From: Audrey Salazar (
Date: Mon Aug 06 2007 - 14:47:08 CDT

Thank you again for your reply and suggestions. I have tried a few
things based on your comments and I am writing this email to report my
results and ask some follow-up questions. Questions will be marked
with ***.

In my previous email (I am hoping that it will be added to the same
string when it is available in the mailing list archive), I described
the .pdb file that i am using and the method that I used to make the
file. The coordinates of the atoms are included in that email.

On 8/6/07, Peter Freddolino <> wrote:
> Hi Audrey,
> a couple of notes; hopefully this is helpful:
> Audrey Salazar wrote:
> >
> > Step 1) I added the topology (top_all27_na.rtf) and parameter
> > (par_all27_na.prm) files that I obtained from
> > I
> > clicked "load". (No problems here)
> >
> Parameter files shouldn't be included here; only topology files

Good to know. Uploading only the topology file does not single
handedly affect the psf builder's output.

> > Step 2) I included "Everything" and clicked "Guess and split chains
> > using current selections." Three lines of text appeared in the "Chains
> > Identified" box:
> > N1 3 1- 62 5TER 3TER RNA
> > N2 3 63- 66 5TER 3TER RNA
> > N3 1 67- 67 5TER 3TER DNA
> > and of the three lines, I kept the ones that had "RNA" in the "Type" column.
> >
> > ** I am unclear as to the meaning of the output above. Does each row
> > correspond to one chain? Is there any documentation as to what the
> > columns headings stand for in the "Chains Identified" box?
> >
> Each row corresponds to one chain. It seems odd in your case that
> multiple separate chains were identified; you should have a look at the
> second and third chain and see what they actually consist of (clicking
> on them in the chains identified box will highlight them in the main vmd
> opengl window). I will try to clarify the docs on the column headings;
> for the record, the ones which might be ambiguous are Length (number of
> residues), Index Range (serial numbers of the atoms in the chain), and
> Nter/Cter (which will be relabeled First/Last in the next release to be
> less protein-centric, and correspond to the patches applied to the first
> and last residues in the chain

Thank you for the clarification. It seems that the hydrogens that I
added to the pdb file by hand are being considered as separate chains.

*** Any idea as to why this is the case?

I don't
> > ** Should I have deleted one of the RNA lines too? I.e. should I only
> > see one line of text if there is only a single molecule in the pdb
> > file?
> >
> Probably. You may need to instead merge them by editing the first chain
> to include all of the atoms, if for some reason a chain break is being
> introduced where there shouldn't be one. You can do this by deleting the
> second and third chain, and then clicking "Edit Chain" with the first
> one highlighted, typing the selection all into the selection box,
> clicking get indices, and then applying these changes.

*** In light of the fact that what the automatic psf builder calls
separate chains are really some hydrogens that have been categorized
separately, a new question comes to mind. Will the psf be written as
if all three chains belong to one chain or will they be written as
three different chains? This question may not be too clear, please let
me know if I can be more specific as to what I mean.

Thank you for your time and help.


Audrey L. Salazar
Amaral Research Group
Dept. of Chemical and Biological Engineering      Phone: 847.491.2188
Northwestern University
Evanston, IL USA