A method and application for analyzing antiviral and virus-promoting bacteria based on metatranscriptome sequencing

Abstract

This invention belongs to the field of biodetection technology and discloses a method and application for analyzing antiviral and proviral bacteria based on metatranscriptome sequencing. Based on metatranscriptome sequencing, this invention utilizes different viral and bacterial marker genes to analyze and calculate the RPM values ​​of viral and bacterial species. Using the RPM values ​​of viruses and bacteria, Spearman correlation analysis is performed to obtain a correlation table showing the interactions between all bacteria and viruses. This invention overcomes the problem of incomplete bacterial genomes in metatranscriptome sequencing, enabling the detection of all viruses and bacteria in a single mosquito sample. The antiviral and proviral effects of bacteria are detected by the correlation between bacterial rpoB genes and viral RdRp load. Furthermore, the large mosquito sample size involved in this invention enhances the statistical reliability and ecological universality of the study, resulting in highly representative research results.

Description

Technical Field

[0001] This invention relates to the field of biological detection technology, specifically to a method and application for analyzing antiviral and virus-promoting bacteria based on metatranscriptome sequencing. Background Technology

[0002] In traditional techniques, the search for antiviral and proviral bacteria primarily relies on laboratory screening and functional validation. These methods typically involve bacterial isolation, purification, virus-host interaction studies, and viral infection experiments. The process usually involves isolating bacteria from the natural environment (such as soil, water, and gut microbiota) or specific hosts (such as healthy or diseased animals), culturing bacterial strains, obtaining candidate strains capable of producing antiviral or proviral active substances, screening for bacteria or their metabolites that can inhibit or promote viral infection or replication using in vitro experiments, and finally performing in vitro or in vivo validation.

[0003] Currently, with the popularization and development of high-throughput sequencing technology, research methods for discovering bacteria with antiviral or proviral effects have shifted from traditional experimental methods to modern methods integrating high-throughput data analysis and functional validation. These new methods offer higher sensitivity and resolution, enabling efficient screening of bacteria with specific functions in complex microbial communities. First, 16S rRNA gene sequencing analysis compares the microbial community structure of healthy and virus-infected groups, screening for bacteria potentially associated with antiviral or proviral effects. The positive and negative correlations between changes in the abundance of specific bacteria and viral infection are also investigated. Second, metagenomic sequencing can directly obtain the genomic information of all microorganisms in environmental samples, used to predict the functional potential of microorganisms and analyze the association between functional genes in bacterial communities (such as genes related to the synthesis of antiviral active substances) and viral infection. Metatranscriptomics analysis of the gene expression profiles of microbial communities in samples reveals the dynamic mechanisms of bacterial-virus interactions.

[0004] However, metagenomic sequencing typically involves complex functional genes and pathways, and the characteristics of metagenomic transcriptomics sequencing often result in the inability to identify complete bacterial genomes, leading to fewer bacterial species being discovered, and usually focusing only on disease-related viruses and related antiviral and proviral bacteria. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method for analyzing antiviral and proviral bacteria based on metatranscriptome sequencing and its application.

[0006] This invention takes arboviruses (viruses that can replicate in the cells of blood-sucking arthropods and vertebrates, infecting humans or other vertebrates, and are transmitted from infected hosts to susceptible hosts via vectors such as mosquitoes, ticks, sandflies, or midges) as its starting point. Based on metatranscriptome sequencing technology, it provides a comprehensive and efficient method for exploring all antiviral and proviral bacteria in mosquitoes. This method can more comprehensively explore the interaction relationships between all viruses and all bacteria in mosquitoes and screen for potential antiviral and proviral bacteria, covering both pathogenic and non-pathogenic viruses, thereby revealing a broader range of microbial community dynamics.

[0007] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0008] In a first aspect, the present invention provides a method for analyzing antiviral and proviral bacteria based on metatranscriptome sequencing, comprising the following steps:

[0009] (1) Collect mosquito samples from no less than 25 provinces and cities, extract nucleic acids, construct transcriptome libraries and perform metatranscriptome sequencing;

[0010] (2) The data from the metatranscriptome sequencing were quality controlled, assembled, and annotated with the viral sequences by alignment with the NR database; the viral sequences were translated and proteins were deduplicated using ORFfinder, and RdRp-related proteins were obtained by alignment with the RdRp database; the RdRp-related proteins were aligned with the reference proteins by multiple sequence alignment, and a phylogenetic tree was constructed after sequence pruning; the corresponding annotated viruses were identified by the virus species confirmed by the phylogenetic tree, mapping quantification was performed, and RPM values ​​were calculated to obtain the identified virus species;

[0011] (3) After quality control of the transcriptome library, ribosomal rRNA is removed, host RNA is removed, and the library is mixed and assembled to obtain a contig sequence. The contig sequence is aligned with the rpoB gene, translated, and deduplicated. It is then aligned with the rpoB gene again to obtain rpoB-related protein sequences. The rpoB-related protein sequences are subjected to HMM iterative search to obtain protein sequences with rpoB protein domains. The protein sequences with rpoB protein domains are compared and annotated with the NR database to screen for bacterial-related proteins. The bacterial-related proteins are compared with reference proteins for multiple sequence alignment, and after sequence pruning, a phylogenetic tree is constructed. The bacterial species identified by the phylogenetic tree are identified, and their corresponding annotated bacteria are determined. Mapping quantification is performed, and RPM values ​​are calculated to obtain the identified bacterial species.

[0012] (4) The nucleic acid sequences of the virus and the bacteria are cleaned, mapped using nohost reads after removing the host, and the RPM values ​​of the virus and bacteria are calculated based on the noRNA reads data after removing the ribosomes.

[0013] (5) Combine the RPM values ​​of viruses and bacteria, screen for bacterial and viral species with a positive value of 20 or more, and obtain a correlation table showing the interaction relationships between all bacteria and viruses through Spearman correlation analysis.

[0014] The method of this invention detects the antiviral and proviral effects of bacteria by correlating specific gene loads. Compared with existing methods, it can screen out all potential antiviral and proviral bacteria. This allows for the identification of all viral and bacterial species in a sample before discovering antiviral and proviral bacteria.

[0015] In a preferred embodiment of the method for analyzing antiviral and proviral bacteria based on metatranscriptome sequencing of the present invention, in step (2), the raw data of the metatranscriptome sequencing is quality controlled by FastP software to obtain clean reads, and ribosomal rRNA is removed using Bowtie2 software to obtain nohost reads; host-related sequences are removed using Bowtie2 software to obtain nohost reads; the clean reads of a single transcriptome library are assembled using Megahit software to obtain contig sequences, and then the assembled contig sequences are annotated using Diamond software through the NR database, and sequences annotated as viruses are selected.

[0016] The virus-related sequences were translated using ORFfinder, and protein sequences with a fragment size of at least 200 bp were selected and compared with the RdRp database to obtain RdRp-related proteins.

[0017] The obtained RdRp-related proteins were deduplicated using cd-hit software with a threshold of 0.9, and then aligned with NCBI reference proteins using multiple sequence alignment. The aligned sequences were pruned using TrimAL software and used for phylogenetic tree construction. For virus species confirmed by the phylogenetic tree, their corresponding contig sequences were identified, the sequences were cleaned, and further quantified using nohostreads data via bowtie2 software. The RPM value was calculated based on the noRNAreads data after ribosome removal.

[0018] As a further preferred embodiment of the method for analyzing antiviral and proviral bacteria based on metatranscriptome sequencing described in this invention, when the screening annotation is a virus-related sequence, preliminary screening is performed based on the keyword "virus", and rows containing the keyword "Ortervirales" are deleted to exclude retroviruses.

[0019] As a preferred embodiment of the method for analyzing antiviral and proviral bacteria based on metatranscriptome sequencing according to the present invention, in step (3), the nohost reads of the transcriptome library are mixed and assembled according to the mosquito genus using Megahit software to obtain a mixed contig sequence; the mixed contig sequence is compared with the bacterial rpoB reference gene in GTDB using Diamond software, and the contig sequence that matches the reference rpoB gene is extracted.

[0020] Bacterial rpoB-related sequences were translated using ORFfinder, and protein sequences with a fragment size of at least 200 bp were selected. The sequences were then deduplicated using cd-hit software with a threshold of 0.98. The deduplicated proteins were then compared with the bacterial rpoB reference gene, and the protein sequences that were aligned with the reference rpoB gene were extracted.

[0021] The obtained rpoB gene-related proteins were searched for protein sequences with rpoB protein domains through two iterative searches using the rpoB gene hmm profile of HMM and GTDB.

[0022] The obtained protein sequences with the rpoB protein domain were compared and annotated with the NR database using Diamond software, and bacterial-related proteins were screened out.

[0023] The bacterial-associated protein was aligned with the rpoB reference protein of GTDB using MAFFT software. The aligned sequences were pruned using TrimAL software and used for phylogenetic tree construction.

[0024] After merging and confirming the final bacterial species, the nohost reads data were mapped and quantified using bowtie2 software, and the RPM value was calculated based on the noRNA reads data after ribosome removal.

[0025] As a further preferred embodiment of the method for analyzing antiviral and proviral bacteria based on metatranscriptome sequencing described in this invention, bacterial protein sequences confirmed by the phylogenetic tree are merged at the species level; the merging criteria include:

[0026] i. Their positions on the phylogenetic tree indicate that they belong to the same type of bacteria;

[0027] ii. Use Diamond MakeDB to construct a database of the protein sequences in “i”, and perform BLASTP alignment of the protein sequences in “i” with this database; find their corresponding CDS sequences, use MakeBlastDB to construct a nucleic acid database, and perform BLASTN alignment of the CDS with this database. If the identity threshold of BLASTP or BLASTN meets 98% and the hit length meets 100bp, they are considered to be the same species.

[0028] iii. After washing the CDS of the bacterial species merged in steps i and ii, nohost reads were mapped using Bowtie2 software, and RPM values ​​were calculated based on the noRNA reads data after ribosome removal. Spearman correlation analysis was performed based on the RPM values ​​to calculate whether there was a correlation between different bacterial species. For bacterial species with a correlation coefficient greater than or equal to 0.8 and a p-value less than 0.05, their positions on the phylogenetic tree were further confirmed to determine whether the high correlation was due to belonging to the same species or a genuine interaction. CDS meeting the above conditions were merged into the same bacterial species.

[0029] As a preferred embodiment of the method for analyzing antiviral and proviral bacteria based on metatranscriptome sequencing according to the present invention, in step (5), the "from" and "to" columns in the table represent different bacterial or viral species, respectively; "cor" represents the correlation coefficient; the "p" value is used to measure the statistical significance of the results, and the results are considered statistically significant when the "p" value is <0.05; the threshold for the "p" value is less than 0.05. When the correlation coefficient between bacteria and viruses is less than 0, the bacteria are potential antiviral bacteria, and the smaller the "cor" correlation coefficient, the greater the correlation. When the correlation coefficient between bacteria and viruses is greater than 0, the bacteria are potential proviral bacteria, and the larger the "cor" correlation coefficient, the greater the correlation.

[0030] Secondly, the present invention provides a method for analyzing antiviral and virus-promoting bacteria, comprising the following steps:

[0031] Obtain the metatranscriptome sequencing data of the sample to be tested; use the above-mentioned method based on metatranscriptome sequencing to analyze antiviral and proviral bacteria to obtain the RPM values ​​of viruses and bacteria in the sample to be tested; merge the RPM values ​​of viruses and bacteria, screen for bacterial and viral species with a positive library of 20 or more, and obtain a correlation table showing the interaction relationships between all bacteria and viruses through Spearman correlation analysis to obtain the types of antiviral and proviral bacteria.

[0032] Thirdly, the present invention provides a virus-promoting bacterium, including Cutibacterium, Flavobacterium, Methylobacterium, Commonas, Pseudomonas, Wolbachia, and Escherichia.

[0033] Fourthly, the present invention provides an antiviral bacterium, including unclassified CAIZLB01, unclassified Patescibacteria, Cutibacterium, unclassified JAAZZY01, unclassified UBA10799, Pseudomonas, Wolbachia, and Escherichia.

[0034] Fifthly, the present invention includes methods for analyzing antiviral and proviral bacteria based on metatranscriptome sequencing, methods for analyzing antiviral and proviral bacteria, said proviral bacteria, and the application of said antiviral bacteria in any of the following fields:

[0035] i. Analyze antiviral and virus-promoting bacteria;

[0036] ii. Detect or develop antiviral probiotics;

[0037] iii. To prepare drugs for the treatment or adjuvant treatment of viral diseases;

[0038] iv. Preparation of antiviral bacterial antibiotics;

[0039] V. Microbiome regulation of proviral bacteria.

[0040] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0041] This invention first identifies viral species through a process involving quality control, assembly, NR database alignment and annotation, viral sequence screening, translation, protein deduplication, RdRp database alignment, multiple sequence alignment, sequence pruning, and phylogenetic tree construction. Bacterial species are identified through a process involving quality control, removal of ribosomal rRNA, host removal, mixed assembly, rpoB gene alignment, translation, protein deduplication, rpoB gene alignment, HMM iterative search, NR database alignment and annotation, bacterial sequence screening, multiple sequence alignment, sequence pruning, and phylogenetic tree construction. The nucleic acid sequences of both viruses and bacteria are washed, and mapping is performed using nohost reads after host removal. The RPM values ​​of the viral and bacterial species are calculated based on the nohost reads data after ribosomal removal. Using the RPM values ​​of viruses and bacteria, Spearman correlation analysis is performed to obtain a correlation table, which shows all the interactions between bacteria and viruses. This invention, based on metatranscriptome data, utilizes different viral and bacterial marker genes to overcome the incompleteness of bacterial genomes in metatranscriptome sequencing. It can identify all viruses and bacteria in a single mosquito sample and detect their antiviral and proviral effects by analyzing the correlation between bacterial rpoB genes and viral RdRp load. Furthermore, the large sample size of mosquitoes involved in this invention, originating from nearly 30 provinces and cities across China, enhances the statistical reliability and ecological universality of the research, resulting in highly representative findings. Attached Figure Description

[0042] Figure 1 This is a schematic diagram of the technical route of the present invention;

[0043] Figure 2 The phylogenetic tree constructed for PhyML;

[0044] Figure 3 An evolutionary tree constructed for the iqtree software;

[0045] Figure 4 This is a visualization of the correlation between bacteria and viruses. Spearman correlation analysis was performed based on RPM values, where cor represents the correlation coefficient. The p-value is used to measure the statistical significance of the results; a p-value < 0.05 is considered statistically significant. The p-value threshold is less than 0.05. When the correlation coefficient between bacteria and viruses is less than 0, the bacteria are potential antiviral bacteria, and the smaller the cor correlation coefficient, the stronger the correlation. When the correlation coefficient between bacteria and viruses is greater than 0, the bacteria are potential virus-promoting bacteria, and the larger the cor correlation coefficient, the stronger the correlation. Different colors in the graph represent the magnitude of the correlation coefficient: red indicates a positive correlation (correlation coefficient between virus-promoting bacteria and viruses), and blue indicates a negative correlation (correlation coefficient between antiviral bacteria and viruses). Different shapes represent different mosquito species. Detailed Implementation

[0046] To better illustrate the objectives, technical solutions, and advantages of this invention, the invention will be further described below with reference to specific embodiments. Those skilled in the art should understand that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0047] Unless otherwise specified, the experimental methods used in the examples are conventional methods; the materials and reagents used are commercially available unless otherwise specified.

[0048] Example:

[0049] 1. Raw data

[0050] Mosquito samples were collected from 24 provinces, autonomous regions, and municipalities, including Anhui, Heilongjiang, Beijing, Yunnan, Guangdong, Gansu, Guangxi, Guizhou, Henan, Hebei, Hainan, Inner Mongolia, Hunan, Jiangxi, Tibet, Zhejiang, Ningxia, Shandong, Shaanxi, Sichuan, Shanghai, Shanxi, Hubei, and Xinjiang. Nucleic acid was extracted from each sample. Transcriptome libraries were constructed and metagenomic sequencing was performed by BGI Genomics Co., Ltd.

[0051] 2. Virus discovery

[0052] (1) The raw data from metatranscriptome sequencing were quality controlled using FastP software to obtain clean reads. Ribosomal rRNA was removed using Bowtie2 software to obtain nohost reads. Host-related sequences were removed using Bowtie2 software to obtain nohost reads.

[0053] (2) Clean reads from a single transcriptome library were assembled using Megahit software to obtain contig sequences. Then, Diamond software was used to annotate the assembled contig sequences using the NR database and to screen out sequences annotated as viruses. Specifically, preliminary screening was performed based on the keyword "virus" and rows containing the keyword "Ortervirales" were deleted to exclude retroviruses.

[0054] (3) Translate virus-related sequences using ORFfinder, screen for protein sequences with a fragment size of at least 200bp, and align them with the RdRp database (diamond software) to obtain RdRp-related proteins.

[0055] (4) The obtained RdRp-related proteins were deduplicated using cd-hit software with a threshold of 0.9, and then subjected to multiple sequence alignment (MAFFT) with reference proteins downloaded from NCBI. The aligned sequences were pruned using TrimAL software and used for phylogenetic tree construction. The phylogenetic tree constructed using PhyML is shown below. Figure 2After identifying the virus species through phylogenetic tree analysis, their corresponding contig sequences were determined. These sequences were then cleaned, and nohost reads were further mapped and quantified using bowtie2 software. The RPM value was calculated based on the noRNA reads data after ribosome removal.

[0056] 3. Bacterial discovery

[0057] (1) The nohost reads of the constructed transcriptome library were mixed and assembled according to the mosquito genus using the Megahit software to obtain the mixed contig sequence.

[0058] (2) Then, the contig sequence was compared with the bacterial rpoB reference gene downloaded from GTDB using Diamond software, and the contig sequence of the reference rpoB gene was extracted.

[0059] (3) Bacterial rpoB-related sequences were translated using ORFfinder, and protein sequences with a fragment size of at least 200 bp were selected. The sequences were then deduplicated using cd-hit software with a threshold of 0.98. The deduplicated proteins were then compared with the bacterial rpoB reference gene (diamond software), and the protein sequences that aligned with the reference rpoB gene were extracted.

[0060] (4) The obtained rpoB-related proteins were searched for protein sequences with rpoB protein domains through two iterations using HMM and the rpoB gene hmm profile downloaded from GTDB.

[0061] (5) Subsequently, the obtained protein sequences with the rpoB protein domain were compared and annotated with the NR database using Diamond software, and bacterial-related proteins were screened out, specifically by screening based on the keyword "Bacteria".

[0062] (6) The bacterial-associated protein and the rpoB reference protein downloaded from GTDB were subjected to multiple sequence alignment using MAFFT software. The aligned sequences were pruned using TrimAL software and used for phylogenetic tree construction. The phylogenetic tree constructed using iqtree software is shown below. Figure 3 .

[0063] (7) Bacterial protein sequences confirmed by the phylogenetic tree (the sequences involved in the phylogenetic tree have a correct tree structure) are merged at the "species" level according to three criteria:

[0064] i. Their positions on the phylogenetic tree indicate that they belong to the same type of bacteria;

[0065] ii. Use Diamond MakeDB to construct a database of the protein sequences in “i”, and perform BLASTP alignment of the protein sequences in “i” with this database; find their corresponding CDS sequences, use MakeBlastDB to construct a nucleic acid database, and perform BLASTN alignment of the CDS with this database. If the identity threshold of BLASTP or BLASTN meets 98% and the hit length meets 100bp, they are considered to be the same species.

[0066] iii. After washing the CDS of the bacterial species merged in the first two steps, nohost reads were mapped using bowtie2 software, and RPM values ​​were calculated based on the noroRNA reads after ribosome removal. Spearman correlation analysis was performed based on the RPM values ​​to calculate whether there was a correlation between different bacterial species. For bacterial species with a correlation coefficient greater than or equal to 0.8 and a p-value less than 0.05, their position on the phylogenetic tree was further confirmed to determine whether the high correlation was due to belonging to the same species or a genuine interaction. CDS meeting the above conditions were merged into the same bacterial species.

[0067] (8) The final bacterial species confirmed by merging were mapped and quantified using nohost reads data through bowtie2 software, and the RPM value was calculated based on the noRNA reads data after ribosome removal.

[0068] 4. Antiviral / proviral analysis

[0069] The quantitative results of viruses and bacteria (i.e., tables of viral RPM values ​​and bacterial RPM values) were merged, and bacterial and viral species with a positive value of 20 or more were screened. The correlation table was obtained by Spearman correlation analysis.

[0070] Table 1. Correlation between bacteria and mosquito-associated virus groups (sum of rpm of all mosquito-associated viruses, reflecting the overall correlation).

[0071] mosquito species Bacteria Virome heart p Culex_tritaeniorhynchus unclassified CAIZLB01 mosquito-related virome -0.58017 0.002267 Culex_tritaeniorhynchus unclassified Patescibacteria mosquito-related virome -0.46697 0.029778 Culex_tritaeniorhynchus Cutibacterium mosquito-related virome 0.251267 0.019845 Mosquito_pipiens Flavobacterium mosquito-related virome 0.427826 0.038162 White_armored Cutibacterium mosquito-related virome 0.473684 0.042171 Culex_tritaeniorhynchus Methylobacterium mosquito-related virome 0.538462 0.049989 Mosquito_pipiens Comamonas mosquito-related virome 0.581739 0.003392 White_armored Pseudomonas mosquito-related virome 0.637363 0.022289

[0072] Table 2 Correlation between bacteria and mosquito-associated viruses

[0073] mosquito species Bacteria Virus heart p Anopheles_sinensis Cutibacterium Mosquito.orthomyxo.like.virus.YN3 -0.88811 9.17E-05 Mosquito_pipiens unclassified JAAZZY01 Mosquito.hepe.like.virus.GD2 -0.7972 0.003161 White_armored unclassified UBA10799 Mosquito.narna.like.virus.SC1 -0.75758 0.015921 Culex_tritaeniorhynchus unclassified JAAZZY01 Mosquito.luteo.like.virus.YN2 -0.74545 0.011866 Culex_tritaeniorhynchus Cutibacterium Mosquito.mono.like.virus.SHX3 -0.67133 0.020442 Mosquito_pipiens unclassified JAAZZY01 Mosquito.narna.like.virus.ZJ1 -0.50762 0.017099 Mosquito_pipiens Pseudomonas Mosquito.orthomyxo.like.virus.SHX2 -0.5049 0.04079 Mosquito_pipiens Wolbachia Mosquito.picorna.like.virus.YN4 -0.40441 0.020298 Culex_tritaeniorhynchus Escherichia Mosquito.mono.like.virus.SHX3 -0.39271 0.015306 Culex_tritaeniorhynchus Escherichia Mosquito.flavi.like.virus.ZJ2 -0.27159 0.033055 White_armored Escherichia Mosquito.narna.like.virus.SC1 -0.22372 0.031309 White_painted_house Wolbachia Mosquito.bunya.like.virus.HEN1 0.190152 0.010442 White_armored Wolbachia Mosquito.all.like.virus.SHX3 0.362021 0.020569 Mosquito_pipiens Wolbachia Mosquito.mono.like.virus.XZ3 0.371378 0.001944 Mosquito_pipiens Wolbachia Mosquito.hepe.like.virus.ZJ2 0.459439 0.000229 Mosquito_pipiens Escherichia Mosquito.narna.like.virus.SC1 0.718182 0.016799

[0074] This table displays all the interactions between bacteria and viruses. The "from" and "to" columns represent different bacterial or viral species, "cor" represents the correlation coefficient, and the p-value measures the statistical significance of the results. A p-value < 0.05 is generally considered statistically significant. The p-value threshold is less than 0.05. When the correlation coefficient between bacteria and viruses is less than 0, the bacteria are potentially antiviral; when the correlation coefficient is greater than 0, the bacteria are potentially proviral.

[0075] In summary, this invention first identifies viral species through quality control, assembly, NR database alignment and annotation, viral sequence screening, translation, protein deduplication, RdRp database alignment, multiple sequence alignment, sequence pruning, and phylogenetic tree construction. Bacterial species are identified through quality control, ribosomal rRNA removal, host removal, mixed assembly, rpoB gene alignment, translation, protein deduplication, rpoB gene alignment, HMM iterative search, NR database alignment and annotation, bacterial sequence screening, multiple sequence alignment, sequence pruning, and phylogenetic tree construction. The nucleic acid sequences of viruses and bacteria are washed, and mapping is performed using nohost reads after host removal. The RPM values ​​of viral and bacterial species are calculated based on the nohost reads after ribosomal removal. Using the RPM values ​​of viruses and bacteria, Spearman correlation analysis is performed to obtain a correlation table, which shows the interactions between all bacteria and viruses. `cor` represents the correlation coefficient, and a p-value < 0.05 is considered statistically significant. The p-value threshold is less than 0.05. When the correlation coefficient between bacteria and viruses is less than 0, the bacteria are potential antiviral bacteria, and the smaller the correlation coefficient, the greater the correlation. When the correlation coefficient between bacteria and viruses is greater than 0, the bacteria are potential proviral bacteria, and the larger the correlation coefficient, the greater the correlation.

[0076] This method allows for the exploration of all viral and bacterial interactions within mosquitoes, and the screening of all potential antiviral and proviral bacteria, including disease-associated viruses and other viruses. It also includes known and unknown viral and bacterial species; for example, *Wolbachia*, *Escherichia*, and *Pseudomonas* exhibit antiviral activity against some viruses, while *Wolbachia*, *Escherichia*, and *Stenotrophomonas* have proviral effects against others. This method significantly reduces the waste of human, material, and financial resources for researchers and greatly expands the database of antiviral and proviral bacteria. Furthermore, the discovery of antiviral and proviral bacteria is crucial for studying viral infection mechanisms and developing novel treatment strategies, providing new perspectives on understanding the occurrence and development of infectious diseases. Practical applications include: developing antiviral probiotics as adjunctive therapy for viral diseases; and designing antibiotics or microbiome modulation technologies targeting proviral bacteria to reduce the severity of viral infections. In this invention, using the above method, 11 bacteria with antiviral or proviral effects were discovered in a library of over 4000 individual mosquitoes from five mosquito species. These bacteria include: Cutibacterium, Flavobacterium, Methylobacterium, Commonas, Pseudomonas, Wolbachia, Escherichia, unclassified CAIZLB01, unclassified Patescibacteria, unclassified JAAZZY01, and unclassified UBA10799. Five of these bacteria showed a positive correlation with mosquito-associated viromes, including Cutibacterium, Flavobacterium, Methylobacterium, Commonas, and Pseudomonas; two showed a negative correlation, including unclassified... CAIZLB01 and unclassified Patescibacteria; two bacteria have proviral activity, including Wolbachia and Escherichia; and six bacteria have antiviral activity, including Cutibacterium, Escherichia, Pseudomonas, Wolbachia, unclassified JAAZZY01, and unclassified UBA10799.

[0077] This invention, based on metatranscriptome data, first identifies all viral and bacterial species in the data according to their respective marker genes, and then cleans the corresponding nucleic acid sequences. Based on the correlation of marker gene load, this invention calculates the load of each viral and bacterial species in each library, and uses Spearman correlation analysis to calculate the correlation coefficient. This invention utilizes different viral and bacterial marker genes to overcome the problem of incomplete bacterial genomes in metatranscriptome sequencing, enabling the discovery of all viruses and bacteria in a single mosquito sample. The antiviral and proviral effects of bacteria are detected by the correlation between bacterial rpoB gene load and viral RdRp load. Furthermore, the large number of mosquito samples involved in this invention, from nearly 30 provinces and cities across China, enhances the statistical reliability and ecological universality of the research, resulting in highly representative findings.

[0078] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the scope of protection of the present invention. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the essence and scope of the technical solutions of the present invention.

Claims

1. A method for analyzing antiviral and virus-promoting bacteria based on metatranscriptome sequencing, characterized in that, Includes the following steps: (1) Take mosquito samples from no less than 25 provinces and cities, extract nucleic acids, construct transcriptome libraries and perform metatranscriptome sequencing; (2) The data of the metatranscriptome sequencing were quality controlled, assembled and annotated with the viral sequence by comparison with the NR database; the viral sequence was translated and proteins were deduplicated by ORFfinder, and RdRp-related proteins were obtained by comparison with the RdRp database; the RdRp-related proteins were compared with the reference protein by multiple sequence alignment, and a phylogenetic tree was constructed after sequence pruning; The virus species identified through the phylogenetic tree are identified by finding their corresponding annotated viruses, performing mapping quantification, calculating RPM values, and obtaining the identified virus species. (3) After quality control of the transcriptome library, ribosomal rRNA is removed, host RNA is removed, and the library is mixed and assembled to obtain a contig sequence; the contig sequence is aligned with the rpoB gene, translated, and deduplicated, and then aligned with the rpoB gene again to obtain rpoB-related protein sequences; the rpoB-related protein sequences are subjected to HMM iterative search to obtain protein sequences with rpoB protein domains; the protein sequences with rpoB protein domains are compared and annotated with the NR database to screen for bacterial-related proteins; the bacterial-related proteins are compared with reference proteins for multiple sequence alignment, and after sequence pruning, a phylogenetic tree is constructed. The bacterial species identified through the phylogenetic tree are then identified by their corresponding annotation bacteria. Mapping is performed to quantify the bacteria, and the RPM value is calculated to obtain the identified bacterial species. (4) The nucleic acid sequences of the virus and the bacteria are cleaned, mapped using nohost reads after removing the host, and the RPM values ​​of the virus and bacteria are calculated based on the noRNA reads data after removing the ribosomes. (5) Combine the RPM values ​​of viruses and bacteria, screen for bacterial and viral species with a positive value of 20 or more, and obtain a correlation table showing the interaction relationships between all bacteria and viruses through Spearman correlation analysis.

2. The method for analyzing antiviral and virus-promoting bacteria based on metatranscriptome sequencing according to claim 1, characterized in that, In step (2), the raw data of the metatranscriptome sequencing is quality controlled by the FastP software to obtain clean reads. Ribosomal rRNA is removed using the Bowtie2 software to obtain nohost reads. Host-related sequences are removed using the Bowtie2 software to obtain nohost reads. The clean reads of a single transcriptome library are assembled using the Megahit software to obtain contig sequences. Then, the assembled contig sequences are annotated using the Diamond software through the NR database, and sequences annotated as viruses are selected. The virus-related sequences were translated using ORFfinder, and protein sequences with a fragment size of at least 200 bp were selected and compared with the RdRp database to obtain RdRp-related proteins. The obtained RdRp-related proteins were deduplicated using cd-hit software with a threshold of 0.9, and then aligned with NCBI reference proteins using multiple sequence alignment. The aligned sequences were pruned using TrimAL software and used for phylogenetic tree construction. For virus species confirmed by the phylogenetic tree, their corresponding contig sequences were identified, the sequences were cleaned, and further quantified using nohost reads data via bowtie2 software. The RPM value was calculated based on the norosome reads data after ribosome removal.

3. The method for analyzing antiviral and virus-promoting bacteria based on metatranscriptome sequencing according to claim 2, characterized in that, When the filtering annotation is a virus-related sequence, preliminary filtering is performed based on the keyword "virus", and rows containing the keyword "Ortervirales" are deleted to exclude retroviruses.

4. The method for analyzing antiviral and virus-promoting bacteria based on metatranscriptome sequencing according to claim 1, characterized in that, In step (3), the nohost reads of the transcriptome library are mixed and assembled according to the mosquito genus using Megahit software to obtain a mixed contig sequence; the mixed contig sequence is compared with the bacterial rpoB reference gene in GTDB using Diamond software, and the contig sequence of the reference rpoB gene that is aligned is extracted. Bacterial rpoB-related sequences were translated using ORFfinder, and protein sequences with a fragment size of at least 200 bp were selected. The sequences were then deduplicated using cd-hit software with a threshold of 0.

98. The deduplicated proteins were then compared with the bacterial rpoB reference gene, and the protein sequences that were aligned with the reference rpoB gene were extracted. The obtained rpoB gene-related proteins were searched for protein sequences with rpoB protein domains through two iterative searches using the rpoB gene hmm profile of HMM and GTDB. The obtained protein sequences with the rpoB protein domain were compared and annotated with the NR database using Diamond software, and bacterial-related proteins were screened out. The bacterial-associated protein was aligned with the rpoB reference protein of GTDB using MAFFT software. The aligned sequences were pruned using TrimAL software and used for phylogenetic tree construction. After merging and confirming the final bacterial species, the nohost reads data were mapped and quantified using bowtie2 software, and the RPM value was calculated based on the noRNA reads data after ribosome removal.

5. The method for analyzing antiviral and virus-promoting bacteria based on metatranscriptome sequencing according to claim 4, characterized in that, Bacterial protein sequences confirmed by the phylogenetic tree are merged at the species level; the merging criteria include: i. Their positions on the phylogenetic tree indicate that they belong to the same type of bacteria; ii. Use Diamond MakeDB to construct a database of the protein sequences in "i", and perform BLASTP alignment of the protein sequences in "i" with this database; find their corresponding CDS sequences, use MakeBlastDB to construct a nucleic acid database, and perform BLASTN alignment of the CDS with this database. If the identity threshold of BLASTP or BLASTN meets 98% and the hit length meets 100bp, they are considered to be of the same species. iii. After washing the CDS of the bacterial species merged in steps i and ii, the nohost reads were mapped using bowtie2 software, and the RPM value was calculated based on the noRNA reads data after ribosome removal. Spearman correlation analysis was performed based on the RPM value to calculate whether there was a correlation between different bacterial species. For bacterial species with a correlation coefficient greater than or equal to 0.8 and a P value less than 0.05, their position on the phylogenetic tree was further confirmed to determine whether their high correlation was due to belonging to the same species or a real interaction relationship. CDS that meet the criteria are merged into the same bacterial species.

6. The method for analyzing antiviral and virus-promoting bacteria based on metatranscriptome sequencing according to claim 1, characterized in that, In step (5), in the table, the from and to columns represent different bacterial or viral species, respectively; cor is the correlation coefficient; the p-value is used to measure the statistical significance of the result. When the p-value is < 0.05, the result is considered to be statistically significant. The p-value threshold is less than 0.

05. When the correlation coefficient between bacteria and viruses is less than 0, the bacteria is a potential antiviral bacterium, and the smaller the cor correlation coefficient, the greater the correlation. When the correlation coefficient between bacteria and viruses is greater than 0, the bacteria is a potential proviral bacterium, and the larger the cor correlation coefficient, the greater the correlation.

7. A method for analyzing antiviral and virus-promoting bacteria, characterized in that, Includes the following steps: Obtain the metatranscriptome sequencing data of the sample to be tested; use the method described in any one of claims 1-6 to obtain the RPM values ​​of the virus and bacteria in the sample to be tested; merge the RPM values ​​of the virus and bacteria, screen for bacterial and viral species with a positive library of 20 or more, and obtain a correlation table showing the interaction relationships between all bacteria and viruses through Spearman correlation analysis to obtain the antiviral and proviral bacterial species.

8. The application of the method according to any one of claims 1-6 and the method according to claim 7 in any of the following fields, characterized in that, include: i. Analyze antiviral and virus-promoting bacteria; ii. Detect or develop antiviral probiotics.

Images