support to cases where the initial structures are multiple PDB files for the same uniprot ID #88
Description
An example of this was ZNF257 where we had to model with AF3 two parts of the protein separately to improve the final accuracy. We have a workaround for cases like this but we are wondering if we could actually implement the full support to it into the automatization snakefile.