Applications and Trends of Machine Learning in Genomics and Phenomics for Next-Generation Breeding

Abstract

Crops are the major source of food supply and raw materials for the processing industry. A balance between crop production and food consumption is continually threatened by plant diseases and adverse environmental conditions. This leads to serious losses every year and results in food shortages, particularly in developing countries. Presently, cutting-edge technologies for genome sequencing and phenotyping of crops combined with progress in computational sciences are leading a revolution in plant breeding, boosting the identification of the genetic basis of traits at a precision never reached before. In this frame, machine learning (ML) plays a pivotal role in data-mining and analysis, providing relevant information for decision-making towards achieving breeding targets. To this end, we summarize the recent progress in next-generation sequencing and the role of phenotyping technologies in genomics-assisted breeding toward the exploitation of the natural variation and the identification of target genes. We also explore the application of ML in managing big data and predictive models, reporting a case study using microRNAs (miRNAs) to identify genes related to stress conditions.

Data synthesis for crop variety evaluation. A review

“…Crop variety evaluation has not always kept pace with the growing complexity of agricultural production and the growing availability of data. As a data-driven type of research, crop variety evaluation can benefit from multiple revolutions occurring in several fields such as genomics, phenomics, big data and machine learning (Bolger et al 2019;Esposito et al 2020;van Etten et al 2017;Tardieu et al 2017). These revolutions are driven by increased data storage and computing capacity, the availability of sensors, improved DNA sequencing technologies and new field data collection approaches, such as high-throughput and high-precision field phenotyping and crowdsourcing (Tardieu et al 2017;Esposito et al 2020;Chawade et al 2019;Reynolds et al 2020;Van Etten et al 2016).…”

“…As a data-driven type of research, crop variety evaluation can benefit from multiple revolutions occurring in several fields such as genomics, phenomics, big data and machine learning (Bolger et al 2019;Esposito et al 2020;van Etten et al 2017;Tardieu et al 2017). These revolutions are driven by increased data storage and computing capacity, the availability of sensors, improved DNA sequencing technologies and new field data collection approaches, such as high-throughput and high-precision field phenotyping and crowdsourcing (Tardieu et al 2017;Esposito et al 2020;Chawade et al 2019;Reynolds et al 2020;Van Etten et al 2016). This has caused not only a quantitative leap in data volumes but also a shift to 'big data' approaches that move beyond small-sample statistics to data analysis based on machine learning (Breiman 2001;Thessen 2016;van Etten et al 2017;Ersoz et al 2020).…”

Identification of Known and Novel MicroRNAs in Raspberry Organs Through High-Throughput Sequencing

“…In recent years, miRNAs have been identified in some plants species through cloning and computational approaches (Saliminejad et al, 2019;Smoczynska et al, 2019), and it has also been shown that miRNAs may be predicted using modern machine learning approaches (Parveen et al, 2019;Esposito et al, 2020). Recent reports have identified hundreds of miRNAs in several species, including Fragaria vesca (Han et al, 2019), cardamom (Anjali et al, 2019), sweet cherry (Wang et al, 2019), and Brazilian pine (Galdino et al, 2019) through highthroughput sequencing.…”

Phenotyping in Arabidopsis and Crops—Are We Addressing the Same Traits? A Case Study in Tomato

“…As the often mentioned phenotyping bottleneck [145] is gradually being overcome, the scientific focus will have to shift towards developing universal phenotyping approaches which integrate results of phenotypic observations across scales, environments, and even across species. In this sense, the advent of phenomics [146] coupled with the newest bioinformatic approaches such as machine learning [147] will probably play a major role in this transition. Still, more traditional phenotyping approaches will always be necessary to some extent.…”