license: mit
language:
- en
metrics:
- r_squared
- mse
- mae
- pearsonr
pipeline_tag: tabular-regression
library_name: transformers
tags:
- biology
- proteomics
- thermostability
- PLM
PPTStab
PPTStab: Prediction and Designing of thermostable proteins with a desired melting temperature
PPTStab is an ANN+MLP based ensemble regressor model developed to predict the melting temperature of long protein sequences and also design thermostable proteins. It integrates state-of-the-art LLMs like ProtBert, ProtT5-XL-UniRef50, Ankh, ProtGPT2, ProstT5 based embeddings into the architecture to advance the predictive modeling capabilities. It is designed for seamless usage, available via a webserver at https://webs.iiitd.edu.in/raghava/pptstab.
Please read/ cite the content about the PPTStab for complete information including working methodology and algorithm behind the approach.
Citation
If you use this method in your research, please cite:
@article
{Tijare2024.09.21.614294,
author = {Tijare, Purva and Kumar, Nishant and Raghava, Gajendra P. S.},
title = {Designing of thermostable proteins with a desired melting temperature},
year = {2024},
doi = {10.1101/2024.09.21.614294},
eprint = {https://www.biorxiv.org/content/early/2024/09/24/2024.09.21.614294.full.pdf},
journal = {bioRxiv}
}
Contact
For questions, do not hesitate to reach out and open an issue on here or on our GitHub.