I would like to obtain a listing of all the single domain chains in the PDB for training an ML model. Currently, it is not clear how I would go about doing this without doing some heavy lifting. Automating the download of PDB files from RCSB and parsing them is not a problem for me. I just don't know a simple way of going about building the list of PDB files for download.

A related question would be how to obtain such a list for chains less than a given number of residues long. Another thing that may help is a list of contiguous domains that I can then chop from chains. Many domain definitions define several segments whereas for my needs I need an unbroken chain of residues, not necessarily with ATOM records, but no holes in the domain definition itself.

Either one of the above will probably serve my needs well. Thanks for any help.