The role of crossover operator in evolutionary-based approach to the problem of genetic code optimization

Area of Science:

• Computational Biology
• Bioinformatics
• Evolutionary Computation

Background:

• The canonical genetic code's structure is hypothesized to minimize errors from nucleotide substitutions and translational mistakes.
• Testing this hypothesis requires finding optimal genetic codes under specific criteria, a computationally intensive task due to the vast search space.

Purpose of the Study:

• To investigate the effectiveness of evolutionary algorithms (EA) in searching for optimal genetic codes.
• To analyze the impact of mutation and crossover operators on EA performance for genetic code optimization.
• To compare the properties of optimized genetic codes with the canonical genetic code.

Main Methods:

• Applied evolutionary algorithms with varying mutation and crossover probabilities to search for optimal genetic codes.
• Adapted a position-based crossover operator for restricted genetic code models and developed a new crossover operator for general models.
• Utilized a fitness function based on the costs of amino acid replacements, considering polarity.

Main Results:

• The use of crossover operators significantly enhanced the quality of the optimized genetic codes.
• Simulations incorporating crossover operators achieved fitness function optimization in fewer generations compared to those without.
• Unrestricted optimal genetic codes demonstrated a 2.7-fold improvement in minimizing amino acid replacement costs compared to the canonical code.

Conclusions:

• Evolutionary algorithms, particularly with effective crossover operators, are powerful tools for exploring the genetic code's solution space.
• Optimized genetic codes exhibit distinct properties, favoring amino acids with polarity values close to the overall average.
• The findings support the theory that the genetic code's structure is optimized for error minimization.