Research map / Probabilistic and scientific modeling / Molecules and proteins
Co-evolution Transformer for Protein Contact Prediction
He Zhang, Fusong Ju, Jianwei Zhu, Liang He, Bin Shao, Nanning Zheng, Tie-Yan Liu
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 |
|---|---|---|---|
| replace average pooling with selective pooling (ablation) | CoT · CASP14 FM | Precision@L (%): 42.7 -> 48.2 | Improved |
| replace average pooling with selective pooling (ablation) | CoT · CAMEO Hard | Precision@L (%): 64.2 -> 66.6 | Improved |
| add co-evolution enhancement (conv) (ablation) | CoT · CASP14 FM | Precision@L (%): 41.6 -> 48.2 | Improved |
| add co-evolution enhancement (conv) (ablation) | CoT · CASP14 TBM | Precision@L (%): 70.3 -> 75.6 | Improved |
| add self-attention module alongside CoA (ablation) | CoT · CASP14 FM | Precision@L (%): 46.4 -> 48.2 | Improved |
| add self-attention module alongside CoA (ablation) | CoT · CASP14 FM/TBM | Precision@L (%): 66.9 -> 66.7 | Mixed |
| use co-evolution attention instead of self-attention (ablation) | CoT · CASP14 FM | Precision@L (%): 41.8 -> 48.2 | Improved |
| use co-evolution attention instead of self-attention (ablation) | CoT · CAMEO Hard | Precision@L (%): 59.8 -> 66.6 | Improved |
| CoT vs CopulaNet | CoT · CASP14 FM | Precision@L (%): 38.5 -> 48.2 | Improved |
| train on smaller PDB30-size data (CoT*) (ablation) | CoT · CASP14 FM | Precision@L (%): 48.2 -> 46.3 | Hurt |
| select most confident prediction over 12 MSAs (CoT†) | CoT · CASP14 FM | Precision@L (%): 48.2 -> 51.6 | Improved |
Closest papers by meaning
- Contact-Distil: Boosting Low Homologous Protein Contact Map Prediction by Self-Supervised Distillation (AAAI 2022 · similarity 0.90)
- MSAGPT: Neural Prompting Protein Structure Prediction via MSA Generative Pre-Training (NeurIPS 2024 · similarity 0.85)
- Ultrafast classical phylogenetic method beats large protein language models on variant effect prediction (NeurIPS 2024 · similarity 0.84)
- Evolution-Inspired Loss Functions for Protein Representation Learning (ICML 2024 · similarity 0.84)
- Pre-Training Protein Bi-level Representation Through Span Mask Strategy On 3D Protein Chains (ICML 2024 · similarity 0.84)
- Protein Multimer Structure Prediction via Prompt Learning (ICLR 2024 · similarity 0.84)
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.
