Ruheng-W / PepBCL
We propose PepBCL, a novel BERT (Bidirectional Encoder Representation from Transformers)-based Contrastive Learning framework to predict the protein-Peptide binding residues based on protein sequences only.
☆14Updated 2 years ago
Alternatives and similar repositories for PepBCL:
Users that are interested in PepBCL are comparing it to the libraries listed below
- ☆32Updated last year
- ☆17Updated last month
- NodeCoder is a general framework based on graph convolutional neural network for protein function prediction.☆33Updated 11 months ago
- This repository contains scripts, data and jupyter notebooks used to produce the evaluation results in the BioPhi 2021 publication☆16Updated 2 years ago
- ☆23Updated 2 years ago
- ☆32Updated 2 years ago
- Antibody paratope prediction using Graph Neural Networks with minimal feature vectors☆36Updated 2 years ago
- ☆25Updated 10 months ago
- Scoring methods for predicting the DDG upon protein mutation☆14Updated 2 years ago
- Predicts the solubility and usability for purification of proteins expressed in E. coli using protein language models☆30Updated 4 months ago
- Evaluating Protein Binding Interfaces with Transformer Networks☆45Updated last year
- ☆20Updated 2 years ago