How to extract all Pfam Sequences with cleared PDB Structure?

Question

[twinstar2]

I downloaded and extracted the Pfam-A.full.ncbi.gz from ftp://ftp.ebi.ac.uk/pub/databases/Pfam/releases/Pfam31.0/

I splittet the huge file into smaller files for each family like: https://www.dropbox.com/sh/t0b173oa8odvsne/AABZLhiaq-jZtE5PrULvBm9na?dl=0

Now i try to extract all Sequences from the Alignment, which have an observed/cleared structures in the pdb database. Luckily Pfam provides the pdbmap file which, links pdb IDs to Pfam IDs. I can extract all Pfam families, which have observed structures in the pdb, but i can not extract the corresponding sequences. So any help on how to accomplish this is very appreciated.

PS: I currently use Python 2.7 with BioPython, but it can't handle the files.

Answer 1

[twinstar2]

Okay for anyone with the same problem, i found the answer!

1. Extract all UniProtKB ids (line 5) from the pdbmap file into a list (you may have to make smaller sub parts of the list due to Server limitations)
2. Upload the List here: http://www.uniprot.org/mapping/ and select "UniProtKB AC/ID" as Input and "EMBL/GenBank/DDBJ CDS" as Output
3. Rejoin the target_id list
4. Iterate over all Pfam files and dump all sequences which are not in your targed_id list
5. Let it run for 2 days
6. ????
7. profit

This will get you ~ 70.000 sequences