Machine-guided design for bioengineering gene therapy vectors: Where are we and what lies ahead?

Area of Science:

• Biotechnology and Bioengineering
• Computational Biology and Bioinformatics
• Gene Therapy and Virology

Background:

• Viral vectors are crucial for therapeutic gene delivery, driving demand for improved variants.
• Traditional engineering methods have limitations in exploring the vast protein sequence space.
• Computational approaches offer powerful solutions for viral vector optimization.

Purpose of the Study:

• To synthesize recent advances in computationally guided viral vector design.
• To bridge algorithmic innovation with practical bioengineering realities.
• To highlight the transformative potential of computational design in viral vector bioengineering.

Main Methods:

• Review and synthesis of machine learning (ML)-based and non-ML computational methods.
• Analysis of applications in viral vector engineering, including stability, structure, and function prediction.
• Exploration of strategies for generating novel viral vector variants.

Main Results:

• Computational approaches significantly enhance the exploration of protein sequence space for viral vector design.
• Both ML and non-ML methods are effective in refining stability, modeling structures, predicting functions, and creating novel variants.
• Successful implementation requires addressing key factors linking computational design with experimental bioengineering.

Conclusions:

• Computational design is a transformative tool for advancing viral vector bioengineering.
• Integrated and holistic design strategies are emerging as the future of the field.
• Researchers are encouraged to adopt computational methods to innovate next-generation viral gene therapies.