Genetic Programming Approach in Better Understanding of the Relationship Between the Number of Viable Cells and Concentration of \({\text{O}}_{2}^{ - }\) , \({\text{NO}}_{2}^{ - }\) and GSH Produced in Cancer Cells Treated with Pd(II) Complexes

Genetic programming (GP) is a powerful tool for creating mathematical models and has been used in various fields to discover complex relationships in data. GP is a method that uses artificial intelligence to automatically create programs, based on the Darwinian principle of natural selection to create a population of better programs over many generations. Therefore, this study investigated the possibility of obtaining a function describing the data obtained from the research work conducted by Petrović et al., by applying genetic programming. Experimental results showed that Pd(II) complexes, labeled as Pd-1, Pd-3, Pd-5 and Pd-6, showed significant cytotoxic effects and extreme oxidative stress in cancer cell lines HCT-116 and MDA-MB-231. Specifically, the aim was to determine whether there is a concentration-dependent relationship between the number of viable cells for all experiments. To create the function, symbolic regression was used. R2 is a useful metric for evaluating the performance of the created function. Overall, the results demonstrate that GP is a powerful tool for creating mathematical models that accurately describe the relationship between concentration and the number of viable cells. In conclusion, it has been shown that GP can be used with high precision on such examples, with R2 ranging from 0.9747 to 0.99948. Such a method of calculating values could facilitate the selection of the required concentration for use in experiments.