Biological Genetics and Its Analogy in Genetic Algorithms

To better understand how Genetic Algorithms (GAs) work, it is useful to draw parallels with concepts from biological genetics. This analogy allows us to view the optimization process as an evolutionary system, where individuals (solutions) are evaluated, selected, recombined, and mutated over generations. The table below summarizes the most relevant biological terms and their corresponding interpretations within the context of GAs:

Comparison Between Biological Genetics and Genetic Algorithms

Biological Term 🧫	GA Equivalent 🤖	Meaning in Genetic Algorithms	Practical Example (GADEMO)
Gene	Variable in the solution	A single unit of encoded information; represents a variable in the objective function	In f(x, y), the genes are x and y
Allele	Value of a gene	The specific value assigned to a variable (gene)	If x = 2.3, then the allele of gene x is 2.3
Chromosome	Individual / Candidate solution	A complete vector containing all genes (i.e., the full solution)	Chromosome = [x, y] = [2.3, 1.1]
Population	Set of candidate solutions	Group of chromosomes (individuals) evaluated in the same generation	Population = 5 chromosomes = 5 possible solutions
Genome	Search space	The range of all possible genes and alleles, defined by the variable bounds	Ranges: x ∈ [0,4], y ∈ [0,4]
Sexual reproduction	Crossover	Combination of genes from two parents to generate offspring	Average of parents: [1.0, 2.0] + [3.0, 0.0] → [2.0, 1.0]
Genetic mutation	Mutation	Random alteration in one or more gene values (alleles)	Change x = 2.0 to x = 2.2
Natural selection	Selection (e.g., tournament, roulette)	Process of choosing parents based on their fitness (quality)	Individuals with higher fitness are more likely to reproduce
Biological fitness	Fitness	Measure of how well a solution solves the problem (objective function value)	f(x, y) = x² + y²; if x = 2, y = 2, then fitness = 8
Evolution	Iteration of generations	Gradual improvement of the population through genetic operations	Each generation tends to perform better than the previous one

Additional Biological Terms in the Context of Genetic Algorithms

Some additional terms from classical genetics are also relevant in the context of Genetic Algorithms, especially when considering encoding strategies:

Supplementary Biological Analogies in GAs

Biological Term	GA Equivalent	Explanation
Locus	Bit or variable position	Specific position of a gene or bit in the chromosome vector; used especially in binary representations
Genotype	Encoded candidate solution	The internal representation of a solution, such as a binary string or real-valued vector
Phenotype	Evaluated solution	The expressed result of the genotype when passed through the objective function, i.e., f(x)