Explore The Underlying Mechanisms Of Microbial Ecology Associated With Fusarium Wilt Base On The Development Of Multiple New Omics-data Analysis Tools

With the development of omics technologies such as metagenomics and metabolomics,the analysis workflow was constantly updated in recent years.Selecting an appropriate method for multi-omics data mining is important for researchers.It is necessary to testing,integration,and re-optimization the omics analysis methods for efficient mining of omics data.The advancement of omics technology has greatly promoted the study of soil microbial ecology.Soil microbial ecology associated with soil-borne diseases was one of the valuable study topics.Fusarium wilt,as typical representatives of soil-borne diseases,was often used for study the relationship between soil microbial ecology and soil-borne diseases.The deterioration of soil microbial community is one of the substantial factors in the outbreak of Fusarium wilt.However,lots of associated studies present quite different results.Search for the general characteristics of the diseased microbial community is crucial for subsequent exploration of the mechanism of the diseased microbial community.Soil microbial community changes are closely related to soil small molecule metabolic components.Metabolomics techniques have facilitated the process of microbial community formation in the diseased soil of Fusarium wilt.It is of great necessity to find key metabolites that mediate the formation of diseased microbial communities for preventing the deterioration of microbial communities.The disease-suppressing soil provides a principal basis for the biological control of soil-borne diseases.Interactions between root exudates and microorganisms is of critical importance in the formation of disease-suppressing soils.The maturity of metagenomics and non-targeted metabolomics technologies has promoted the understanding of both the formation and the function in disease-suppressing soils.Therefore,this paper firstly optimized the common omics analysis methods in soil microbial ecology research,and developped more efficient omics analysis pipelines;secondly,these optimized methods were used for relevant microbial community characteristics of Fusarium wilt disease;Finally,the formation and functional characteristics of diseased/disease-suppressed soil were deeply excavated.The main results are presented as follows:1.To better evaluation and integration of existing methods for omics analysis,four omics analysis tools（Easy Amplicon,gg Cluster Net,Easy Metabolome and Easymeta Pro）were developed.Easy Amplicon provides the mining of microbiome sequence data,with installed and used on multiple platforms（Windows,Mac,Linux）freely.The content of analysis mainly includes raw data processing,community diversity analysis,multivariate statistics and modeling,graph drawing,etc.Overall,it has the advantages of faster analysis speed,fruitful and efficient methods.gg Cluster Net provides the great mass of content required in microbiome network analysis,including:correlation matrix calculation,network construction,network and node attribute analysis,network layout and visualization process.gg Cluster Net was advantaged in the modular analysis and contains ten layout algorithm for visualization the modularity relationships of biological Networks.Easy Metabolome provides efficient mining of untargeted metabolomics and the contents mainly include:data normalization,metabolite annotation,differential analysis visualization,biomarker analysis,metabolic network analysis,and exploration of the interaction between metabolites and microbiome.Easymeta Pro provides mining of metagenomics data processing,and the analysis contents mainly include raw data quality control,sequence assembly,species and function annotation,species data mining,functional data mining and species-function joint analysis,etc.Easymeta Pro provides a framework that integrating species data and functional data and thus analysis of relationships between specific microbes and functions.2.By integrating Fusarium wilt-related studies,37 independent studies with the samples of 1105 bacterial sequences data and 26 independent studies with the samples of444 fungal sequences data were collected.The diversity of bacterial community revealed diseased soil microbiomes harbored higher abundances of Proteobacteria,Actinobacteria and Acidobacteria,while healthy soil microbiome contained more Firmicutes and Bacteroidetes,Choroflexi and Gemmatimonadetes;The diversity of fungal community showed diseased soil microbiomes harbored higher abundances of Ascomycota and Basidiomycota,while healthy soils microbiome contained more Mortierellomycota and Mucoromycota High-accuracy classification models by machine learning could be constructed with bacterial/fungal community at the OTU level.The accuracies of bacterial and fungal community model were 95%and 92%,respectively.Thirteen independent datasets containing both diseased and healthy samples were sued to validated the bacterial community model and showed the mean accuracy was 94.46%.six independent datasets including diseased and healthy samples were used to validated the fungal community model and showed the mean accuracy was 93.05%.Further validation with five groups of sampling data indicated that the mean accuracy of the bacterial model was 90%,while the fungal model was 87.2%.forty-five bacterial characteristic OTUs and 40 fungal characteristic OTUs were obtained by in-depth analysis of models.Thirty-three of the 45bacterial signature OTUs were more abundant in diseased soils,and only 12 OTUs were more abundant in healthy soils.Twenty-six of the 40 fungal characteristic OTUs were more abundant in diseased soils,and only 14 OTUs were more abundant in healthy soils.According to the characteristic bacterial OTU co-occurrence network analysis,more connections,higher network average degree,centralization-closeness and clustering coefficient values could be found in healthy networks.The fungal characteristic OTU network indicated that more connections,higher network average degree,centralization-closeness,and clustering coefficient in diseased network.3.By integrating metadata analysis,sampling from different fields and soil simulation experiments,the bacterial community assembly process associated with Fusarium wilt disease h was investigated.The results showed that the diseased bacterial community assembly process was mainly dominated by deterministic processes.The untargeted metabolomics was employed to analyze the metabolomic characteristics of field samples.Results showed the deterministic assembly process which associated with diseased rhizosphere microbiomes was significantly correlated to five metabolites（tocopherol acetate,citrulline,galactitol,octadecylglycerol and behenic acid）.And these metabolites drove the deterministic assembly process of rhizosphere microbial community was verified by soil simulation experiments.The soils conditioned by these metabolites resulted in severe Fusarium wilt occurrence（average occurrence:60%）.The shotgun metagenome sequencing was used to reveal the mechanisms of metabolite-mediated process after soil conditioning.Results showed that the ability of autotoxin degradation（eg.nitrotoluene degradation,arachidonic acid metabolism,polycyclic aromatic hydrocarbon degradation,naphthalene degradation,xylene degradation,toluene degradation,styrene degradation and dioxin degradation）were significantly down-regulated by special metabolite application.Those down-regulated pathways were primarily mediated though a feature microbial group（FM1）including Bradyrhizobium,Streptomyces,Variovorax,Pseudomonas and Sphingomonas.Furthermore,the metabolism pathway of small-molecule sugars（fructose and mannose metabolism）,organic acids and amino acids（citrate cycle metabolism;fatty acid degradation;pyruvate metabolism;valine,leucine and isoleucine degradation;cysteine and methionine metabolism;glycine,serine and threonine metabolism）were significantly up-regulated by metabolites application.Those up-regulated pathways were primarily mediated though another feature microbial group（FM2）including Anaeromyxobacter,Bdellovibrio,Conexibacter,Gemmatimonas and Flavobacterium.The continuous cropping experiment also showed that the autotoxin degradation metabolism was significantly down-regulated in diseased soil,while the metabolic pathways of small molecular carbohydrates,organic acids and amino acids were significantly enriched.4.The disease-suppressive rhizosphere soil of cucumber Fusarium wilt was formed at the end of the 8^th generation(the disease incidence of 8^th generation was 15%,and the control was 100%).We found the key microbes（Bacillus and Sphingomonas）were enriched with the increase of continuous cropping times by microbiome sequencing data.At the end of the 8^th generation,the relative abundances of Bacillus and Sphingomonas in the rhizosphere were 40%and 10%,respectively.Isolators with six Bacillus and five Sphingomonas were isolated from 8^th generation rhizosphere soil.Further experiment showed key microbes（Bacillus and Sphingomonas）could induce ROS（mianly OH·）burst in root and thus protect cucumber reject Fusarium wilt.Moreover,the content of ROS was significant increase in the cucumber root planted in disease-suppressing soil.More threonic acid and lysine were found in cucumber root exudate by Fusarium oxysporum infection and could be the main factors to recruit the key microbes.Shotgun metagenomics sequencing showed the key microbes could activate pathways（Two-component system,Bacterial secretion system and Flagellar assembly）to induce ROS burst in root.In summary,we developed four tools（Easy Amplicon,gg Cluster Net,Easy Metabolome and Easymeta Pro）by optimized by amplicon data analysis,non-targeted metabolomics and metagenomics data analysis process.By using such tools,we characterized the diseased/healthy microbial community of Fusarium wilt,and explored the formation and characterization of the diseased microbial community and the cucumber disease-suppressive microbial community.This paper explored the diseased/healthy microbial community of Fusarium wilt,and provides important theoretical basis for the biological control of soil-borne diseases in the future.