Eftekhari M1, Ma C2, Yuriy L. Orlov3,4,5*
1 Department of Horticultural Sciences, Faculty of Agriculture, Tarbiat Modares University, Tehran, Iran
2 Center of Bioinformatics, College of Life Sciences, Northwest A&F University, Shaanxi, China
3 Sechenov First Moscow State Medical University of the Russian Ministry of Health (Sechenov University), Moscow, Russia
4 Agrarian and Technological Institute, Peoples’ Friendship University of Russia, Moscow, Russia
5 Institute of Cytology and Genetics SB RAS, Novosibirsk, Russia
m.eftekhari [at] modares.ac.ir, chuangma2006 [at] gmail.com, orlov [at] d-health.institute
Abstract
The field of plant breeding has witnessed a paradigm shift driven by advancements in artificial intelligence (AI) technologies, including machine learning (ML) and deep learning (DL) technologies. These cutting-edge techniques have transformed our understanding of plant biology, reshaping the landscape of plant breeding. AI-assisted omics techniques offer insights into plant-pathogen interactions and facilitate the identification of stress-responsive genes.
We have organized thematic journal issue at Frontiers in Plant Science – Research Topic “Applications of artificial intelligence, machine learning, and deep learning in plant breeding”. Here we present the trends of AI applications and the prediction problems challenged in plant science.
We collected research papers on the topic of AI applications in plant biology in areas of sequencing data analysis, image recognition, technology process optimization. It includes description of the potential of AI algorithms, particularly ML and DL, in decoding complex omics data to elucidate the molecular foundations of plant defense. Climate change poses significant threats to agricultural systems, emphasizing the importance of elucidating cold defense mechanisms in crops. From the methodical point of view, the Self Organizing Maps (SOM)-based ML methods can decipher gene expression patterns in response to different temperature regimes. The YOLO (You Only Look Once) architecture, known for its real-time object detection capabilities, is employed for object detection in plant image analysis. The efficacy of the CFNet-VoVGCSP-LSKNet-YOLOv8s model in accurately identifying cotton pests and diseases amidst challenging environmental conditions is shown. Another key phenotypic trait in plants – pubescence, correlates with stress resistance, particularly in wheat. Overall, AI, ML, and DL techniques offer unique opportunities from deciphering complex omics data to automating phenotypic trait analysis and disease detection to revolutionize breeding practices, develop stress-tolerant and high-yielding crop varieties, and contribute to global food security in the face of escalating environmental challenges.
Keywords: agrobiology, AI, Machine Learning, omics data
Acknowledgements: The study is supported by Russian Science Foundation (grant project 23-44-00030).