Enhancing the reverse transcriptase function in Taq polymerase via AI-driven multiparametric rational design

Area of Science:

• Biotechnology and Bioengineering
• Molecular Biology
• Enzyme Engineering

Background:

• Enzyme modification is crucial for bioengineering novel functionalities.
• Taq polymerase is a key enzyme with potential for enhanced reverse transcriptase (RTase) activity.
• Balancing RTase activity with fidelity and substrate tolerance is a significant challenge.

Purpose of the Study:

• To develop Taq polymerase mutants with improved RTase activity.
• To retain essential properties like fidelity and exonuclease activity.
• To integrate AI-driven rational design with experimental validation.

Main Methods:

• Selected candidate mutations based on literature and initial wet-lab assessment.
• Employed protein language models and regression (Ridge, Gaussian process) for predictive modeling.
• Performed in silico screening of millions of mutations followed by wet-lab validation.

Main Results:

• Identified 18 enzyme variants with significantly enhanced RTase activity.
• Observed a balance between improved RTase activity, lower K^d, and moderate fidelity changes.
• Validated enzyme utility in real-time reverse-transcription PCR applications.

Conclusions:

• AI-driven design effectively enhances enzyme functionality.
• Developed a robust framework for designing tailored enzyme mutants.
• The identified Taq polymerase mutants are suitable for advanced molecular diagnostic tools.