OpenMM Roadmap
- Most critical (Q3/Q4 2013)
CustomNonbondedForce analytical dispersion corrections (necessary for explicit solvent free energy calculations)
- faster postprocessing (faster querying of multi-poles)
- Drude oscillators (CHARMM Langevin Drude integrator)
- RPMD ring contraction (to improve performance)
Migrate OpenMM to GitHub to facilitate community contributions
- OpenMM 5.x (Q1/Q2 2014)
- report energy in parts (solute, solvent, solute-solvent, electrostatics, vdw, etc)
- Improve convergence algorithm for AMOEBA
- Multiple time step integrators (RESPA script)
- Virtual sites (more types, achieve larger timesteps)
- Colored noise integrator (TM)
- Custom electrostatics force class (include ability for charge penetration and damping, DPP2 functional)
- package/distribute test binaries with release for easy testing
- Faster AMBER prmtop/inpcrd reader (faster unit conversion?)
- repex.py integration in OpenMM
- Online tutorials
- radius optimizer for direct vs reciprocal space PME
CustomIntegrator access to box vectors for stochastic barostats
- forcefield conversion
- more forcefields (recent AMBER forcefields, CHARMM, OPLS)
- provenance data (where it came from and when it was converted) created in XML file during conversion
- Longer term (2015)
- Normal mode analysis script
- AMOEBA OpenCL implementation
- Constant pH implementation (JDC)
- YANK release soon (JDC)
- test/validate ABSINTH implicit solvent
- More modeling tools within OpenMM app
- Further development Rosetta force field
- Triclinic boxes
- A more accurate SASA calculation for use with GB models
- Parameterize GB/VI at different temperatures
- CHARMM27 force field
- Thermodynamic ensemble validation tests
- PME for Lennard-Jones
- Much longer term