Supported Applications

DeepRank-GNN-esm

Description

a graph neural network framework for scoring protein-protein interfaces that incorporates ESM-2 language model embeddings. The tool provides both a command-line interface for direct complex scoring and a Python framework for custom applications, generating predictions of interface quality (fnat values) from PDB structures. It includes pre-trained models and supports both CPU and GPU environments for analyzing protein-protein binding interfaces.
Usage
To list all executables provided by DeepRank-GNN-esm, run: $ sbgrid-list deeprank-gnn-esm
Installation
Use the following command to install this title with the CLI client: $ sbgrid-cli install deeprank-gnn-esm Available operating systems: Linux 64
Primary Citation*

X. Xu and A. M. J. Bonvin. 2024. Ranking protein-protein models with large language models and graph neural networks. arxiv.
- *Full citation information available through

Keywords

Protein-Protein Interaction Prediction
Default Versions

Linux 64: 0.2.3 (15.8 GB)
Other Versions
Developers

Anna Kravchenko, Alexandre MJJ Bonvin, Rodrigo Vargas Honorato, Xiaotong Xu