Deep learning for genomics and epi-genomics in horticulture crops improvement

Official URL: https://dx.doi.org/10.1016/B978-0-443-13229-2.00029-6

As the world population grows and the environment changes, the constant development of more resilient crops is required. Although studies to improve crop tolerance and yield are becoming more important in agricultural production, most agricultural traits evolve owing to complex genetic factors. In contrast to traditional machine learning, deep learning algorithms can automatically detect complex features, and profound learning models based on convolutional neural networks (CNNs) allow predictions to be interpreted biologically and the use of deep learning methods in plant biology to accelerate. Variety identification, yield prediction, quality, stress phenotypic detection, plant development monitoring, and many more areas are among the agricultural applications of deep learning technologies. This review explains the importance of deep learning in agricultural applications, including what is known about the effects of genomic and epigenetic changes on plant biology, and explains what complex issues deep learning should be used to answer in horticultural crop production.