Research map / Probabilistic and scientific modeling / Molecules and proteins
Generating Novel, Designable, and Diverse Protein Structures by Equivariantly Diffusing Oriented Residue Clouds
Yeqing Lin, Mohammed AlQuraishi
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 Cartesian-coordinate diffusion with Frenet–Serret frame reasoning in the denoiser | Genie · Short proteins (50–128 residues) | confident designability: 3.2% and 17.7% -> 58.3% | Improved |
| Use Cartesian-coordinate diffusion with Frenet–Serret frame reasoning in the denoiser | Genie · Short proteins (50–128 residues) | maximum TM score to generated set: 0.583 ± 0.115 and 0.668 ± 0.178 -> 0.561 ± 0.086 | Improved |
| Use Cartesian-coordinate diffusion with Frenet–Serret frame reasoning in the denoiser | Genie · Long-protein generation | F1: 0.531 -> 0.654 | Improved |
| Train Genie on the larger AlphaFold-SwissProt dataset instead of SCOPe | Genie · Long-protein generation | designability: 0.586 -> 0.790 | Improved |
| Train Genie on the larger AlphaFold-SwissProt dataset instead of SCOPe | Genie · Long-protein generation | diversity: 0.738 -> 0.642 | Hurt |
| Reduce the sampling noise scale (ablation) | Genie · Long-protein generation | designability | Improved |
| Reduce the sampling noise scale (ablation) | Genie · Long-protein generation | diversity | Hurt |
| Set sampling noise scale to 0.4 (ablation) | Genie · Long-protein generation | F1 | Improved |
| Replace OmegaFold with ESMFold in the short-model designability pipeline (ablation) | Genie and short-model baselines · Short proteins (50–128 residues) | overall evaluation trends | No clear effect |
Closest papers by meaning
- Proteina: Scaling Flow-based Protein Structure Generative Models (ICLR 2025 · similarity 0.90)
- CarbonNovo: Joint Design of Protein Structure and Sequence Using a Unified Energy-based Model (ICML 2024 · similarity 0.90)
- Variable-Length Generative Protein Design via Generalized Poisson Flow (NeurIPS 2026 · similarity 0.89)
- Proteus: Exploring Protein Structure Generation for Enhanced Designability and Efficiency (ICML 2024 · similarity 0.89)
- Protein Design with Dynamic Protein Vocabulary (NeurIPS 2025 · similarity 0.88)
- Diffusion on Language Model Encodings for Protein Sequence Generation (ICML 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.
