Parallel Sampling of Protein-Ligand Dynamics
Masters, M. R.; Mahmoud, A. H.; Lill, M. A.
Show abstract
AO_SCPLOWBSTRACTC_SCPLOWMolecular dynamics (MD) simulations of protein-ligand complexes are essential for computer-aided drug design. In particular they enable the calculation of free energies and thus binding affinities. However, these simulations require significant computational resources and can take days to weeks to achieve relatively short timescales compared to biologically relevant timescales. To address this issue, we introduce a method for non-sequential generation of MD samples using a generative deep neural network trained on a large corpus of protein-ligand complex simulations. The method generates accurate protein-ligand complexes with full protein and ligand flexibility and is able to recapitulate the conformation space sampled by MD simulations with high coverage. This development is a step forward towards one-shot molecular sampling that can be utilized in the calculation of protein-ligand free energies.3
Matching journals
The top 3 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- An Orientationally Averaged Version of the Rotne-Prager-Yamakawa Tensor Provides A Fast But Still Accurate Treatment Of Hydrodynamic Interactions In Brownian Dynamics Simulations Of Biological Macromolecules 96%
- AlphaMut: a deep reinforcement learning model to suggest helix-disrupting mutations 96%
- Biasing RNA coarse-grained folding simulations with Small--Angle X--ray Scattering (SAXS) data 95%
Similar papers in this journal
Similar papers in this journal
- Interpreting Molecular Dynamics Forces as DeepLearning Gradients Improves Quality Of PredictedProtein Structures 97%
- Molecular insights from conformational ensembles via Machine Learning 96%
- ATOMDANCE: kernel-based denoising and allosteric resonance analysis for functional and evolutionary comparisons of protein dynamics 95%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.