Research map / Probabilistic and scientific modeling / Molecules and proteins
Efficient Protein Optimization via Structure-aware Hamiltonian Dynamics
Jiahao Wang, Shuangjia Zheng
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 |
|---|---|---|---|
| add Hamiltonian dynamics sampling instead of random mutation (ablation) | HADES · GB1 | cumulative max fitness: 0.84 -> 1.00 | Improved |
| add Hamiltonian dynamics sampling instead of random mutation (ablation) | HADES · GB1 | fDiv: 0.73 -> 0.84 | Improved |
| add Hamiltonian dynamics sampling instead of random mutation (ablation) | HADES · PhoQ | cumulative max fitness: 0.75 -> 0.80 | Improved |
| add structure decoder (ablation) | HADES · GB1 | cumulative max fitness: 0.95 -> 1.00 | Improved |
| add structure decoder (ablation) | HADES · GB1 | fDiv: 0.84 -> 0.84 | No clear effect |
| add structure decoder (ablation) | HADES · PhoQ | cumulative max fitness: 0.74 -> 0.80 | Improved |
| add ensemble uncertainty (UCB) selection (ablation) | HADES · GB1 | cumulative max fitness: 0.93 -> 1.00 | Improved |
| add ensemble uncertainty (UCB) selection (ablation) | HADES · GB1 | mean fitness: 0.51 -> 0.59 | Improved |
| add ensemble uncertainty (UCB) selection (ablation) | HADES · PhoQ | mean fitness: 0.19 -> 0.22 | Improved |
| add virtual barrier (ablation) | HADES · GB1 | cumulative max fitness: 0.97 -> 1.00 | Improved |
| add virtual barrier (ablation) | HADES · PhoQ | cumulative max fitness: 0.63 -> 0.80 | Improved |
| use Hamiltonian dynamics instead of Langevin dynamics | HADES · GB1 | cumulative max fitness: 0.93 -> 1.00 | Improved |
Closest papers by meaning
- A survey and benchmark of high-dimensional Bayesian optimization of discrete sequences (NeurIPS 2024 · similarity 0.88)
- Steering Generative Models with Experimental Data for Protein Fitness Optimization (NeurIPS 2025 · similarity 0.87)
- Improving protein optimization with smoothed fitness landscapes (ICLR 2024 · similarity 0.87)
- A Variational Perspective on Generative Protein Fitness Optimization (ICML 2025 · similarity 0.87)
- Approximation-Aware Bayesian Optimization (NeurIPS 2024 · similarity 0.87)
- Tree Search-Based Evolutionary Bandits for Protein Sequence Optimization (AAAI 2024 · similarity 0.87)
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.
