Unlocking wheat drought tolerance : the synergy of omics data and computational intelligence

Abstract

Currently, approximately 4.5 billion people in developing countries consider bread wheat (Triticum aestivum L.) as a staple food crop, as it is a key source of daily calories. Wheat is, therefore, ranked the second most important grain crop in the developing world. Climate change associated with severe drought conditions and rising global mean temperatures has resulted in sporadic soil water shortage causing severe yield loss in wheat. While drought responses in wheat crosscut all omics levels, our understanding of water-deficit response mechanisms, particularly in the context of wheat, remains incomplete. This understanding can be significantly advanced with the aid of computational intelligence, more often referred to as artificial intelligence (AI) models, especially those leveraging machine learning and deep learning tools. However, there is an imminent and continuous need for omics and AI integration. Yet, a foundational step to this integration is the clear contextualization of drought—a task that has long posed challenges for the scientific community, including plant breeders. Nonetheless, literature indicates significant progress in all omics fields, with large amounts of potentially informative omics data being produced daily. Despite this, it remains questionable whether the reported big datasets have met food security expectations, as translating omics data into pre-breeding initiatives remains a challenge, which is likely due to data accessibility or reproducibility issues, as interpreting omics data poses big challenges to plant breeders. This review, therefore, focuses on these omics perspectives and explores how AI might act as an interface to make this data more insightful. We examine this in the context of drought stress, with a focus on wheat.