Build a peptide from a starting residue

[brianjimenez]

Hi,

First of all, congratulations on this nice piece of software. I would like to know if it's possible to use PeptideBuilder to build a peptide starting from a PDB file which contains a single residue and use its atom coordinates in the process.

Thank you in advance for your time and help!

[mtien]

Hi,

Can you clarify the statement "and use its atom coordinates in the process"?

[brianjimenez]

Sure, I mean using the only residue from the PDB file as an anchor and build the rest of the peptide given a sequence.

[mtien]

PeptideBuilder cannot modify an existing PDB file. If the peptide anchor coordinates in (x,y,z) space is important for your purposes, you can investigate the method "def initialize_res(residue: Union[Geo, str]) -> Structure." An override of this method will allow you to specify the (x,y,z) coordinates of your anchor residue instead of how it is currently written to start at (0,0,0). You will be able to take your PDB file and create a new PDB that can build peptides of your given sequence.

I hope I am understanding your question correctly?

[brianjimenez]

Amazing, that is basically what I needed :)
Thanks a lot for your quick and useful answer!

[clauswilke]

PeptideBuilder cannot modify an existing PDB file.

Matthew, is this fully true? If I understand correctly how PeptideBuilder works, it appends residues to the end of chain A model 0 of a biopython structure object. Shouldn't it be possible to read a PDB file that has only a single chain, A, (containing one or more residues) and then append to that chain?

[mtien]

Claus, you are correct, in that this is not "fully true". From the tester scripts we have provided, one can "read" the geometries of a PDB file (lets call it "A.pdb") and then start the construction of a biopython structure object. This will not modify the original read A.pdb file, but will create an object that can be used to append residues to their specified anchor during the reading. The created object during reading and construction will anchor the first read residue at (0,0,0). If the user wants to retain the (x,y,z) coordinates of A.pdb, then that will take a more sophisticated approach to the "initialize_res" method. However, if the user is ok with this nuance, they in essence can modify an existing A.pdb, and should write the modified biopython PDB as "A_modified.pdb" for instance.

[clauswilke]

Ah, I see.

@brianjimenez There's also always the option of first constructing the peptide you need anchored at (0, 0) and then using the PDB module of biopython to translate/rotate the peptide in space to where you need it.

[brianjimenez]

Thanks a lot for the answers and discussion. @clauswilke , we plan to use PeptideBuilder starting from an anchor residue which has been predicted to be in a specific binding position with a receptor protein molecule (kind of threading, but in complex context). For that reason the approach of building and after superimposing won't be good enough.