Protposer: The web server that readily proposes protein stabilizing mutations with high PPV.

Area of Science:

• Protein engineering
• Computational biology
• Biotechnology

Background:

• Protein stability is crucial for biotechnological and medical applications.
• Natural proteins often exhibit insufficient stability ex vivo, necessitating protein engineering efforts.
• Current protein stability prediction tools lack sufficient accuracy, limiting their utility in rational design.

Purpose of the Study:

• To develop a novel computational tool for identifying protein-stabilizing mutations.
• To provide accurate predictions of mutation success rates for protein stabilization.
• To offer a user-friendly platform for rational protein design.

Main Methods:

• Utilizing structure- and sequence-based screening modules to identify candidate mutations.
• Employing a logistic regression model, trained to prevent overfitting, for mutation evaluation.
• Analyzing Protein Data Bank (PDB) files to find stabilization opportunities.

Main Results:

• The tool provides a ranked list of promising mutations with estimated success rates (eSR).
• Excellent agreement was observed between eSRs and actual positive predictive values (PPV) on external datasets.
• The tool achieved a PPV above 0.7 with an optimal threshold and outperformed other leading software (FoldX, Rosetta, PoPMusiC).

Conclusions:

• The developed tool is a distinct, straightforward, and highly successful method for designing protein-stabilizing mutations.
• It accurately predicts the likelihood of a mutation increasing protein stability.
• The tool is freely available for academic use, facilitating advancements in protein engineering.