Genetic Algorithm for Forecasting Bioinformatic Outcomes of Mutation-induced Cowpeas for Sustainable Development

Abstract

The application of data engineering techniques like a genetic algorithm in forecasting outcomes in plant genetics and breeding can help solve the twin problems of food insecurity and insufficiency. To demonstrate the practicality of using artificial intelligence (AI) to address these problems, the genetic algorithm is applied to genetic engineering (genetic mutation) of cowpea in a crop improvement program to generate useful bioinformatic information for further improvement of the crop. The aim of this work is to address malnutrition, immune deficiency, hunger, and poverty as canvassed in United Nations Sustainable Development Goals 1 and 2 (SDGs 1 and 2). Three genotypes (specifies) of cowpea obtained from Kontagora in Niger State of Nigeria were treated with chemical and physical mutagens: 200, 400, 600, and 800 of ethyl methane sulphonate (EMS) and 0.372gy of gamma rays. The study applied genetic algorithm as a stochastic optimizer using Python programming to determine the convergence pattern for obtaining an optimal cowpea solution that combines high yield and drought-tolerance. Huge data was generated in three iterative experiments. The outcomes of the three experiments showed that in experiment 1, the convergence occurred in the 9412th generation while in experiment 2, we obtained convergence in the 899th generation of the cowpea. Experiments show that the genetic mutation resulted in phenotypic traits in the first-generation offspring. The result of the third experiment indicated that the optimal cowpea solution was obtained in the 14338th generation. This implies that the use of AI (genetic algorithm) in ensuring food security and sufficiency may be time-consuming but would result in the desired traits in crops for meeting the 4 pillars of sustainability (human, social, economic and environmental).