Machine learning applications to improve flavor and nutritional content of horticultural crops through breeding and genetics

Ferrão, Luís Felipe Ventorim; Dhakal, Rajeev; Dias, R.; Tieman, Denise M.; Whitaker, Vance M.; Gore, Michael A.; Messina, Carlos; Resende, M. D. V. de

doi:10.1016/j.copbio.2023.102968

Cited by 15 publications

(2 citation statements)

References 72 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In biology, for instance, ML techniques are instrumental in genomic sequencing, where they aid in predicting gene functions and understanding complex genetic networks. Similarly, in the realm of astrophysics, machine learning methods are employed to analyze astronomical data, facilitating the detection of celestial objects and the discernment of cosmic structures that elude traditional observational techniques [3].…”

Section: Machine Learning As a Methodsmentioning

confidence: 99%

Machine Learning - Method or Solution?

Tymoteusz

2023

Scientific Practice: Modern and Classical Research Methods

View full text Add to dashboard Cite

This article explores the dualistic nature of machine learning (ML) in scientific research, dissecting its role as both a methodological tool and a direct solution to complex problems. It begins by defining ML and its foundational principles, algorithms, and statistical models. The versatility of ML is highlighted through its applications across various disciplines, such as biology, astrophysics, healthcare, environmental science, and business. The article then delves into the challenges and limitations of ML, including data quality issues, ethical concerns, and the risk of over-reliance on automated decisions. It advocates for the integration of ML with classical research methods to ensure robust and ethical scientific practice. Concluding with future perspectives, the article speculates on the evolving role of ML in advancing both methodology and problem-solving in diverse fields, emphasizing the importance of responsible and effective utilization of ML in scientific research.

show abstract

Section: Machine Learning As a Methodsmentioning

confidence: 99%

Machine Learning - Method or Solution?

Tymoteusz

2023

Scientific Practice: Modern and Classical Research Methods

View full text Add to dashboard Cite

show abstract

“…Incorporating transcriptomics, proteomics and metabolomics can provide in-depth pathway information for tropical fruit flavour. This may be required in some cases because many of the reported metabolic pathways have been characterised for model species such as tomato and may be different in tropical fruit species (Ferrão et al 2023 ). Supporting the need for such studies, Jamil et al ( 2021 ) provided evidence in mangosteen, where the enzymes responsible for xanthone biosynthesis (benzophenone synthases) were not functioning as expected from model species.…”

Section: Introductionmentioning

confidence: 99%

Multi-omic applications for understanding and enhancing tropical fruit flavour

Lomax,

Ford,

Bar

2024

Plant Mol Biol

View full text Add to dashboard Cite

Consumer trends towards nutrient-rich foods are contributing to global increasing demand for tropical fruit. However, commercial cultivars in the breeding pipeline that are tailored to meet market demand are at risk of possessing reduced fruit flavour qualities. This stems from recurrent prioritised selection for superior agronomic traits and not fruit flavour, which may in turn reduce consumer satisfaction. There is realisation that fruit quality traits, inclusive of flavour, must be equally selected for; but currently, there are limited tools and resources available to select for fruit flavour traits, particularly in tropical fruit species. Although sugars, acids, and volatile organic compounds are known to define fruit flavour, the specific combinations of these, that result in defined consumer preferences, remain unknown for many tropical fruit species. To define and include fruit flavour preferences in selective breeding, it is vital to determine the metabolites that underpin them. Then, objective quantitative analysis may be implemented instead of solely relying on human sensory panels. This may lead to the development of selective genetic markers through integrated omics approaches that target biosynthetic pathways of flavour active compounds. In this review, we explore progress in the development of tools to be able to strategically define and select for consumer-preferred flavour profiles in the breeding of new cultivars of tropical fruit species.

show abstract

A comparison of genomic and phenomic selection methods for yield prediction in Coffea canephora

Adunola,

Tavares Flores,

Riva‐Souza

et al. 2024

The Plant Phenome Journal

View full text Add to dashboard Cite

Genomic prediction has been proposed as the standard method to predict the genetic merit of unphenotyped individuals. Despite the promising results reported in the plant breeding literature, its routine implementation remains difficult for some crops. This is the case with Coffea canephora, in which costs and availability of molecular tools are major challenges for most breeding programs. To circumvent this, the use of near‐infrared spectroscopy (NIR) has been recently proposed as an alternative to complement marker‐assisted selection. The so‐called phenomic selection relies on the reflectance spectrum to capture similarities between individuals and emerges as a valid approach for prediction. With promising results reported in multiple annual crops, we hypothesize that phenomic prediction could be a cost‐efficient approach to incorporate into a practical coffee breeding program. To test it, we relied on a diverse population of C. canephora, evaluated for yield production, in two geographical locations over four harvest seasons. Our contributions in this paper are twofold: (i) We compared phenomic and genomic selection results, and showed large predictive abilities when NIR is used as a predictor for within and across‐location predictions, and (ii) we presented a critical view of how both information sets could be combined into a contemporaneous coffee breeding program. Altogether, our results show how multi‐omic information could be integrated in the same framework to leverage genetic gains in the long term.

show abstract

Machine learning applications to improve flavor and nutritional content of horticultural crops through breeding and genetics

Cited by 15 publications

References 72 publications

Machine Learning - Method or Solution?

Machine Learning - Method or Solution?

Multi-omic applications for understanding and enhancing tropical fruit flavour

A comparison of genomic and phenomic selection methods for yield prediction in Coffea canephora

Contact Info

Product

Resources

About