Glacier potential risk microorganism identification method

Abstract

The invention relates to the technical field of identification methods, in particular to a glacier potential risk microorganism identification method. S11, an antibiotic resistance gene (ARGs) identification method, S12, a metagenome analysis method, S13, a virulence factor (VFs) identification method, S14, a pathogenic drug-resistant bacterium identification method, and S15, movable ARGs analysis. According to the method, by using a metagenome analysis method, microbial communities can be directly and comprehensively analyzed without culture, so that more unknown or difficult-to-culture microbial species can be revealed, and antibiotic resistance genes and a propagation mechanism thereof can be accurately detected by using an ARGs identification method and movable ARGs analysis; an important basis is provided for evaluating potential risks of microorganisms, and the pathogenicity and drug resistance of the microorganisms can be more accurately evaluated by detecting specific genetic markers such as virulence factors and resistance genes of the microorganisms by adopting a VFs recognition method and a pathogenic drug-resistant bacterium recognition method.

Description

Technical Field

[0001] The present invention relates to the technical field of identification methods, and in particular to a method for identifying potential risk microorganisms on glaciers. Background Art

[0002] Current microbial identification methods mainly rely on classical microbiology and molecular biology techniques, including microscopic observation, microbial culture and separation, which play an important role in microbial classification, identification and risk assessment.

[0003] Some current identification methods mainly rely on the cultivation and isolation of microorganisms, but many glacial microorganisms are difficult or uncultivable. Therefore, traditional methods cannot fully reflect the diversity of glacial microorganisms. Even within the range of cultivable microorganisms, traditional methods may ignore certain important microbial species or subtypes due to technical limitations. Although traditional morphological observation and physiological and biochemical characteristic identification are intuitive and rapid, they have low resolution and are difficult to distinguish between microbial species with similar morphology or similar genetic backgrounds. The cultivation and isolation of microorganisms takes a long time and requires complex experimental operations and strict sterile conditions, which consumes a lot of manpower and material resources.

[0004] Therefore, the above-mentioned identification methods mainly rely on the cultivation and isolation of microorganisms, but many glacial microorganisms are difficult or uncultivable. Although traditional morphological observation and physiological and biochemical characteristics identification are intuitive and rapid, the resolution is low and it is difficult to distinguish between microbial species with similar morphology or similar genetic background. At the same time, the cultivation and isolation of microorganisms takes a long time. In this way, a method for identifying glacier potential risk microorganisms with higher recognition efficiency, accuracy and resolution can be designed to better meet the needs of glacier microbial risk assessment. Summary of the invention

[0005] In order to overcome the problems of existing microbial identification methods that mainly rely on the cultivation and isolation of microorganisms, many glacial microorganisms are difficult or uncultivable. Although traditional morphological observation and physiological and biochemical characteristic identification are intuitive and rapid, they have low resolution and the cultivation and isolation of microorganisms takes a long time.

[0006] The technical solution of the present invention is: a method for identifying potential risk microorganisms in glaciers, and the method steps are as follows;

[0007] S11: Identification method of ARGs

[0008] ARGs-OAPv2.0 online analysis tool was used to quickly identify ARGs and quantitatively analyze gene abundance. The webpage address is: https: / / github.com / xiaole99 / ARGs-OAP-v2.0. The analysis process is as follows;

[0009] S111; Pre-screen potential ARG sequences from clean reads using UBLAST;

[0010] S112; Annotate the sequences using UBLAST;

[0011] S113; Obtain the normalized abundance of ARG types and subtypes in each sample according to the annotation results, in ppm;

[0012] S114; Use ARGs-OAP v2.0, web address:

[0013] http: / / bacmet.biomedicine.gu.se / , refer to the BacMet database to annotate and quantify ARGs types and subtypes ( Mao et al., 2023 );

[0014] S115;ARGs abundance was calculated with reference to 16S rRNA according to the following formula (Yin et al., 2018);

[0015]

[0016] Among them, N 16S sequence Mapped reads to the reference database; N ARG―like sequence Reference sequence length; L reads Read length; N 16S sequence 16 S rRNA clean reads number; L 16S sequence 16SrRNA average read length;

[0017] Reads assigned to 16S rRNA by ARG-OAP v2.0;

[0018] S12: Metagenomic Analysis Methods

[0019] MEGAHIT (v1.1.3) (Li et al., 2016) software was used to assemble single samples and mixed samples from the same location and sampling time to obtain contigs. MetaBAT2 (Kang et al., 2019), MaxBin 2 (Wu et al., 2016) and CONCOCT (Alneberg et al., 2014) were used to calculate the correlation between contigs, cluster and bin them, and obtain metagenome assembled genomes (MAGs). dRep (Olm et al., 2017) was used to identify highly similar MAGs groups in samples and obtain non-redundant MAGs sets.

[0020] The abundance of MAGs in the sample was calculated according to the following formula:

[0021]

[0022] Among them, reads mapping percentage refers to the percentage of reads in the sample that are aligned with the non-redundant MAGs set to the total number of reads in the sample, MAG coverage refers to the coverage of MAG in the sample, and total MAG coverage refers to the sum of the coverage of all MAGs in the sample;

[0023] The glacier hosts of ARGs, drug-resistant bacteria, were identified based on the metagenomic assembly method, and the identification steps were as follows;

[0024] S121: First, the sequencing data was quality controlled, and the sequences were assembled and genes were predicted using bioinformatics analysis software. The predicted genes were compared with the NCBI nr database, the functional gene database KEGG and the ARGs database CARD, the pathogenic gene PHI-base database, and the transposon ISfinder database for sequence alignment and functional annotation;

[0025] S122: Construct glacier microbial assembled genomes, i.e., MAGs, screen out microbial assembled genomes containing ARGs in glacier samples, i.e., drug-resistant bacteria, and combine and count the abundance of drug-resistant bacteria carrying the same ARGs type, and then compare the spatial distribution pattern of drug-resistant bacteria in glacier habitats;

[0026] S13: VFs identification method

[0027] The identification steps are as follows:

[0028] S1311: Use the bbwrap.sh script of BBMap (v38.43) software in slow mode to compare with the core VFs sequences in the VFs database (Virulence Factors Database, VFDB) (Chen et al., 2005), namely the nucleic acid sequence VFDB_setA_nt.fas and the protein sequence VFDB_setA_pro.fas, to identify and align the corresponding VFs;

[0029] S1312: Estimate the coverage of the reference sequence in each sample using the in-house script pile.sh with default parameters;

[0030] The steps for VFs classification are as follows;

[0031] S1321: Use the tool GTDB-Tk (v0.3.2) to classify the VFs carried by MAGs based on the VFDB classification database;

[0032] S1322: According to the classification criteria in the VFDB database and the functions of VFs, they are divided into offensive, motile, defensive, immune, toxic proteins, and unknown types;

[0033] S14: Methods for identifying pathogenic and drug-resistant bacteria

[0034] ARGs and VFs are considered to be microbial marker genes associated with health risks. The steps for identifying the coexistence risk of ARGs and VFs are as follows:

[0035] S141: First, predict the open reading frames, i.e., ORFs, in all contigs of each MAG;

[0036] S142: In order to identify potential ORFs belonging to ARGs, BLASTP was used to compare the ORFs with the ARGs database, and the screening conditions were: similarity>80%, the length of its top hit in the database>70%, and all ORFs considered to be potential ARGs sequences were extracted;

[0037] S143: Using the same method, BLASTP was used to align the ORFs with the VFDB protein sequence VFDB_setA_pro.fas, and all ORFs considered as potential VFs sequences were identified;

[0038] S144: MAGs containing both ARGs sequences and VFs sequences are considered ARGs-VFs paired MAGs;

[0039] S145: Calculate the abundance of all ARGs-VFs paired MAGs in each sample and perform species annotation of MAGs;

[0040] S15: Analysis of movable ARGs

[0041] To further estimate the mobility of ARGs, the pathfact pipeline (v.1.0) with default instructions was used to identify MGEs, including plasmids and phages (de Nies et al., 2021), and further co-linear analysis of ARGs and MGEs was performed to determine whether these genes were co-located on the same MAGs.

[0042] Preferably, the metagenomic analysis method can be used to comprehensively analyze the microbial community directly without cultivation, thereby revealing more unknown or difficult-to-cultivate microbial species. The ARGs identification method and mobile ARGs analysis can accurately detect antibiotic resistance genes and their transmission mechanisms, providing an important basis for assessing the potential risks of microorganisms. The VFs identification method and the identification method of pathogenic and resistant bacteria can more accurately assess the pathogenicity and drug resistance of microorganisms by detecting specific genetic markers such as virulence factors and resistance genes of microorganisms. Although it requires certain experimental technology and equipment support, once the corresponding analysis process and database are established, a high-throughput, automated analysis process can be achieved, thereby greatly improving work efficiency and accuracy. It can also reveal the complex relationships and potential risks between microorganisms by analyzing the genetic information and interaction networks of microbial communities.

[0043] As a preference, the parameters in S112 are: >75bp align length, >80% identity and e-value <10 -7 .

[0044] Preferably, the bin_refinement module in the MetaWRAP pipeline (Uritskiy et al., 2018) was used in S12 to screen MAGs based on completeness > 50% and contamination rate < 10%.

[0045] Preferably, the NCBI nr database webpage address in S121 is:

[0046] https: / / www.ncbi.nlm.nih.gov, the webpage address of the functional gene database KEGG is: http: / / www.genome.jp / kegg / , the webpage address of the ARGs database CARD is: http: / / arpcard.mcmaster.ca / ?q=CARD / ontology / 35506, the webpage address of the pathogenic gene PHI-base database is: http: / / www.phi-base.org / , and the webpage address of the transposon ISfinder database is: https: / / www-is.biotoul.fr / .

[0047] Preferably, the download address of the core VFs sequence in S1311 is:

[0048] http: / / www.mgc.ac.cn / VFs / download.htm.

[0049] Preferably, the VFDB database used for VFs classification in S1321 comes from: http: / / www.mgc.ac.cn / cgi-bin / VFs / v5 / main.cgi.

[0050] Preferably, the main functions of the offensive, motile, defensive, immune, toxic proteins and unknown types in S1322 are as follows;

[0051] Offensive: Pathogens carrying offensive VFs can easily invade the host and infect it;

[0052] Motile: Pathogens carrying motile VFs actively attach to ecological niches suitable for their reproduction, and most of them are prone to forming biofilms;

[0053] Defensive type: Pathogens carrying defensive VFs can resist the stimulation of the external environment and increase their drug resistance;

[0054] Immune type: Pathogens carrying immune type VFs can enhance their tolerance to the host immune response;

[0055] Toxic protein: Toxic protein factors can regulate pathogens to secrete specific toxins to attack different tissue cells of the host;

[0056] Unknown type: The specific mechanism of action of unknown type VFs is still unclear.

[0057] As a preference, pathfact pipeline (v.1.0) was chosen in S15 because of its better performance compared with other similar tools.

[0058] Beneficial effects of the present invention:

[0059] 1. This method can conduct a comprehensive analysis of microbial communities directly without cultivation by using metagenomic analysis methods, thereby revealing more unknown or difficult-to-cultivate microbial species. The ARGs identification method and mobile ARGs analysis can accurately detect antibiotic resistance genes and their transmission mechanisms, providing an important basis for assessing the potential risks of microorganisms. The VFs identification method and the identification method of pathogenic and resistant bacteria can more accurately assess the pathogenicity and drug resistance of microorganisms by detecting specific genetic markers such as virulence factors and resistance genes of microorganisms. Although certain experimental techniques and equipment support are required, once the corresponding analysis process and database are established, a high-throughput, automated analysis process can be achieved, thereby greatly improving work efficiency and accuracy. The complex relationships and potential risks between microorganisms can also be revealed by analyzing the genetic information and interaction networks of microbial communities. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1Shown is a flow chart of the method for identifying potentially risky microorganisms in glaciers of the present invention. DETAILED DESCRIPTION

[0061] The present invention will be further described below in conjunction with the embodiments.

[0062] The present invention provides an embodiment: a method for identifying glacier potential risk microorganisms, the method steps are as follows;

[0063] Step 1: Dataset Description

[0064] 85 metagenomes and genomes of 883 strains from 21 glaciers were downloaded from the Tibetan Glacier Genomes and Genes (TG2G) catalog. The metagenome-assembled genomes of 85 Tibetan glacier metagenomes were combined with the genomes of 883 cultured isolates to generate a dataset. The TG2G catalog contains 26 snow samples from 14 glaciers, 39 ice samples (-5 cm depth) from 15 glaciers, and 20 ice dust samples from 7 glaciers. The 85 samples were collected in different years from November 2016 to September 2020. However, the sampling date had no significant effect on the abundance and composition of ARGs. The 883 bacterial strains were cultured in R2A broth at 4°C and centrifuged at 4000 rpm for 10 min for DNA extraction.

[0065] Step 2: Identification of ARGs

[0066] The ARGs-OAPv2.0 online analysis tool was used to quickly identify ARGs and quantify gene abundance. The analysis process is as follows;

[0067] 1); Pre-screen potential ARG sequences from clean reads using UBLAST;

[0068] 2) Use UBLAST to annotate the sequences;

[0069] 3); According to the annotation results, the normalized abundance of ARG types and subtypes in each sample was obtained in ppm;

[0070] 4) Use ARGs-OAP v2.0, web address:

[0071] http: / / bacmet.biomedicine.gu.se / , refer to the BacMet database to annotate and quantify ARGs types and subtypes ( Mao et al., 2023 );

[0072] 5); ARGs abundance was calculated with reference to 16S rRNA according to the following formula (Yin et al., 2018);

[0073]

[0074] Among them, N 16S sequence Mapped reads to the reference database; N ARG―like sequence Reference sequence length; L reads Read length; N 16S sequence 16 S rRNA clean reads number; L 16S sequence 16SrRNA average read length;

[0075] Reads assigned to 16S rRNA by ARG-OAP v2.0;

[0076] Step 3: Metagenomic analysis methods

[0077] MEGAHIT (v1.1.3) (Li et al., 2016) software was used to assemble single samples and mixed samples from the same location and sampling time to obtain contigs.

[0078] 2 (Kang et al., 2019), MaxBin 2 (Wu et al., 2016) and CONCOCT (Alneberg et al., 2014) were used to calculate the correlation between contigs, perform clustering and binning, and obtain metagenome assembled genomes (MAGs). dRep (Olm et al., 2017) was used to identify highly similar MAGs groups in the samples and obtain non-redundant MAGs sets.

[0079] The abundance of MAGs in the sample was calculated according to the following formula:

[0080]

[0081] Among them, reads mapping percentage refers to the percentage of reads in the sample that are aligned with the non-redundant MAGs set to the total number of reads in the sample, MAG coverage refers to the coverage of MAG in the sample, and total MAG coverage refers to the sum of the coverage of all MAGs in the sample;

[0082] The glacier hosts of ARGs, drug-resistant bacteria, were identified based on the metagenomic assembly method, and the identification steps were as follows;

[0083] 1): First, the sequencing data was quality controlled, and the sequences were assembled and genes were predicted using bioinformatics analysis software. The predicted genes were compared with the NCBI nr database, the functional gene database KEGG and the ARGs database CARD, the pathogenic gene PHI-base database, and the transposon ISfinder database for sequence alignment and functional annotation;

[0084] 2): Construct glacier microbial assembled genomes, i.e., MAGs sets, screen out microbial assembled genomes containing ARGs in glacier samples, i.e., drug-resistant bacteria, and combine and count the abundance of drug-resistant bacteria carrying the same ARGs type, and then compare the spatial distribution pattern of drug-resistant bacteria in glacier habitats;

[0085] Step 4: Identification of potential PARBs

[0086] In order to identify potential pathogenic drug-resistant bacteria (PARB), the VFs identification method was used, and the identification steps were as follows:

[0087] 1) Using the bbwrap.sh script of BBMap (v38.43) software in slow mode, the core VFs sequences in the VFs database (VFDB) (Chen et al., 2005), namely the nucleic acid sequence VFDB_setA_nt.fas and the protein sequence VFDB_setA_pro.fas, were compared to identify and align the corresponding VFs;

[0088] 2): Estimate the coverage of the reference sequence in each sample using the in-house script pile.sh with default parameters;

[0089] The steps for VFs classification are as follows;

[0090] 1): Use the tool GTDB-Tk (v0.3.2) to classify the VFs carried by MAGs based on the VFDB classification database;

[0091] 2): According to the classification standards in the VFDB database and the functions of VFs, they are divided into offensive, motile, defensive, immune, toxic proteins and unknown types;

[0092] Step 5: Identification of pathogenic and drug-resistant bacteria

[0093] ARGs and VFs are considered to be microbial marker genes associated with health risks. The steps for identifying the coexistence risk of ARGs and VFs are as follows:

[0094] 1): First, predict the open reading frames, i.e., ORFs, in all contigs of each MAG;

[0095] 2): In order to identify potential ORFs belonging to ARGs, BLASTP was used to compare the ORFs with the ARGs database, and the screening conditions were: similarity>80%, the length of its top hit in the database>70%, and all ORFs considered to be potential ARGs sequences were extracted;

[0096] 3): Using the same method, BLASTP was used to align the ORF with the VFDB protein sequence VFDB_setA_pro.fas to identify all ORFs that were considered potential VFs sequences;

[0097] 4): MAGs containing both ARGs sequences and VFs sequences are considered ARGs-VFs paired MAGs;

[0098] 5): Calculate the abundance of all ARGs-VFs paired MAGs in each sample and perform species annotation of MAGs:

[0099] Step 6: Analysis of movable ARGs

[0100] To further estimate the mobility of ARGs, the pathfact pipeline (v.1.0) with default instructions was used to identify MGEs, including plasmids and phages ( de Nies et al., 2021 ), and further co-linear analysis of ARGs and MGEs was performed to determine whether these genes were simultaneously located on the same MAGs;

[0101] Step 7: Statistical Analysis

[0102] The multiple regression tree (MRT) method was used to reveal the cutoff values ​​of the glacier position variables associated with the ARG distribution;

[0103] Step 8: Results

[0104] 1. Metagenomics reveals the prevalence of ARGs in Tibetan glaciers

[0105] A total of 375 resistance-related gene types were detected, and these ARGs were associated with resistance to 29 classes of antibiotics. There were significantly more ARGs in snow than in ice and ice dust.

[0106] 2. Comparison of ARGs between monsoon-dominated and westerly-dominated glaciers

[0107] The multiple regression tree (MRT) model shows that The threshold is the main factor affecting the distribution of ARGs. Twelve of the 21 glaciers have latitude values ​​below this threshold, while the other nine glaciers have latitude values ​​equal to or above this threshold.

[0108] The distribution of ARGs in snow, ice, and cryoconite is different between monsoon-dominated and westerly-dominated glaciers, with the relative abundance of ARGs in snow samples from monsoon-dominated glaciers being significantly higher than that from westerly-dominated glaciers;

[0109] 3. Host bacteria of ARGs in Tibetan glaciers and their potential pathogenicity

[0110] A total of 3241 MAGs were obtained from all samples, of which 2047 MAGs did not carry ARGs. Among the other 1194 MAGs, 96 MAGs had only one ARG, which was resistant to only one drug class. The average relative abundance of MAGs from snow samples was higher in monsoon-dominated glaciers than in westerly-dominated glaciers, but there was no significant difference between ice and ice dust samples.

[0111] 4. Potential risks of horizontal transfer of ARGs in Tibetan glaciers

[0112] To further evaluate the potential mobility of ARGs, we investigated the co-occurrence of MGEs and ARGs at the MAGs level, and found MGE-ARG co-occurrence patterns in 45 MAGs. 90 ARGs co-occurred with plasmids and phages, accounting for 3.3% of the total ARGs. Among the MAGs carrying both ARGs and MGEs, 90.9% and 9.1% of the MEGs were identified as plasmids and phages, respectively;

[0113] Step 9: Conclusion

[0114] In this study, the profiles of ARGs in 21 Tibetan glaciers were investigated, and the results showed that the propagation of glacial ARGs was affected by atmospheric circulation, and different ARG patterns were observed in glacier surface habitats. ARGs were delivered to glaciers in a native state. The characteristics of the Indian monsoon and the potential sources of exotic ARGs determined the different geographical patterns of ARGs in Tibetan glaciers, that is, the relative abundance of glacial ARGs and their hosts was higher in the monsoon than in those glaciers dominated by westerlies. The potential risks of glacial ARGs include potential pathogen hosts and migration of ARGs through MGES. Due to global warming and glacier melting caused by human activities, ARGs may be released to downstream ecosystems, so careful monitoring is needed to further study the possible impacts of ARGs on downstream cities dependent on Tibetan glacier water.

[0115] Through the above steps, this method can use the metagenomic analysis method to directly conduct a comprehensive analysis of the microbial community without cultivation, thereby revealing more unknown or difficult-to-cultivate microbial species. The ARGs identification method and mobile ARGs analysis can accurately detect antibiotic resistance genes and their transmission mechanisms, providing an important basis for assessing the potential risks of microorganisms. The VFs identification method and the identification method of pathogenic and resistant bacteria can more accurately assess the pathogenicity and drug resistance of microorganisms by detecting specific genetic markers such as virulence factors and resistance genes of microorganisms, so as to solve the problem that the identification method mainly relies on the cultivation and isolation of microorganisms, but many glacier microorganisms are difficult or uncultivable, and the traditional morphological observation and physiological and biochemical characteristics identification are intuitive and rapid, but the resolution is low, and the cultivation and isolation of microorganisms requires a long time.

[0116] The embodiments of the present invention are described in detail above, but the present invention is not limited to the above embodiments, and various changes can be made within the knowledge scope of those skilled in the art without departing from the spirit of the present invention.

Claims

1. A method for identifying potentially risky microorganisms on glaciers, characterized in that: The method steps are as follows; S11: Identification method of ARGs ARGs-OAPv2.0 online analysis tool was used to quickly identify ARGs and quantitatively analyze gene abundance. The webpage address is: https: / / github.com / xiaole99 / ARGs-OAP-v2.

0. The analysis process is as follows; S111; Pre-screen potential ARG sequences from clean reads using UBLAST; S112; Annotate the sequences using UBLAST; S113； The normalized abundance of ARG types and subtypes in each sample was obtained based on the annotation results, in ppm; S114; Use ARGs-OAP v2.0, web address: http: / / bacmet.biomedicine.gu.se / , refer to the BacMet database to annotate and quantify ARGs types and subtypes ( Mao et al., 2023 ); S115;ARGs abundance was calculated with reference to 16S rRNA according to the following formula (Yin et al., 2018); Among them, N 16Ssequence Mapped reads to the reference database; N ARG―likesequence Reference sequence length; L reads Read length; N 16Ssequence 16S rRNA clean reads number; L 16Ssequence 16SrRNA average read length; Reads assigned to 16S rRNA by ARG-OAP v2.0; S12: Metagenomic Analysis Methods MEGAHIT (v1.1.3) (Li et al., 2016) software was used to assemble single samples and mixed samples from the same location and sampling time to obtain contigs. MetaBAT 2 (Kang et al., 2019), MaxBin 2 (Wu et al., 2016) and CONCOCT (Alneberg et al., 2014) were used to calculate the correlation between contigs, cluster and bin them, and obtain metagenome assembled genomes (MAGs). dRep (Olm et al., 2017) was used to identify highly similar MAGs groups in samples and obtain non-redundant MAGs sets. The abundance of MAGs in the sample was calculated according to the following formula: Among them, reads mapping percentage refers to the percentage of reads in the sample that are aligned with the non-redundant MAGs set to the total number of reads in the sample, MAG coverage refers to the coverage of MAG in the sample, and total MAG coverage refers to the sum of the coverage of all MAGs in the sample; Based on the metagenomic assembly method, the ARGs glacier host-resistant bacteria were identified, and the identification steps were as follows; S121: First, the sequencing data was quality controlled, and the sequences were assembled and genes were predicted using bioinformatics analysis software. The predicted genes were compared with the NCBInr database, the functional gene database KEGG and the ARGs database CARD, the pathogenic gene PHI-base database, and the transposon ISfinder database for sequence alignment and functional annotation; S122: Construct glacier microbial assembled genomes, i.e., MAGs, screen out microbial assembled genomes containing ARGs in glacier samples, i.e., drug-resistant bacteria, and combine and count the abundance of drug-resistant bacteria carrying the same ARGs type, and then compare the spatial distribution pattern of drug-resistant bacteria in glacier habitats; S13: VFs identification method The identification steps are as follows: S1311: Use the bbwrap.sh script of BBMap (v38.43) software in slow mode to compare with the core VFs sequences in the VFs database (Virulence Factors Database, VFDB) (Chen et al., 2005), namely the nucleic acid sequence VFDB_setA_nt.fas and the protein sequence VFDB_setA_pro.fas, to identify and align the corresponding VFs; S1312: Estimate the coverage of the reference sequence in each sample using the in-house script pile.sh with default parameters; The steps for VFs classification are as follows; S1321: Use the tool GTDB-Tk (v0.3.2) to classify the VFs carried by MAGs based on the VFDB classification database; S1322: According to the classification criteria in the VFDB database and the functions of VFs, they are divided into offensive, motile, defensive, immune, toxic proteins, and unknown types; S14: Methods for identifying pathogenic and drug-resistant bacteria ARGs and VFs are considered to be microbial marker genes associated with health risks. The steps for identifying the coexistence risk of ARGs and VFs are as follows: S141: First, predict the open reading frames, i.e., ORFs, in all contigs of each MAG; S142: In order to identify potential ORFs belonging to ARGs, BLASTP was used to compare the ORFs with the ARGs database, and the screening conditions were: similarity>80%, the length of its top hit in the database>70%, and all ORFs considered as potential ARGs sequences were extracted; S143: Using the same method, BLASTP was used to align the ORFs with the VFDB protein sequence VFDB_setA_pro.fas, and all ORFs considered as potential VFs sequences were identified; S144: MAGs containing both ARGs sequences and VFs sequences are considered ARGs-VFs paired MAGs; S145: Calculate the abundance of all ARGs-VFs paired MAGs in each sample and perform species annotation of MAGs; S15: Analysis of movable ARGs To further estimate the mobility of ARGs, the pathfact pipeline (v.1.0) with default instructions was used to identify MGEs, including plasmids and phages (de Nies et al., 2021), and further co-linear analysis of ARGs and MGEs was performed to determine whether these genes were co-located on the same MAGs.

2. The method for identifying glacier potential risk microorganisms according to claim 1, characterized in that: The parameters in S112 are: >75bp alignlength, >80% identity and e-value <10 -7 .

3. The method for identifying glacier potential risk microorganisms according to claim 1, characterized in that: In S12, the bin_refinement module in the MetaWRAP pipeline ( Uritskiy et al., 2018 ) was used to screen MAGs based on completeness > 50% and contamination rate < 10%.

4. The method for identifying glacier potential risk microorganisms according to claim 1, characterized in that: The web address of the NCBInr database in S121 is: https: / / www.ncbi.nlm.nih.gov, the web address of the functional gene database KEGG is: http: / / www.genome.jp / kegg / , the web address of the ARGs database CARD is: http: / / arpcard.mcmaster.ca / ?q=CARD / ontology / 35506, the web address of the pathogenic gene PHI-base database is: http: / / www.phi-base.org / , and the web address of the transposon ISfinder database is: https: / / www-is.biotoul.fr / .

5. The method for identifying glacier potential risk microorganisms according to claim 1, characterized in that: The download address of the core VFs sequence in S1311 is: http: / / www.mgc.ac.cn / VFs / download.htm.

6. The method for identifying glacier potential risk microorganisms according to claim 1, characterized in that: The VFDB database used for VFs classification in S1321 comes from: http: / / www.mgc.ac.cn / cgi-bin / VFs / v5 / main.cgi.

7. The method for identifying glacier potential risk microorganisms according to claim 1, characterized in that: The main functions of the offensive, motile, defensive, immune, toxic proteins, and unknown types in S1322 are as follows; Offensive: Pathogens carrying offensive VFs can easily invade the host and infect it; Motile: Pathogens carrying motile VFs actively attach to ecological niches suitable for their reproduction, and most of them are prone to forming biofilms; Defensive type: Pathogens carrying defensive VFs can resist the stimulation of the external environment and increase their drug resistance; Immune type: Pathogens carrying immune type VFs can enhance their tolerance to the host immune response; Toxic protein: Toxic protein factors can regulate pathogens to secrete specific toxins to attack different tissue cells of the host; Unknown type: The specific mechanism of action of unknown type VFs is still unclear.

8. The method for identifying glacier potential risk microorganisms according to claim 1, characterized in that: Pathfact pipeline (v.1.0) was chosen in S15 because of its better performance compared with other similar tools.

Images