Hi Xiaonan,
What you need to do here is just a basic decimal search until you find the
buffer size needed to give you the number of waters you want. Since leap can
be automated by reading an input file (tleap -f input_file) you should be
able to write a simple perl script (or similar scripting language of your
choice) that creates an input_file for leap with an initial guess at the
buffer. It then runs tleap and checks the output to see how many waters were
added. Then the script can just adjust the buffer size up and down by
varying fractions based on the number of excess or deficient waters and then
it repeats the process until you get the exact number of waters you want or
the number to within some simple tolerance.
Being able to write such scripts is not optional if you plan a career in
computational chemistry / molecular biology so if you don't know how to do
it now would probably be a good time to learn. I would suggest getting a
book such as "Teach yourself perl in 24 hours..."
http://www.amazon.com/gp/product/0672327937?ie=UTF8&tag=freelydownloa-20&lin
kCode=as2&camp=1789&creative=9325&creativeASIN=0672327937
Good luck,
Ross
> -----Original Message-----
> From: amber-bounces.ambermd.org [mailto:amber-bounces.ambermd.org] On
> Behalf Of gmail
> Sent: Thursday, February 05, 2009 9:21 PM
> To: AMBER Mailing List
> Subject: Re: [AMBER] How to build reference coordinate file for targeted
> MD
>
> Dear Dr. Simmerling
> Thank you for your suggestions. But I still did not figure out how to
> mannually adjust the buffer
> size in order to match the water molecules in starting coordinate.I used
> solvateoct command and carefully
> adjust the buffer size, but as my protein is quite big and surrounded by
> thousands of waters. I find it
> quite difficult to adjust to the exact number of waters.
> Is that a way that can mannually revise the topology or the
> coordinate file? I do not know the
> exact organization format of inpcrd file so no clue for this.
>
>
> Thank you!
>
> Xiaonan
>
>
>
>
> --------------------------------------------------
> From: "Carlos Simmerling" <carlos.simmerling.gmail.com>
> Sent: Thursday, February 05, 2009 7:19 PM
> To: "AMBER Mailing List" <amber.ambermd.org>
> Subject: Re: [AMBER] How to build reference coordinate file for targeted
> MD
>
> > if they are the same protein sequence but different atom order in side
> > each residue, leap will fix that.
> > for the water molecules you might need to manually adjust the buffer
> > size until you get the number from before.
> >
> >
> >
> >
> > On Thu, Feb 5, 2009 at 1:44 AM, gmail <heptoking.gmail.com> wrote:
> >> Dear All
> >> I am trying to do a targeted MD for a big protein. There is a
> question on how to build a reference
> >> coordinate file. As the manual pointed out that the reference
> coordinate are expected to match the prmtop
> >> file of the starting structure. However, the starting and final
> structures are two different pdb files,
> >> which although have the same amino acid sequence, the exact atom
> sequence are somewhat different. In addition,
> >> I am thinking that if I create the solvated reference prmtop and
> inpcrd files, the number of water molecules
> >> will be slightly different to those generated with starting pdb file.
> That will most probably cause failure of
> >> recognition.
> >>
> >>
> >> With no definite solution in my mind, I have to ask experts for
> suggestions.
> >>
> >>
> >> Thank you in advance.
> >>
> >>
> >> Cheers!
> >>
> >>
> >> Xiaonan Zhang
> >>
> >> Shanghai public health clinical center
> >> _______________________________________________
> >> 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
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Feb 06 2009 - 01:27:45 PST
