Research map / Image generation / Fast diffusion sampling
BioDynaSpec: Harmonic-Guided Spatio-Spectral Autoregressive Diffusion for Protein Dynamics Generation
Mujie Lin, Yutian Liu, Yudi Guo, Yanzhen Hou, Yiheng Tao, Ruochong Zheng, Kaiwen Cheng, Xin Shan, Youdong Mao, Jie Chen
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 |
|---|---|---|---|
| use BioDynaSpec instead of ProAR | trajectory generation · ATLAS test (250-frame trajectory) | R250 Calpha-RMSE (A): 3.529 -> 1.509 | Improved |
| use BioDynaSpec instead of MDGEN | trajectory generation · ATLAS test (250-frame trajectory) | R250 Calpha-RMSE (A): 3.813 -> 1.509 | Improved |
| use BioDynaSpec instead of ProAR for short horizon | trajectory generation · ATLAS test (250-frame trajectory) | R50 Calpha-RMSE (A): 3.120 -> 1.256 | Improved |
| use BioDynaSpec instead of ProAR | stepwise displacement · ATLAS test (PCA-2D, Calpha coord.) | Stepwise Pearson correlation: -0.003 -> 0.452 | Improved |
| use BioDynaSpec instead of ProAR | long-range displacement · ATLAS test (PCA-2D, Calpha coord.) | Reference-displacement Hausdorff distance: 1.014 -> 2.505 | Hurt |
| use BioDynaSpec instead of MDGEN | local displacement · ATLAS test (PCA-2D, Calpha coord.) | Stepwise Hausdorff distance: 0.721 -> 0.487 | Improved |
| use BioDynaSpec instead of ConfRover | conformational sampling · ATLAS test (conformational sampling) | Root mean W2-dist.: 2.62 -> 1.31 | Improved |
| use BioDynaSpec instead of ConfRover | conformational sampling · ATLAS test (conformational sampling) | MD PCA W2-dist.: 1.39 -> 0.90 | Improved |
| use BioDynaSpec instead of ConfDiff | conformational sampling · ATLAS test (conformational sampling) | Joint PCA W2-dist.: 2.27 -> 1.19 | Improved |
| use BioDynaSpec instead of ProAR | local dynamics · ATLAS test (local dynamics, 250 frames) | PCA-PSD-LogCorr: 0.670 -> 0.817 | Improved |
| use BioDynaSpec instead of MDGEN | covariance consistency · ATLAS test (local dynamics, 250 frames) | CFRE: 1.563 -> 0.989 | Improved |
| use BioDynaSpec instead of ProAR | local dynamics · ATLAS test (local dynamics, 250 frames) | tICA-PSD-LogCorr: 0.662 -> 0.661 | No clear effect |
Closest papers by meaning
- TEMPO: Temporal Multi-scale Autoregressive Generation of Protein Conformational Ensembles (NeurIPS 2025 · similarity 0.90)
- Simultaneous Modeling of Protein Conformation and Dynamics via Autoregression (NeurIPS 2025 · similarity 0.87)
- Spectral Feedback for Test-Time Alignment of Protein Diffusion Models (NeurIPS 2026 · similarity 0.87)
- NMA-tune: Generating Highly Designable and Dynamics Aware Protein Backbones (ICML 2025 · similarity 0.85)
- Equivariant Blurring Diffusion for Hierarchical Molecular Conformer Generation (NeurIPS 2024 · similarity 0.85)
- Rao-Blackwell Gradient Estimators for Equivariant Denoising Diffusion (NeurIPS 2025 · similarity 0.85)
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.
