APPLICATIONS OF ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING IN PLANT TISSUE CULTURE

Abstract

Plant tissue culture is a cornerstone of modern plant biotechnology, enabling rapid clonal propagation, germplasm conservation, and the controlled study of morphogenetic and stress-related responses. However, the multifaceted and highly nonlinear nature of in vitro systems, involving complex interactions among genotype, culture media composition, plant growth regulators, and environmental conditions, often limits the efficiency and reproducibility of conventional experimental approaches. In recent years, artificial intelligence (AI) and machine learning (ML) techniques have emerged as powerful tools for modeling, predicting, and optimizing tissue culture responses, complementing and, in some contexts, outperforming traditional statistical methods. This review provides a comprehensive and critical synthesis of AI- and ML-based applications in plant tissue culture, with particular emphasis on artificial neural networks, support vector machines, random forest models, k-nearest neighbors, genetic algorithms, and hybrid optimization frameworks. We examine their use across key application domains, including micropropagation efficiency prediction, culture medium and hormone optimization, assessment of abiotic stress tolerance, secondary metabolite production, image-based phenotyping, and control of automated and bioreactor-based culture systems. The strengths, limitations, and data requirements of each modeling approach are discussed in the context of biological interpretability, model robustness, and experimental reproducibility. In addition, the review explores emerging directions such as quantum machine learning and explainable AI, highlighting their potential contributions while critically addressing current technical and practical constraints. Finally, we outline future perspectives for integrating AI-driven decision support systems into plant tissue culture research and production, aligning these advances with the broader framework of Plant Biotechnology 5.0. By consolidating methodological insights and identifying existing gaps, this review aims to guide researchers toward more efficient, transparent, and data-driven in vitro culture strategies.