Hello everyone,

I have two issues on which I would seek your advice.  I am using PyMol 1.4.1.

The first concerns creating protein structures from scratch.  I select Build > Residue from the menu, and then I select my desired secondary structure.  Then I use the Alt shortcut keys to add amino acids.  Then I select Display > Sequence.  The first amino acid I get will sometimes be numbered 2, or 3, and in one case the numbering began at residue 22!  But never has the first residue received the CORRECT number, which is of course, 1.  This happens even when I start PyMol fresh, and there are no other structures loaded which might influence the numbering.

Is this a bug?  Or am I doing something wrong?

My second issue concerns edited proteins, and it also relates, in part, to numbering.  Recently, I have been constructing deletion mutants.  Typing "help editing" into the command line gives you a passage which begins as follows:

> PyMOL has a rudimentary, but quite functional molecular structure
> editing capability.  However, you will need to use an external mimizer
> to "clean-up" your structures after editing.

(As a quick aside, I'm wondering what a "mimizer" might be.  If this was a typographical error, and the word "minimizer" was meant instead, that still doesn't help me to understand the meaning of this sentence.)

According to various PyMol displays, removing an amino acid does nothing more than that.  It does not create two objects where there was one.  Now, if I save a PDB file of an edited polypeptide, things are not so clear.  A TER record is inserted where the residue was deleted.  The numbering of residues on each side of the TER is unaltered.  So if you deleted residue 8, for example, then the TER record is preceded by residue 7 and followed by residue 9.  Using the Select Molecules option will give you one fragment or the other, but there are apparently no separate names for these fragments which would allow one to use a join command.  Using Select Objects gives you both fragments as one object.

Looking deeper into the PDB file format specification, I think I see what the problem might be. 


In ATOM records, column 22 can specify an optional "chain identifier".  It can be left blank, according to the specification, but if it is left blank, the assumption is that only a single polypeptide is present in the file.

So, which way is PyMol treating this collection of amino acids after a deletion?  As one polypeptide, as suggested by the blank chain identifier field and the Select Objects result?  Or, as two polypeptides, as suggested by the presence of a TER record and the Select Molecules result?

I know that other programs that I work with, downstream of PyMol, have trouble with this ambiguity.  For now, I am hand-editing PDB files.  I'm a decent programmer. I can write a clean-up routine which would delete undesired TER record from PDB files, and renumber amino acids.  But if these functions can (and/or should) be handled from inside PyMol, I would appreciate knowing how.