You're using an outdated browser. Please upgrade to a modern browser for the best experience.
Please note this is a comparison between Version 1 by Ramon Guevara-Gonzalez and Version 3 by Conner Chen.
High-throughput analyses increase the chances to elucidate physiological processes and ecological interactions of plants from the broadened perspective of systems biology. The generation of big data sets from the simultaneous analysis of an extensive collection of biomolecules corresponding to a definite category (genes, transcripts, proteins, and metabolites) has led to the so-called omics approach, which is the primary tool of systems biology. Furthermore, a multi-omics approach makes it possible to obtain a more detailed snapshot of a plant system by simultaneously analyzing its whole genome, proteome, transcriptome, and metabolome. Moreover, the multi-omics approach applied to single-cell functional analyses can simplify data processing and modeling to accurately depict many biological processes in plants.
- crop improvement
- Artificial Intelligence
1. Genomics and Transcriptomics
1. Genomics and Transcriptomics
Genomics refers to the sequencing, assembly, and functional analysis of the genome of a plant, and it has advanced more rapidly than any other omics in plant science [44]. Only in the last two decades, the sequences of more than 100 plant genomes have been published, and further technological advances in genomics have increased our understanding of plant biology leading to substantial agricultural progress [45].
Genome sequencing has several applications in plant stress science. The structural analysis of DNA is not only fundamental for classifying organisms but also for identifying stress-driven mutations, which occur in plants under heat [46], drought [47], and other abiotic stresses [48]. Moreover, DNA structural variations occurring under low-dose stress can be linked to gene function using gene ontology analyses to reveal the genetic basis of hormesis [49]. DNA sequence variations such as Single Nucleotide Polymorphisms (SNPs) are also helpful for understanding the molecular mechanisms underlying hormetic responses when analyzed along with phenotypic traits as in Genome Wide Association Studies (GWAS) [50,51]. GWAS analyses consider big data sets to identify and predict gene candidates and quantitative trait loci accountable for stress responses [52].
SNPs genotyping in combination with other data sets from high-throughput analyses such as phenomics or enviromics has also led to the development of genomic selection for optimizing crop breeding [53]. With this strategy, it is possible to improve physiological traits with hormetic behavior in crops, such as yield, pest resistance, and environmental stress tolerance, to shorten breeding cycles and decrease the need for continuous phenotyping [54]. Additionally, the advent of outstanding new genome-editing techniques, such as the Zinc Finger Nucleases (ZFN), the Transcription Activator-Like Effector Nucleases (TALENs), and the Clustered Regulatory Interspaced Short Palindromic Repeats (CRISPR) systems, implies, along with transcriptomics, the most significant advance in the development of stress-resistant crops [55].
The rapid advances in sequencing technologies and bioinformatics have also substantially impelled RNA analyses [56]. The synthesis of RNA is dynamic, depending on the activation of a gene to occur. Therefore, transcriptomics is the key to investigating gene function in targeted physiological mechanisms qualitatively and quantitatively [57]. Detecting hormetic stimulation at the transcript level can be achieved by analyzing the differential expression of known genes on small (~20) [58,59,60] and very large scales using microarray technology (~50,000) [61] or by completely sequencing the RNA from a sample as in next-generation and third-generation sequencing [62,63,64].
The computational analysis of transcriptomic datasets precedes the reconstruction of gene regulatory networks and the crosstalk by which they interconnect during specific physiological processes [65]. In particular, machine learning algorithms can infer interactions between genes with great accuracy [66]. Nevertheless, gene regulation during hormetic responses involves biological processes other than transcription, such as epigenome dynamics, which depends on chromatin structural changes, namely DNA methylation and histone modifications [67]. Therefore, integrating additional types of datasets and adding spatial and temporal information is fundamental to increasing model resolution and depicting the mechanisms of hormesis truthfully [68].
Genomics refers to the sequencing, assembly, and functional analysis of the genome of a plant, and it has advanced more rapidly than any other omics in plant science [1]. Only in the last two decades, the sequences of more than 100 plant genomes have been published, and further technological advances in genomics have increased our understanding of plant biology leading to substantial agricultural progress [2].
2. Proteomics
Proteins are the main regulatory molecules in every cell process. Therefore, the ensemble of differentially translated proteins in response to a given stimulus is an additional dataset that contains essential information for ascertaining hormetic cellular mechanisms [69]. Moreover, many studies show that RNA quantity does not proportionally relate to protein abundance [70]. The latter occurs mainly due to additional regulation steps between transcription and protein synthesis and the stability of the end products [71]. For that reason, transcriptomics and proteomics, or other high-throughput analyses should be simultaneously conducted to validate and reconstruct entire regulatory networks [72].
Detecting differential changes in plant proteome is especially useful for studying plant stress responses since relatively small variations in the dose of a stressor result in a significant difference in the proteome at both mild and severe stressor incidence [73]. Furthermore, under stress conditions, the plant cell upregulates the expression of proteins associated with primary metabolic processes such as photosynthesis, redox homeostasis, energy metabolism, nitrogen absorption, and the biosynthesis of signaling molecules [74,75]. Therefore, plant proteomics can help researchers detect stress at a molecular level earlier than it would be possible by analyzing changes in observable phenotypic traits and for both stress-susceptible and tolerant genotypes [76].
Proteome analyses make it possible to identify and characterize novel proteins, and along with bioinformatics, proteomics enables tracking variations in protein abundance, form, cellular location, and activity following a stressors incidence [77]. Additionally, proteome research has proven helpful for clarifying cellular organelle function, post-translational modifications, and protein–protein interactions, providing a more in-depth insight into the stress-driven molecular mechanisms of plant cells [78]. Proteomics technologies range from the classic gel-based and the Liquid Chromatography coupled to Mass Spectrometry (LC-MS) approaches to the modern Mass Spectrometry Imaging (MSI), and combined with additional high-throughput analyses, these still underexploited tools are among the most powerful methods for unraveling the molecular mechanism of hormetic stress responses in plants [79].
Genome sequencing has several applications in plant stress science. The structural analysis of DNA is not only fundamental for classifying organisms but also for identifying stress-driven mutations, which occur in plants under heat [3], drought [4], and other abiotic stresses [5]. Moreover, DNA structural variations occurring under low-dose stress can be linked to gene function using gene ontology analyses to reveal the genetic basis of hormesis [6]. DNA sequence variations such as Single Nucleotide Polymorphisms (SNPs) are also helpful for understanding the molecular mechanisms underlying hormetic responses when analyzed along with phenotypic traits as in Genome Wide Association Studies (GWAS) [7][8]. GWAS analyses consider big data sets to identify and predict gene candidates and quantitative trait loci accountable for stress responses [9].
3. Metabolomics
A number of the differentially expressed proteins resulting from stress incidence are regulators that activate and shape specialized metabolic pathways inside the cell [80]. Metabolomics is the study of all the small molecules in a tissue, which, in the case of plants, possess a unique structural and functional complexity [81]. Moreover, due to their sessility, plants depend on chemical signaling to maintain homeostasis and ecological interactions at intra- and interspecies levels. Plant specialized metabolism is evolutionarily shaped by environmental pressures to synthesize chemical compounds with an enormous structural and functional diversity and capable of interacting with living tissues [82].
Many plant specialized metabolites are an active part of plant internal signaling pathways or exert bioactivity on other organisms [83]. Furthermore, every plant organism is a genuine biological factory capable of synthesizing an estimated 30,000 metabolites [84]. Hence, plants are a significant source of chemical compounds with pharmacological properties and are particularly valuable among natural products for drug discovery purposes [85]. In addition, plant metabolites are fundamental for maintaining human health by conferring nutritional, functional, and nutraceutical value to food products [86].
Being an adaptive response, the activation of plant metabolism also exhibits a hormetic behavior [87,88,89], and deliberately exposing crops to low-dose stress is a convenient means for stimulating metabolites accumulation [90]. Moreover, the metabolomics analysis of plant stress response along with bioinformatics makes it possible to find candidate markers for directing crop breeding and predicting crop performance under environmental stress [91].
Given the structural diversity of plant metabolites, the main limitation of metabolomics resides in developing comprehensive extraction techniques and analytical methods to detect a big heterogeneous ensemble of chemical compounds simultaneously. Nevertheless, thanks to the recent advances in coupled analytical technologies and bioinformatics, particularly Mass Spectrometry (MS), Nuclear Magnetic Resonance (NMR), and hybrid MS/NMR methods [92], it is now possible to separate and detect the whole metabolome from a biological sample quickly and affordably [93]. Moreover, many intrinsic experimental conditions for metabolome analysis are compatible with other omics studies, making metabolomics a convenient foundation for designing and fulfilling multi-omics experiments and an effective tool for systems biology research [94].
SNPs genotyping in combination with other data sets from high-throughput analyses such as phenomics or enviromics has also led to the development of genomic selection for optimizing crop breeding [10]. With this strategy, it is possible to improve physiological traits with hormetic behavior in crops, such as yield, pest resistance, and environmental stress tolerance, to shorten breeding cycles and decrease the need for continuous phenotyping [11]. Additionally, the advent of outstanding new genome-editing techniques, such as the Zinc Finger Nucleases (ZFN), the Transcription Activator-Like Effector Nucleases (TALENs), and the Clustered Regulatory Interspaced Short Palindromic Repeats (CRISPR) systems, implies, along with transcriptomics, the most significant advance in the development of stress-resistant crops [12].
The rapid advances in sequencing technologies and bioinformatics have also substantially impelled RNA analyses [13]. The synthesis of RNA is dynamic, depending on the activation of a gene to occur. Therefore, transcriptomics is the key to investigating gene function in targeted physiological mechanisms qualitatively and quantitatively [14]. Detecting hormetic stimulation at the transcript level can be achieved by analyzing the differential expression of known genes on small (~20) [15][16][17] and very large scales using microarray technology (~50,000) [18] or by completely sequencing the RNA from a sample as in next-generation and third-generation sequencing [19][20][21].
The computational analysis of transcriptomic datasets precedes the reconstruction of gene regulatory networks and the crosstalk by which they interconnect during specific physiological processes [22]. In particular, machine learning algorithms can infer interactions between genes with great accuracy [23]. Nevertheless, gene regulation during hormetic responses involves biological processes other than transcription, such as epigenome dynamics, which depends on chromatin structural changes, namely DNA methylation and histone modifications [24]. Therefore, integrating additional types of datasets and adding spatial and temporal information is fundamental to increasing model resolution and depicting the mechanisms of hormesis truthfully [25].
2. Proteomics
Proteins are the main regulatory molecules in every cell process. Therefore, the ensemble of differentially translated proteins in response to a given stimulus is an additional dataset that contains essential information for ascertaining hormetic cellular mechanisms [26]. Moreover, many studies show that RNA quantity does not proportionally relate to protein abundance [27]. The latter occurs mainly due to additional regulation steps between transcription and protein synthesis and the stability of the end products [28]. For that reason, transcriptomics and proteomics, or other high-throughput analyses should be simultaneously conducted to validate and reconstruct entire regulatory networks [29].
Detecting differential changes in plant proteome is especially useful for studying plant stress responses since relatively small variations in the dose of a stressor result in a significant difference in the proteome at both mild and severe stressor incidence [30]. Furthermore, under stress conditions, the plant cell upregulates the expression of proteins associated with primary metabolic processes such as photosynthesis, redox homeostasis, energy metabolism, nitrogen absorption, and the biosynthesis of signaling molecules [31][32]. Therefore, plant proteomics can help researchers detect stress at a molecular level earlier than it would be possible by analyzing changes in observable phenotypic traits and for both stress-susceptible and tolerant genotypes [33].
Proteome analyses make it possible to identify and characterize novel proteins, and along with bioinformatics, proteomics enables tracking variations in protein abundance, form, cellular location, and activity following a stressors incidence [34]. Additionally, proteome research has proven helpful for clarifying cellular organelle function, post-translational modifications, and protein–protein interactions, providing a more in-depth insight into the stress-driven molecular mechanisms of plant cells [35]. Proteomics technologies range from the classic gel-based and the Liquid Chromatography coupled to Mass Spectrometry (LC-MS) approaches to the modern Mass Spectrometry Imaging (MSI), and combined with additional high-throughput analyses, these still underexploited tools are among the most powerful methods for unraveling the molecular mechanism of hormetic stress responses in plants [36].
3. Metabolomics
A number of the differentially expressed proteins resulting from stress incidence are regulators that activate and shape specialized metabolic pathways inside the cell [37]. Metabolomics is the study of all the small molecules in a tissue, which, in the case of plants, possess a unique structural and functional complexity [38]. Moreover, due to their sessility, plants depend on chemical signaling to maintain homeostasis and ecological interactions at intra- and interspecies levels. Plant specialized metabolism is evolutionarily shaped by environmental pressures to synthesize chemical compounds with an enormous structural and functional diversity and capable of interacting with living tissues [39].
Many plant specialized metabolites are an active part of plant internal signaling pathways or exert bioactivity on other organisms [40]. Furthermore, every plant organism is a genuine biological factory capable of synthesizing an estimated 30,000 metabolites [41]. Hence, plants are a significant source of chemical compounds with pharmacological properties and are particularly valuable among natural products for drug discovery purposes [42]. In addition, plant metabolites are fundamental for maintaining human health by conferring nutritional, functional, and nutraceutical value to food products [43].
Being an adaptive response, the activation of plant metabolism also exhibits a hormetic behavior [44][45][46], and deliberately exposing crops to low-dose stress is a convenient means for stimulating metabolites accumulation [47]. Moreover, the metabolomics analysis of plant stress response along with bioinformatics makes it possible to find candidate markers for directing crop breeding and predicting crop performance under environmental stress [48].
Given the structural diversity of plant metabolites, the main limitation of metabolomics resides in developing comprehensive extraction techniques and analytical methods to detect a big heterogeneous ensemble of chemical compounds simultaneously. Nevertheless, thanks to the recent advances in coupled analytical technologies and bioinformatics, particularly Mass Spectrometry (MS), Nuclear Magnetic Resonance (NMR), and hybrid MS/NMR methods [49], it is now possible to separate and detect the whole metabolome from a biological sample quickly and affordably [50]. Moreover, many intrinsic experimental conditions for metabolome analysis are compatible with other omics studies, making metabolomics a convenient foundation for designing and fulfilling multi-omics experiments and an effective tool for systems biology research [51].
4. Phenomics
Plant phenotyping is the measurement of phenotypic traits either at the cell, organ, or whole plant level for understanding the underlying mechanisms of the interactions of plants with their environment [52]. Molecular responses drive phenotypic change, and for that reason, the developmental traits of plants, such as growth, seed germination, photosynthesis, transpiration, stomatal conductance, and pigmentation, among others, also display hormetic behavior [53]. As a result, various sensors can be used for differentially analyzing physiological plant hormetic responses to stress-related events [54].
Image-based phenotyping is useful to detect leaf morphological variations in plants [55]. Red-green-blue (RGB) imaging uses Charge Coupled Device (CCD) or Complementary Metal Oxide Semiconductor (CMOS) sensors to detect color changes related to plant stress responses. Such sensors work within the visible range of the electromagnetic spectrum and are convenient to diagnose nitrogen (N), phosphorous (P), potassium (K), magnesium (Mg), calcium (Ca), and iron (Fe) deficiency symptoms [56]. Detecting nutrient deficit can also help identify environmental stress incidence. For example, Martinez et al. (2020) [57] reported that water deficit modifies nitrate uptake by altering the expression of genes related to nitrate assimilation in the roots and the shoot. Moreover, changes in pigment content can be related to visible stress symptoms in such a detailed manner, that it is possible to discriminate between their biotic or abiotic origin [58].
Yellowing is the most notable symptom of leaf senescence, and it appears due to seasonal developmental processes, pathogen attack, and abiotic stressors incidence, indicating a decrease in the photosynthetic rate [59]. Chlorophyll metabolism is regulated in a hormetic manner, and therefore it can perform as a biomarker to identify other metabolic changes resulting from low-dose stress incidence [60][61]. Many imaging techniques focus on detecting chlorophyll fluctuations with convenient results for biotic and abiotic stress phenotyping, such as chlorophyll fluorescence. This technique is relevant to determining overall crop fitness, and due to its high sensitivity, it has been extensively applied for the early detection of stress incidence [62].
Imaging techniques can also be used to analyze plant physiological processes and identify stress even in the absence of symptoms unobservable for the unaided eye. Magnetic resonance imaging can be applied to plant systems to elucidate plant-water relationships and as a post-harvest control to determine maturity and mechanical damage of agricultural products [63]. Thermography uses optical sensors that detect radiation outside the visual range of the electromagnetic spectrum, and it has been used to detect plant interactions with biotic and abiotic stressors and monitor environmental stress susceptibility and resistance [64][65]. Mild-stress responses can also be detected using multispectral and hyperspectral imaging. Multispectral imaging considers only specific bands of electromagnetic spectra, whereas hyperspectral imaging increases the resolution of the wavelengths. These technologies can identify plant diseases such as yellow rust and powdery mildew in wheat and leaf rust in sugar beet from early stages [66]. Moreover, multispectral imaging works on large scales by employing uncrewed aerial vehicles and satellites. Therefore, multispectral and hyperspectral imaging, along with other omics techniques, could be used to develop hormesis management protocols at a crop scale.
Given the intricacy of physiological responses, the elucidation of the adaptive mechanisms of plants to low-dose stress must be carried out from a multidimensional approach, utilizing comprehensive analyses for detecting the differential changes stimulated in different layers of the stress response. Understanding such mechanisms and, in particular, characterizing the hormetic dose-response curve allows eustress treatments to be implemented to enhance stress tolerance and increase food production and quality [67]. Nevertheless, integrating multiple-layer datasets gives rise to additional challenges beyond data collection and storage, including data management and processing [68]. Therefore, handling and modeling hormetic responses from multi-omics data requires computational methods for transforming data into knowledge.
4. Phenomics
Plant phenotyping is the measurement of phenotypic traits either at the cell, organ, or whole plant level for understanding the underlying mechanisms of the interactions of plants with their environment [95]. Molecular responses drive phenotypic change, and for that reason, the developmental traits of plants, such as growth, seed germination, photosynthesis, transpiration, stomatal conductance, and pigmentation, among others, also display hormetic behavior [96]. As a result, various sensors can be used for differentially analyzing physiological plant hormetic responses to stress-related events [97].
Image-based phenotyping is useful to detect leaf morphological variations in plants [98]. Red-green-blue (RGB) imaging uses Charge Coupled Device (CCD) or Complementary Metal Oxide Semiconductor (CMOS) sensors to detect color changes related to plant stress responses. Such sensors work within the visible range of the electromagnetic spectrum and are convenient to diagnose nitrogen (N), phosphorous (P), potassium (K), magnesium (Mg), calcium (Ca), and iron (Fe) deficiency symptoms [99]. Detecting nutrient deficit can also help identify environmental stress incidence. For example, Martinez et al. (2020) [100] reported that water deficit modifies nitrate uptake by altering the expression of genes related to nitrate assimilation in the roots and the shoot. Moreover, changes in pigment content can be related to visible stress symptoms in such a detailed manner, that it is possible to discriminate between their biotic or abiotic origin [101].
Yellowing is the most notable symptom of leaf senescence, and it appears due to seasonal developmental processes, pathogen attack, and abiotic stressors incidence, indicating a decrease in the photosynthetic rate [102]. Chlorophyll metabolism is regulated in a hormetic manner, and therefore it can perform as a biomarker to identify other metabolic changes resulting from low-dose stress incidence [103,104]. Many imaging techniques focus on detecting chlorophyll fluctuations with convenient results for biotic and abiotic stress phenotyping, such as chlorophyll fluorescence. This technique is relevant to determining overall crop fitness, and due to its high sensitivity, it has been extensively applied for the early detection of stress incidence [105].
Imaging techniques can also be used to analyze plant physiological processes and identify stress even in the absence of symptoms unobservable for the unaided eye. Magnetic resonance imaging can be applied to plant systems to elucidate plant-water relationships and as a post-harvest control to determine maturity and mechanical damage of agricultural products [106]. Thermography uses optical sensors that detect radiation outside the visual range of the electromagnetic spectrum, and it has been used to detect plant interactions with biotic and abiotic stressors and monitor environmental stress susceptibility and resistance [107,108]. Mild-stress responses can also be detected using multispectral and hyperspectral imaging. Multispectral imaging considers only specific bands of electromagnetic spectra, whereas hyperspectral imaging increases the resolution of the wavelengths. These technologies can identify plant diseases such as yellow rust and powdery mildew in wheat and leaf rust in sugar beet from early stages [109]. Moreover, multispectral imaging works on large scales by employing uncrewed aerial vehicles and satellites. Therefore, multispectral and hyperspectral imaging, along with other omics techniques, could be used to develop hormesis management protocols at a crop scale.
Given the intricacy of physiological responses, the elucidation of the adaptive mechanisms of plants to low-dose stress must be carried out from a multidimensional approach, utilizing comprehensive analyses for detecting the differential changes stimulated in different layers of the stress response. Understanding such mechanisms and, in particular, characterizing the hormetic dose-response curve allows eustress treatments to be implemented to enhance stress tolerance and increase food production and quality [7]. Nevertheless, integrating multiple-layer datasets gives rise to additional challenges beyond data collection and storage, including data management and processing [110]. Therefore, handling and modeling hormetic responses from multi-omics data requires computational methods for transforming data into knowledge.