SCONES : Self-Consistent Neural Network for Protein Stability Prediction Upon Mutation

SCONES is a robust interpretable minimalistic structure-based method for estimating relative protein stabilities for missense mutations.

The repository is organized into four sections:

• DataGeneration: tools to create and build datasets
• Analysis: tools to analyze predictions and datasets
• DataCollection: tools to retrive predictions from existing methods
• SCONES: code for training SCONES

Each section is independent and contains separate installation and usage instructions.

Citing

If you found this work useful, please cite the following article:

@article{doi:10.1021/acs.jpcb.1c04913,
	author = {Samaga, Yashas B. L. and Raghunathan, Shampa and Priyakumar, U. Deva},
	title = {SCONES: Self-Consistent Neural Network for Protein Stability Prediction Upon Mutation},
	journal = {The Journal of Physical Chemistry B},
	volume = {0},
	number = {0},
	pages = {null},
	year = {0},
	doi = {10.1021/acs.jpcb.1c04913},
	note ={PMID: 34546056},
}