[gmx-users] gmx pdb2gmx for synthetic nucleic acid sequences

[Justin Lemkul]

On 3/12/16 6:55 PM, fulvio ciriaco wrote:
> Hello,
> I have to simulate some non natural sequences containing sometimes both
> T and U nucleic acids.
> However pdb2gmx incorrectly detects the terminus, i.e., for a sequence like
> TCAUA
> it says
> Identified residue DT1 as a starting terminus.
> Warning: Residue RU4 in chain has different type (RNA) from starting residue DT1
> (DNA).
> Warning: Residue DA5 in chain has different type (DNA) from starting residue DT1
> (DNA).
> Identified residue DA3 as a ending terminus.
>
> Then it does not treat DA5 as a ending terminus, i.e. , with amber99sb-ildn ff
> Fatal error:
> Atom H3T in residue DA 5 was not found in rtp entry DA with 32 atoms
> while sorting atoms.
>
> Is there some way to circumvent the excessive knowledge pdb2gmx has of biological
> systems?
>

Make a local copy of residuetypes.dat and set RU to DNA.  pdb2gmx functions to 
prevent people from mixing things that shouldn't be mixed; if you are doing 
something strange, this is the only way to convince pdb2gmx to listen to you 
rather than its programming :)

> I attach the tertiary structure for one such sequence.
>

The mailing list does not accept attachments, but none is necessary here.

-Justin

-- 
==================================================

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

==================================================