DeepREx-WS

A Deep-learning based method for the prediction of Residue solvent Exposure starting from protein sequence

New version available here based on Protein Language Models for faster and more accurate predictions!

Standalone version and training sets

Docker container
Source code
Download training set
Download blind test set
Download cross-validation split

Both the training set and the blind test set are available as a tsv file made of 5 columns where each row corresponds to a single residue. The columns store in order: the pdb id; the position; the residue type; the classification (Buried/Exposed); the rsa. Inside the cross_validation folder you will find 10 files (split0-split9), each containing the pdb ids belonging to the corresponding subset used in cross-validation.