Research map / Label-efficient and robust learning / Self-supervised and contrastive
Protein Representation Learning by Geometric Structure Pretraining
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 multiview contrastive pretraining on AlphaFoldDB (805K) | GearNet-Edge · EC | F_max: 0.810 to 0.874 | Helped |
| Add multiview contrastive pretraining on AlphaFoldDB | GearNet-Edge-IEConv · Fold classification (avg) | accuracy (%): 72.7 to 78.1 | Helped |
| Add edge message passing layer (GearNet-Edge vs GearNet) | GearNet · EC | F_max: 0.730 to 0.810 | Helped |
| Replace relational graph convolution with single-kernel GCN (ablation) | GearNet-Edge · EC | F_max: 0.810 to 0.752 (6 layers) | Hurt |
| Use only subsequence cropping with identity noise instead of random sampling of cropping and noise (ablation) | Multiview Contrast · EC | F_max: 0.874 to 0.866 | Hurt |
| Use subspace cropping with random edge masking only (fixed augmentation combo) (ablation) | Multiview Contrast · EC | F_max: 0.874 to 0.876 | No clear effect |
| Add IEConv layer to GearNet-Edge | GearNet-Edge · Fold classification (avg) | accuracy (%): 69.9 to 72.7 | Helped |
| Add IEConv layer to GearNet-Edge | GearNet-Edge · EC | F_max: 0.810 to 0.810 | No clear effect |
Closest papers by meaning
- Evaluating Representation Learning on the Protein Structure Universe (ICLR 2024 · similarity 0.92)
- Protein 3D Graph Structure Learning for Robust Structure-Based Protein Property Prediction (AAAI 2024 · similarity 0.90)
- Pre-Training Protein Bi-level Representation Through Span Mask Strategy On 3D Protein Chains (ICML 2024 · similarity 0.90)
- Learning Complete Protein Representation by Dynamically Coupling of Sequence and Structure (NeurIPS 2024 · similarity 0.88)
- Multi-Scale Representation Learning for Protein Fitness Prediction (NeurIPS 2024 · similarity 0.88)
- Enhancing Graph Contrastive Learning for Protein Graphs from Perspective of Invariance (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.