex. Please, should I use two force fields (for example, glycam06 and amber.ff99) for protein-carbohydrate complex ?
>
> thank you in advance!
>
> john
>
>
>> Date: Sun, 15 Jan 2012 14:07:58 -0500
>> From: lf.list.gmail.com
>> To: amber.ambermd.org
>> Subject: Re: [AMBER] Using glycam06 under Amber8
>>
>> Carbohydrate alone, or carbohydrate and protein?
>>
>>
>> On Sun, Jan 15, 2012 at 1:45 PM, john Kerk <gor63.hotmail.de> wrote:
>> >
>> > Dear Amber users
>> > I am using Amber8 version in my MD simulations but I do not know how can implant glycam06 force field in Amber8 version. Please any suggestions (in detail, if possible) will be helpful.
>> > John!
>> >
>> > _______________________________________________
>> > AMBER mailing list
>> > AMBER.ambermd.org
>> > http://lists.ambermd.org/mailman/listinfo/amber
>>
>>
>>
>> --
>> :-) Lachele
>> Lachele Foley
>> CCRC/UGA
>> Athens, GA USA
>>
>> _______________________________________________
>> 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
--
:-) Lachele
Lachele Foley
CCRC/UGA
Athens, GA USA
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Jan 16 2012 - 11:00:02 PST
