A comprehensive guide to machine learning-mediated optimization of medium composition for enhanced in vitro performance

The development and optimization of plant tissue culture media plays a pivotal role impacting the success and efficiency of in vitro culture systems. A well-designed culture medium provides essential nutrients and growth factors necessary for plant cell proliferation, differentiation, and organogenesis. Optimizing the medium composition can significantly enhance growth rates, improve organogenesis/embryogenesis efficiency, and increase the overall yield of healthy plantlets. Despite the critical importance of basal media composition, tissue culture methodologies are commonly conducted using a selection of formulations that were initially developed for a small number of species and modified for others on an ad-hoc basis. Although this approach is not ideal, fully optimizing tissue culture mediums for each species is a long and tedious process and is often not practical using traditional methods. Recent developments in computational power and artificial intelligence provides new opportunities to revisit this practice by dramatically improving the speed, precision, and efficiency of culture media optimization. This paper provides a review of the fundamental components of plant culture media, including macronutrients, micronutrients, amino acids, and vitamins. Additionally, the application of machine learning (ML) techniques is presented as a more efficient approach to develop and optimize species specific plant tissue culture media as a stepwise process, from experimental design and data collection to model training and validation. Ultimately, this work highlights ML as a transformative tool to enhance the precision and efficiency of plant tissue culture media optimization as a pathway to improve plant biotechnology. • Optimization of culture media enhances in vitro plant growth and development. • Traditional media formulations are often adapted from a few initial species. • Machine learning accelerates and improves tissue culture media optimization. • ML-driven approaches refine media composition for species-specific needs.