Measuring pairwise distances between residues/atoms in structural alignment of protein family

Question

[Jonathan Lefebre]

Hi all,

I analyzed binding pockets across X-ray structures within a protein family and aligned them based on their structure. Now I would like to compare the pockets to make a statement about structural conservation of these pockets. For this, I would like to measure the distances between the centers, so that I can create a matrix of pairwise distances of each pocket and then cluster them. I am new to python/biopython and only know how to measure distances between single residues in different proteins in pymol, but this would take forever to do it for 60 structures with an avereage of 8 pockets per protein. Is there a way in biopython to approach this? The pocket centers are defined as heteroatoms in a different chain than the protein (protein = chainA; pockets = chainZ).

Any kind of help is appreciated!

Best

Jonathan

[Woa]

check this toolkit:

https://www.biorxiv.org/content/10.1101/605907v2.full.pdf

[Jonathan Lefebre]

Dear Woa,

I really appreciate your suggestion, but the bitbucket link in the preprint does not work and I cannot find any further information on the toolkit. It also seems like no one else really used it, since it was not published since 2019. But I might get something wrong.

Best

Jonathan

Answer 1

[jgreener]

There are efficient and flexible distance calculators available in BioStructures.jl if you are okay with using Julia.

Answer 2

[Woa]

haven't used Pymol for a while but I think you can write a script to automate the analysis rather than doing interactively

Answer 3

[Jonathan Lefebre]

Thanks for the suggestions! I actually found a lazy solution to my problem :). In ChimeraX there is a clashes command that is supposed to find atom clashes, but if you set the threshold for a clash to lets say 100 A, every atom-atom distance will be recognized as clash. The respective distance of each "clash" can be saved in a output txt file. Like this you get pairwise distances between each and every atom in your input file.

If anyone of you has a good tool to plot and cluster the pairwise distances in form of a matrix, suggestions are also very welcome!

Thanks a lot.

Jonathan

[Joe]

Maybe this would be helpful? https://warwick.ac.uk/fac/sci/moac/people/students/peter_cock/python/protein_contact_map/