Extract and analyze beta-sandwich features.
- Author
- Doo Nam Kim (doona.nosp@m.m.ki.nosp@m.m@gma.nosp@m.il.c.nosp@m.om, started from Tim Jacobs' code) @overview @ task 0: Determine whether we deal with given pdb file @ task 1: Identify all beta-strands @ task 2: Identify all beta-sheets with these strands @ task 3: Identify all beta-sandwiches with these sheets @ task 3-1: Merge sheets to each other if possible @ task 3-2: Make beta-sandwiches with sheets that are ideal only @ task 3-2-1: Exclude if this_sheet_is_surrounded_by_more_than_1_other_sheet @ task 3-2-2: Exclude sheets that are too close to each other @ task 3-2-3: Exclude sheets that are too distant to each other @ task 3-2-4: Exclude sheets that do not face each other @ task 3-2-4-1: Exclude sheets that do not face each other by an angle with two terminal residues and one central residue @ task 3-2-4-2: Exclude sheets that do not face each other by an angle with four terminal residues in two edge strands @ task 3-4: Test canonical sandwich test @ task 3-4-1: Canonical sandwiches need to have low number of helix or strand residues in any loop (beta-hairpin-loop or inter-sheet-loop) @ task 3-4-2: Canonical sandwiches need to not have same-direction-strands as connecting two beta-sheets @ task 3-4-3: Canonical sandwiches should not be beta-barrel obviously @ task 4: Write beta-sandwiches that passed canonical tests into database @ task 4-1: Write hairpin_loop and inter-sheet loop @ task 4-2: Write starting_loop and endng_loop @ task 4-3: Write ratio_hydrophobic_philic/net_charge @ task 4-4: Write total size of sandwich @ task 5: Write resfiles automatically