Research map / Probabilistic and scientific modeling / Molecules and proteins
RankFlow: Property-aware Transport for Protein Optimization
Lu Yu, Wei Xiang, Kang Han, Gaowen Liu, Ramana Rao Kompella
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 rank-consistent loss LRFlow to energy-guided flow matching (ablation) | RankFlow · ProteinGym Fitness | Spearman: 0.702 -> 0.742 | Improved |
| add rank-consistent loss LRFlow to energy-guided flow matching (ablation) | RankFlow · ProteinGym Activity | Spearman: 0.680 -> 0.722 | Improved |
| add flow-matching energy loss LPFM to rank loss (ablation) | RankFlow · ProteinGym Fitness | Spearman: 0.624 -> 0.742 | Improved |
| use rank-consistent loss instead of kinetic regularizer (ablation) | RankFlow · ProteinGym Fitness | Spearman: 0.714 -> 0.742 | Improved |
| use rank-consistent loss instead of pairwise hinge (ablation) | RankFlow · ProteinGym Fitness | Spearman: 0.709 -> 0.742 | Improved |
| use energy weight lambda=0.5 instead of 0 (ablation) | RankFlow · ProteinGym Fitness | Spearman: 0.718 -> 0.742 | Improved |
| use energy weight lambda=0.5 instead of 1 (ablation) | RankFlow · ProteinGym Fitness | Spearman: 0.738 -> 0.742 | Improved |
| use cosine schedule instead of linear (ablation) | RankFlow · ProteinGym Fitness | Spearman: 0.736 -> 0.742 | Improved |
| use combined rank-and-normalized energy instead of negative label energy (ablation) | RankFlow · ProteinGym Fitness | Spearman: 0.738 -> 0.742 | Improved |
| add structure information and fusion encoder (ablation) | RankFlow · ProteinGym (Fitness, Activity) | Spearman | Improved |
| add property-guided steering gate (ablation) | RankFlow · ProteinGym | Spearman | Improved |
| use RankFlow instead of Kermut on contiguous split | RankFlow · ProteinGym Contiguous | Spearman: 0.591 -> 0.589 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- ProteinNPT — ProteinNPT: Improving Protein Property Prediction and Design with Non-Parametric Transformers (NeurIPS 2023) · better on 5 of 5 shared results
- Kermut — Kermut: Composite kernel regression for protein variant effects (NeurIPS 2024) · better on 3 of 4 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- A Variational Perspective on Generative Protein Fitness Optimization (ICML 2025 · similarity 0.89)
- Improving protein optimization with smoothed fitness landscapes (ICLR 2024 · similarity 0.88)
- Steering Generative Models with Experimental Data for Protein Fitness Optimization (NeurIPS 2025 · similarity 0.88)
- Contrastive losses as generalized models of global epistasis (NeurIPS 2024 · similarity 0.88)
- Evolution-Inspired Loss Functions for Protein Representation Learning (ICML 2024 · similarity 0.88)
- DePLM: Denoising Protein Language Models for Property Optimization (NeurIPS 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.
