Method for regulating and controlling highland barley rhizosphere microflora

Abstract

The invention discloses a highland barley rhizosphere microbial community regulation and control method, which comprises the following steps: clustering sequencing sequences of microbial data in rhizosphere soil data of mature highland barley to obtain taxonomic information and an operable classification unit, performing zero-model analysis on the soil data and the operable classification unit according to planting data, and determining whether the soil data and the operable classification unit are in zero-model analysis. The method comprises the following steps: acquiring a ratio of nitrogen fertilizer data to environmental screening contribution and drift contribution of microbial community composition, performing diversity analysis on an operable classification unit according to taxonomy information to obtain dominant species data, extracting a nitrogen fertilizer amount and a fertilization period corresponding to the dominant species data, and performing parameter optimization through a Bayesian optimization algorithm. The rhizosphere soil is fertilized according to the fertilization period and the optimization parameters of the planting area, and the regulation and control result of the microflora is obtained. According to the method, the optimal nitrogen application amount gradient is determined through zero model analysis and Bayesian parameter optimization, so that the rhizosphere microorganism function adapts to the growth requirement of highland barley, and an interpretable basis is provided for optimizing nitrogen fertilizer management.

Description

Technical Field

[0001] This invention relates to the field of microbial regulation, and more particularly to a method for regulating the rhizosphere microbial community of highland barley. Background Technology

[0002] Barley, belonging to the genus *Haloxylon* of the Poaceae family, is a variety of cultivated barley. It is also known as naked barley because the palea and lemma separate from the caryopsis at maturity, exposing the grain. Barley thrives in cool, high-altitude climates, is highly cold-resistant, has a short growing season, is high-yielding, early-maturing, and widely adaptable. Barley production not only drives economic development in Tibet but also plays a vital role in social stability. As the most important grain crop in Tibet, barley grains are rich in nutrients and physiologically active substances, possessing significant nutritional value and medicinal and health benefits. The straw is also a major forage crop. However, in actual barley cultivation, nitrogen fertilizer utilization is not the highest. Insufficient nitrogen fertilizer application leads to lower barley yields, while excessive nitrogen application affects nitrogen fertilizer utilization efficiency, pollutes the environment, and increases farmers' costs.

[0003] The rhizosphere microbiome is closely related to the healthy development of plants. Microorganisms are the most active components in the soil and are sensitive to the external environment. They are easily affected by fertilizer application, type, and different cultivation practices. They not only participate in the transformation and cycling of soil nutrients, but also play an irreplaceable role in maintaining soil ecological functions. To a certain extent, they determine the speed of material cycling and can reflect soil conditions, such as the decomposition of organic matter, nutrient cycling, and biological nitrogen fixation. The application of nitrogen fertilizer can improve the diversity of soil microorganisms. Therefore, how to effectively regulate the rhizosphere microbial community of barley has become an urgent problem to be solved. Summary of the Invention

[0004] The purpose of this invention is to provide a method for regulating the rhizosphere microbial community of highland barley.

[0005] To achieve the above objectives, the present invention is implemented according to the following technical solution: This invention provides a method for regulating the rhizosphere microbial community of highland barley, comprising: Rhizosphere soil data of mature highland barley were preprocessed to obtain planting data, microbial data, and soil data; Clustering of the sequencing sequences of the microbial data to obtain taxonomic information and operational taxonomic units; zero-model analysis of soil data and operational taxonomic units based on planting data to obtain the environmental screening contribution and drift contribution ratio of nitrogen fertilizer data to the composition of microbial communities. Diversity analysis was performed on operable taxonomic units based on taxonomic information. Principal component analysis was used to obtain the difference contribution rate based on bacterial diversity. The difference contribution rate and community composition were used to screen bacterial species with differences in abundance changes, and dominant species data under taxonomic information were obtained. The nitrogen fertilizer amount and fertilization cycle corresponding to the dominant species data are extracted, and parameters are optimized using a Bayesian optimization algorithm based on the environmental screening contribution ratio. Fertilization is carried out on the rhizosphere soil according to the fertilization cycle and the optimized parameters of the planting area to obtain the regulation results of the microbial community.

[0006] Furthermore, the method for obtaining the taxonomic information and operable taxonomic units includes: Based on the microbial data, sequencing sequences and species databases of samples were obtained. Sequencing sequences with a sequence similarity of ≥97% were clustered into operational taxonomic units. The sequencing sequence with the highest frequency of occurrence in each operational taxonomic unit was selected as the representative sequence. Species annotation was performed on the representative sequence based on the species database. Multiple sequence alignment was conducted on the representative sequence based on base pairs to obtain the phylogenetic relationship of the representative sequence. Based on the taxonomic information of the annotations, the community composition and abundance data of samples under each taxonomy were statistically analyzed. Based on the abundance data, the sample with the largest amount of data was selected as the abundance standard. The remaining samples were normalized using the abundance standard to obtain the abundance matrix of the operational taxonomic units.

[0007] Furthermore, the method for obtaining the ratio of environmental screening contribution to drift contribution includes: Nitrogen application information is extracted from the planting data. Based on the phylogenetic relationships, abundance matrix and rhizosphere soil physicochemical data in the soil data, the minimum average phylogenetic distance between samples is calculated based on the abundance matrix and phylogenetic tree to obtain the standard phylogenetic distance. Keeping the total number of sequences in the sample unchanged, and keeping the order of ammonium nitrogen content and nitrate nitrogen content in the sample unchanged, the abundance distribution of operable taxa within the same nitrogen level is randomly permuted according to the abundance matrix to obtain the randomly simulated community structure. The minimum mean phylogenetic distance of the randomly simulated community structure is used as the simulated phylogenetic distance. The z-score is standardized according to the standard phylogenetic distance, the simulated phylogenetic distance and the standard deviation of the simulated phylogenetic distance to obtain the phylogenetic dissimilarity index of the sample. The environmental screening contribution index of nitrogen fertilizer was calculated using phylogenetic anisotropy, and the formula for the environmental screening contribution index is as follows: in Contribution index to environmental screening of nitrogen fertilizers This represents the total number of rhizosphere soil samples. To ensure similarity in nitrogen application rates, , For the sample The amount of nitrogen applied, For the sample The amount of nitrogen applied, To the maximum nitrogen application rate, To minimize nitrogen application, For the sample and The phylogenetic dissimilarity index, The phylogenetic distances between samples in the phylogenetic relationships were obtained through null model randomization. For the sample and The nitrogen distance gradient residual was obtained through fitting. For indicator functions, if When the value is greater than 0, the indicator function takes the value of 1; otherwise, it takes the value of 0. Variance decomposition and null model analysis were performed on the non-nitrogen data in the rhizosphere soil physicochemical data to obtain the remaining screening contribution index of the other environmental factors. The drift contribution index was obtained by using the environmental screening contribution index of nitrogen fertilizer and the remaining screening contribution index through the residual difference method, and the ratio of environmental screening contribution to drift contribution was calculated.

[0008] Furthermore, a method for obtaining said bacterial diversity includes: Based on the operational taxonomic units of the samples, a species accumulation box plot is drawn. The change trend of the species accumulation box plot is used to check whether the sample size is sufficient. If the position of the box plot changes to a flat point, it is determined that the sample size is sufficient and the bacterial community diversity is rich. Based on the nitrogen application information of the samples in the planting data, operational taxonomic units are divided to obtain the set of operational taxonomic units under different nitrogen application rates. The relative abundance and relative abundance map of dominant bacteria under phylum and genus in taxonomic information are obtained from the abundance matrix of operational taxonomic units. Based on the relative abundance map, the Shannon index, Chao1 index, observed species and PD whole tree of bacterial communities under different nitrogen application levels are used as Alpha diversity indices and box plots are drawn respectively. The significance difference of different sets of operational taxonomic units is calculated based on the box plots to obtain bacterial community richness and bacterial diversity. The unweighted UniFrac distance and the weighted UniFrac distance are calculated based on the phylogenetic relationship between the sets of operable taxa, and a Beta diversity index heatmap is plotted. The optimal nitrogen application gradient is obtained based on the diversity index heatmap. The sets of operable taxa include common operable taxa, unique operable taxa, the most operable taxa, and the least operable taxa.

[0009] Furthermore, the method for obtaining the dominant species data includes: Principal component analysis is performed based on the abundance matrix of operable taxonomic units to obtain the difference contribution rate of the first and second principal components to the samples. Clustering circles are added to the samples in the difference contribution rate map. The similarity and specificity of the sample community composition are obtained according to the difference trend of the samples. The operable taxonomic unit sets with high similarity and specificity are used as difference combinations. A T-test was performed on the community composition of the differential combinations. Based on the significance of the differences, the comparative combinations with a significance of less than 0.05 were selected. Based on taxonomic information, bacterial species with different abundance changes in the comparative combinations were selected at the phylum and genus levels. The bacterial species were ranked and standardized according to a preset number. Based on the species and sample levels, cluster analysis and heatmaps were generated for the standard bacterial species. Based on the heatmaps, the dominant species data under taxonomic information were obtained.

[0010] Furthermore, the method for parameter optimization includes: Based on the dominant species data, the corresponding nitrogen fertilizer amount and fertilization cycle are extracted, and a comprehensive objective function is constructed according to the environmental screening contribution ratio. The formula of the comprehensive objective function is as follows: in Nitrogen application rate Fertilization cycle The overall objective function is as follows: nitrogen application rate Fertilization cycle These are candidate parameters for the comprehensive objective function. For the first The relative abundance of the dominant species, The total number of dominant species is obtained from dominant species data. Contribution ratio to nitrogen fertilizer environmental screening The influence weight of nitrogen residual, This represents the average value of the nitrogen distance gradient residuals. The community linear diversity index is obtained by linearly weighting the Shannon index and the Chao1 index. The distance is the unweighted UniFrac distance. For the first The weights of beta-like diversity indicators For the first The actual values ​​of the Beta-like diversity index; Based on the nitrogen application rate, fertilization cycle, and sample data, the probability distribution of the comprehensive objective function is fitted using a Gaussian process regression model. The predicted mean and standard deviation of the candidate parameters in the comprehensive objective function are obtained. Based on the predicted mean and standard deviation, the selection of new candidate parameters is guided by an upper confidence function. The predicted values ​​of the comprehensive objective function corresponding to the new candidate parameters are added to the sample set. The Gaussian process regression model is iteratively updated based on the sample set until the change in the predicted value of the comprehensive objective function is less than 5% after three consecutive iterations. At this point, the optimization is stopped, and the optimal nitrogen application rate and optimal fertilization cycle are obtained.

[0011] Furthermore, the method for obtaining the regulation result includes: Optimization parameters are obtained based on dominant species data. These parameters are then assigned to different plots within the planting area based on fertilization cycles. Different gradients of basal fertilizer application rates are set according to the optimization parameters and the optimal nitrogen application rate gradient. The pre-set basal fertilizer (phosphate and potassium) is applied to each plot in one application. Barley is used for manual row sowing in each plot. Topdressing is applied based on the pre-set basal fertilizer at a ratio of 7:3, with the barley jointing stage as the topdressing period. Microbial community regulation samples are obtained based on the planting area using a multi-point mixed sampling method. New dominant species data and new abundance data are obtained through diversity analysis of operable taxa based on the regulation samples. These new dominant species data and new abundance data are used as the regulation results. The basal fertilizer application rate includes nitrogen application gradients of basal fertilizer at 10% below the optimal amount, the optimal amount, and 10% above the optimal amount.

[0012] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects: This invention uses zero-model analysis to distinguish the contribution ratio of nitrogen fertilizer to environmental screening and drift in rhizosphere microbial communities, and to quantify the intensity of nitrogen fertilizer's targeted regulation of microorganisms. This avoids the problem of blind nitrogen application due to the lack of clear regulatory mechanisms, and provides a clear basis for parameter optimization. A comprehensive objective function is constructed based on the abundance of dominant species, community diversity index, and nitrogen fertilizer screening ratio. Combined with Bayesian optimization algorithm, parameters are optimized to ensure the enrichment effect of dominant species while reducing the impact of insufficient nitrogen application on microbial community regulation. The optimal nitrogen application gradient is determined by Beta diversity index heatmap, and dominant species are screened by principal component analysis and T-test. This achieves synergistic regulation of community diversity maintenance and targeted enrichment of dominant species, avoiding the disruption of microbial ecological balance by a single regulatory target. It ensures that the function of rhizosphere microorganisms is adapted to the growth needs of barley, and provides a feasible technical path for improving nitrogen utilization efficiency and farmland ecological stability in barley. Attached Figure Description

[0013] Figure 1 This is a flowchart illustrating the steps of the method for regulating the rhizosphere microbial community of highland barley in this embodiment of the invention. Detailed Implementation

[0014] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0015] Reference Figure 1 As shown, this invention provides a method for regulating the rhizosphere microbial community of highland barley, comprising: Rhizosphere soil data of mature highland barley were preprocessed to obtain planting data, microbial data, and soil data; In the actual assessment, a major barley-producing area was selected. This area has a typical plateau temperate monsoon semi-arid climate with an average annual temperature of 5.9℃ and annual precipitation of 360mm, mainly concentrated from May to October, primarily at night. The annual sunshine duration is 3200 hours, the frost-free period is 120-140 days, and the windy season is approximately 100 days. The soil physicochemical properties are as follows: pH 8.2, classifying it as alkaline soil; the topsoil organic matter content is 19.3g / kg, total nitrogen content is 0.1%, total phosphorus content is 0.72g / kg, total potassium content is 17.1g / kg, hydrolyzable nitrogen content is 76.7mg / kg, available phosphorus content is 11.3mg / kg, and available potassium content is 116mg / kg. Five gradients of basal fertilizer application rates were set: CK (0 kg / hm2), N1 (43.125 kg / hm2), N2 (86.25 kg / hm2), and N2 (86.25 kg / hm2). The base fertilizer (N3) and topdressing (N4) were applied at a ratio of 7:3. The topdressing was applied at the jointing stage. Each treatment was repeated three times. Each plot was 25 m2 (2 m × 12.5 m), with a total of 15 plots. The seeding rate was 225 kg / hm2. The sowing method was manual row sowing with a row spacing of 25 cm. Before sowing, phosphate fertilizer and potassium fertilizer were applied as base fertilizer in one application, with 90 kg / hm2 of phosphate fertilizer and 45 kg / hm2 of potassium fertilizer. Other field management practices are the same as for general high-yield fields. Soil samples were collected using a multi-point pooled sampling method, with three sampling points randomly selected from each treatment plot. Soil samples were taken from a depth of 15 cm at each sampling point. After collection, the samples were placed in sterile tubes, labeled, and stored in liquid nitrogen. The samples were then sieved through a 20-mesh sieve to remove impurities such as gravel and plant roots. The samples were then aliquoted into 10 ml sterile tubes, sealed, and stored at -80°C. Genomic DNA was extracted from the samples using the CTAB method. The purity and concentration of the DNA were then detected using agarose gel electrophoresis. An appropriate amount of sample was placed in a centrifuge tube and diluted with sterile water to 1 ng / μl. Using the diluted genomic DNA as a template, specific primers with barcodes were used according to the selected sequencing region. New England Biolabs' Phusion® High-Fidelity PCR Master Mix with GC... Buffer and high-efficiency, high-fidelity enzymes were used for PCR to record information such as nitrogen application rate, fertilization cycle of 25-40 days, and planting density in each plot to obtain planting data. The planting data, microbial data, and soil data were then integrated to form rhizosphere soil data. Clustering of the sequencing sequences of the microbial data to obtain taxonomic information and operational taxonomic units; zero-model analysis of soil data and operational taxonomic units based on planting data to obtain the environmental screening contribution and drift contribution ratio of nitrogen fertilizer data to the composition of microbial communities. In the actual evaluation, Uparse software was used to cluster all clean reads of all samples. By default, sequences were clustered into operational taxonomic units based on 97% identity similarity. Representative sequences of operational taxonomic units were selected, and the most frequent sequence in each operational taxonomic unit was selected as the representative sequence according to the algorithm principle. Species annotation was performed on the representative sequences, and the Mothur method was used to perform species annotation analysis with the SILVA SSUrRNA database to obtain taxonomic information at each taxonomic level (kingdom, phylum, class), order, family, genus, and species. The community composition and abundance of each sample were also statistically analyzed. MUSCLE software was used for fast multiple sequence alignment, and FastTree was used to construct a phylogenetic tree to obtain the phylogenetic relationships of all representative sequences. Finally, the data of each sample were normalized, with the sample with the largest amount of data being the standard for normalization, to obtain the abundance matrix. Subsequent Alpha diversity analysis and Beta diversity analysis were based on the normalized data. In the actual assessment, the regulatory effect of nitrogen fertilizer on the microbial community was calculated based on the abundance matrix and phylogenetic tree. Nitrogen application information for 15 plots was obtained. The minimum mean phylogenetic distance between samples was calculated based on the abundance matrix and phylogenetic tree to obtain the standard developmental distance. Non-nitrogen physicochemical indicators such as pH, organic matter, and available phosphorus were extracted for subsequent variance decomposition. For three plots with the same nitrogen application gradient of 150 kg / hm², keeping the total sequence number and nitrogen level order unchanged, the abundance distribution was randomly permuted and repeated 1000 times to generate a simulated community structure. The simulated developmental distance was calculated and normalized by z-score. The phylogenetic dissimilarity index was obtained. The results showed that the mean phylogenetic dissimilarity index under the 150 kg / hm² gradient was 2.1, which is greater than 2, indicating that nitrogen fertilizer-driven environmental screening was significant. The nitrogen application rate similarity was the mean of 0.85 within the 150 kg / hm² gradient. The nitrogen distance gradient residual was fitted by linear regression and the R² was 0.72 with a mean of 0.12. Substituting this into the calculation of the environmental screening contribution index, we got 0.45. Using the residual difference method, the drift contribution ratio of 1 - nitrogen fertilizer screening contribution - non-nitrogen factor screening contribution was 0.32. Therefore, the nitrogen fertilizer environmental screening contribution ratio was determined to be 0.45. Diversity analysis was performed on operable taxonomic units based on taxonomic information. Principal component analysis was used to obtain the difference contribution rate based on bacterial diversity. The difference contribution rate and community composition were used to screen bacterial species with differences in abundance changes, and dominant species data under taxonomic information were obtained. In the actual evaluation, a total of 14 samples were sequenced. Within a certain range, as the number of samples increased, the box plot reflected the rate of emergence of new operational taxa under continuous sampling. A sharp rise in the box plot position indicated that a large number of species were discovered in the community, while a flattening position indicated that the number of species discovered would not increase with further increases in sample size. Therefore, the sequencing data was reasonable and the bacterial community was highly diverse. After clustering all high-quality sequences, a total of 9679 operational taxa were obtained. Under the four nitrogen application levels, there were 3399 common operational taxa. Among them, N2 had the most operational taxa with 400, while N1 had the fewest with only 145. N2 and N4 had a total of The maximum number of operable taxa is 611. Based on abundance statistics of bacteria and microorganisms in the soil, at the phylum level, the top ten dominant phyla are: Proteobacteria (27.83%), Acidobacteria (24.04%), Planctomycetes (8.97%), Bacteroidetes (8.59%), Gemmatimonadetes (7.80%), Chloroflexi (6.25%), and Actinobacteria (…). The most abundant phyla were: tinobacteria (4.02%), Verrucomicrobia (2.85%), Armatimonadetes (1.59%), and Latescibacteria (1.13%). At the genus level, the top ten most abundant genera were: Acidobacteria (2.60%), Sphingomonas (1.89%), Stenotrophobacter (1.55%), and Gammaproteobacteria (1%). The bacterial community diversity indices of barley root soil were calculated under different nitrogen application levels, and box plots were plotted. According to the box plots, under the four nitrogen level treatments, N3 had the highest Shannon index, Chao1 index, observed species, and PD whole tree. Among the four indices, N3 showed significant differences from N1, N2, and N4, respectively, with p < 0.05, where the Shannon index of N1 was significantly different from that of N2 and N4, respectively (p<0.05). The results indicate that during barley cultivation, at the nitrogen application level of N3, the bacterial community richness and diversity in the non-rhizosphere soil were the highest, while N1 had the lowest, followed by N2 and N4. There was no significant difference in community diversity. Using the phylogenetic relationships between operational taxa, the unweighted Unifrac distance and Unweighted Unifrac were further calculated. Using the abundance information of operational taxa, the unweighted Unifrac distance was further used to construct the weighted Unifrac distance, and a box plot was drawn. Under the Unweighted Unifrac distance, N3 was significantly different from N1 and N2, respectively (p<0.05). Under the weighted Unifrac distance, N3 was extremely significantly different from N1 and N2, respectively (p<0.01). Using the weighted Unifrac distance and Unweighted Unifrac distance... The Beta diversity index heatmap was plotted using two indicators: Unifrac distance and the dissimilarity coefficient between two samples. A smaller value indicates a smaller difference in species diversity between the two samples. Analysis of the Beta diversity index heatmap showed that treatments N3 and N4 had the smallest difference in species diversity, followed by N1 and N3. The largest difference was between N1 and N2. The optimal nitrogen application gradient was determined from the diversity index heatmap. When the nitrogen application rate was within the gradient range of 120-180 kg / hm², the color was lightest among different replicates at the same nitrogen application rate, with an unweighted Unifrac distance of 0.32 and a weighted Unifrac distance of 0.28. This indicates that the microbial community structure had the best repeatability and stability within this gradient. Therefore, 120-180 kg / hm² was determined as the optimal nitrogen application gradient. In the actual assessment, PCA analysis was used to reflect the differences between samples to the greatest extent. After adding cluster circles to each sample, it was found that samples N1 and N3 were the closest, indicating that the community composition of these two groups was the most similar. N2 was the farthest from the other three groups on the coordinate axis, indicating that its community composition was the most unique. The first principal component contributed 10.95% to the sample differences, and the second principal component contributed 8.68%. Operational taxonomic units with high similarity and specificity were used as difference combinations. T-tests were used to identify differences in community structure among treatments and to identify species with different abundance changes among groups. At the phylum level, the nitrogen application rate of N2 showed community differences compared to groups N1, N3, and N4 (p<0.05), while differences among other groups were not significant. Based on the species annotations and abundance of all samples at the genus level, the top 35 genera by abundance were selected for UPGMA analysis at both the species and sample levels. Clustering and heatmap plotting revealed that the genera with relatively high abundance differed significantly under different nitrogen application treatments. Under nitrogen application level N1, the genera with high abundance were: *Pirellula*, *Lysobacter*, *Azoarcus*, and *Dechloromonas*. Under nitrogen application level N2, the genera with the most abundance were: *Stenotrophobacter*, *Sphingomonas*, *Flavitalea*, *Flavisolibacter*, *Adhaeribacter*, and *Massilia*. Under nitrogen application level N3, *Aquicella* was the most abundant genera. Under nitrogen application level N4, the genera with high abundance were: *Arenimonas* and *Dongia*. These data were used as the dominant species data for taxonomic information. The nitrogen fertilizer amount and fertilization cycle corresponding to the dominant species data are extracted, and parameters are optimized using a Bayesian optimization algorithm based on the environmental screening contribution ratio. Fertilization is carried out on the rhizosphere soil according to the fertilization cycle and the optimized parameters of the planting area to obtain the regulation results of the microbial community.

[0016] In the actual assessment, the corresponding nitrogen fertilizer amount and fertilization cycle were extracted based on the dominant species data, and a comprehensive objective function was constructed according to the environmental screening contribution ratio. The relative abundance of six target species under different nitrogen application rates and cycles was obtained. Among the diversity parameters, the Shannon index was 3.6, the Chao1 index was 4.04, the linearly weighted community linear diversity index was 3.82, and the unweighted UniFrac index was... The maximum distance is 0.35, the two types of Beta diversity indices S1=0.68 and S2=0.72, the weight is preset to 0.5, the average nitrogen residual is 0.12, and 10 sets of rhizosphere soil data are selected as sample data, where the nitrogen application rate x1 is 75-225 kg / hm², and the fertilization cycle x2 is 25-40 days. The probability distribution of the comprehensive objective function is fitted by the RBF kernel function of the Gaussian process regression model to obtain the predicted mean and standard deviation. Points are selected according to UCB(x)=μ(x)+2.0σ(x). When the comprehensive objective function value f changes by less than 5% for three consecutive times in the 15th iteration, the optimal parameters are output, x1. * =155kg / hm², x2 * =32 days; In the actual evaluation, fertilization was carried out in the planting area according to the optimal parameters and the optimal nitrogen application gradient. Nitrogen was applied at 77.5 kg / hm² at 32 days of tillering and 64 days of jointing, and then covered with soil after trenching and strip application. Phosphorus and potassium fertilizers were applied according to the preset base fertilizer standards. Based on the planting area, the microbial community regulation samples were obtained by multi-point mixed sampling. Based on the regulation samples, the new dominant species data and new abundance data were obtained by diversity analysis of operable taxa. The new dominant species data and new abundance data were used as the regulation results. The total abundance of the target dominant species increased by 28% compared with that before optimization, the environmental screening contribution index was 0.47, the environmental screening contribution index was ≥0.4, and the Alpha diversity fluctuation was ≤8%.

[0017] In this embodiment, the method for obtaining the taxonomic information and operable taxonomic units includes: Based on the microbial data, sequencing sequences and species databases of samples were obtained. Sequencing sequences with a sequence similarity of ≥97% were clustered into operational taxonomic units. The sequencing sequence with the highest frequency of occurrence in each operational taxonomic unit was selected as the representative sequence. Species annotation was performed on the representative sequence based on the species database. Multiple sequence alignment was conducted on the representative sequence based on base pairs to obtain the phylogenetic relationship of the representative sequence. Based on the taxonomic information of the annotations, the community composition and abundance data of samples under each taxonomy were statistically analyzed. Based on the abundance data, the sample with the largest amount of data was selected as the abundance standard. The remaining samples were normalized using the abundance standard to obtain the abundance matrix of the operational taxonomic units.

[0018] In this embodiment, the method for obtaining the ratio of environmental screening contribution to drift contribution includes: Nitrogen application information is extracted from the planting data. Based on the phylogenetic relationships, abundance matrix and rhizosphere soil physicochemical data in the soil data, the minimum average phylogenetic distance between samples is calculated based on the abundance matrix and phylogenetic tree to obtain the standard phylogenetic distance. Keeping the total number of sequences in the sample unchanged, and keeping the order of ammonium nitrogen content and nitrate nitrogen content in the sample unchanged, the abundance distribution of operable taxa within the same nitrogen level is randomly permuted according to the abundance matrix to obtain the randomly simulated community structure. The minimum mean phylogenetic distance of the randomly simulated community structure is used as the simulated phylogenetic distance. The z-score is standardized according to the standard phylogenetic distance, the simulated phylogenetic distance and the standard deviation of the simulated phylogenetic distance to obtain the phylogenetic dissimilarity index of the sample. The environmental screening contribution index of nitrogen fertilizer was calculated using phylogenetic anisotropy, and the formula for the environmental screening contribution index is as follows: in Contribution index to environmental screening of nitrogen fertilizers This represents the total number of rhizosphere soil samples. To ensure similarity in nitrogen application rates, , For the sample The amount of nitrogen applied, For the sample The amount of nitrogen applied, To the maximum nitrogen application rate, To minimize nitrogen application, For the sample and The phylogenetic dissimilarity index, The phylogenetic distances between samples in the phylogenetic relationships were obtained through null model randomization. For the sample and The nitrogen distance gradient residual was obtained through fitting. For indicator functions, if When the value is greater than 0, the indicator function takes the value of 1; otherwise, it takes the value of 0. Variance decomposition and null model analysis were performed on the non-nitrogen data in the rhizosphere soil physicochemical data to obtain the remaining screening contribution index of the other environmental factors. The drift contribution index was obtained by using the environmental screening contribution index of nitrogen fertilizer and the remaining screening contribution index through the residual difference method, and the ratio of environmental screening contribution to drift contribution was calculated.

[0019] In this embodiment, the method for obtaining the bacterial diversity includes: Based on the operational taxonomic units of the samples, a species accumulation box plot is drawn. The change trend of the species accumulation box plot is used to check whether the sample size is sufficient. If the position of the box plot changes to a flat point, it is determined that the sample size is sufficient and the bacterial community diversity is rich. Based on the nitrogen application information of the samples in the planting data, operational taxonomic units are divided to obtain the set of operational taxonomic units under different nitrogen application rates. The relative abundance and relative abundance map of dominant bacteria under phylum and genus in taxonomic information are obtained from the abundance matrix of operational taxonomic units. Based on the relative abundance map, the Shannon index, Chao1 index, observed species and PD whole tree of bacterial communities under different nitrogen application levels are used as Alpha diversity indices and box plots are drawn respectively. The significance difference of different sets of operational taxonomic units is calculated based on the box plots to obtain bacterial community richness and bacterial diversity. The unweighted UniFrac distance and the weighted UniFrac distance are calculated based on the phylogenetic relationship between the sets of operable taxa, and a Beta diversity index heatmap is plotted. The optimal nitrogen application gradient is obtained based on the diversity index heatmap. The sets of operable taxa include common operable taxa, unique operable taxa, the most operable taxa, and the least operable taxa.

[0020] In this embodiment, the method for obtaining the dominant species data includes: Principal component analysis is performed based on the abundance matrix of operable taxonomic units to obtain the difference contribution rate of the first and second principal components to the samples. Clustering circles are added to the samples in the difference contribution rate map. The similarity and specificity of the sample community composition are obtained according to the difference trend of the samples. The operable taxonomic unit sets with high similarity and specificity are used as difference combinations. A T-test was performed on the community composition of the differential combinations. Based on the significance of the differences, the comparative combinations with a significance of less than 0.05 were selected. Based on taxonomic information, bacterial species with different abundance changes in the comparative combinations were selected at the phylum and genus levels. The bacterial species were ranked and standardized according to a preset number. Based on the species and sample levels, cluster analysis and heatmaps were generated for the standard bacterial species. Based on the heatmaps, the dominant species data under taxonomic information were obtained.

[0021] In this embodiment, the parameter optimization method includes: Based on the dominant species data, the corresponding nitrogen fertilizer amount and fertilization cycle are extracted, and a comprehensive objective function is constructed according to the environmental screening contribution ratio. The formula of the comprehensive objective function is as follows: in Nitrogen application rate Fertilization cycle The overall objective function is as follows: nitrogen application rate Fertilization cycle These are candidate parameters for the comprehensive objective function. For the first The relative abundance of the dominant species, The total number of dominant species is obtained from dominant species data. Contribution ratio to nitrogen fertilizer environmental screening The influence weight of nitrogen residual, This represents the average value of the nitrogen distance gradient residuals. The community linear diversity index is obtained by linearly weighting the Shannon index and the Chao1 index. The distance is the unweighted UniFrac distance. For the first The weights of beta-like diversity indicators For the first The actual values ​​of the Beta-like diversity index; Based on the nitrogen application rate, fertilization cycle, and sample data, the probability distribution of the comprehensive objective function is fitted using a Gaussian process regression model. The predicted mean and standard deviation of the candidate parameters in the comprehensive objective function are obtained. Based on the predicted mean and standard deviation, the selection of new candidate parameters is guided by an upper confidence function. The predicted values ​​of the comprehensive objective function corresponding to the new candidate parameters are added to the sample set. The Gaussian process regression model is iteratively updated based on the sample set until the change in the predicted value of the comprehensive objective function is less than 5% after three consecutive iterations. At this point, the optimization is stopped, and the optimal nitrogen application rate and optimal fertilization cycle are obtained.

[0022] In this embodiment, the method for obtaining the regulation result includes: Optimization parameters are obtained based on dominant species data. These parameters are then assigned to different plots within the planting area based on fertilization cycles. Different gradients of basal fertilizer application rates are set according to the optimization parameters and the optimal nitrogen application rate gradient. The pre-set basal fertilizer (phosphate and potassium) is applied to each plot in one application. Barley is used for manual row sowing in each plot. Topdressing is applied based on the pre-set basal fertilizer at a ratio of 7:3, with the barley jointing stage as the topdressing period. Microbial community regulation samples are obtained based on the planting area using a multi-point mixed sampling method. New dominant species data and new abundance data are obtained through diversity analysis of operable taxa based on the regulation samples. These new dominant species data and new abundance data are used as the regulation results. The basal fertilizer application rate includes nitrogen application gradients of basal fertilizer at 10% below the optimal amount, the optimal amount, and 10% above the optimal amount.

[0023] The above description is merely an example and illustration of the structure of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the structure of the invention or exceed the scope defined in the claims, all of which should fall within the protection scope of the present invention.

Claims

1. A method for regulating the rhizosphere microbial community of highland barley, characterized in that, The method comprises the following steps: The rhizosphere soil data of mature highland barley is preprocessed to obtain planting data, microbial data and soil data; The sequencing sequences of the microbial data are clustered to obtain taxonomic information and operational classification units, and the soil data and the operational classification units are analyzed by zero model according to the planting data to obtain the proportion of environmental filtering contribution and drift contribution of the nitrogen fertilizer data to the microbial community composition; The diversity of the operational classification units is analyzed according to the taxonomic information, the difference contribution rate is obtained by principal component analysis based on bacterial diversity, the bacterial species with different abundance changes are screened by using the difference contribution rate and the community composition, and the dominant species data under the taxonomic information are obtained; The nitrogen fertilizer amount and the fertilization cycle corresponding to the dominant species data are extracted, the parameters are optimized by a Bayesian optimization algorithm according to the environmental filtering contribution proportion, the rhizosphere soil is fertilized according to the optimization parameters of the fertilization cycle and the planting area, and the regulation result of the microbial community is obtained.

2. The method for regulating the rhizosphere microbial community of highland barley according to claim 1, characterized in that, The method for obtaining the taxonomic information and the operational classification units comprises: According to the microbial data, the sequencing sequences and the species database of the sample are obtained, the sequencing sequences with a sequence similarity of greater than or equal to 97% are clustered into operational classification units, the sequencing sequence with the highest frequency of occurrence is taken as a representative sequence according to the operational classification units, the representative sequence is annotated according to the species database, and the phylogenetic relationship of the representative sequence is obtained by multiple sequence alignment based on base pairs, the community composition and the abundance data of the sample under each classification are counted based on the annotated taxonomic information, the sample with the largest data amount is taken as an abundance standard according to the abundance data, and the remaining samples are uniformly processed by using the abundance standard, and the abundance matrix of the operational classification units is obtained.

3. The method for regulating the rhizosphere microbial community of highland barley according to claim 1, characterized in that, The method for obtaining the proportion of the environmental filtering contribution and the drift contribution comprises: The nitrogen application amount information of the sample is extracted based on the planting data, the phylogenetic relationship, the abundance matrix and the rhizosphere soil physicochemical data in the soil data of the sample are extracted, the minimum average phylogenetic distance between the samples is calculated according to the abundance matrix and the phylogenetic tree, and the standard phylogenetic distance is obtained; The total sequence number of the sample is kept unchanged, the grade order of the ammonium nitrogen content and the nitrate nitrogen content in the sample is kept unchanged, the abundance distribution of the operational classification units in the samples with the same nitrogen level is randomly replaced according to the abundance matrix, the community structure of the random simulation is obtained, the minimum average phylogenetic distance of the random simulation community structure is taken as a simulation phylogenetic distance, the phylogenetic dissimilarity index of the sample is obtained by z-score standardization according to the standard phylogenetic distance, the simulation phylogenetic distance and the standard deviation of the simulation phylogenetic distance; The environmental filtering contribution index of the nitrogen fertilizer is calculated by using the phylogenetic dissimilarity, and the environmental filtering contribution index formula of the nitrogen fertilizer is: wherein is the environmental screening contribution index of nitrogen fertilizer, is the total sample number of rhizosphere soil, is the nitrogen application amount similarity, , is the nitrogen application amount of sample , is the nitrogen application amount of sample , is the maximum nitrogen application amount, is the minimum nitrogen application amount, is the phylogenetic dissimilarity index of samples and , is obtained by zero model randomization according to the phylogenetic distance between samples in the phylogenetic relationship, is the nitrogen distance gradient residual of samples and , obtained by fitting, is the indicator function, if >0, the indicator function takes the value 1, otherwise takes the value 0; The non-nitrogen data in the rhizosphere soil physicochemical data is subjected to variance decomposition and zero model analysis to obtain the remaining filtering contribution index of the remaining environmental factors, the drift contribution index is obtained by using the environmental filtering contribution index of the nitrogen fertilizer and the remaining filtering contribution index through the residual difference method, and the proportion of the environmental filtering contribution and the drift contribution is calculated.

4. The method for regulating the rhizosphere microbial community of highland barley according to claim 1, characterized in that, The method for obtaining the bacterial diversity comprises: The sample-based operable classification unit draws a species cumulative box plot, checks whether the sample quantity is sufficient according to the change trend of the species cumulative box plot, determines that the sample quantity is sufficient and the bacterial community diversity is rich when the box plot position changes to be flat, divides the operable classification units according to the nitrogen application quantity information of the samples in the planting data, and obtains the operable classification unit set under different nitrogen application quantities; The relative abundance and relative abundance graph of the dominant bacteria under the phylum and genus classification in the taxonomic information are obtained according to the abundance matrix of the operable classification unit, the Shannon index, the Chao1 index, the Observed species and the PD Whole tree of the bacterial community under different nitrogen levels are taken as Alpha diversity indexes based on the relative abundance graph, and the box plots are drawn respectively, the significant differences of different operable classification unit sets are calculated according to the box plots, and the bacterial community richness and bacterial diversity are obtained; The non-weighted UniFrac distance and the weighted UniFrac distance are calculated based on the phylogenetic relationship between the operable classification unit sets, the Beta diversity index heat map is drawn, and the optimal nitrogen application quantity gradient is obtained according to the diversity index heat map, wherein the operable classification unit set includes the common operable classification unit, the unique operable classification unit, the most operable classification unit and the least operable classification unit.

5. The method for regulating the rhizosphere microbial community of highland barley according to claim 1, characterized in that, The method for obtaining the dominant species data comprises the following steps: Principal component analysis is performed based on the abundance matrix of the operable classification unit, the difference contribution rate of the first principal component and the second principal component to the samples is obtained, a clustering circle is added to the sample in the difference contribution rate graph, the similarity and the particularity of the sample community composition are obtained according to the difference trend, and the operable classification unit set with large similarity and particularity is taken as a difference combination; T test is performed on the community composition of the difference combination, the comparison combination with a significance less than 0.05 is screened out according to the difference significance, the bacterial species with different abundance changes in the comparison combination are screened out according to the taxonomic information under the phylum and genus classification, the bacterial species are ranked and standardized according to a preset number, clustering analysis and heat map drawing are performed on the standard bacterial species based on the species and sample levels, and the dominant species data under the taxonomic information is obtained according to the heat map.

6. The method of claim 1, wherein the method is characterized by, The method for obtaining the parameter optimization comprises the following steps: The corresponding nitrogen fertilizer quantity and fertilization period are extracted based on the dominant species data, and a comprehensive objective function is constructed according to the environmental screening contribution proportion, and the comprehensive objective function formula is as follows: wherein is the amount of nitrogen applied , the fertilization cycle , the amount of nitrogen applied , the fertilization cycle is a candidate parameter of the comprehensive objective function, is the relative abundance of the th dominant species, is the total number of dominant species, obtained from the dominant species data, is the contribution proportion of nitrogen fertilizer environment screening, is the influence weight of nitrogen residual error, is the average value of nitrogen distance gradient residual error, is the community linear diversity index, obtained by linear weighting from the Shannon index and the Chao1 index, is the unweighted UniFrac distance, is the weight of the th Beta diversity index, is the actual value of the th Beta diversity index; The probability distribution of the comprehensive objective function is fitted through a Gaussian process regression model according to the nitrogen application quantity, the fertilization period and the sample data, the prediction mean and the prediction standard deviation of the candidate parameters in the comprehensive objective function are obtained, the selection of new candidate parameters is guided through a confidence upper limit function according to the prediction mean and the prediction standard deviation, the prediction value of the comprehensive objective function corresponding to the new candidate parameters is supplemented to the sample set, and the Gaussian process regression model is iteratively updated according to the sample set, until the change amount of the prediction value of the comprehensive objective function in three consecutive iterations is less than 5%, the optimization is stopped, and the optimal nitrogen application quantity and the optimal fertilization period are obtained.

7. The method of claim 1, wherein the method is characterized by, The method for obtaining the regulation result comprises the following steps: Based on the advantage species data, the optimization parameter is obtained, the optimization parameter is divided into each plot of the planting area through the fertilization period, different gradient base fertilizer application rates are set according to the optimization parameter and the optimal nitrogen application gradient, the preset base fertilizer of phosphorus fertilizer and potassium fertilizer is applied to each plot at one time, the highland barley is artificially strip seeded in each plot, the topdressing is set according to the preset base fertilizer at a ratio of 7:3, the jointing stage of the highland barley is taken as the topdressing period, the adjustment sample of the microbial community is obtained based on the planting area through the multi-point mixed sampling method, the new advantage species data and the new abundance data are obtained through the diversity analysis of the operable classification unit according to the adjustment sample, and the new advantage species data and the new abundance data are taken as the regulation result, and the base fertilizer application rate includes the base fertilizer nitrogen gradient lower than the optimal amount by 10%, the optimal amount and higher than the optimal amount by 10%.

Images