Nanotope: A Graph Neural Network-Based Model for Nanobody paratope prediction
Meng, X.; Li, S.; Li, R.; Huang, B.; Wang, X.
Show abstract
Nanobodies are artificial antibodies derived from the immune system of camelid species, obtained through artificial processing and isolation of antigen-binding proteins. Their small molecular size and high specificity endow nanobodies with extensive potential applications in various fields. However, determining nanobody paratopes through experimental methods is both costly and time-consuming, while traditional computational approaches often lack sufficient accuracy. To enhance the prediction accuracy, we propose Nanotope, a structurebased model capable of efficiently and accurately predicting paratopes, utilizing graph neural networks and an antibody pretrained language model. Our approach primarily leverages sequence features obtained from the AntiBERTy antibody pretrained language model and three-dimensional spatial structure features of nanobodies as inputs. Employing one-dimensional convolution, EGConv graph convolution, and GATConv convolution, we predict the probability of nanobody paratopes with a much higher accuracy. Our method significantly improves the prediction of nanobody paratopes and requires only nanobody sequences and structural information, without the need for additional antigen information. Source code is freely available at https://github.com/WangLabforComputationalBiology/Nanotope
Matching journals
The top 4 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Learning Context-aware Structural Representations to Predict Antigen and Antibody Binding Interfaces 97%
- Identifying B-cell epitopes using AlphaFold2 predicted structures and pretrained language model 96%
- Paragraph - Antibody paratope prediction using Graph Neural Networks with minimal feature vectors 96%
Similar papers in this journal
- EGRET: Edge Aggregated Graph Attention Networks and Transfer Learning Improve Protein-Protein Interaction Site Prediction 95%
- Interpretable and Generalizable Attention-Based Model for Predicting Drug-Target Interaction Using 3D Structure of Protein Binding Sites: SARS-CoV-2 Case Study and in-Lab Validation 95%
- GraphGPSM: a global scoring model for protein structure using graph neural networks 95%
Similar papers in this journal
- Physical-aware model accuracy estimation for protein complex using deep learning method 95%
- DeepNeuropePred: a robust and universal tool to predict cleavage sites from neuropeptide precursors by protein language model 95%
- SpatialPPI: three-dimensional space protein-protein interaction prediction with AlphaFold Multimer 94%
Similar papers in this journal
- nanoBERT: A deep learning model for gene agnostic navigation of the nanobody mutational space 96%
- SAINT-Angle: self-attention augmented inception-inside-inception network and transfer learning improve protein backbone torsion angle prediction 94%
- AbLang: An antibody language model for completing antibody sequences 94%
Similar papers in this journal
- Multi-Head Attention-based U-Nets for Predicting Protein Domain Boundaries Using 1D Sequence Features and 2D Distance Maps 96%
- Struct2Graph: A graph attention network for structure based predictions of protein-protein interactions 95%
- DISTEMA: distance map-based estimation of single protein model accuracy with attentive 2D convolutional neural network 95%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.