AMBER Archive (2009)

Subject: RE: [AMBER] database

From: FyD (fyd_at_q4md-forcefieldtools.org)
Date: Fri Sep 18 2009 - 19:49:17 CDT

Dear Steve,

> Thank you so much Prof. Case. The "pennie dropped" as they say when
> I throughly read your email. You were correct. The solution was
> making my mol2 one residue, calling it P2M, and then coordinating
> that with how I introducted it into leap containing the newdCTP.pdb
> file with residue P2M. I took my files and wrote leap.in as:
>
> P2M=loadmol2 newdCTP.mol2
> DNA_pol=loadPdb newdCTP.pdb
> Saveamberparm DNA_pol DNA_pol.prmtop and DNA_Pol.inpcrd
>
> Thanks Prof. Case and Franscois for having patience in showing me
> how to do this!!!

In these cases, sometimes this is better to take a break, and work
again on it only after a good night of sleep ;-)

I played with dCTP2 represented by a single residue (instead of 4) + a
tripeptide NALA-GLY-CALA (representation of your protein) and it was
fortunately OK.

We obtained these co-factors libraries from building blocks generated
using R.E.D. IV. R.E.D. does not generate dCTP2.mol2 directly but the
P2M, P1 & C5 building blocks in the mol2 file format. dCTP2.mol2 was
constructed in a second step using LEaP and these elementary building
blocks. We thought it would be more convenient to use pre-built
co-factors instead of these individual P2M, P1 and C5 FF libraries.
Thanks to your report & the help of Dr Case, it turns out it might not
be true and directly using the elementary building blocks might be a
best bet; Anyway R.E.DD.B. will contain the building blocks + the
pre-built co-factor version; Thus, users will be able to construct
their co-factors by themself using these building blocks or to use our
pre-defined version... Thanks for this discussion.

We tried here to play with t/xLEaP to re-generate the P2M, P1 & C5
building blocks from the dCTP2.mol2 (or dCTP2.off) file without success:

P2M = copy dCTP2.1 seems to work
desc P2M seems to work as well
  but
savemol2 P2M P2M.mol2 1 does not work
edit P2M does not work

It might be normal as P2M is a "constituent" of dCTP2. However, this
would be convenient if such a copy would be possible. We did not try
sleap...

regards, Francois

_______________________________________________
AMBER mailing list
AMBER_at_ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber