A metagenomics-based method and system for detecting resistance genes

Abstract

This invention discloses a metagenomics-based method and system for detecting resistance genes, belonging to the field of resistance gene detection. The method includes the following steps: constructing a microbial genome sequence based on metagenomic reads; performing species annotation and resistance gene identification based on the microbial genome sequence to obtain annotated resistance gene sequences; analyzing the evolutionary relationships of resistance genes in the annotated resistance gene sequences; and monitoring and providing early warning of dynamic changes in resistance gene expression based on the evolutionary relationships and annotation information of the resistance gene sequences. This invention achieves a complete analysis process from raw data to the distribution, evolution, and propagation patterns of resistance genes, providing strong data support for the tracing of resistance genes and risk assessment.

Description

Technical Field

[0001] This invention belongs to the field of resistance gene detection technology, and particularly relates to a method and system for detecting resistance genes based on metagenomics. Background Technology

[0002] Metagenomic data contains a large number of resistance gene fragments, which are typically short and scattered across different parts of the dataset, posing challenges to the identification and analysis of resistance genes. Since these resistance gene fragments overlap, they can be assembled into complete resistance gene sequences using sequence splicing. However, due to the sheer volume of metagenomic data, traditional single-machine storage and computation methods are insufficient for data analysis needs. Therefore, distributed storage and parallel computing are required to improve data processing efficiency. Furthermore, resistance gene sequences exhibit homology and variability; that is, there are highly similar sequences that do not originate from the same gene, as well as variant sequences resulting from mutations of the same gene in different environments. This necessitates considering the evolutionary relationships between sequences during the splicing process, using clustering methods to categorize them into different gene families or variant classes to achieve comprehensive analysis and tracing of resistance genes. Simultaneously, since functional annotation of resistance genes relies on alignment with known sequences, a high-quality reference database is needed to integrate existing resistance gene data and provide reliable annotation information for newly discovered resistance genes. Summary of the Invention

[0003] To address the aforementioned technical problems, this invention proposes a metagenomics-based method and system for detecting resistance genes, thereby resolving the issues present in the prior art.

[0004] To achieve the above objectives, the present invention provides a metagenomics-based method for detecting resistance genes, comprising:

[0005] Constructing microbial genome sequences based on metagenomic reads;

[0006] Based on the genome sequence of the microorganism, species annotation and resistance gene identification were performed to obtain the annotated resistance gene sequence;

[0007] The evolutionary relationships of resistance genes in the annotated resistance gene sequences are analyzed, and the dynamic changes in resistance gene expression are monitored and warned based on the evolutionary relationships of resistance genes and the annotation information of resistance gene sequences.

[0008] Optionally, the process of constructing the genome sequence of a microorganism includes:

[0009] Quality control was performed on metagenomic reads to obtain clean reads;

[0010] The clean reads are mapped onto the host reference genome to obtain reads after removing host gene contamination.

[0011] The reads, after removing host gene contamination, were assembled to obtain the genome sequence of the microorganism.

[0012] Optionally, the process of annotating the genome sequence of the microorganism and identifying resistance genes includes:

[0013] The genome sequence of the microorganisms was annotated with microbial domain affiliation to obtain species annotation results;

[0014] Open reading frames are used to predict genes from the genome sequences of microorganisms annotated with microbial domain affiliation.

[0015] The predicted genes were annotated using a resistance gene library to obtain resistance gene sets belonging to different microbial domains;

[0016] The abundance of resistance genes belonging to different microbial domains was obtained by aligning clean reads to the resistance gene sets belonging to different microbial domains.

[0017] Optionally, the process of aligning clean reads to resistance gene sets belonging to different microbial domains includes:

[0018] Clean reads were aligned to the resistance gene set using bwa or bowtie2, the abundance of resistance genes was calculated using Salmon, and the abundance of the resistance genes was normalized.

[0019] Optionally, the process of performing evolutionary relationship analysis on the resistance genes in the annotated resistance gene sequence includes:

[0020] Sequence data of different gene families and variants were obtained based on the annotated resistance gene sequences;

[0021] The sequence data are compared and analyzed using a multiple sequence comparison algorithm to obtain the comparison results;

[0022] Based on the comparison results, the evolutionary distance between different gene sequences was calculated, and a distance matrix was constructed;

[0023] Construct a phylogenetic tree based on the distance matrix;

[0024] The confidence level of the evolutionary relationship is obtained by performing a reliability assessment based on the phylogenetic tree.

[0025] Optionally, the process of monitoring and providing early warning of dynamic changes in resistance gene expression includes:

[0026] A resistance gene dataset was constructed based on annotation information and evolutionary relationship confidence of resistance genes;

[0027] The feature vectors of the resistance gene dataset are extracted to obtain the resistance gene feature matrix;

[0028] The frequent co-occurrence patterns of resistance genes in different environments are obtained by analyzing the feature matrix of resistance genes based on the frequent pattern mining algorithm.

[0029] Based on the frequent co-occurrence pattern, an association rule mining algorithm is used to mine the association rules between resistance genes to obtain the distribution characteristics of resistance genes;

[0030] Based on the distribution characteristics of the resistance genes, the distribution trend of the resistance genes in different environments is determined;

[0031] The dynamic expression of resistance genes is determined based on their distribution characteristics and distribution trends under different environments.

[0032] This invention also provides a metagenomics-based resistance gene detection system, comprising:

[0033] The data processing module is used to acquire and preprocess metagenomic data;

[0034] A resistance annotation module, which is used to construct resistance gene sets under different microbial domains;

[0035] An evolutionary analysis module, used to infer the evolutionary relationships between gene families and variants;

[0036] A data mining and visualization module is used to discover distribution patterns and generate visualization results.

[0037] The present invention also provides a computer terminal device, comprising:

[0038] One or more processors;

[0039] A memory, coupled to the processor, for storing one or more programs;

[0040] When the one or more programs are executed by the one or more processors, the one or more processors implement a metagenomics-based method for detecting resistance genes.

[0041] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a metagenomics-based method for detecting resistance genes.

[0042] Compared with the prior art, the present invention has the following advantages and technical effects:

[0043] This invention provides a metagenomics-based method for detecting antibiotic resistance genes. This method constructs the genome sequence of microorganisms, then performs species annotation and resistance gene identification, ultimately obtaining the annotated resistance gene sequences. This method offers significant technical advantages. First, it comprehensively covers the microbial genome in the sample, ensuring the comprehensiveness and accuracy of resistance gene detection. Second, through species annotation, this method can accurately distinguish resistance genes from different microbial sources, providing accurate species background information for subsequent analysis of resistance gene evolutionary relationships. Furthermore, by analyzing the evolutionary relationships of the annotated resistance gene sequences, this method can monitor and provide early warning of dynamic changes in resistance gene expression, thus providing a scientific basis for antibiotic resistance research and management, effectively predicting and controlling the spread of resistance genes, and playing a crucial role in combating antibiotic resistance. Finally, based on the annotation information of the resistance gene sequences, this method can monitor and provide early warning of dynamic changes in resistance gene expression, providing a scientific basis for antibiotic resistance management, and helping to reduce antibiotic abuse and control the spread of resistance genes. In summary, the method of this invention has significant application value in improving the accuracy of resistance gene detection, providing early warning of the risk of resistance gene spread, and guiding the rational use of antibiotics. Attached Figure Description

[0044] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0045] Figure 1 This is a flowchart of the metagenomics-based resistance gene detection method according to an embodiment of the present invention;

[0046] Figure 2 This is a structural diagram of a metagenomics-based resistance gene detection system according to an embodiment of the present invention. Detailed Implementation

[0047] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0048] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0049] Example 1

[0050] like Figure 1As shown, this embodiment provides a method and system for detecting resistance genes based on metagenomics, including the following steps:

[0051] Step 1: Constructing the microbial genome sequence based on metagenomic reads. The specific implementation process includes: quality control of the metagenomic reads to obtain clean reads; mapping the clean reads onto the host reference genome to obtain reads after removing host gene contamination; and assembling the reads after removing host gene contamination to obtain the microbial genome sequence.

[0052] Furthermore, Trimmomatic or Fastp software is used to perform quality control on metagenomic reads, removing sequencing adaptors. Low-quality reads are called clean reads to ensure the reliability of downstream analysis data.

[0053] Using short sequence alignment software such as bwa / bowtie2, clean reads are mapped to the host (mainly large vertebrates such as humans, mice, and pigs) reference genome to remove host gene contamination, with the aim of eliminating the influence of host genes on downstream analysis;

[0054] The reads after removing the host were assembled using short sequence assembly software such as Megahit / metaSpades, and contig sequences that were too short (usually 500bp) were filtered out. The assembly results were then evaluated using MetaQUAST to obtain the genome sequence of the microorganism.

[0055] Step 2: Based on the genome sequence of the microorganism, perform species annotation and resistance gene identification to obtain the annotated resistance gene sequence; the specific implementation process includes: performing microbial domain classification annotation on the genome sequence of the microorganism to obtain species annotation results; performing open reading frame prediction on the genome sequence of the microorganism after microbial domain classification annotation to obtain predicted genes; using a resistance gene library to annotate the predicted genes to obtain a resistance gene set; and aligning cleanreads to the resistance gene sets belonging to different microbial domains to obtain the abundance results of resistance genes belonging to different microbial domains.

[0056] Furthermore, it is recommended to directly binning, using GTDB_Tk to annotate the binning results for species. Subsequent gene annotations, such as resistance genes, are all based on the binning results. For contig sequences that cannot be binned, use software such as Taxometer / MMseqs2 / RAT to identify bacterial and archaea sequences (custom database BAV). Further, use geNomad / PlasFlow to distinguish plasmids and chromosomes, use VirSorter2, Deepvirfinder, and VIBRANT to identify viral sequences, and use CheckV to evaluate viral sequences.

[0057] The contigs were predicted using open reading frames (ORF) using software such as Prodigal / MetaGeneMark. The predicted genes were then annotated using Blastp / DIAMOND and the resistance gene database NCRD. Annotation results with similarity and coverage greater than 80% were retained to construct a non-redundant resistance gene set.

[0058] Short sequence alignment software such as bwa / bowtie2 was used to map the quality-controlled reads onto the assembled contig sequence and to retain read pairs where both ends of the reads were mapped onto the contig sequence. The reads were divided into three categories: bacteria, archaea, and viruses, and the number of reads in each category was counted.

[0059] Short sequence alignment software such as bwa / bowtie2 was used to align reads that had already been classified into species to construct a non-redundant set of resistance genes. Salmon was used to calculate the abundance of resistance genes, and TPM was used to standardize the gene abundance. The species classification information corresponding to each read was then matched.

[0060] Step 3: Using evolutionary analysis methods, the annotated resistance gene sequences are analyzed to infer the evolutionary relationships between different gene families and variants, construct a phylogenetic tree, and reveal the evolutionary history and propagation patterns of resistance genes.

[0061] Furthermore, based on the sequence annotation information of the resistance gene, sequence data of different gene families and variants are obtained; multiple sequence alignment algorithms, such as Clustal or MUSCLE, are used to perform alignment analysis on the obtained gene sequences; the evolutionary distance between different gene sequences is calculated based on the alignment results, and a distance matrix is ​​constructed; based on the distance matrix, a phylogenetic tree is constructed using methods such as the neighbor-joining method or the maximum likelihood method; the reliability of the constructed phylogenetic tree is assessed, such as through bootstrap analysis, to determine the confidence level of the evolutionary relationship; by analyzing the topological structure and branch length of the phylogenetic tree, the evolutionary history and dissemination direction of the resistance gene among different species or strains are inferred; combining the variation information of the gene sequence and the phylogenetic tree, the key mutation sites and evolutionary driving forces of the resistance gene are identified, revealing its adaptive evolutionary laws.

[0062] Step 4: Based on the annotation information and evolutionary relationships of resistance genes, an association rule mining algorithm is used to discover the distribution patterns and co-occurrence rules of resistance genes in different environments, and to predict the spread trend and potential risks of resistance genes.

[0063] Furthermore, annotation information and evolutionary relationship data of resistance genes were obtained to construct a resistance gene dataset. Data preprocessing techniques were employed to clean, integrate, and standardize the resistance gene dataset, resulting in high-quality resistance gene data. Based on the annotation information and evolutionary relationships of resistance genes, feature vectors were extracted to construct a resistance gene feature matrix. Frequent pattern mining algorithms were used to analyze the resistance gene feature matrix, revealing frequent co-occurrence patterns of resistance genes in different environments. Association rule mining algorithms were then used to mine these frequent co-occurrence patterns, obtaining association rules between resistance genes and revealing their distribution characteristics. Clustering algorithms were used to perform cluster analysis on the resistance genes, and the distribution trends of resistance genes in different environments were determined based on the clustering results. Combining association rules and clustering results, a comprehensive analysis of the propagation trends and potential risks of resistance genes was conducted, providing decision support for the monitoring and control of resistance genes.

[0064] As a specific implementation method in this embodiment, the analysis of resistance gene data aims to reveal the evolutionary history, propagation patterns, and potential risks of resistance genes. First, it is necessary to obtain annotation information of resistance genes from gene databases or literature, such as gene names, functional annotations, variation information, and evolutionary relationship data, such as gene family classifications and phylogenetic trees. This information is used to construct a resistance gene dataset, providing a foundation for subsequent analysis. For example, the sequences and annotation information of resistance genes can be downloaded from the CARD database and combined with published phylogenetic trees to construct a dataset. After obtaining the raw data, data preprocessing is required. This includes data cleaning, such as removing repetitive sequences and handling missing values; data integration, such as merging data from different sources; and data standardization, such as converting gene sequences to a unified format. For example, gene sequences shorter than 100 bp can be removed, missing values ​​can be filled with the mode or median, and all gene sequences can be converted to FASTA format. High-quality data is a prerequisite for accurate analysis. Next, it is necessary to extract feature vectors of resistance genes and construct a feature matrix. Feature vectors can include gene sequence length, GC content, amino acid composition, etc., and can also include gene functional annotations, variation information, etc. For example, the GC content of each gene sequence is calculated, the frequency of different amino acids in each gene sequence is counted, and this information is combined into a feature vector. Rows in the feature matrix represent different genes, and columns represent different features. Frequent pattern mining algorithms are used to discover frequent co-occurrence patterns of resistance genes under different environments. For example, it is found that certain resistance genes always appear simultaneously in environments with high concentrations of antibiotics. Frequent pattern mining algorithms can help understand which resistance genes are more likely to co-occur, thereby inferring possible interactions or co-evolutionary relationships between them. Suppose frequent pattern mining finds that gene A and gene B always co-occur, this suggests that these two genes may be located on the same mobile genetic element, or their expression may be affected by the same regulatory mechanism. Association rule mining algorithms can further reveal the association rules between resistance genes. For example, it is found that the occurrence of gene A is always accompanied by the occurrence of gene B, with a confidence level of 90%. This indicates a strong association between gene A and gene B, possibly suggesting that the resistance mechanism of gene B depends on the presence of gene A. Association rule mining can help understand the interdependencies between resistance genes and their roles in the propagation process. Clustering algorithms can perform cluster analysis on resistance genes. For example, resistance genes with similar functions or evolutionary relationships can be clustered together. By analyzing the clustering results, the distribution trends of resistance genes in different environments can be determined. For instance, it can be found that some types of resistance genes are more likely to spread in hospital environments, while others are more likely to spread in farm environments.Suppose cluster analysis groups several resistance genes together, and these genes all encode β-lactamases. This suggests that these genes may share similar resistance mechanisms and co-evolve under the selective pressure of antibiotics. Combining association rules and clustering results, a comprehensive analysis of the spread trends and potential risks of resistance genes can be conducted. For example, if certain high-risk resistance genes are frequently co-occurring and rapidly spreading in specific environments, appropriate monitoring and control measures are needed. This comprehensive analysis provides decision support for the prevention and control of resistance genes and helps delay the development of antibiotic resistance. For instance, if genes encoding carbapenemases and genes encoding aminoglycoside-modifying enzymes are frequently co-occurring and rapidly spreading in hospital settings, this suggests that these genes may spread between different bacteria through mechanisms such as horizontal gene transfer, leading to the emergence of multidrug-resistant bacteria. Strengthening hospital infection control measures is necessary to limit the spread of these genes.

[0065] Step 5: Present the analysis results in a visual manner, generating distribution maps, phylogenetic trees, and association networks of resistance genes. This will help researchers intuitively understand and interpret the composition, distribution, and evolution of resistance genes, providing data support for tracing the origins and assessing the risks of resistance genes.

[0066] Furthermore, the analysis results of resistance genes are obtained, and the data is preprocessed to ensure data quality and integrity by standardizing the data format. Based on the composition information of resistance genes, clustering algorithms are used to group resistance genes into different categories. For each category of resistance genes, their geographical distribution characteristics are analyzed to generate a distribution map, visually displaying the distribution in different regions. Evolutionary relationships of resistance genes are analyzed using a phylogenetic tree algorithm to construct a phylogenetic tree, revealing the evolutionary patterns and paths of resistance genes. Association rule mining algorithms are used to discover association patterns between different resistance genes, generating an association network diagram of resistance genes. By combining the analysis results of the distribution map, phylogenetic tree, and association network, the origin of resistance genes is determined, identifying their origin and propagation path. Based on the distribution, evolution, and association characteristics of resistance genes, their potential risks are assessed, providing data support and decision-making basis for the prevention and control of resistance genes. Risk assessment can provide data support and decision-making basis for the prevention and control of resistance genes.

[0067] This invention discloses a metagenomics-based method for detecting resistance genes. The method first performs distributed storage and parallel preprocessing of metagenomic data to obtain high-quality resistance gene fragments. Then, sequence alignment and splicing algorithms are used to assemble the fragments into longer sequences, and cluster analysis is employed to classify the sequences into different gene families and subclasses. Next, a resistance gene reference database is constructed, and the newly spliced ​​sequences are functionally annotated. Furthermore, evolutionary analysis is used to infer the evolutionary relationships between gene families, and association rule mining algorithms are used to discover the distribution patterns and propagation trends of resistance genes. Finally, the analysis results are visualized. This invention achieves a complete analysis process from raw data to the distribution, evolution, and propagation patterns of resistance genes, providing strong data support for the tracing of resistance genes and risk assessment.

[0068] Example 2

[0069] like Figure 2 As shown, this embodiment also provides a metagenomics-based resistance gene detection system, including:

[0070] The data processing module is used to acquire and preprocess metagenomic data.

[0071] The resistance annotation module is used to construct resistance gene sets under different microbial domains;

[0072] The evolutionary analysis module is used to infer the evolutionary relationships between gene families and variants.

[0073] The Data Mining and Visualization module is used to discover distribution patterns and generate visualization results.

[0074] Example 3

[0075] This embodiment also provides a computer terminal device, including:

[0076] One or more processors;

[0077] Memory, coupled to the processor, is used to store one or more programs;

[0078] When one or more programs are executed by one or more processors, the one or more processors implement a metagenomics-based method for detecting resistance genes.

[0079] This embodiment also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements a metagenomics-based method for detecting resistance genes.

[0080] The above are merely preferred embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for detecting resistance genes based on metagenomics, characterized in that, Includes the following steps: Constructing microbial genome sequences based on metagenomic reads; Based on the genome sequence of the microorganism, species annotation and resistance gene identification were performed to obtain the annotated resistance gene sequence; The process of species annotation and resistance gene identification of the microbial genome sequence includes: performing microbial domain classification annotation on the microbial genome sequence to obtain species annotation results; performing open reading frame prediction on the microbial genome sequence after microbial domain classification annotation to obtain predicted genes; using a resistance gene library to annotate the predicted genes to obtain resistance gene sets belonging to different microbial domains; and aligning clean reads to the resistance gene sets belonging to different microbial domains to obtain the abundance results of resistance genes belonging to different microbial domains. Analyze the evolutionary relationships of resistance genes in the annotated resistance gene sequences, and monitor and provide early warning of dynamic changes in resistance gene expression based on the evolutionary relationships of resistance genes and the annotation information of resistance gene sequences; The process of performing evolutionary relationship analysis on the annotated resistance gene sequences includes: obtaining sequence data of different gene families and variants based on the annotated resistance gene sequences; performing comparative analysis on the sequence data using a multiple sequence alignment algorithm to obtain comparison results; calculating the evolutionary distance between different gene sequences based on the comparison results and constructing a distance matrix; constructing a phylogenetic tree based on the distance matrix; and performing a reliability assessment based on the phylogenetic tree to obtain the confidence level of the evolutionary relationship. The process of monitoring and early warning of dynamic changes in the expression of resistance genes includes: constructing a resistance gene dataset based on the annotation information and evolutionary relationship confidence of the resistance genes; extracting feature vectors from the resistance gene dataset to obtain a resistance gene feature matrix; analyzing the resistance gene feature matrix using a frequent pattern mining algorithm to obtain frequent co-occurrence patterns of resistance genes in different environments; mining association rules between resistance genes using an association rule mining algorithm based on the frequent co-occurrence patterns to obtain the distribution characteristics of resistance genes; determining the distribution trend of resistance genes in different environments based on the distribution characteristics of resistance genes; and determining the dynamic expression of resistance genes based on the distribution characteristics and distribution trends of resistance genes in different environments.

2. The metagenomics-based method for detecting resistance genes according to claim 1, characterized in that, The process of constructing the genome sequence of microorganisms includes: Quality control was performed on metagenomic reads to obtain clean reads; The clean reads are mapped onto the host reference genome to obtain reads after removing host gene contamination. The reads, after removing host gene contamination, were assembled to obtain the genome sequence of the microorganism.

3. The metagenomics-based method for detecting resistance genes according to claim 1, characterized in that, The process of aligning cleanreads to resistance gene sets belonging to different microbial domains includes: Clean reads were aligned to the resistance gene set using bwa or bowtie2, the abundance of resistance genes was calculated using Salmon, and the abundance of the resistance genes was normalized.

4. A metagenomics-based resistance gene detection system, characterized in that, The system for implementing the method as described in any one of claims 1-3 includes: The data processing module is used to acquire and preprocess metagenomic data; A resistance annotation module, which is used to construct resistance gene sets under different microbial domains; An evolutionary analysis module, used to infer the evolutionary relationships between gene families and variants; A data mining and visualization module is used to discover distribution patterns and generate visualization results.

5. A computer terminal device, characterized in that, include: One or more processors; A memory, coupled to the processor, for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the metagenomics-based resistance gene detection method as described in any one of claims 1-3.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the metagenomics-based resistance gene detection method as described in any one of claims 1-3.

Images