DNNGP, a deep neural network-based method for genomic prediction using multi-omics data in plants

Molecular Plant, Год журнала: 2022, Номер 16(1), С. 279 - 293

Опубликована: Ноя. 10, 2022

Genomic prediction is an effective way to accelerate the rate of agronomic trait improvement in plants. Traditional methods typically use linear regression models with clear assumptions; such are unable capture complex relationships between genotypes and phenotypes. Non-linear (e.g., deep neural networks) have been proposed as a superior alternative because they can non-additive effects. Here we introduce learning (DL) method, network genomic (DNNGP), for integration multi-omics data We trained DNNGP on four datasets compared its performance built five classic models: best unbiased (GBLUP); two based machine (ML) framework, light gradient boosting (LightGBM) support vector (SVR); DL selection (DeepGS) genome-wide association study (DLGWAS). novel ways. First, it be applied variety omics predict Second, multilayered hierarchical structure dynamically learns features from raw data, avoiding overfitting improving convergence using batch normalization layer early stopping rectified activation (rectified unit) functions. Third, when small were used, produced results that competitive other methods, showing greater accuracy than large-scale breeding used. Fourth, computation time required by was comparable commonly used up 10 times faster DeepGS. Fifth, hyperparameters easily tuned local machine. Compared GBLUP, LightGBM, SVR, DeepGS DLGWAS, these existing widely (GS) methods. Moreover, generate robust assessments diverse datasets, including quickly incorporate large into usable models, making promising practical approach straightforward GS platforms.

Язык: Английский

---

A comprehensive overview of cotton genomics, biotechnology and molecular biological studies

Science China Life Sciences, Год журнала: 2023, Номер 66(10), С. 2214 - 2256

Опубликована: Март 6, 2023

Язык: Английский

---

How Plants Tolerate Salt Stress

Current Issues in Molecular Biology, Год журнала: 2023, Номер 45(7), С. 5914 - 5934

Опубликована: Июль 15, 2023

Soil salinization inhibits plant growth and seriously restricts food security agricultural development. Excessive salt can cause ionic stress, osmotic ultimately oxidative stress in plants. Plants exclude excess from their cells to help maintain homeostasis stimulate phytohormone signaling pathways, thereby balancing tolerance enhance survival. Continuous innovations scientific research techniques have allowed great strides understanding how plants actively resist stress. Here, we briefly summarize recent achievements elucidating homeostasis, regulation, hormonal responses under Such lay the foundation for a comprehensive of salt-tolerance mechanisms.

Язык: Английский

---

Enhancing climate change resilience in agricultural crops

Current Biology, Год журнала: 2023, Номер 33(23), С. R1246 - R1261

Опубликована: Дек. 1, 2023

Climate change threatens global food and nutritional security through negative effects on crop growth agricultural productivity. Many countries have adopted ambitious climate mitigation adaptation targets that will exacerbate the problem, as they require significant changes in current agri-food systems. In this review, we provide a roadmap for improved production encompasses effective transfer of knowledge into plant breeding management strategies underpin sustainable agriculture intensification resilience. We identify main problem areas highlight outstanding questions potential solutions can be applied to mitigate impacts Although translation scientific advances lags far behind technology, consider holistic approach, combining disciplines collaborative efforts, drive better connections between research, policy, needs society.

Язык: Английский

---

Machine Learning-Assisted Approaches in Modernized Plant Breeding Programs

Genes, Год журнала: 2023, Номер 14(4), С. 777 - 777

Опубликована: Март 23, 2023

In the face of a growing global population, plant breeding is being used as sustainable tool for increasing food security. A wide range high-throughput omics technologies have been developed and in to accelerate crop improvement develop new varieties with higher yield performance greater resilience climate changes, pests, diseases. With use these advanced technologies, large amounts data generated on genetic architecture plants, which can be exploited manipulating key characteristics plants that are important improvement. Therefore, breeders relied high-performance computing, bioinformatics tools, artificial intelligence (AI), such machine-learning (ML) methods, efficiently analyze this vast amount complex data. The bigdata coupled ML has potential revolutionize field increase review, some challenges method along opportunities it create will discussed. particular, we provide information about basis bigdata, AI, ML, their related sub-groups. addition, bases functions learning algorithms commonly breeding, three common integration strategies better different datasets using appropriate algorithms, future prospects application novel equip efficient effective tools development improve efficiency process, tackling facing agriculture era change.

Язык: Английский

---

Towards sustainable agriculture: Harnessing AI for global food security

Artificial Intelligence in Agriculture, Год журнала: 2024, Номер 12, С. 72 - 84

Опубликована: Апрель 30, 2024

The issue of food security continues to be a prominent global concern, affecting significant number individuals who experience the adverse effects hunger and malnutrition. finding solution this intricate necessitates implementation novel paradigm-shifting methodologies in agriculture sector. In recent times, domain artificial intelligence (AI) has emerged as potent tool capable instigating profound influence on sectors. AI technologies provide advantages by optimizing crop cultivation practices, enabling use predictive modelling precision techniques, aiding efficient monitoring disease identification. Additionally, potential optimize supply chain operations, storage management, transportation systems, quality assurance processes. It also tackles problem loss waste through post-harvest reduction, analytics, smart inventory management. This study highlights that how utilizing power AI, we could transform way produce, distribute, manage food, ultimately creating more secure sustainable future for all.

Язык: Английский

---

Temperature‐smart plants: A new horizon with omics‐driven plant breeding

Physiologia Plantarum, Год журнала: 2024, Номер 176(1)

Опубликована: Янв. 1, 2024

Abstract The adverse effects of mounting environmental challenges, including extreme temperatures, threaten the global food supply due to their impact on plant growth and productivity. Temperature extremes disrupt genetics, leading significant issues eventually damaging phenotypes. Plants have developed complex signaling networks respond tolerate temperature stimuli, genetic, physiological, biochemical, molecular adaptations. In recent decades, omics tools other strategies rapidly advanced, offering crucial insights a wealth information about how plants adapt stress. This review explores potential an integrated omics‐driven approach understanding temperatures. By leveraging cutting‐edge methods, genomics, transcriptomics, proteomics, metabolomics, miRNAomics, epigenomics, phenomics, ionomics, alongside power machine learning speed breeding data, we can revolutionize practices. These advanced techniques offer promising pathway developing climate‐proof varieties that withstand fluctuations, addressing increasing demand for high‐quality in face changing climate.

Язык: Английский

---

Revisiting growth–defence trade‐offs and breeding strategies in crops

Plant Biotechnology Journal, Год журнала: 2024, Номер 22(5), С. 1198 - 1205

Опубликована: Фев. 27, 2024

Plants have evolved a multi-layered immune system to fight off pathogens. However, activation is costly and often associated with growth development penalty. In crops, yield the main breeding target usually affected by high disease resistance. Therefore, proper balance between defence critical for achieving efficient crop improvement. This review highlights recent advances in attempts designed alleviate trade-offs resistance crops mediated (R) genes, susceptibility (S) genes pleiotropic genes. We also provide an update on strategies optimizing growth-defence breed future desirable yield.

Язык: Английский

---

Functional genomics of Brassica napus: Progress, challenges, and perspectives

Journal of Integrative Plant Biology, Год журнала: 2024, Номер 66(3), С. 484 - 509

Опубликована: Март 1, 2024

ABSTRACT Brassica napus , commonly known as rapeseed or canola, is a major oil crop contributing over 13% to the stable supply of edible vegetable worldwide. Identification and understanding gene functions in B. genome crucial for genomic breeding. A group genes controlling agronomic traits have been successfully cloned through functional genomics studies . In this review, we present an overview progress made including availability germplasm resources, omics databases genes. Based on current progress, also highlight main challenges perspectives field. The advances contribute better genetic basis underlying complex will expedite breeding high quality, resistance yield varieties.

Язык: Английский

---

Applications of Artificial Intelligence in Climate-Resilient Smart-Crop Breeding

International Journal of Molecular Sciences, Год журнала: 2022, Номер 23(19), С. 11156 - 11156

Опубликована: Сен. 22, 2022

Recently, Artificial intelligence (AI) has emerged as a revolutionary field, providing great opportunity in shaping modern crop breeding, and is extensively used indoors for plant science. Advances phenomics, enviromics, together with the other "omics" approaches are paving ways elucidating detailed complex biological mechanisms that motivate functions response to environmental trepidations. These have provided researchers precise tools evaluate important agronomic traits larger-sized germplasm at reduced time interval early growth stages. However, big data relationships within impede understanding of behind genes driving agronomic-trait formations. AI brings huge computational power many new strategies future breeding. The present review will encompass how applications technology, utilized current breeding practice, assist solve problem high-throughput phenotyping gene functional analysis, advances technologies bring opportunities make envirotyping widely Furthermore, methods, linking genotype phenotype remains massive challenge impedes optimal application field phenotyping, genomics, enviromics. In this review, we elaborate on be preferred tool increase accuracy genotyping, data; moreover, explore developing challenges multiomics computing integration. Therefore, integration can allow rapid identification eventually accelerate crop-improvement programs.

Язык: Английский

---