Predicting Genetic Variability in Susceptibility to Carcinogens in Aquatic Species with the GeneticSuscepAquaNet Algorithm

Abstract

Carcinogenic pollutants pose serious hazards for aquatic populations, and aquatic mammals seem to have differing levels of susceptibility, as do other aquatic species. The process to ascertain the factors that elicit this variability is paramount to improve environmental risk assessments and the accompanying conservation efforts. This study outlines the use of the GeneticSuscepAquaNet algorithm, a machine learning algorithm we designed to predict genetic variability in aquatic species' susceptibility to carcinogens. The GeneticSuscepAquaNet model integrates genomics and environmental information and produces results that can be interpreted to understand which genetic information is likely to be associated with an organism’s capacity to repair DNA damage as result of exposure to acceptable levels of pollutants. Our results reveal we detected a range of variability associated with genetic susceptibility related to carcinogenic pollutants, along with their likely molecular mechanisms, while we did not deal specifically with carcinogenesis in the aquatics specifically. This paper will contemplate the capacity for the GeneticSuscepAquaNet model to predict cancer risk in aquatic species, therefore contributing a computational application towards our understanding and prevention of cancer, as well as a method to monitor their health at the ecosystem level by monitoring pollution levels affecting risk to aquatic organisms.