Authors: Taotao Wang, Wenjing Hu, Weifeng Song, Xiwen Liao, Hongjian Zheng, Xingping Zhang, Xiufang Xin, Pawan Kumar Singh, Yuan Chen, Yunbi Xu
Categories: Review Article, plant disease management, disease pyramid network, hostomics, pathomics, microbiomics, enviromics
Source: Plant Communications
Authors: Taotao Wang, Wenjing Hu, Weifeng Song, Xiwen Liao, Hongjian Zheng, Xingping Zhang, Xiufang Xin, Pawan Kumar Singh, Yuan Chen, Yunbi Xu
Understanding plant disease development requires moving beyond the classic disease triangle, which considers the host, pathogen, and environment. Recent advances in multi-omics have highlighted the importance of a disease pyramid that integrates the host, pathogen, microbiome, and environment to capture the complex interactions among these core biological/ecological components. This pyramid framework emphasizes how host genetic architecture, pathogen traits, microbiome dynamics, and environmental conditions collectively and interactively shape disease outcomes, plant phenotypes, and adaptive potential. The conceptual expansion from the disease triangle to a pyramid model reflects this shift, providing a more holistic and dynamic view of plant disease ecology. Environmental factors regulate host susceptibility and restructure both pathogenic and non-pathogenic microbial communities, thereby influencing infection severity and disease progression. Multi-omics approaches—encompassing hostomics, pathomics, microbiomics, and enviromics—hold strong promise for dissecting these interactions, enabling predictive disease modeling and the development of sustainable management strategies. Moreover, integrating enviromics data into resistance breeding enables the identification of key environmental variables and their interactions with host genotypes and pathogenic and non-pathogenic microbes, thereby supporting the deployment of durable resistance across diverse agroecosystems. Together, these perspectives advance a systems-level understanding of plant health and open new avenues for disease management through omics-driven breeding, microbiome-informed strategies, and environmentally responsive interventions.
Crop pathogens lead to losses in both yield and quality, exacerbating challenges to global food security and nutrition. With ongoing reductions in arable land and freshwater resources and a continuously growing population, we urgently need to increase food production while controlling agricultural greenhouse gas emissions. Achieving this two-part goal is important for mitigating climate change, protecting natural habitats, and maintaining biodiversity (Richard et al., 2022).
The core challenge of plant disease management lies in analyzing the complex relationships among host plants, pathogenic microorganisms, and their environment. This is the foundation of the “disease triangle” (Figure 1). Modern plant pathology recognizes that this framework can be expanded by explicitly incorporating plant-associated microbes as a distinct biological component. Four corresponding omics approaches are advancing our understanding of the four elements involved in disease hostomics (the genetic and physiological basis of host resistance) (Zhu et al., 2025), pathomics (pathogen virulence and effector repertoires) (Gupta et al., 2019), microbiomics (the structure and function of non-pathogenic microbial communities) (Compant et al., 2025; Tsigehana, 2025), and enviromics (the systematic quantification of environmental factors that influence plant–microbe interactions) (Newman and Furbank, 2021; Xu et al., 2022; Napier et al., 2023; Resende et al., 2025). A comprehensive analysis of host–pathogen–microbe–environment interactions (HPMEIs) can clarify how environmental conditions shape host-related microbial communities and thereby regulate plant phenotypes and adaptability (Figure 1). This four-component network, which we refer to as the disease pyramid network, is particularly important, as it reveals how biological traits, pathogen virulence, microbiome functions, and environmental factors jointly and dynamically determine plant disease outcomes. In the host plant system, the study of HPMEIs provides a key perspective for understanding plant adaptation, evolution, and health status (Saltz et al., 2018; Trivedi et al., 2020; Blonde et al., 2025).Figure 1Transition from the traditional host–pathogen–environment (h–p–e) plant disease triangle to the modern hostomics–pathomics–microbiomics–enviromics (H–P–M–E) pyramid network.Left: the traditional incomplete disease triangle (host–pathogen–environment, h–p–e) with limited, weak h–p–e interactions at the level of individuals at a specific time point. Middle: disease pyramid network (h–p–m–e) with microbes (non-pathogens) (m) incorporated. Right: modern complete disease pyramid network (hostomics–pathomics–microbiomics–enviromics) with strong interaction or interplay (H–P–M–E) at the level of omics across spatiotemporal scales. Bidirectional arrows indicate multidirectional interactions among all four elements, emphasizing that disease outcomes emerge from their dynamic interplay.
In the study of genotype–environment interactions (GEIs) or HPMEIs, environmental factors have been viewed as a “black box” and a mixture of complex components. Enviromics has emerged as a discipline that aims to systematically quantify and analyze all the environmental factors that affect organisms (Roden and Ingle, 2009; Resende et al., 2021; Xu et al., 2022; Elmerich et al., 2023), providing tools to open this “black box” for hosts, pathogenic microbes, and non-pathogenic microbes alike.
In response to the challenges posed by diseases such as Fusarium head blight (FHB), there is an urgent need for innovative plant disease and pathogen management strategies. By gaining a comprehensive understanding of GEI mechanisms, host–pathogen interactions, microbiome ecology, and the factors that influence disease dynamics, we can leverage the potential of HPMEI to drive the transformation of disease prevention and control (Figure 1). This article provides a detailed roadmap for the study and application of HPMEIs in disease control. We first propose an expansion of the classic disease triangle into a pyramid model that explicitly includes plant-associated microbes as core components and positions enviromics as a key analytical approach. We then dissect how environmental variables regulate pathogen incidence, disease progression, and virulence, and we highlight the transformative potential of enviromics, from high-resolution sensing to artificial intelligence (AI)-guided analytics, to advance the prediction, monitoring, and control of plant diseases. Finally, we confront the key challenges that still impede implementation and outline forward-looking strategies to turn insights into resilient, field-ready solutions.
The plant canopy and root system host different but interconnected microbial communities that together influence plant health. Plant diseases caused by the spread of pathogens through soil, air, and insects are not isolated events; they are the result of ongoing interactions among the host plant, the pathogen, and the environment, known as the disease triangle (Figure 1). Building on the classic disease triangle framework (Scholthof, 2007), this section synthesizes recent advances in research on air- and soilborne diseases and proposes a transition toward a more integrated network the disease pyramid.
The rhizosphere, which is the area surrounding the roots of plants, harbors various microbial communities that play especially critical roles in regulating plant growth, nutrient availability, abiotic stress tolerance, and disease dynamics (Li et al., 2025) (Figure 2). The rhizosphere and its associated endosphere harbor taxonomically and functionally distinct communities shaped by root exudates and soil physicochemical gradients (Mendes et al., 2011). Combining HPMEI analysis with rhizosphere microbiome studies reveals how these microorganisms contribute to disease resistance and highlights beneficial taxa that enhance immunity or suppress pathogens, providing microbiome-based strategies for sustainable disease management.Figure 2Atmosphere, canopy (carposphere, caulosphere, and phyllosphere), and rhizosphere microbiomics–enviromics interactions and their effects on host plants, pathogens, and non-pathogenic microbes.Plant growth environments can be categorized into three distinct the atmosphere, canopy, and rhizosphere. Each layer has unique functional characteristics and exhibits specific relationships with plant diseases. Atmospheric/above-canopy this zone is characterized by airborne microorganisms and is influenced by atmospheric climate factors (e.g., airflow, rainfall, radiation, and carbon dioxide) and pollutants. Typical diseases in this zone include powdery mildew, rusts, gray mold, viral diseases, downy mildew, and anthracnose. Canopy the microclimate of the plant canopy, determined by factors such as leaf moisture, temperature, light intensity, and human activities, directly affects the infection and spread of seedborne and fruit rot diseases. Rhizosphere the soil environment around roots is shaped by soil properties (e.g., water content, pH, nutrients, and salt) and root exudates. This zone harbors soilborne microorganisms and is the primary site for diseases such as root rot, Fusarium wilt, damping off, bacterial wilt, white silk disease, Sclerotinia, and root-knot nematode disease. Detailed annotations of microbiome–enviromics interactions can be found in Table 1.
In the phyllosphere, different types of bacteria and fungi form biofilms on the surface of leaves (Liu et al., 2024) (Figure 2). Airborne pathogens can spread over long distances through air currents, causing epidemics that affect aerial organs such as stems, leaves, flowers, and seeds/fruits (Aylor, 2003). Classic examples include wheat rust (Puccinia spp.), wheat powdery mildew (Blumeria graminis), rice blast (Magnaporthe oryzae), maize northern and southern leaf blights (Setosphaeria turcica and Cochliobolus heterostrophus), potato late blight (Phytophthora infestans), apple scab (Venturia inaequalis), and cucurbit downy mildew (Pseudoperonospora cubensis). Under suitable meteorological conditions, these pathogens can cause significant reductions in yield (Dean et al., 2012).
Soilborne pathogens (including fungi, bacteria, actinomycetes, and nematodes) can survive in plant debris or soil organic matter and infect the roots or stem bases when conditions are favorable (Schuster and Coyne, 1974; Punja, 1985; Weller et al., 2002; Loria et al., 2006; Hamada et al., 2020) (Figure 2). Fungi dominate this guild and are classified as either non-obligate root invaders (e.g., Pythium spp. that cause damping-off and Rhizoctonia solani, which causes seedling blight) or obligate vascular pathogens (e.g., Fusarium oxysporum, which induces wilt, and Verticillium albo-atrum, which induces vascular discoloration and stunting). The severity of root diseases is strongly modulated by the chemical composition and concentration of root exudates, which serve both as nutrients for pathogens and as signals for spore germination (Hinsinger et al., 2009). Consequently, suppression of pathogen activity in the rhizosphere is the cornerstone of soilborne disease management but must be framed within the constraints imposed by soil physiochemical properties on the tripartite interaction among plants, soil microbes, and root pathogens (Janvier et al., 2007).
The triad of disease factors determines whether an infection develops and the severity of the the host provides a biological environment for establishment of the pathogen; the pathogen must be able to invade and proliferate; and the environment affects the susceptibility of the host and the survival ability of the pathogen. When appropriate conditions of all three disease factors coincide, outbreaks occur. Given the multifactorial nature of epidemics, disease management must integrate host resistance, soil health, and micro-climate regulation (Cook, 2000; Strange and Scott, 2005). Host resistance acts as the “intrinsic core” that determines a plant’s inherent defense potential and includes vertical, horizontal, and quantitative resistance. Soil health serves as the “fundamental support,” and optimization of the soil physical environment, nutrient supply, and microbial community can reinforce host resistance and suppress soil-borne pathogens. Microclimate regulation functions as the “external safeguard”: modulation of temperature, humidity, light levels, and air circulation disrupts pathogen transmission chains and reduces the chances of infection (Yusuf et al., 2025). Together, these three components constitute the plant disease-resistance system.
Host characteristics, including genetics, physiological status, and stress levels, determine disease susceptibility. In plants, resistance genes govern vulnerability to disease infections. In addition to genetics, host defense capacity is also shaped by age, nutrition, and pre-existing conditions. Host resistance is fundamentally governed by genetic factors and can be enhanced by combining multiple durable resistance traits, which further strengthens plant defenses and reduces susceptibility to pathogens. The plant immune system can be activated by the application of synthetic elicitors, which are used to study immune mechanisms and develop crop protection strategies (Beckers and Conrath, 2007).
Plant pathogens deploy diverse virulence factors, including toxins, adhesion molecules, and immune evasion mechanisms, to colonize their hosts and overcome plant immunity. Among bacterial pathogens, Pseudomonas syringae pv. tomato DC3000, a Gram-negative bacterium used as a model pathogen for the last two decades, secretes effectors such as HopM1 and AvrPto/AvrPtoB to interfere with host trafficking and pattern recognition receptor (PRR)-mediated signaling, respectively, as well as the phytotoxin coronatine, which mimics jasmonate to promote stomatal reopening (Shan et al., 2008; Arnaud et al., 2017). Likewise, oomycete pathogens such as P. infestans and Phytophthora sojae secrete large repertoires of RXLR and CRN effectors that either suppress immunity or induce host cell death (King et al., 2014; Du et al., 2015; Zhang et al., 2015; Ren et al., 2019). For example, P. infestans Avr3a suppresses hypersensitive cell death, whereas P. sojae Avr1b dampens defense responses; in both cases, coevolution with host resistance genes has shaped effector recognition (Yaeno et al., 2011; King et al., 2014). These pathogens also secrete cutinases, glucanases, and other cell wall–degrading enzymes to facilitate tissue penetration, and P. sojae further produces glucanase inhibitors to block host antifungal enzymes (King et al., 2014). In fungi, the cereal pathogen Fusarium graminearum relies heavily on mycotoxins, particularly trichothecenes such as deoxynivalenol synthesized via the TRI5 gene, which inhibit host protein synthesis and weaken immunity; additional toxins, such as zearalenone, further enhance virulence (Huang et al., 2024; Cai et al., 2025). Similarly, M. oryzae, the causal agent of rice blast, deploys a large repertoire of effectors, including AVR-Pita, to suppress host defenses and promote colonization (Liu et al., 2024). The maize pathogen Ustilago maydis secretes numerous effector proteins that interfere with host peroxidase activity, thereby suppressing early immune responses (Ma et al., 2018). Plant viruses have evolved distinct molecular strategies to manipulate host defenses and vector interactions. For instance, the rice stripe virus recruits E3 ubiquitin ligases via its viral proteins to target the chitin-binding lectin OsChtBL1 for degradation, thereby enhancing vector feeding and facilitating viral transmission (Yang et al., 2025).
Environmental conditions influence host susceptibility, pathogen survival, and the surrounding microbial community (Figure 2). Temperature, humidity, and soil composition determine whether a pathogen proliferates or remains dormant. Precision irrigation minimizes prolonged leaf wetness, sensor-guided nitrogen management optimizes nutrient supply at fine scales, and climate-driven forecasting models translate anticipated weather conditions into proactive management decisions (Tsong and Khor, 2023).
For instance, prolonged leaf wetness fosters fungal infections, as illustrated by a 10-fold increase in the density of Plasmopara viticola sporangia when vineyard canopy humidity increased from 60% to 85% (Rossi et al., 2008). Similarly, nutrient balance plays a pivotal excessive nitrogen input (200 kg N ha^−1^) aggravates wheat Septoria tritici blotch, whereas optimized N:Si ratios reduce disease severity by 45% (Savary et al., 2017). Climate change further complicates these interactions by expanding the range of vector-borne diseases and altering host–pathogen relationships; for example, elevated CO2 (700 ppm) delays sugar beet senescence, intensifying Cercospora leaf spot (Vallad and Goodman, 2004).
Interactions among the host, pathogen, and environment (the disease triangle) often amplify disease outcomes. When optimal conditions coincide, infection rates accelerate, severity intensifies, and transmission potential increases. A mildly virulent pathogen under normal conditions may trigger an epidemic when paired with a stressed host and favorable environmental conditions. Such synergy is evident in wheat rust outbreaks, in which pathogen spread accelerates under warm, humid conditions in susceptible wheat fields, with climate change projected to expand disease-risk zones (Aloyce, 2025). Powdery mildew in cucurbits is more severe during dry summers or in late fall seasons characterized by low light intensity and high humidity (Pawełkowicz et al., 2025).
Environmental shifts can tip the balance toward widespread disease. Sudden drought weakens plant defenses, increasing vulnerability, while pathogens exploit these conditions by producing spores or toxins more efficiently (Lahlali et al., 2024). Extreme high temperatures (>45°C) suppress both the powdery mildew pathogen and cucurbit plants (Pawełkowicz et al., 2025). Virus-induced damage to the host may be reduced under conditions that favor host growth (Jeger, 2023).
Host–pathogen co-evolution further influences disease dynamics, particularly when environmental stressors accelerate genetic adaptation (Liu et al., 2025). Under rapidly changing climates, pathogens may evolve more aggressive traits, whereas hosts often lag behind. This imbalance is particularly concerning for vector-borne diseases, as environmental shifts introduce new hosts and shorten parasite replication cycles, thereby increasing transmission rates (Lahlali et al., 2024). Non-linear interactions frequently precipitate outbreaks. For example, the large-scale Chinese stripe rust epidemic in 2016 resulted from the convergence of the PstS7 race with the highly susceptible wheat cultivar Zhoumai 22 under favorable climatic conditions, in particular, a mild winter (Hovmøller et al., 2016). Studies in maize have shown that ZmWRKY79 regulates drought tolerance by modulating abscisic acid (ABA) biosynthesis, highlighting the dual role of ABA in coordinating both phytoalexin production and abiotic stress responses (Gulzar et al., 2021). Collectively, these findings underscore the multifaceted impacts of climate change on plant disease dynamics and the urgent need for strategies to understand and manage these processes under current and future climate scenarios (Singh et al., 2023; Lahlali et al., 2024).
Late blight of potato is a classic example of the “disease triangle.” Phytophthora can overcome the defenses of resistant potato varieties under cold, humid conditions. The Great Irish Famine in the 19th century resulted from the convergence of a susceptible host, rapid pathogen proliferation, and prolonged rainy weather. Even today, modern outbreaks of late blight remain closely linked to environmental conditions that favor pathogen proliferation. Therefore, enhanced monitoring and optimized deployment of disease-resistant crop varieties are essential (Dong and Zhou, 2022). Root rot caused by Fusarium is particularly severe under conditions of soil waterlogging due to excessive irrigation or poor drainage, highlighting the critical role of environmental management in controlling fungal diseases.
The disease triangle was first proposed by Huetzel in 1918 and systematically elaborated by Gäumann in 1951 (Oliver, 2024); it posits that plant diseases occur only when a susceptible host, a virulent pathogen, and a favorable environment overlap in time and space (Figure 1). Contemporary plant pathology has extended this concept to a “disease tetrahedron,” identifying human interventions as a fourth dimension that interacts with the host, pathogen, and environment. These interventions include cultivar selection, chemical control, and agronomic practices (Scholthof, 2007). Although valuable, the tetrahedron model focuses primarily on human management as an external force. By contrast, the HPME pyramid network proposed here is distinguished by its explicit recognition of plant-associated microbes (both pathogenic and non-pathogenic microbes) as intrinsic biological components with agency. This microbe component actively competes and collaborates with other components (host and environment) and provides feedback, rather than being a passive aspect of the environment or merely a target for human intervention. This distinction is crucial for understanding natural disease suppression mechanisms such as microbial competition. Supplemental Table S1 provides a comparative analysis of these conceptual models.
The classic disease triangle simplifies the complex nature of plant–pathogen–environment interactions. It considers the host, pathogen, and environment as largely independent variables, overlooking the complex feedback loops and the agency of plant-associated microbes as key biological players (Figure 1). For example, wheat scab is caused by Fusarium, but the disease triangle cannot fully explain the significant variations in disease severity observed under seemingly similar environmental conditions. The presence or absence of certain beneficial microorganisms in the soil can greatly influence the outcome of the disease, but this is not accounted for in the traditional model (Leveau, 2024). Although the disease tetrahedron usefully incorporates human intervention, it still treats the broader microbial community largely as a part of the “environment” rather than as a distinct biological entity capable of active interactions with the host and pathogen.
The rationale for separating microbes (M) from the broader environment (E) lies in their unique biological properties and their direct, dynamic interactions with the host and pathogen. Plant-associated microbes (including rhizosphere, phyllosphere, and endosphere communities) are not merely part of the external environment but constitute living communities that (1) co-evolve with the host plant, (2) engage in direct molecular dialogue with the host (e.g., through signal exchange and induced systemic resistance), (3) compete directly with pathogens for resources and space (e.g., via niche exclusion and antibiosis), and (4) can be actively recruited and shaped by host genotype and physiology through root exudates and related mechanisms (Tiwari et al., 2025). By contrast, the environment (E) in our pyramid network refers primarily to abiotic factors (e.g., temperature, humidity, light, and soil chemistry) that establish the physical and chemical context that regulates—but does not directly participate in—the biological interactions among H, P, and M. This distinction is critical for modeling disease dynamics, as it separates the players (H, P, and M) from the stage and conditions (E) that modulate their performance.
A key consideration in defining non-pathogenic microbes as a core component is the boundary of the “disease system.” Antagonistic interactions between non-pathogenic microbes (e.g., antibiosis and competition) occur ubiquitously in soil and on plant surfaces, often independently of a living host. If a beneficial microbe suppresses a pathogen in bulk soil, does this constitute part of the disease triangle? This question challenges a strictly plant-centric perspective. We propose that the pyramid network is most informative when focused on host-associated microbes—communities whose assembly, structure, and function are directly influenced by the living host (through exudates, immune signals, and related processes) and that directly interact with both the host and invading pathogens in the rhizosphere, phyllosphere, or endosphere. Within this interaction zone, microbial competition becomes a direct mediator of host health. However, this framework also acknowledges that the broader environmental microbial pool (as part of E) serves as a reservoir from which host-associated microbiota are recruited, and its inherent suppressive capacity reflects environmental and management history. Thus, the pyramid network, to be discussed below, integrates both the immediate, host-modulated microbiome and the broader microbial context from which it is derived.
A more holistic model for understanding plant disease integrates four core H, P, M, and E. Here, we propose a pyramid network model (Figure 1) to represent the dynamic, multidirectional interactions among these entities. This model emphasizes that disease outcomes arise from complex, nonlinear interactions, in which no single factor is universally “dominant.” The environment acts as a critical modulator and integrator, simultaneously influencing and being influenced by the biological interactions among H, P, and M (Figure 1). For example, atmospheric CO2 concentration and phosphorus availability jointly regulate microbial community dynamics in the wheat rhizosphere. Clarifying the mechanisms by which these factors interact is crucial for predicting ecosystem functions under climate change (Jin et al., 2022). Some plants are disease-resistant under cold conditions but exhibit susceptible phenotypes in warm and humid environments, indicating that host–pathogen interactions are environmentally dependent (Cheng et al., 2019). The pyramid network model provides a more refined framework for plant disease management. To fully realize the predictive ability of this model, it will be necessary to characterize the precise mechanisms driving disease dynamics through integrated omics approaches (hostomics, pathomics, microbiomics, and enviromics) that study each component.
In recent years, the advent of multi-omics and metagenomics approaches, grounded in microbial enviromics, has revolutionized the study of GEI (López-Mondéjar et al., 2017; Doni et al., 2022). These research methods can be expanded to comprehensively analyze organisms (including host plants, pathogens, and non-pathogenic microorganisms) and the mechanisms by which they respond to environmental factors, thereby deepening our understanding of disease dynamics. By integrating multi-omics and metagenomic technologies, scientists can reveal the complex pyramid network that affects an organism’s response to environmental signals and pathogen invasion. Enviromics can decipher various environmental variables that affect plant growth and development. Analysis of the complex interactions among genotypes, the environment, and management measures (G × E × Mg) can significantly enhance the efficiency of crop improvement and enable crops to better adapt to the challenges of climate change (Resende et al., 2025). These omics techniques provide valuable insights into the molecular mechanisms of disease resistance, pathogen virulence, and complex HPMEIs. The core entities in our model are H, P, M, and E, which form a disease pyramid, at the top of which the environment interacts with H, M, and P. The corresponding “omics” disciplines (hostomics, pathomics, microbiomics, and enviromics) are the analytical toolkits used to measure, characterize, and understand the states and interactions of these entities (Figure 1). These technologies can also help researchers identify potential disease control targets and enhance plant health by regulating microbial communities, thereby reducing the negative impacts of pathogens.
To study the rhizosphere microbiome of tetraploid wheat (wild and cultivated varieties), researchers integrated multi-omics and metagenomic technologies. 16S/ITS (internal transcribed spacer) sequencing analysis indicated that host genotype had a higher explanatory power for bacterial diversity (R^2^ = 17.2%, p < 0.001) than for fungal diversity and that the domestication process had significantly reshaped fungal community structure, transforming it from a fungal-dominant type to a bacterial-dominant type. Metagenomic data showed that wild wheat has a wider variety of functional genes (e.g., those related to carbon fixation and nitrogen conversion), whereas cultivated varieties use root secretions to recruit microorganisms carrying genes related to nitrogen fixation and inorganic phosphorus solubilization. Microbiome-wide association study (MWAS) linked exudates to keystone taxa and phenotypes, as confirmed by Microbacterium mitrae inoculation, supporting the role of the microbiome in nutrient cycling and crop growth (Liu et al., 2024).
The disease pyramid network has another obvious characteristic, which is its dynamic nature. Disease expression is an emergent property of the host (genotype, phenology, and physiological state), pathogens (inoculum density and virulence spectrum), non-pathogenic microorganisms, and the environment (soil moisture, temperature, radiation, wind, and cultural practices). The interplay among these factors determines the spatiotemporal distribution and ultimate yield loss of an epidemic (Savary et al., 2017). Pathogens require living hosts to complete their life cycles, and hosts require specific environmental cues for optimal defense; environmental extremes can either foster or suppress pathogen multiplication. For instance, air-borne epidemics are strongly governed by rainfall, temperature, humidity, illumination, wind speed, and solar radiation (De Wolf and Isard, 2007; Monteil et al., 2014; Meyer et al., 2017). Mechanical wounding of plants increases the number of pathogen entry points, underscoring the importance of micro-climate management to restrict secondary spread. Maintenance of optimal humidity and temperature is essential in this regard. Warm, humid conditions accelerate sporulation of some foliar fungi, whereas drought can restrict pathogen development and reduce disease pressure.
Recent breakthroughs in microbiomics and metagenomics technologies have provided new perspectives on the complex interactions between plants and their symbiotic microbial communities within the disease pyramid network (Smith et al., 2025; Wani et al., 2025; Yousuf et al., 2025). In-depth analysis of the intrinsic mechanisms by which these microorganisms affect plant health is valuable for enhancing the nutrient utilization efficiency of crops, strengthening their resistance to adverse conditions, and optimizing their responses to disease.
Microorganisms in the environment play a crucial role in the nutrient supply of plant ecosystems. Through processes such as nitrogen fixation, phosphorus solubilization, and activation of micronutrients, microorganisms can not only improve soil nutrient conditions but also enhance the absorption of nutrients by plants (Figure 2). For instance, certain rhizosphere bacteria can establish a symbiotic relationship with plant roots, forming nitrogen-fixing root nodules that convert atmospheric nitrogen into forms that plants can use (van Velzen et al., 2019). This symbiotic relationship ensures a sustainable supply of nitrogen, especially under nitrogen-deficient conditions. In addition, some soil bacteria and fungi have the ability to solubilize inorganic phosphates, significantly enhancing the bioavailability of phosphorus to plants (Li et al., 2024). Microbes can decompose complex organic matter to release key nutrients needed for plant growth, thereby promoting plant nutrient absorption, growth, and development. Research also shows that microorganisms play a significant role in the bioenrichment and activation of trace elements, which can further enhance plant nutrient levels (Li et al., 2025). Through these mechanisms of action, both aerial and soil-associated microbes play a significant role in enhancing nutrient availability and promoting overall plant nutrition.
Importantly, this nutrient–microbiome interaction is microbes enhance plant nutrient uptake, and the plant’s nutrient status actively shapes microbial recruitment, colonization, and community assembly. Recent mechanistic studies have shown that plant-derived nutrients and signaling molecules, such as sugars, amino acids, and secondary metabolites, serve as key determinants of microbiome composition and function (Pang et al., 2021; Uribe-Acosta et al., 2025). A landmark study by Tsai et al. (2025) further characterized the spatiotemporal precision of this process, demonstrating that the root endodermis functions as a selective barrier to control the leakage of photoassimilates (e.g., glutamine) into the rhizosphere at specific sites along lateral roots and pre-Casparian strip junctions (Tsai et al., 2025). In addition, phosphorus-deficient plants can exude specific organic acids that selectively enrich phosphate-solubilizing microbes, thereby creating a feedback loop that fine-tunes the rhizosphere microbiome to meet plant nutritional demands (Abbasi, 2023). This reciprocal interplay underscores the need to consider nutrient availability not merely as an outcome of microbial activity, but as a central driver of microbiome assembly within the pyramid network.
Plants actively reshape their microbial structure and composition through root exudates, which function as signals in the rhizosphere, capable of activating plant defense signaling pathways and enhancing stress tolerance (Wang and Song, 2022) (Figure 2). Under stress conditions, plants can communicate with rhizosphere-associated microbes to solicit assistance—a process commonly referred to as the “plant-to-microbe distress signal” or the “cry for help.” Throughout evolution, plants have developed sophisticated mechanisms to perceive and respond to diverse biotic and abiotic stresses. Plant genotypes shape the assembly of rhizosphere microbial communities, but the specificity and underlying mechanisms are not fully understood. Although the “cry for help” concept has attracted considerable attention, recent evidence emphasizes the complexity of plant–microbe interactions. For example, resistant tomato varieties harbor a more robust rhizosphere microbiota, selectively recruiting beneficial bacteria like Sphingomonas sp. Cra20 and Pseudomonas malodora KT2440 to enhance resistance in susceptible plants (Yin et al., 2022). In addition, combined inoculation with Trichoderma harzianum Tg-M33 and Bacillus velezensis SQR9 increased the control effect on tomato Fusarium wilt (Li et al., 2025).
Although hostomics, pathomics, microbiomics, and enviromics are the four core analytical pillars of the pyramid network, a broader suite of functional genomics and systems biology tools—including transcriptomics, metabolomics, single-cell omics, and spatial omics—provides the mechanistic depth needed to dissect specific HPMEIs at molecular and cellular resolution. Advances in gene sequencing and omics technologies have improved the analysis of HPMEIs. Omics approaches can comprehensively analyze the genetic composition, gene expression patterns, metabolic characteristics, and epigenetic modifications of microorganisms under different environmental conditions (Mu et al., 2024; Dai et al., 2025; Xu et al., 2025). Take RNA sequencing (RNA-seq) as an example. This technology can simultaneously monitor changes in gene expression of the host, pathogen, and associated microbes during the infection process. This dynamic monitoring is particularly important for studying fungal diseases such as FHB: changes in environmental stress factors such as soil nitrogen availability cause corresponding changes in the fungal gene expression profile. Double RNA-seq technology can simultaneously map the gene expression profiles of both the host and pathogen, thereby revealing mechanistic details of the host–pathogen interaction (Westermann et al., 2012). These studies highlight the key regulatory role of plant species in the construction and function of rhizosphere microbial communities, providing a new perspective for understanding the dynamics of plant–microbial interactions. Through gene editing technology, knockout of the MLO gene endowed barley with broad-spectrum resistance to Blumeria graminis f. spp. Hordei (Acevedo-Garcia et al., 2017). With the development of single-cell sequencing technology (Liang et al., 2023; Tang et al., 2023), all components of the pyramid network can now be analyzed through single-cell omics at the spatiotemporal scale, thereby revealing the complete characteristics of HPMEIs at the cellular level (Figure 2).
In addition to single-cell transcriptomics, emerging spatiotemporal omics technologies are revolutionizing our ability to contextualize molecular events within the architecture of plant tissues. Spatial transcriptomics techniques such as multiplexed error-robust fluorescence in situ hybridization (MERFISH) can map the expression of hundreds of genes simultaneously, preserving their spatial relationships during infection (Saarenpää et al., 2024). Imaging mass spectrometry enables the distribution of metabolites and small molecules to be visualized at cellular or even subcellular levels, linking metabolic shifts to specific microenvironments within the plant. Furthermore, single-cell multi-omics platforms, such as those that combine single-cell assay for transposase-accessible chromatin using sequencing (scATAC-seq) with transcriptomics (single-cell RNA-seq [scRNA-seq]), enable the concurrent profiling of chromatin accessibility and gene expression in the same cell (Schäfer et al., 2024). This provides unprecedented insights into the regulatory mechanisms that govern cell-type-specific responses to pathogens and environmental cues within the pyramid network.
The rise of microbiome research and metagenomic technologies has fundamentally changed our understanding of the complexity and functional potential of air and soil microbial communities (Nayfach et al., 2021; Deng et al., 2024). These microorganisms, especially rhizosphere microorganisms, promote plant growth and development by providing nutrients, enhancing plant tolerance to abiotic stress, and inhibiting diseases (Mu et al., 2024; Dai et al., 2025; Xu et al., 2025). The latest metagenomic technology has revealed the complex mechanisms of interaction between microorganisms and plants, providing a new perspective for analyzing their beneficial effects (Newman et al., 2024). For instance, metatranscriptomic studies have shown that plant species significantly influence the composition and function of rhizosphere microbial communities (Turner et al., 2013). A deep understanding of the dynamic HPME interplay will open new avenues for the development of sustainable agricultural practices and the optimization of plant disease control strategies.
Although we advocate for treating microbes as a distinct core component within the extended disease pyramid, it is important to acknowledge a conceptual nuance. In natural ecosystems, microbial communities engage in complex competitive and cooperative interactions independently of plants. For instance, antibiotic production, resource competition, and niche exclusion among soil microbes can suppress pathogen populations, even in the absence of a host plant (Sharma et al., 2023; Čaušević et al., 2024). This raises an important if microbial suppression occurs in bulk soil without direct plant involvement, should the microbiome be considered an intrinsic component of the disease pyramid or as a part of the biotic environment that modulates pathogen access to the host?
The proposed pyramid network distinguishes microbes as a separate component (M) primarily because of their direct and dynamic interactions with both hosts and pathogens that differ qualitatively from the modulating effects of abiotic factors (E). However, we recognize that the boundary between “microbes as active players” and “microbes as environmental filters” can be porous. Future models may benefit from distinguishing between plant-associated microbes (rhizosphere, endosphere, and phyllosphere) that are directly shaped by the host and bulk soil microbes that function more as external biotic filters. This distinction does not diminish the importance of microbial communities in disease outcomes but rather encourages more precise modeling of their roles in different ecological contexts.
In recent years, the development of enviromics and microbiomics (especially rhizosphere microbiomics) has provided a new perspective for studying interactions among the four components of the disease pyramid (Figure 3A). This section focuses on the mechanisms by which environmental factors influence disease occurrence. By integrating the pyramid network, we systematically analyze typical cases and successful practices.Figure 3The interplay between environmental factors, the microbiome, and host plant immunity determines disease outcomes.(A) A conceptual framework illustrating the core interacting enviromics (abiotic factors), microbiomics (non-pathogenic microbial communities, including beneficial and harmful microorganisms), hostomics (host plant genetics and defense capabilities), and pathomics (pathogen virulence factors). The outcome of plant health or disease is governed by complex interactions among these four spheres.(B) The disease outcome (resistance vs. susceptibility) is influenced by interactions between the host genotype (e.g., overexpression of the resistance gene Sr6) and environmental conditions (e.g., temperature and plant growth stage). This highlights the fact that genetic resistance can be temperature sensitive or stage dependent.(C) Application of microbiome data for disease forecasting. Through field sampling and metagenomic sequencing of the microbial community, it is possible to predict disease outbreaks (e.g., bacterial wilt) weeks before they occur, enabling proactive crop protection strategies.
Susceptibility or resistance of the host to pathogens is determined by both the genotype and the environmental background. For example, plants may resist fungal infections in cooler climates but lose their resistance in warm and humid environments (Cheng et al., 2019). Environmental factors can also directly regulate the growth, reproduction, and virulence of pathogens, thereby influencing the overall severity of disease. Conditions such as temperature, humidity, rainfall, and nutrient supply can change the abundance and infection potential of pathogens. In the case of wheat scab, the environmental conditions from the flowering period to the grain development stage have a significant effect on the degree of disease occurrence (Boenisch and Schäfer, 2011; Imboden et al., 2018; Fulcher et al., 2019) (Figure 3B).
The prediction of plant diseases on the basis of interactions between microorganisms and the environment is a new frontier in precision agriculture. Recent studies have shown that subtle changes in microbial community structure of the rhizosphere or phyllosphere often precede a significant increase in the abundance of pathogenic bacteria, serving as an earlier and more sensitive warning of disease outbreaks. Environmental factors enhance or weaken the predictive potential of these microbial communities by regulating their composition and function. Under the theoretical framework of the “pyramid network,” the microbiome and environmental variables are regarded as the two key pillars for precise early prediction of plant diseases (Leveau, 2024) (Figure 3C).
Environmental factors influence the immune mechanisms of plants, affecting the balance between resistance and susceptibility to pathogens (Table 1). Enviromics, as a discipline that studies environmental factors, plays a crucial role in the pyramid network and indirectly influences plant defense mechanisms (Figure 4A and 4B).Table 1Environmental factors that affect hosts, pathogens, and non-pathogenic microbes.Environmental factorInfluence on hostsInfluence on pathogensInfluence on non-pathogenic microbesRepresentative referencesAtmosphereLightinteraction between hosts and pathogenssurvival of pathogensgrowth, metabolism, and interaction with hostsPierik and Ballaré (2021); Griebel and Zeier (2008)Temperaturegrowth and immunity of plantsreproduction and spread of pathogensdiversity and stability of community structureClarke et al. (2004); Larkindale et al. (2005)Air humidityphysiological states and biological processespathogenicity and virulencegrowth, spread, and community structurePanchal et al. (2016); Zhou et al. (2004)Rainfallgrowth of roots and leavespromoting spread and diffusion of pathogenspromoting structure and diversityKim et al. (2019)Airflowgrowth of plants by regulating temperature and humiditydistribution and spread of pathogensspread, survival, and reproductionElbers et al. (2022)CO2plant immunitystructure and metabolism of microbial communitiesgrowth strategy, physiological metabolism, community structureZhou et al. (2017); Zhang et al. (2015)Radiationinducing plants to produce defense metabolitesinhibiting pathogensinhibition of metabolism and growth, leading to a decline in diversityKunz et al. (2006)Carposphere, caulosphere, and phyllosphereLight intensityresistance of plants to different pathogenspathogen infectionecological functions and community dynamics of non-pathogensBallaré (2014); Cheng et al. (2021)Temperature fluctuationPTI and ETIvirulence and colonizationecological functions and survival strategiesZhu et al. (2010)Canopy humidityimmune response of plantsenhancing pathogenicity of pathogensaltering composition of microbial communitiesFall et al. (2015)Leaf surface moistureexcessive water suppresses plant immune systemgermination of fungal spores and spread of bacteriacolonization, community structure, and functional activityAung et al. (2018)Canopy airflowadjusting temperature and humidity of the canopyspread of pathogensdiffusion, propagation, attachment, and colonizationDouma and Noordhoek (2025)Human activitieschanging growth environment and increasing probability of disease occurrencedistribution, composition, and function of microorganismscomposition and function of microbial communitiesSantini et al. (2018)RhizosphereSoil temperaturegrowth of root system and absorption of nutrientsreproduction and activity of pathogensgrowth rate and activity of non-pathogensGonzález-Garcia et al. (2023); Metze et al. (2024)Water contentplant immunity suppressed by flooding or droughtactivity of pathogenssurvival, metabolism, and niche competition of microorganismsIrulappan et al. (2022); Mooney et al. (2024)Soil pHaltering disease resistance of plantssurvival and reproduction of pathogenscomposition, abundance, interaction, and ecological functionLi et al. (2023)Nutrientspromoting growth and enhancing immunitypromoting pathogen growthregulating community structure, function, and metabolic activitiesCao et al. (2024); Ding et al. (2021)Salt contentenhancing or weakening disease resistancepathogen growthdiversity, symbiotic networks, and stabilityMahmud et al. (2024)Soil microorganismspromoting growth through symbiosisrestricting pathogen growthsoil ecosystem through competition, symbiosis, and antagonismZhang et al. (2024)Figure 4Schematic representation of plant immune responses and their modulation by environmental factors.(A) The diagram illustrates the key components of plant immunity, including pattern-triggered immunity (PTI) and effector-triggered immunity (ETI), which are activated upon recognition of pathogen-associated molecular patterns (PAMPs) and effectors, respectively. Signaling cascades such as the mitogen-activated protein kinase cascade, calcium (Ca^2+^) fluxes, and reactive oxygen species production are central to immune activation, leading to defense gene expression and hormone signaling (e.g., salicylic acid [SA], jasmonic acid [JA], and ethylene).(B) Environmental factors, such as temperature, humidity, light, and soil conditions, significantly influence the outcome of plant–pathogen–microbe interactions, as indicated by the disease-specific environmental optima listed for airborne and soilborne pathogens.(C) Airborne and soilborne pathogens thrive under distinct optimal environmental conditions for infection.(D) Application of multi-omics for understanding host–pathogen–microbe interactions.
Climatic factors profoundly affect the immune response of plants and the pathogenicity of pathogenic microorganisms. High temperatures can inhibit plant immunity by disrupting the activity of nucleotide-binding leucine-rich repeat (NLR) proteins and suppressing the biosynthesis of salicylic acid (SA) (Clarke et al., 2004; Larkindale et al., 2005; Zhu et al., 2010) (Table 1). For example, EDS1 and PAD4, two key regulators of both basal and R gene-mediated disease resistance, exhibit higher steady-state expression at 22°C than at 28°C (Yang and Hua, 2004). A temperature change from 20°C to 35°C not only accelerates the replication and proliferation of viruses within the plant but also increases the reproduction rate and activity range of virus-carrying insects like aphids, whiteflies, and thrips, thereby enhancing the chances of virus transmission (Tsai et al., 2022). Under heat stress, rice prioritizes thermotolerance through the transcription factor OsHSFA4d, but this trade-off results in diminished disease resistance (Fang et al., 2025). Compared with heat stress, cold stress can have different influences on host–pathogen interactions and plant immune responses. During cold stress, ZmDREB1A enhances cold tolerance in maize by co-regulating raffinose synthesis and SA metabolism, but this occurs at the cost of reduced immune responsiveness, revealing a molecular trade-off between cold resistance and immunity (Han et al., 2020; Zhang et al., 2025). By contrast, low temperatures can enhance pathogen resistance, as the transcription factor ICE1 interacts with key components of the SA signaling pathway (such as NPR1 and TGA3) to co-activate the expression of immune-related genes like PR1, thereby strengthening plant defenses (Li et al., 2024).
High humidity can attenuate stomatal immunity by upregulating hormonal signaling in guard cells of legume plants (Zhou et al., 2004; Fall et al., 2015; Panchal et al., 2016) (Table 1). In the rice–M. oryzae interaction, elevated humidity reduces accumulation of the ethylene precursor 1-aminocyclopropane-1-carboxylic acid and expression of ethylene-responsive genes, thereby facilitating pathogen infection (Qiu et al., 2022). Drought stress exacerbates fungal colonization and endophytic invasion while suppressing defense responses, thus increasing the incidence of dry root rot in chickpea (Irulappan et al., 2022). Drought stress weakens the immune system of Korean pine (Pinus koraiensis), facilitating the growth of Cenangium ferruginosum, which transitions from an endophyte to a pathogenic agent that causes shoot blight (Ryu et al., 2018). Flooding and elevated atmospheric CO2 concentrations can also impair plant immunity by altering hormonal homeostasis (Zhang et al., 2015; Zhou et al., 2017; Martínez-Ferri et al., 2019; Mooney et al., 2024) (Table 1).
Light also plays a pivotal role in regulating plant immunity and disease outcomes (Griebel and Zeier, 2008; Ballaré, 2014; Cheng et al., 2021; Pierik and Ballaré, 2021). The photoreceptor phytochrome B (PhyB) has been shown to enhance both rice yield and resistance to blast disease by repressing the transcription factors GT1 and PIL15 and their downstream signaling cascades (Li et al., 2024). By contrast, the blue-light receptor phototropin 1 (PHOT1) negatively regulates INF1-triggered immune responses, thereby promoting Phytophthora infection in potato (Naqvi et al., 2022; Lin et al., 2024), suggesting that blue-light signaling may have contrasting effects in different host–pathogen systems. In addition, rain, airflow, radiation, human activities, and soil properties such as pH, nutrients, salt content, and soil microorganisms can all affect plant immune responses to pathogens (Kunz et al., 2006; Aung et al., 2018; Bai et al., 2018; Santini et al., 2018; Kim et al., 2019; Ding et al., 2021; Elbers et al., 2022; González-García et al., 2023; Li et al., 2023; Cao et al., 2024; Metze et al., 2024; Zhang et al., 2024; Douma and Noordhoek, 2025) (Table 1).
Enviromics provides important insights into how environmental factors affect pathogen adaptability. Different pathogens have specific environmental requirements for growth, reproduction, and virulence expression. F. graminearum exhibits optimal perithecium production at 21.7°C and requires a relative humidity (RH) ≥ 75.5% (water activity [aw] > 0.75), with maturation and ascospore production occurring only at 20°C–25°C and RH ≥ 85% (aw > 0.82). Lower temperatures (<15°C) or reduced humidity (RH < 75%) significantly suppress perithecium development, and no perithecia form at 35°C–40°C or RH ≤ 62.5% (aw ≤ 0.62) (Manstretta and Rossi, 2016) (Figure 4C). At the same time, host resistance is not a fixed characteristic but is significantly influenced by environmental factors.
The role of the microbiome in HPMEIs is strongly regulated by the environment. This regulation occurs through two primary direct effects of abiotic conditions on microbial physiology and community assembly and indirect modulation via host-mediated changes in response to environmental stress. For instance, soil pH is a master regulator that can fundamentally shift microbiome composition and function (Lee et al., 2025). In addition to modulating community composition, enviromics helps to reveal how environmental factors mechanistically alter microbiome functions critical for disease suppression. A seminal example involves soil acidification, which not only shifts bacterial community structure but also specifically downregulates sulfur metabolism genes in the microbiome (Li et al., 2023). This functional impairment was correlated with a reduced capacity to inhibit Fusarium spp., linking a specific environmental driver (pH) to a defined metabolic pathway (sulfur compound synthesis) and a disease outcome. Similarly, drought stress has been shown to enrich specific bacterial taxa that possess genes for exopolysaccharide production and osmotic stress tolerance, which can indirectly influence pathogen competition for water and niches in the rhizosphere (Li et al., 2025; Xiang et al., 2025). These studies move beyond correlation to establish causal chains within the pyramid network, where E shapes the functional gene repertoire of M, which in turn modulates the outcome for P and H.
Building upon the four core omics disciplines introduced earlier, this section focuses on how the integration of functional genomics approaches (e.g., the combination of transcriptomics, proteomics, and metabolomics) can be used to dissect the mechanistic basis of HPMEIs. Advances in omics technology have greatly enhanced our ability to characterize the complex molecular and cellular mechanisms that underlie HPMEIs, enabling us to explore in high resolution the genetic, epigenetic, and biochemical processes involved in the signaling pathways that control plant immune responses, nutrient exchange, and pathogen-triggered responses. These integrated research methods help identify key biomarkers, signal transduction networks, and immune response mechanisms, providing a comprehensive theoretical framework for understanding plant immunity at the systems level (Figure 4D).
Numerous disease-resistance genes in plants have been identified by genomic approaches (Tong et al., 2024). Transcriptomics and proteomics analyses have revealed a significant correlation between gene expression dynamics and protein interaction networks in plant defense mechanisms (Wang et al., 2020; Chen et al., 2023). The integration of metabolomics and epigenomics has further expanded research dimensions. By identifying characteristic metabolite profiles and heritable epigenetic modifications, it has clarified how these factors regulate the interaction processes between plants and pathogens (Kumar et al., 2020; Shilpa et al., 2022). The rise of single-cell transcriptome and spatial transcriptome technologies has provided new tools for analyzing the defense responses of specific cell types and accurately mapping the spatial distribution of host and pathogen activities (Cao et al., 2023; Liang et al., 2023; Tang et al., 2023; Zhu et al., 2023; Wang et al., 2025). By integrating these multi-dimensional data, researchers can reconstruct the complex regulatory networks that underlie resistance, facilitating the development of new crop varieties with broad-spectrum and long-lasting disease resistance (Zhu et al., 2023).
The power of omics lies in its integration. Systems biology approaches are now revealing the HPME interactome. For instance, integrated transcriptomics and metabolomics of the wheat F. graminearum pathosystem under different nitrogen regimes revealed that the wheat auxin receptor TIR1 negatively regulates defense against F. graminearum (Su et al., 2021). Spatial multi-omics is another frontier. Techniques like spatial transcriptomics (e.g., MERFISH) coupled with imaging mass spectrometry can simultaneously map host immune gene expression, pathogen biomass, and defensive metabolites within a single infected leaf sample, providing unprecedented spatial resolution for the study of HPMEIs (Saarenpää et al., 2024). However, key challenges the high dimensionality and heterogeneity of multi-omics data require advanced computational frameworks, such as machine learning–based data fusion and network inference algorithms, to derive biologically meaningful insights (Baião et al., 2025). Future efforts must focus on developing standardized pipelines for multi-omics integration specific to plant disease systems.
The disease triangle, a classic model in plant pathology, offers simplicity but has notable limitations in explaining plant disease mechanisms. It fails to fully capture the complex interactions between environmental factors and the other three components (H, P, and M). This theoretical limitation requires innovation in the plant disease paradigm, prompting researchers to systematically restructure the traditional concept of disease. Our increasing understanding of the complex interplay among the pyramid components and the crucial role of associated microbiota necessitates a more comprehensive and dynamic approach. Enviromics, which focuses on the systematic quantification of environmental cues and their interactions with plants, pathogens, and non-pathogenic microbiota, has emerged as a powerful framework that can revolutionize plant disease management.
The classic framework for managing plant diseases initially consisted of four core isolation (exclusion) to prevent the introduction of pathogens, eradication to eliminate existing infections, protection through the establishment of physical or chemical barriers, and immunization to enhance the host’s inherent resistance. This system was subsequently expanded to incorporate two key avoidance through adjustment of planting times and locations to avoid diseases and treatment to cure infected plants. These six strategies together form the complete strategic foundation of modern plant disease management. Modern management approaches were developed (Baker, 1968) around three main (1) reducing the initial amount of inoculum or preventing disease development, (2) managing the pathogen–host–environment interaction, and (3) preventing the spread or survival of the pathogen. These objectives are operationalized through six strategic measures (Figure 5) that integrate and extend the earlier (1) avoidance, planting crops in areas with lower pathogen presence or at specific times (e.g., site selection and adjustment of sowing dates), (2) exclusion, preventing introduction of the pathogen (e.g., using certified seeds, implementing quarantine, and controlling vectors), (3) eradication, eliminating established pathogens (e.g., removal of infected plants, sanitation practices, crop rotation, and thermal/chemical/biological soil treatment), (4) protection, safeguarding infected or vulnerable areas (e.g., chemical barriers and environmental or nutritional management), (5) host resistance through deployment of genetic resistance or induced resistance, and (6) treatment to cure valuable plants by physical or chemical means after infection.Figure 5Six major functional modules for plant disease prevention and control, with enviromics-driven disease management strategies as the core.The central circle of this figure illustrates the interactions among host plants, pathogens, non-pathogenic microbes, and the environment. Enviromics-driven disease management strategies are further expanded to include host resistance, environmental impacts, microbial balance (e.g., crop rotation of different plant species), and disease control methods. These enviromics-driven strategies are depicted as radiating outward to outline six approaches for disease control. (1) Avoidance: by integrating the four aspects of hostomics, pathomics, microbiomics, and enviromics, this approach aims to separate host susceptibility from the environmental conditions favorable to pathogens. (2) Exclusion: thermal seed treatment (treatment in warm water at 45°C for 2–3 h) reduces the infection rate of the target spp. (specific microbial genus) by up to 98%, and this method relies on enviromics thresholds to prevent pathogen introduction. (3) Eradication: through rotation of different plant species, the soil environment is modified to make it less favorable for soilborne pathogens, thereby reducing their presence. (4) Protection: enviromics is used to create conditions unfavorable to pathogens and in combination with two models—the De Wolf model (accuracy: 70%–84%) and DONcast (accuracy: 80%–85%)—temperature, humidity, and rainfall during the anthesis period (7–10 days before and after anthesis) are integrated to predict the risk of Fusarium head blight, enabling timely fungicide application. (5) Host a thorough understanding of gene–environment interactions, combined with multi-gene pyramiding and phenotypic evaluation across multiple environments, facilitates the development of genotypes with durable resistance under diverse environmental conditions. (6) Therapy: thermal therapy and hormone therapy (SA and PAA) significantly reduce the titer of CLas, thereby alleviating the severity of citrus HLB. F.g, Fusarium graminearum; SA, salicylic acid; PAA, phenylacetic acid; HLB, Huanglongbing; and PSS, percentage of scabbed spikelets.
Disease avoidance strategies seek to prevent overlap between host susceptibility and conditions favorable for pathogen infection. Enviromics approaches now enable precise identification of high-risk areas and vulnerable time windows. Consider FHB in by integrating satellite-derived soil moisture data with historical rainfall patterns, researchers can pinpoint geographic areas where the wheat flowering period (when plants are most vulnerable) coincides with conditions ideal for F. graminearum infection (temperatures of 20°C–25°C and humidity >85%) (Figure 5). Adjustment of wheat sowing date affects disease incidence in a region-specific in a study of Argentine bread wheat (Azul, Argentina, 2018/2019 growing season), delayed sowing increased FHB severity (2.2-fold after F. graminearum inoculation) and mycotoxin contamination (3.2-fold for deoxynivalenol, 8-fold for nivalenol, and 5-fold for zearalenone), whereas early sowing did not increase FHB risk (Arata et al., 2022). This pattern is not universal, as early sowing may reduce FHB infection in regions with favorable rainfall patterns, and delayed sowing can increase the incidence of leaf spot or wheat blast (e.g., Indo-Gangetic Plain), precluding generalizations across environments. In addition, microclimate modifications, such as adjustment of planting density in rice to enhance canopy ventilation and reduce humidity, can optimize the microenvironment for grain filling, particularly in dense populations (Yang et al., 2021). Erect-panicle rice varieties (e.g., R499) maintain lower humidity and higher light intensity in the lower canopy at high densities, thereby producing higher yields compared with curved-panicle types.
Modern exclusion strategies now use precise environmental thresholds to optimize pathogen prevention. By analyzing key variables like temperature, vapor pressure deficit, and wind speed, researchers have developed quarantine protocols and seed–soil disinfestation schedules that dynamically adapt to actual conditions (Novick et al., 2024). This data-driven approach has proven more effective than traditional calendar-based methods at reducing initial inoculum pressure. Temperature critically influences the biology of invasive pests like the Asian citrus psyllid (ACP), the vector of Huanglongbing (HLB) disease. Although ACP can complete development at high temperatures (28°C–43°C), prolonged exposure to extreme heat (43°C for 6+ h) reduces survival and slows development (Antolínez et al., 2022), suggesting that climate-driven temperature increases may limit its invasive potential in regions with intensifying heatwaves. Seed and soil disinfestation methods can also be optimized using enviromics data. Thermal seed treatments—in particular, warm water at 45°C for 2–3 h—reduced Microdochium spp. infections in wheat by up to 98% (from 52% to 0%–2% infection) while minimizing phytotoxic effects (Bänziger et al., 2022). Steam treatments at 68°C–70°C were also effective but required precise parameters to avoid seed damage, highlighting warm water as the most consistent method for disease control under varying field conditions (Figure 5).
Advances in environmental profiling are refining eradication approaches. By analyzing how environmental conditions affect both pathogen persistence and beneficial soil communities, agronomists can now design crop rotation systems that more effectively suppress disease recurrence. In FHB-prone regions, crop rotation can suppress F. graminearum: rotations without wheat (e.g., flax–alfalfa) reduce its presence compared with those that include wheat (e.g., wheat monocropping) (Champeil et al., 2004) (Figure 5). However, this effect varies among Fusarium species and is not dependent on soil texture or pH (Marburger et al., 2015). Soil treatments, such as solarization (Abed Gatea Al-Shammary et al., 2020), are effective against F. oxysporum, with soil moisture significantly enhancing its effects by improving heat conduction.
Contemporary plant protection strategies increasingly incorporate environmental monitoring to disrupt pathogen development. Field-specific microclimate data now inform precision spray systems, which automatically calibrate fungicide application parameters—including dosage, delivery method, and treatment timing—in response to real-time environmental fluctuations (Ratajkiewicz et al., 2016). Targeted sprays can be triggered by enviromics models to create unfavorable conditions for pathogens. Weather-based models, such as De Wolf models (accuracy: 70%–84%) and DONcast (accuracy: 80%–85%), integrate temperature, humidity, and rainfall during the anthesis period (7–10 days pre- and post-flowering) to predict FHB risk, enabling timely fungicide applications (Matengu et al., 2024). By optimizing application timing—given that efficacy is substantially reduced when fungicides are applied either before or after anthesis—these models can reduce the incidence of wheat FHB while minimizing unnecessary chemical inputs under low-risk conditions (Figure 5).
The development of disease-resistant crop varieties depends critically on analysis of GEIs. By systematically testing genetic lines across diverse growing conditions, plant breeders can pinpoint resilient gene combinations that maintain stable resistance despite environmental fluctuations. Combining this information with advanced breeding techniques facilitates the identification and incorporation of resistance genes into breeding programs, as well as the pyramiding of multiple genes or alleles for resistance to different pathogens/races into one genetic background (Figure 5).
Breeding personnel determine the priority order of screening locations by mapping out an environmental profile that is prone to causing diseases. Take the citrus disease-resistance gene CsNPR1 as an example. Its expression is dynamically regulated by environmental cues, including hormonal signals and abiotic stresses. SA and salt stress induce its expression, whereas gibberellin and drought stress suppress it (Wu et al., 2021). It is precisely these characteristics that enable CsNPR1 to play a prominent role in integrating stress responses and pathogen defense pathways (Figure 5).
Enviromics data enable specialists to more precisely refine disease treatment protocols. For instance, the efficacy of heat therapy against HLB varies significantly depending on specific environmental conditions, highlighting the need for tailored implementation strategies. In previous related reports, heat therapy was shown to reduce the titer of Candidatus Liberibacter asiaticus (CLas) (Hoffman et al., 2013; Doud et al., 2017), but the specific parameters of this technique, such as treatment temperature and duration, require further optimization and improvement. Application of SA and phenylacetic acid to the soil is also an environmentally friendly method for early management of HLB (Widyawan et al., 2023). These two hormones can activate defense-related pathways within citrus seedlings, significantly reducing the titer of CLas and thereby reducing disease severity (Figure 5).
The role of enviromics in plant disease management has fundamentally transformed in recent years. Originally considered a supplementary approach, it has now emerged as a primary research methodology capable of sophisticated quantitative analyses that support decision-making in disease control strategies. Through this quantitative perspective, researchers can more accurately assess the efficacy of prevention and treatment strategies, more precisely define their implementation scope, and better take into account ecological sustainability. Enviromics helps optimize multiple aspects of disease prevention and treatment, including timing, target areas, and environmental impacts, but its full potential will be realized only when integrated into holistic approaches that combine multiple tactics and are supported by technological advances.
Translating the pyramid network into breeding programs requires moving beyond broad concepts to actionable strategies. A key approach is enviromics-informed germplasm evaluation. Through systematic multi-environment trials in which genetic lines/varieties are evaluated across diverse conditions, breeders can pinpoint stable resistance traits mediated by particular genetic architectures that perform consistently under environmental fluctuations. Combining this information with advanced breeding techniques facilitates the identification and incorporation of resistance genes into breeding programs, as well as the pyramiding of multiple resistance genes/alleles into one genetic background (Figure 6).Figure 6Enviromics-driven disease management breeding for disease resistance.The diagram illustrates how achievements in enviromics can be translated. It covers the process from determination of soil microorganisms to environmental prediction, then to plant-level resistance through identification of multiple pathogens, adaptation of disease-resistant varieties via multiple breeding methods, and finally translation of achievements into improved disease-resistant varieties by use of the H–P–M–E pyramid. H–P–M–E, host–pathogen–microbe–environment; MAS, marker-assisted selection; and DH, doubled haploid.
A prime example of translating HPME management into practice is the use of genomics–enviromics predictions to select wheat varieties with stable FHB resistance. FHB severity is highly dependent on GEIs, particularly during the flowering period when temperature and humidity critically influence F. graminearum infection and mycotoxin production. In a pioneering study, a prediction model for a diverse wheat panel was developed by integrating high-density SNP markers (genomics) with high-resolution environmental covariates (enviromics), including site-specific rainfall, temperature, and humidity data during anthesis. This genomics–enviromics model significantly outperformed traditional genomic prediction models in accurately ranking varieties for FHB resistance across geographically and climatically distinct testing sites. The model successfully identified lines carrying known FHB resistance QuantitativeTrait Loci (QTL) (e.g., Fhb1) that maintained their resistance in both high- and low-disease-pressure environments, while also revealing novel genetic backgrounds with stable performance (Xu et al., 2022; Brault et al., 2025). This approach enables breeders to pre-select promising candidates for target environments, thereby optimizing resource allocation in multi-location trials and accelerating the development of locally adapted varieties with durable FHB resistance.
HPMEIs form the scientific foundation for modern gene pyramiding approaches in plant breeding. These strategies systematically combine multiple resistance genes into elite varieties to achieve comprehensive disease resistance. Within contemporary disease management systems, enviromics analysis has transitioned from a supplementary technique to a core, quantitative methodology driving resistance development. Various disease management strategies can achieve clear, precise, and timely effectiveness, as well as clear, targeted spatial positioning through the meticulous selection of genotypes. When this approach is combined with systematic evaluations in multiple environments, researchers can develop genotypes with broad-spectrum resistance. These disease-resistance genes possess two advantages. They not only reduce the risk of disease outbreaks but also decrease the reliance on chemical pesticides in agricultural production. In wheat, for example, disease-resistance genes tested under different environmental conditions differed significantly in their effects on wheat FHB resistance and yield performance (Mengesha et al., 2022). These findings suggest that effective control of wheat FHB can be achieved by screening suitable genotypes for local environments and combining them with scientific cultivation and management measures.
Microbiome-aware selection is another emerging approach. Future crop improvement programs will increasingly integrate environmental and microbial datasets to enable the identification of varieties with enhanced disease resistance and improved environmental stress tolerance, as illustrated by recent studies on wheat–rhizosphere microbiome interactions under drought conditions (Hafeez et al., 2023). By incorporating information on microbial communities into breeding research, scientists can also identify host genetic loci associated with the recruitment of beneficial microbes (e.g., disease-suppressive rhizosphere communities). Incorporating these loci as indirect selection targets could enable the breeding of plants that are “programmed” to cultivate their own defensive microbiota (Zhang et al., 2023). However, major hurdles the plasticity of microbes in different soils and the often context-dependent benefit of microbial taxa pose challenges for the development of universally effective microbiome-informed breeding strategies. This highlights the need for large-scale, multi-environment phenotyping of the holobiont (host + microbes).
Modern breeding techniques (such as doubled haploid breeding, speed breeding, genome editing, etc.) are highly practical tools for disease control (Paspureddy et al., 2025) (Figure 6). In actual breeding projects, breeders can use doubled haploid technology to accelerate the production of homozygous lines, more efficiently integrating disease-resistance traits into new varieties. Speed-breeding techniques, which shorten plant life cycles through environmental and genetic manipulation, significantly reduce the time needed to develop new varieties. Genome editing enables precise modifications to plant genomes, helping to develop genotypes with enhanced resistance to specific pathogens. These tools can also be used to breed varieties with broad-spectrum resistance to multiple diseases or pathogen races.
Although traditional methods remain fundamental, modern tools such as enviromics and precision agriculture now provide critical support for the design of effective, targeted control strategies. These technologies include genomics (of both hosts and pathogens), enviromics, and microbiomics. By characterizing the complex interactions among pyramid components, researchers can develop more targeted and sustainable approaches to mitigate the impact of disease outbreaks.
Effective plant disease research and management require comprehensive data integration. By combining multi-omics datasets with detailed phenotypic observations, researchers can enhance diagnostic precision, design more robust breeding programs, and improve the development of targeted control measures (Mohamedikbal et al., 2025). This integrated approach is particularly valuable for addressing complex pathosystems in which traditional single-method analyses have proven insufficient.
Various methods of ecological engineering have been used to regulate and optimize beneficial microbial communities. These include introducing beneficial microorganisms, promoting the proliferation of plant-associated microorganisms, and using specific microbial combinations to establish diverse and stable microbial communities (Jain et al., 2013). By establishing more favorable plant–microbe interactions, this approach aims to inhibit colonization by pathogenic microorganisms and thus protect plant health.
Recent discoveries enable precise enhancement of disease resistance through targeted genetic modifications, offering plant breeders more effective tools for crop protection. After key disease-resistance genes are identified, genome editing techniques such as CRISPR-Cas9 are used for precise modification, thereby strengthening crop resistance to specific pathogenic bacteria (Gomez et al., 2019). This molecular breeding method overcomes the limitations of traditional breeding techniques, more precisely and efficiently introducing target traits and significantly shortening the breeding cycle of disease-resistant varieties.
Recent advances in enviromics have significantly enhanced our understanding of plant defense signaling networks. Through comprehensive analysis of these intricate pathways and their environmental interactions, scientists are now developing innovative disease resistance strategies that target specific nodes in plant signaling cascades. Specifically, precise regulation of key hormone signaling pathways (e.g., SA, jasmonic acid, and ethylene) can significantly enhance plant resistance to pathogens (Li et al., 2004). Further studies have shown that the introduction of specific signal response elements into the crop genome enables precise regulation of defense responses and establishes an adaptive control system that can respond dynamically to environmental changes.
Precise environmental classification is a necessary prerequisite for in-depth study of the new disease prevention and control strategies mentioned above. This process entails systematic characterization of the environmental factors that affect disease occurrence and development, followed by identification of key environmental signals and their patterns of correlation with disease response. By analyzing the complex pyramid network, a comprehensive enviromics map can be established, enabling researchers to determine the optimal prevention and control plans for specific environmental conditions.
Multiple technological breakthroughs have played a decisive role in transforming enviromics-driven disease control from theory to practice. Research has shown that IoT (Internet of Things) devices can effectively monitor and regulate the microclimate in the field (Shamshiri et al., 2020). One case study demonstrated this integration by combining IoT-enabled microclimate monitoring with predictive modeling to enable precise fungicide application against FHB in wheat. Field-deployed IoT sensors continuously collected real-time data on canopy temperature, relative humidity, leaf wetness duration, and rainfall. These high-resolution microclimatic data were fed into validated epidemiological models, such as the De Wolf model (accuracy: 70%–84%) or the DONcast system (a forecasting model for deoxynivalenol toxin risk, 80%–85%). These models integrated historical weather patterns with the critical anthesis period (7–10 days pre- and post-flowering) to calculate a daily disease risk index. When the predicted risk exceeded a predefined threshold (e.g., indicating conditions conducive to F. graminearum infection and mycotoxin accumulation), an automated alert was triggered. This signal could be integrated with precision spray systems, enabling growers to target fungicide applications within a narrow, high-risk window, which usually coincides with early anthesis. This enviromics-guided approach was shown to reduce unnecessary fungicide applications by up to 50% in low-risk seasons while maintaining or even improving FHB control efficacy compared with calendar-based spraying. By aligning management actions with real-time, environmentally driven disease risk, this strategy optimizes resource use, minimizes environmental impact, and enhances economic sustainability (Matengu et al., 2024).
At the same time, microbial communities with soil pH response characteristics provide a new idea for implementing precise disease control. Bio-enhancement technology based on nanocarriers also shows good application prospects. In practical applications, beneficial bacteria such as Pseudomonas viridiflava are embedded within chitosan-based nanocarriers that can protect them from various soil stresses, improving their survival and colonization in the soil environment (Hu et al., 2014; Saberi Riseh et al., 2022). This method has the potential to improve the delivery of antifungal metabolites while inhibiting soilborne pathogens such as F. oxysporum. The integration of multiple disease management approaches guided by enviromics data thus offers promising potential for enhanced disease control.
The pyramid model also suggests new approaches for plant disease assessment, breeding, and management, although such approaches will depend on a thorough understanding of environment–microbial interactions. Environmental conditions directly regulate the structure and activity of microbial communities, determining whether they will be effective in promoting plant health and disease resistance. By characterizing features of the microbial community and their changes in response to environmental signals, researchers can identify beneficial microorganisms with biological control potential. Microbial community engineering or the application of beneficial microorganisms can then be used for disease management, helping to reduce reliance on agricultural chemicals.
The development of environmentally friendly and sustainable solutions for disease control has been identified as an important goal in modern agriculture. Excessive use of traditional chemical pesticides has led to widespread concerns about their environmental and health hazards, making the integration of biological control agents with agronomic measures particularly important. Biological control agents based on beneficial microorganisms have shown promising application prospects (Villavicencio-Vásquez et al., 2025). These agents can not only inhibit the activity of pathogenic bacteria but also activate the defense systems of plants. For example, species of Trichoderma can enhance plant defense mechanisms, as evidenced by their ability to induce SA- and jasmonic-acid-mediated resistance pathways in Arabidopsis leaves and to promote the accumulation of hydrogen peroxide and camphorine. At the same time, rotation systems, intercropping patterns, and standardized field hygiene management can reduce the probability of disease occurrence by interfering with the life cycle of pathogenic bacteria (Singh et al., 2023).
Contemporary disease management strategies also increasingly recognize the importance of socioeconomic and ecological considerations. Of particular concern is how patented synthetic microbial community (SynCom) technologies may inadvertently widen existing disparities in global agricultural innovation. Small farmers in developing countries often find it difficult to afford such advanced technologies (Piesse and Thirtle, 2010). Only by narrowing the gap between research achievements and practical applications can innovative prevention and control technologies be widely adopted. The promotion of open-source microbial genome databases can, to some extent, alleviate the problem of inequitable access to this technology; at the same time, the carbon footprint of disease management measures cannot be ignored.
Soil microorganisms influence biogeochemical cycles, helping to maintain soil fertility and thus supporting plant growth and the health of the ecosystem as a whole. In climate-sensitive environments, they play a crucial role in enhancing carbon storage, promoting nutrient cycling, and improving ecosystem resilience. With continued research, soil microbial communities could potentially be leveraged to help mitigate the impacts of climate change (Jansson and Hofmockel, 2020).
Empirically validated integrative frameworks that assimilate data spanning the microenvironment, microbiome composition, host genetic factors, and pathogen dissemination patterns are increasingly advancing predictive capacity. For example, Kwak et al. (2018) demonstrated that shifts in rhizosphere bacterial community structure could predict the occurrence of tomato bacterial wilt up to 2 weeks before symptom onset, with microbiome composition serving as a more sensitive indicator than direct pathogen quantification (Kwak et al., 2018). These predictive models can also simulate the disease transmission process, evaluate the actual effects of different management practices, and consider economic costs and environmental impacts. Such models can be developed into practical decision-support systems, helping growers to select the best and most sustainable control strategies on the basis of specific production conditions.
In plant disease research and control, the main task of predictive systems biology is the integration of multi-omics data to enable accurate prediction of biological effects. The key lies in the use of interpretable machine learning methods to ensure prediction accuracy while revealing underlying mechanisms. Multi-omics datasets are the primary input for such research. Before actual analysis, these datasets must be preprocessed through dimensionality reduction and feature selection to address issues of high dimensionality and sparsity (Xu et al., 2022; Liu et al., 2024). Machine learning models show considerable promise for plant disease control. They can not only predict pathogen behavior and disease dynamics from complex multi-omics data but also, when designed for interpretability, reveal key molecular mechanisms to guide targeted intervention (Sidak et al., 2022). Combining interpretable machine learning with systems biology enables researchers to predict disease development and obtain practical and actionable insights from which to formulate disease management strategies.
Leveraging interactions between microorganisms and the environment to predict plant diseases is a new frontier in precision agriculture. An increasing body of research indicates that subtle changes in microbial community structure at the root or leaf interface often occur before a significant increase in pathogen abundance; these changes can therefore serve as earlier and more sensitive indicators for predicting disease outbreaks (Gu et al., 2022). Environmental factors regulate the composition and function of these microbial communities, and this regulation can either enhance or weaken their predictive potential (Leveau, 2024). In the classic “pathogen triangle” framework, microbial community characteristics and environmental variables have become two core elements for achieving early and precise warning of plant diseases.
In one practical case study, shifts in the rhizosphere microbiome were used as an early warning of tomato bacterial wilt caused by Ralstonia solanacearum. Researchers performed longitudinal 16S rRNA gene sequencing of the rhizosphere microbiome of tomato plants grown in disease-conducive soil. Specific and reproducible changes in bacterial community structure occurred as early as 2 weeks before the onset of visible wilting symptoms and a detectable surge in R. solanacearum pathogen titers. These changes included a decline in the relative abundance of certain Bacillus and Pseudomonas taxa and an increase in Oxalobacteraceae. A machine learning model trained on these pre-symptomatic microbial profiles predicted subsequent disease onset with high accuracy (>80%). This “microbiome dysbiosis” signature was a more sensitive and earlier predictor than direct measurement of pathogen load. This case demonstrates that the plant microbiome, as a biosensor integrated within the pyramid network, can reflect subtle, pre-pathogenic changes in the rhizosphere environment and plant health, providing a critical time window for interventions such as specific biocontrol agents or soil amendments to steer the microbes back to a suppressive state (Kwak et al., 2018) (Figure 3C).
Field-based investigations identified siderophore-mediated microbial competition in the rhizosphere as a reliable biological marker for soil suppressiveness against plant pathogens (Gu et al., 2020). When ambient humidity increased or available iron declined, the abundance of bacteria capable of producing specific siderophores (which cannot be exploited by pathogens) increased accordingly, leading to a significant reduction in Ralstonia solanacearum populations. Therefore, if the competitive functions of microorganisms and environmental driving variables can be integrated into prediction models, this will not only help to extend the time window for disease prediction but also provide guidance for the precise application of microbial preparations or targeted interventions in the field microenvironment.
Plant breeders use genomic data and statistical models to predict the growth performance of individual plants or germplasm lines under specific environmental conditions. Genomics–enviromics predictions obtained by integrating environmental information can further enhance the accuracy of such predictions (Xu et al., 2022). Using a selection method that incorporates environmental information, breeders can more precisely select plants with ideal disease-resistance traits, improving overall breeding efficiency. Practice has shown that this approach can significantly enhance breeding efficiency and greatly shorten the research and development cycle for new varieties with high disease resistance.
Enviromics-driven prediction models have shown significant value in assessing pathogen distributions. For instance, researchers integrated spatial geographic data and neural network algorithms, combined the temperature adaptability of the wheat stripe rust pathogen (Puccinia striiformis f. sp. tritici) with the distribution of wheat planting areas, and redefined the overwintering regions of this pathogen in China. Their optimized model introduced a 24°C temperature threshold and the concept of core overwintering areas, providing a scientific basis for precise prevention and control. Advanced modeling techniques can thus improve the accuracy with which pathogen transmission dynamics are predicted against the background of climate change (Zhang et al., 2025), providing new possibilities for more precise and efficient disease prevention and control systems.
The transition from the classic disease triangle to the dynamic pyramid network marks a paradigm shift in understanding plant disease ecology. However, as highlighted throughout this review, realizing the full potential of this integrative network for predictive modeling and sustainable management requires overcoming significant scientific and translational challenges. Only by fully understanding these challenges and planning a clear direction can the subsequent work be steadily advanced.
The field of microbiome analysis currently lacks unified standardized methods, making it difficult to directly compare results across studies and hindering research integration. Another major problem lies in the transformation of laboratory results into practical applications on a large scale in the field. The laboratory environment is relatively controllable, whereas the field environment is complex and variable (e.g., differences in soil fertility, climate fluctuations, etc.), and these differences often cause difficulties in implementing laboratory results in the field. Climate change has also brought new variables to disease management. It not only continuously changes the distribution range, population size, and pathogenic ability of many plant pathogens but also has a significant impact on environmental functions and the functions of plant-associated microbiomes (Chakraborty and Newton, 2011; Ristaino et al., 2021). These changes further increase the complexity of formulating and implementing disease management strategies. To propel this field forward and address the need for science and development, we propose the following research priorities.
Examples include how transpiration alters leaf microclimates for pathogens, how pathogens manipulate host metabolite transport (Freyberg and Harvill, 2017), and how roots and microbes reshape the rhizosphere (Pantigoso et al., 2022). Measuring these feedbacks reproducibly in both controlled and field settings is essential for building predictive models and translating insights into practice.
To implement this systems-based approach, close collaboration is needed among researchers, farmers, and policymakers. Researchers should provide the models and tools, farmers must test and adapt the strategies in real field conditions, and policymakers can enable scaling through supportive platforms and resources. This integrated effort is crucial for transforming our understanding of the pyramid network into resilient, sustainable disease management systems that work under climate change and align with broader soil health and biodiversity goals.
Current meteorological station networks, soil databases, and geographic information systems can support large-scale collection of climate and soil parameters. By combining satellite remote sensing, ground sensors, unmanned aerial vehicles, and high-throughput environmental monitoring facilities, researchers can obtain enviromics data, including microclimate characteristics and biogeochemical gradients, at multiple temporal and spatial scales. This process requires strict standardization and systematic operational procedures. The resulting datasets are characterized by high resolution, multi-dimensionality, and wide coverage, providing unprecedented data support for disease prevention and control.
Researchers must now develop analytical frameworks capable of processing these complex, high-dimensional datasets that incorporate both spatial and temporal biological information. Integrating these heterogeneous streams into predictive models will push the boundaries of both data stewardship and algorithmic design. Meeting this challenge will require AI-driven model architectures coupled with the extraordinary computational power promised by quantum computing, ultimately enabling real-time forecasting and early-warning systems for plant disease management. In addition, the high cost and technical complexity of multi-omics approaches remain prohibitive for routine application in many agricultural settings, particularly in low-resource regions. Although these technologies generate unprecedented insights, their translation into scalable, affordable tools for growers and breeders requires substantial technological simplification and cost reduction.
From the classic disease triangle to the pyramid network, new ideas have been opened up for plant disease research. Disease prevention and control methods driven by enviromics provide a new approach for achieving more precise and sustainable plant protection. By integrating multiple control strategies and making full use of technological progress, this approach offers significant advantages for improving the efficacy of disease prevention and control. However, to fully realize the potential of enviromics in plant health management, key technical challenges need to be addressed.
The reproducibility of field microbiome data poses a major hurdle. Microbial community composition is notoriously variable across space and time, influenced by soil heterogeneity, weather fluctuations, and crop management practices. This context-dependent plasticity means that a microbiome signature identified as “protective” in one field trial may not generalize to another location or season, complicating the development of universally applicable microbiome-based diagnostics or biocontrol products. A closely related challenge is model transferability across different cropping systems. Predictive models trained on specific host–pathogen–microbeenvironment combinations often fail when applied to new genetic backgrounds, pathogen races, or agroecological zones. Addressing these interconnected challenges requires not only larger and more diverse training datasets but also a deeper mechanistic understanding that can inform model parameterization across diverse systems.
We reiterate that the inclusion of M as a core component is not merely semantic but reflects a shift from a plant-centric to a holobiont-centric view of disease. Future models may need to explicitly partition “host-associated microbiomes” (M) from the “environmental microbial pool” (E) to avoid over-extending the system boundary.
Over the next decade, plant disease management will undergo a fundamental transformation driven by the integration of multi-dimensional technologies within the classic disease pyramid framework. This model comprises the host, pathogen, non-pathogenic microbes, and the environment and serves as the foundation for new strategies. By leveraging advanced omics, AI, and molecular tools, the field is moving toward precise and sustainable control methods to safeguard global food security against evolving pathogens and climate change.
A critical step in this evolution is systematic quantification of the pyramid components through genotyping, phenotyping, and envirotyping. Envirotyping provides a theoretical basis by linking external conditions to phenotypic variations (Xu, 2016). High-throughput omics generate detailed genomics identifies resistance genes and pathogen virulence factors, phenomics captures disease-related traits via non-invasive imaging, and enviromics maps disease-influencing environmental conditions. Practical applications include unmanned aerial vehicle (UAV) -based phenomics combined with enviromics data for evaluation of rice bacterial blight resistance (Bai et al., 2023) and satellite-enabled enviromics (Resende et al., 2024) for large-scale environmental monitoring to guide regional disease prevention. In addition, microbiome–transcriptome integration has revealed how resistant maize recruits beneficial Bacillus communities to enhance defense against Fusarium stress (Xia et al., 2024).
The massive datasets generated by these omics approaches are increasingly managed through AI-driven modeling. Building on genomics–enviromics prediction models (Xu et al., 2022), the field is expanding into disease forecasting by synthesizing multi-omics and environmental information. AI technologies harmonize heterogeneous data to create high-precision tools, ranging from machine learning algorithms for late blight prediction to optical sensors and robotic systems that facilitate timely field monitoring (Mahlein et al., 2024). Deep learning algorithms like AlphaFold are facilitating the structural engineering of immune receptors (e.g., NLRs) to expand pathogen recognition spectra (Li et al., 2025), and generative models are streamlining the virtual screening of bioactive compounds, significantly accelerating the development of novel pesticides (Djoumbou-Feunang et al., 2023). In addition to monitoring, AI is revolutionizing molecular intervention strategies, particularly in the structural engineering of immune receptors. A recent breakthrough in potato NLRome analysis demonstrated that grafting an HMA “plug-in” domain from Rpi-brk1 onto the functionally compromised R1 receptor successfully restored its disease resistance (Wang et al., 2026). This modular engineering approach represents only the tip of the iceberg; when integrated with newly developed machine learning-based prediction models, future strategies are expected to leverage AI to guide the directed evolution of structural domains, enabling recognition of a much broader spectrum of pathogen effectors.
Ultimately, these diagnostic and predictive insights will inform tangible solutions through synthetic biology and molecular breeding. Genome editing technologies like CRISPR-Cas9 enable the creation of novel resistance traits, such as broad-spectrum powdery mildew resistance without growth penalty in wheat (Li et al., 2022). Simultaneously, synthetic biology is enhancing biocontrol efficacy, including the development of volatile-emitting fungi for postharvest protection. These molecular advances support integrated management strategies that combine improved crops with cultural practices to minimize chemical inputs and delay pathogen resistance. By embedding omics, AI, and biotechnology within the disease pyramid concept, agricultural systems can shift from reactive measures to proactive, resilience-based management.
We propose the disease pyramid and its network as a robust and necessary scaffold for a systems-level understanding of plant health. By directing research toward key priority areas, including data standardization, AI-driven integration, cross-system comparison, socio-economic integration, and mechanistic enviromics, this framework can evolve from a conceptual model into a predictive and actionable science. Such progress will not only advance our fundamental understanding of HPMEIs but also enable the development of resilient crops and the design of adaptive, precise, and sustainable disease management strategies for agriculture under global climate change.
The authors would like to thank Dr. Xinyao He (International Maize and Wheat Improvement Center) for his critical reading and comments and Die Zhao and Weidong Zhu for their literature search. The relevant projects and preparation of this article are supported by the following Key R&D Program of Shandong Province, China (2024CXPT034); Provincial Technology Innovation Program of Shandong, Agricultural Science and Technology Innovation Program (ASTIP) of the Chinese Academy of Agricultural Sciences; Shenzhen Science and Technology Program (KQTD202303010928390070); the 10.13039/501100001809National Natural Science Foundation of China (32341037); and the Scientific Research Special Fund of Lixiahe Institute of Agricultural Sciences (SJ (22)101). No conflict of interest declared.
Y.X. and Y.C. developed the scope and outline. Y.C., T.W., W.H., W.S., and X.L. performed the literature search and screening. Y.C., Y.X., T.W., W.H., W.S., and X.L. drafted the manuscript. Y.X., Y.C., T.W., and W.H. designed and prepared the figures. X.X., X.Z., H.Z., and P.K.S. provided constructive comments, suggestions, and discussions that significantly improved the manuscript draft. All authors read and approved the final version of the manuscript.