mlsanityData to 7 Oct 2026

SurfDesign: Effective Protein Design on Molecular Surfaces

arXiv 2606.07567 ↗First posted 25 May 2026q-bio.BM cs.AI cs.CEcited by 0

Not on the map. Why?

Abstract (arXiv)

Protein function is largely determined by molecular surface geometry and physicochemical complementarity, yet most protein design methods condition only on backbone structure. We introduce SurfDesign, a surface-conditioned protein design framework that models molecular surfaces as continuous geometric manifolds and integrates them with pretrained protein language models. SurfDesign employs surface-based equivariant message passing to capture surface normals, curvature, and directional geometry, together with a parameter-efficient fine-tuning strategy. Focusing on functional protein design, we show that SurfDesign consistently outperforms prior surface-conditioned and backbone-only methods on de novo binder and enzyme design benchmarks. We also report strong performance on inverse-folding benchmarks as a diagnostic of structural compatibility. Our results highlight manifold-aware surface representations as a principled foundation for functional protein and enzyme design. Code is available at https://github.com/smiles724/SurfDesign.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 47 of its 105 references that are in the map. First posted on arXiv 25 May 2026.

Builds on

47 of 105 references.

References · 47

  1. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,084
  2. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 8,964
  3. 2020Scaling Laws for Neural Language Modelscited by 6,411
  4. 2021The Power of Scale for Parameter-Efficient Prompt Tuningcited by 3,769
  5. 2019DeepSDF: Learning Continuous Signed Distance Functions for Shape Representationcited by 2,682
  6. 2021E(n) Equivariant Graph Neural Networkscited by 951
  7. 2020Directional Message Passing for Molecular Graphscited by 556
  8. 2020Learning from Protein Structure with Geometric Vector Perceptronscited by 308
  9. 2020Fast and Uncertainty-Aware Directional Message Passing for Non-Equilibrium Moleculescited by 254
  10. 2019Evaluating Protein Transfer Learning with TAPEcited by 229
  11. 2022Diffusion probabilistic modeling of protein backbones in 3D for the motif-scaffolding problemcited by 229
  12. 2023SE(3) diffusion model with application to protein backbone generationcited by 214
35 more
  1. 2020ProtTrans: Towards Cracking the Language of Life's Code Through Self-Supervised Deep Learning and High Performance Computingcited by 162
  2. 2022Protein Representation Learning by Geometric Structure Pretrainingcited by 161
  3. 2022ProGen2: Exploring the Boundaries of Protein Language Modelscited by 156
  4. 2018Activation Functions: Comparison of trends in Practice and Research for Deep Learningcited by 130
  5. 2021Step-unrolled Denoising Autoencoders for Text Generationcited by 102
  6. 2024Diffusion Language Models Are Versatile Protein Learnerscited by 93
  7. 2022PiFold: Toward effective and efficient protein inverse foldingcited by 82
  8. 2022OntoProtein: Protein Pretraining With Gene Ontology Embeddingcited by 70
  9. 2023Structure-informed Language Models Are Protein Designerscited by 68
  10. 2022Multi-Scale Representation Learning on Proteinscited by 52
  11. 2021Molformer: Motif-based Transformer on 3D Heterogeneous Molecular Graphscited by 44
  12. 2023Graph Denoising Diffusion for Inverse Protein Foldingcited by 44
  13. 2022AlphaDesign: A graph protein design method and benchmark on AlphaFoldDBcited by 39
  14. 2023GeoMAE: Masked Geometric Target Prediction for Self-supervised Point Cloud Pre-Trainingcited by 38
  15. 2022DiffMD: A Geometric Diffusion Model for Molecular Dynamics Simulationscited by 24
  16. 2022Exploring evolution-aware & -free protein language models as protein function predictorscited by 20
  17. 2022Pre-training of Equivariant Graph Matching Networks with Conformation Flexibility for Drug Bindingcited by 15
  18. 2022Generative De Novo Protein Design with Global Contextcited by 15
  19. 2023EquiPocket: an E(3)-Equivariant Geometric Graph Neural Network for Ligand Binding Site Predictioncited by 15
  20. 2022Integration of Pre-trained Protein Language Models into Geometric Deep Learning Networkscited by 14
  21. 2024SurfPro: Functional Protein Design Based on Continuous Surfacecited by 14
  22. 2023A Hierarchical Training Paradigm for Antibody Structure-sequence Co-designcited by 13
  23. 2025DeepSearch: Overcome the Bottleneck of Reinforcement Learning with Verifiable Rewards via Monte Carlo Tree Searchcited by 11
  24. 2022Improving Molecular Representation Learning with Metric Learning-enhanced Optimal Transportcited by 9
  25. 2024UniIF: Unified Molecule Inverse Foldingcited by 9
  26. 2023PDB-Struct: A Comprehensive Benchmark for Structure-based Protein Designcited by 6
  27. 2024Simple, Efficient and Scalable Structure-aware Adapter Boosts Protein Language Modelscited by 6
  28. 2026Proteo-R1: Reasoning Foundation Models for De Novo Protein Designcited by 6
  29. 2024Reinforcement learning on structure-conditioned categorical diffusion for protein inverse foldingcited by 5
  30. 2024Mask prior-guided denoising diffusion improves inverse protein foldingcited by 5
  31. 2025Protein Inverse Folding From Structure Feedbackcited by 5
  32. 2025PRISM: Enhancing Protein Inverse Folding through Fine-Grained Retrieval on Structure-Sequence Multimodal Representationscited by 5
  33. 2026Surface-based Molecular Design with Multi-modal Flow Matchingcited by 3
  34. 2026A Semi-supervised Molecular Learning Framework for Activity Cliff Estimationcited by 3
  35. 2026Dynamics-inspired Structure Hallucination for Protein-protein Interaction Modelingcited by 3

Built on by

Not cited yet.