Hey! I'm trying to count the strands in a beta-barrel, so that i can group proteins based on their barrel size (not strand amount). It seems like a simple enough issue, but I can't quite get it to work. I'm using 3slj and 2qom as test structures (both are of the same protein at different stages, and is considered a 12 stranded barrel) and will scale up to a much larger database of pdbs. If i use DSSP to count strands by stringing together beta-strand residues and adding it to a counter i run into problems when the structure has small undefined regions inserted into what should be a single strand! I also thought about using turns to count the strands as the amount of turns between beta strands should correlate to the amount of strands, however this method falls on it's face when it has to deal with structures like 2qom which has turns regions frickin' missing from the structure. I'm quite at a loss about what to do, is there a database i can reference to get this info, or another method? I really don't fancy having to count the barrel sizes for each protein. any help is very welcome!!!
