ProtRAP-LM: Protein Relative Accessibility Prediction through protein Language Model embeddings

Introduction

We present a novel transformer-based model, ProtRAP-LM, utilizing language model embeddings as input features, to quickly and accurately predict membrane contact probability and relative accessibility for each residue of a given protein sequence.

Here is a server of ProtRAP-LM, please feel free to use it.

Submit

Query protein sequence, one for a time.

Result

The figure of ProtRAP-LM prediction results (png and svg)

The detailed ProtRAP-LM prediction results (csv)

The value of predicted MCP, RLA, and RSA shown on the protein sequence (png)

Code and Weight

https://github.com/ComputBiophys/ProtRAP-LM

Contact

Feel free to contact us if you have any question. Thank you for using ProtRAP-LM.

Email: wanglei_cqb@pku.edu.cn

Song Group website

Reference

Wang, L.; Zhang, J.; Wang, D.; Song, C.* Membrane Contact Probability: An Essential and Predictive Character for the Structural and Functional Studies of Membrane Proteins. PLoS Comput. Biol. 2022, 18, e1009972. Link

Kang, K.#; Wang, L.#; Song, C.* ProtRAP: Predicting Lipid Accessibility Together with Solvent Accessibility of Proteins in One Run. J. Chem. Inf. Model. 2023, 63, 1058-1065. Link