Integrative Multi-omics and Artificial Intelligence (AI)-driven Approaches for Superior Nutritional Quality and Stress Resilience in Crops

Simardeep Kaur¹Naseeb Singh^1,2*Gurjeet Singh³RAKESH BHARDWAJ⁴

• ¹The ICAR Research Complex for North Eastern Hill Region (ICAR RC NEH), Umiam, India
• ²Michigan State University, East Lansing, United States
• ³Texas A&M University, AgriLife Research Center, Beaumont, Texas-77713, USA, Beaumon, United States
• ⁴ICAR - National Bureau of Plant Genetic Resources, New Delhi, India

The increasing demand for sustainable agriculture and nutrient-rich crops has driven the need for innovative strategies that go beyond the conventional breeding approaches. The Research Topic (RT) titled "Integrative Multi-omics and Artificial Intelligence (AI)-driven Approaches for Superior Nutritional Quality and Stress Resilience in Crops", published in Frontiers in Nutrition, sought to address the growing demand for nutritious and climate-resilient crops by leveraging cutting-edge tools such as multi-omics and AI. The multi-omics approach integrating genomics, transcriptomics, proteomics, metabolomics, and phenomics offers a comprehensive view of plant systems and allows for the identification of key genes, regulatory pathways, and metabolites involved in nutrition and stress adaptation. AI further enhances these efforts by enabling predictive modeling, big data analysis, and real-time decision-making.Together, these technologies offer a powerful platform for understanding the complex interactions that govern desirable traits and for accelerating crop improvement. Despite considerable progress, translating omics insights into field-ready innovations remains challenging. This RT provides an interdisciplinary platform to address this gap, focusing on crops like wheat, chickpea, taro, fenugreek, cassava, and pigeon pea, as well as technological solutions such as drones and remote sensing. The collection comprises nine articles (seven original research articles and two reviews), each offering novel insights and innovative applications of multi-omics and AI. Collectively, these studies advance the understanding and practical implementation of integrative technologies in plant science, contributing to a more sustainable and food-secure future. This study reported that 15% organic fertilizer substitution (OFS) in wheat significantly enhanced grain micronutrient bioavailability without reducing yield. While average yield was 9.06 Mg/ha, the highest (9.58 Mg/ha) occurred under 15% OFS. Grain iron and zinc increased by 24.7% and 19.2%, respectively, with no major change in soil micronutrient levels, indicating improved plant uptake and reduced phytate interference. OFS also lowered PA:Fe and PA:Zn ratios, enhancing bioavailability. Health impact modeling suggested reductions in Fe and Zn deficiency by up to 3.94% and 7.15%, respectively. Random forest analysis identified phytate content, soil organic carbon, and yield as key predictors. This study evaluated taro genotypes from the Eastern Himalaya, revealing significant variation in agronomic, nutritional, and mineral traits. 'Tamachongkham' and 'Tamitin' excelled in total and cormel yield, respectively, while 'Megha Taro 1' and 'Megha Taro 2' showed high cormel number and yield. Crude protein ranged from 3.25% to 7.10%, with notable differences in fiber, ash, starch, and sugar contents across genotypes. Antioxidant traits correlated positively with phenolic and anthocyanin levels. 'Tamitin' was rich in N, K, Zn, Cu, and Mn; 'Tagitung White' in P; and 'BCC 2' in Fe and Ca+Mg. PCA identified sugar, starch, fiber, anthocyanin, and FRAP as key contributors to diversity, highlighting several genotypes for nutritional breeding. This study highlighted the transformative potential of integrating remote sensing and artificial intelligence (AI) in agricultural pest management. By analyzing multispectral data through AI algorithms, pest damage could be detected at early stages, well before visible symptoms emerge, allowing for timely and localized intervention. The fusion of remote sensing with weather and phenology datasets enabled dynamic modeling of pest spread, thereby improving the resilience of pest forecasting under changing climatic conditions. AI-driven precision pest control strategies were shown to optimize resource use by significantly reducing pesticide application and labor costs while enhancing treatment efficacy. Despite these advancements, the study also identified key challenges to adoption, including data heterogeneity, lack of algorithm transparency, high implementation costs, and the need for adequate farmer training and capacity building. This study identified and characterized methyltransferase and demethylase gene families in pigeon pea (Cajanus cajan [L.] Millspaugh), with a focus on their expression profiles under various biotic and abiotic stress conditions. Notably, the demethylase gene CcALKBH10B exhibited strong upregulation in response to drought, salinity, and pest stress, while CcALKBH8 showed peak expression under heat stress, indicating their stress-responsive roles. Tissue-specific expression patterns suggested that these genes may also be involved in developmental and organ-level regulation. Phylogenetic analysis revealed that m6A-related proteins in pigeon pea cluster closely with those of other legumes, pointing to conserved evolutionary functions and potential cross-species functional relevance.