How to taxonomically subsample a proteomes file ?

I have a proteomes .faa file with all the protein sequences encoded by LUCA. I want to create a file with the proteomes of only a few specific eukaryotes, few specific archaea and all bacteria. How do I do this ?

I have tried downloading the taxdmp.zip file from the NCBI database and but the nodes.dmp and names.dmp file is not making sense to me. I am an undergrad and I would appreciate any help