Generative Modeling of Molecular Dynamics Trajectories
About
Molecular dynamics (MD) is a powerful technique for studying microscopic phenomena, but its computational cost has driven significant interest in the development of deep learning-based surrogate models. We introduce generative modeling of molecular trajectories as a paradigm for learning flexible multi-task surrogate models of MD from data. By conditioning on appropriately chosen frames of the trajectory, we show such generative models can be adapted to diverse tasks such as forward simulation, transition path sampling, and trajectory upsampling. By alternatively conditioning on part of the molecular system and inpainting the rest, we also demonstrate the first steps towards dynamics-conditioned molecular design. We validate the full set of these capabilities on tetrapeptide simulations and show that our model can produce reasonable ensembles of protein monomers. Altogether, our work illustrates how generative modeling can unlock value from MD data towards diverse downstream tasks that are not straightforward to address with existing methods or even MD itself. Code is available at https://github.com/bjing2016/mdgen.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Protein Conformation Generation | MD-Cath 320K 1.0 (500 conformations) | Pairwise RMSD0.79 | 8 | |
| Protein Conformation Generation | MD-Cath 450 K (test) | Pairwise RMSD JSD0.22 | 8 | |
| Protein Conformation Generation | MD-CATH | FNC JSD0.2 | 8 | |
| Protein conformational sampling | MD-Cath 450K S20 homology level (test) | Pairwise RMSD (Pearson r)0.43 | 8 | |
| Trajectory Generation | ATLAS 14 protein monomers (test) | JSD (Rg)0.493 | 6 | |
| Protein Trajectory Generation | 4dhkB00 (159-residue protein) (test) | Wall-clock Time (s)31.7 | 5 | |
| Conformational Distribution Matching | Tetrapeptides | Torsions (bb) JSD0.13 | 5 | |
| Molecular Dynamics Trajectory Generation | MDGen Tetrapeptide | Torsion BB0.13 | 2 |