Machine learning algorithms: their applications in plant omics and agronomic traits’ improvement

Abstract

Agronomic traits of plants especially those of economic or aesthetic importance are threatened by climatic and environmental factors such as climate change, biotic, and abiotic stresses. These threats are now being mitigated through the analyses of omics data like genomics, transcriptomics, proteomics, metabolomics, and phenomics. The emergence of high-throughput omics technology has led to an avalanche of plant omics data. Plant research demands novel analytical paradigms to extract and harness large plant omics data for plant improvement effectively and efficiently. Machine learning algorithms are well-suited analytical and computational approaches for the integrative analysis of large unstructured, heterogeneous datasets. This study presents an overview of omics approaches to improve plant agronomic traits and crucial curated plant genomic data sources. Furthermore, we summarize machine learning algorithms and software tools/programming packages used in plant omics research. Lastly, we discuss advancements in machine learning algorithms' applications in improving agronomic traits of economically important plants. Extensive application of machine learning would advance plant omics studies. These advancements would consequently help agricultural scientists improve economically important plants’ quality, yield, and tolerance against abiotic and biotic stresses and other plant health-threatening issues.