Research map / Probabilistic and scientific modeling / Molecules and proteins
BioMD: All-atom Generative Model for Biomolecular Dynamics Simulation
Bin Feng, Jiying Zhang, Xinni Zhang, Zijing Liu, Yu Li
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| generate auto-regressively (AR-5) instead of all-at-once (ablation) | BioMD-rel · DD-13M | Unbinding success@10: .0294 -> .9706 | Improved |
| generate auto-regressively (AR-5) instead of all-at-once (ablation) | BioMD-abs · DD-13M | Unbinding success@10: .0588 -> .7941 | Improved |
| generate auto-regressively (AR-5) instead of all-at-once (ablation) | BioMD-rel · DD-13M | Unbinding success@1: .0029 -> .7088 | Improved |
| generate auto-regressively (AR-5) instead of all-at-once (ablation) | BioMD-abs · DD-13M | Unbinding path RMSD: .6802 -> .5645 | Improved |
| generate auto-regressively (AR-5) instead of all-at-once (ablation) | BioMD-rel · DD-13M | Unbinding path RMSD: .6845 -> .7055 | Hurt |
| generate auto-regressively (AR-5) instead of all-at-once (ablation) | BioMD-rel · DD-13M | Bond geometry MAE: .0308 -> .0580 | Hurt |
| generate auto-regressively (AR-5) instead of all-at-once (ablation) | BioMD-abs · DD-13M | Angle geometry MAE: .0545 -> .0802 | Hurt |
| generate auto-regressively (AR-5) instead of all-at-once (ablation) | BioMD-rel · DD-13M | Protein-ligand steric clashes: .0004 -> .6375 | Hurt |
| predict absolute coordinates instead of relative changes | BioMD · MISATO | Protein RMSF correlation: .5945 -> .6854 | Improved |
| predict relative coordinate changes instead of absolute | BioMD · MISATO | Protein-ligand steric clashes: .0023 -> .0006 | Improved |
| predict relative coordinate changes instead of absolute (AR-5) | BioMD · DD-13M | Unbinding success@10: .7941 -> .9706 | Improved |
| BioMD instead of NeuralMD-ODE | MISATO ligand dynamics · MISATO | Ligand RMSF correlation: .3405 -> .4861 | Improved |
Closest papers by meaning
- Generative Modeling of Molecular Dynamics Trajectories (NeurIPS 2024 · similarity 0.90)
- Integrating Protein Dynamics into Structure-Based Drug Design via Full-Atom Stochastic Flows (ICLR 2025 · similarity 0.89)
- FlashMD: long-stride, universal prediction of molecular dynamics (NeurIPS 2025 · similarity 0.89)
- UniSim: A Unified Simulator for Time-Coarsened Dynamics of Biomolecules (ICML 2025 · similarity 0.89)
- Atomic Trajectory Modeling with State Space Models for Biomolecular Dynamics (NeurIPS 2026 · similarity 0.89)
- Generative Modeling of Full-Atom Protein Conformations using Latent Diffusion on Graph Embeddings (NeurIPS 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
