COStar: a D-star Lite-based dynamic search algorithm for codon optimization.

Area of Science:

• Bioinformatics
• Molecular Biology
• Computational Biology

Background:

• Codon optimization enhances gene synthesis and expression in host organisms by modifying codon usage.
• The complexity of codon optimization lies in balancing multiple, often conflicting, objectives.
• Identifying optimal sequences from numerous synonymous codon combinations is computationally challenging.

Purpose of the Study:

• To introduce COStar, a novel dynamic search algorithm for efficient codon optimization.
• To address the challenge of finding optimal codon sequences for de novo gene synthesis and heterologous gene expression.

Main Methods:

• COStar employs a D-star Lite-based dynamic search algorithm.
• The codon optimization problem is modeled as a weighted directed acyclic graph using a sliding window approach.
• The D-star Lite algorithm finds the shortest path within the generated graph to determine the optimal gene sequence.

Main Results:

• In silico experiments demonstrated COStar's effectiveness in optimizing various genes, including those from the human genome.
• The algorithm successfully converts gene optimization into a real-time shortest path search problem.
• COStar shows promise for improving de novo gene synthesis and heterologous gene expression.

Conclusions:

• COStar offers a computationally efficient and effective solution for codon optimization.
• The algorithm facilitates both de novo gene synthesis and enhanced heterologous gene expression.
• COStar represents a significant advancement in computational tools for molecular biology.