Dear CCTBX developers, I am a postdoc at EPFL working with HHpred for homology modeling of membrane proteins. I have been trying to write my own HHpred alignment parser until I found the python script under “cctbx_fork/iotbx/bioinformatics/__init__.py/” that contains an HHpred parser. My goal is to correctly parse the raw HHpred output file (.hhr), which involves unwrapping every alignment, parsing out a lot of text to finally obtain something like this:
pdb_name
query-sequence column score Example:
4U15
VYGFIGGIFGFMSIMTMAMISIDRYNVIGRPMAASKKMSHRRAFIMIIFVWLWS +........+..+..++|+++|++++.++.+.++++ +..+.++.+|+++|++.++...+........ +...|.. Being somewhat new to python, I was wondering whether the people who wrote this script are still around and could help me figure out whether the parser could be implemented in such a way. Thanks for any help you can provide! Best, Louis D