I am using the Rosetta software to make comparisons with a peptide folded by a molecular dynamics protocol. Although, I am encountering some issues as, for example, the sheets modelling.
In my molecular dynamics assay I found a b-hairpin-like conformation for the peptide evaluated, but Rosetta only gives me structures featuring alpha-helixes.
I read some discussions pointing to a deficience in Rosetta’s algorithm on beta-sheet formation so I wanted to know if there is a practical way of favouring a beta-hairpin conformation without biasing my results.
I’m not really clear on what you’re doing with Rosetta here. Which executeable (or is it custom?) Are you doing structure prediction of just the peptide, or is it in complex with something?
Three ways I can think of off the top of my head to tweak this secondary structure formation:
A) If you are using fragments, check what fragments you have. If you’ve only helical fragments then that’s probably your problem; perhaps a more diverse group of fragments will help. (I don’t know how to get custom fragments, unfortunately).
Atom-pair distance constraints could be used creatively to favor one hydrogen bonding network over another. This is clearly biasing your results…
C) You could try tweaking the scorefunction. The backbone-backbone hydrogen bonding scores are split up into long-range (beta sheet) and short range (helix) terms with separate weights. I’m not sure where a hairpin would fall here; it might be too short and still fall in the helix bin. In any case you could try upweighting the longer-range BB hbonding term.