This is a place to jot down project ideas, and link to WikiWords^? for projects where the ideas are more fleshed out.

• OptimizingRosetta for ideas specifically pertaining to the Rosetta code.
• MageFlips^?: Jane Richardson wants us to calculate the cost of flips in mage so that it can be integrated into MolProbity^?.
• Fast pruning of a tree-structured rotamers. We've implemented a RotamerTrie for Rosetta, now can we look at the general problem of rotamers that are structured as overlapping trees:
  • How can you quickly prune out those branches that will have intersections with backbone? Are there quick rejection tests that show that branches of a tree cannot fit into a small space? perhaps by computing the neighborhood of one atom and using that to prune sets of nodes from the tree.
  • If you place two tries next to each other, is there a way other than brute force to determine how they can fit tightly together -- perhaps by exploring separating planes?
  • A completely different way is to think of rotamer atoms as point clouds and evaluate energies of one point cloud against another by a WellSeparatedPairDecomposition. Does this give us energies any faster?
• Baker's outside-in rotamers: How would we do rotamer tries starting from the tips (Baker's theozymes?) Given a collection of possible triples of Calpha positions of rotamers with fixed tips, how do we find a scaffold that can match the positions without clashing with the rotamers?
• Ramachandran plots for RNA: what are the allowable ranges for angle parameters in the hard sphere model, and how do they compare to those observed by Berman's and the Richardsons' labs?
• RigidityQuestions there are a number of hard rigidity questions, including the MolecularConjecture^?.

-- JackSnoeyink - 02 Apr 2005
to top