Method for detecting pathogenic fungus combined pollution based on 18S rRNA and ITS amplicon sequencing

Abstract

The invention discloses a method for detecting pathogenic fungus combined pollution based on 18S rRNA and ITS amplicon sequencing. The method comprises the following steps: S1, constructing a non-redundant pathogenic fungus database; s2, extracting DNA (deoxyribonucleic acid) of the sample; s3, carrying out 18S rRNA or ITS gene sequencing on the sample to obtain sequencing off-machine data; s4, acquiring a representative sequence based on the sequencing offline data; and S5, comparing the representative sequence with the pathogenic fungus database and annotating pathogenic fungi to obtain an ASV species table only containing potential pathogenic fungi, then performing species diversity and community structure analysis, and predicting a combined pollution risk index. Traditional fungal detection utilizes an isolated culture method which is long in time consumption, tedious in operation and low in sensitivity, needs several weeks and is easy to ignore some trace pathogenic fungi. According to the method, separation culture is not needed, and the detection efficiency can be greatly improved directly through DNA sequencing and analysis.

Description

Technical Field

[0001] The present invention relates to the fields of ecology and bioinformatics, and specifically to the construction of a non-redundant database of plant and animal pathogenic fungi and the optimization of the existing high-throughput sequencing data comparison and annotation process to achieve accurate identification and quantification of pathogenic fungal complex pollution in the environment-plant-animal ecosystem. Background Art

[0002] Fungal diseases cause 10%-23% annual crop losses worldwide (Stukenbrock & Gurr, 2023). With the development of globalization and the intensification of global warming, the pathogenicity and pathogenic potential of fungi have increased, and new pathogenic fungi continue to emerge, seriously threatening global food security and human health (Casadevall, 2023; Li et al., 2023). The complexity and diversity of fungi restrict the development of fungal detection technology. Currently, the most commonly used detection methods are culture, PCR and metagenomic sequencing (Wang Yuehua, 2020). The culture method is time-consuming and has limited detection accuracy; although the PCR method has the advantages of high specificity and high sensitivity, it requires the search for specific primers and is prone to false positives (Sun, Hoy, & Heitman, 2020); metagenomic sequencing can accurately identify pathogenic fungi and their genetic information, but its high price and complex data processing are not suitable for large-scale detection (Zhou et al., 2022). At present, the technology dedicated to the detection of pathogenic fungi is still immature. The databases for high-throughput sequencing annotation are mainly UNITE and Silva, which can identify fungal species but do not contain pathogenic information. Although the FUNGuild and FungalTraits databases can identify some pathogenic fungi, the species that can be identified are limited and cannot achieve good detection results (Nguyen et al., 2016; et al., 2021). Therefore, by constructing a new method for detecting pathogenic fungi based on high-throughput sequencing, thousands of pathogenic fungi can be detected quickly and sensitively, providing an important technology for reducing the complex pollution of pathogenic fungi in the ecosystem and ensuring food security and human health. Summary of the invention

[0003] The present invention addresses the problems existing in the background technology by constructing a new method for detecting pathogenic fungi based on high-throughput sequencing, which can quickly and sensitively detect thousands of pathogenic fungi, providing an important technology for reducing the complex pollution of pathogenic fungi in the ecosystem and ensuring food security and human health.

[0004] The present invention integrates relatively complete pathogenic fungal species information through literature and network retrieval methods, and constructs a non-redundant pathogenic fungal database for high-throughput sequencing analysis by comparing the sequence resources of the latest UNITE, Silva, NT, 18S, and ITS databases, thereby realizing the analysis of pathogenic fungal types and abundance based on 18S rRNA and ITS gene sequencing, providing technical support for ensuring ecological stability and human health.

[0005] 1) Construction of a non-redundant database of pathogenic fungi

[0006] Through online retrieval, 91 web pages and literature resources containing fungal pathogenicity information were analyzed and integrated. After removing duplicates, 8057 fungal species-level Latin names and 34 sequencing source files of different types of samples containing potential pathogenic fungi were obtained. While summarizing the sequences of different new pathogenic fungi, the fungal 18SrRNA and ITS sequences in the UNITE, Silva, NT, 18S, and ITS databases were merged to remove redundancy. Subsequently, the Latin names of pathogenic fungi were matched with the sequence information library to obtain the corresponding subspecies and strain information. The database finally obtained contained 10,013 sequences of 6,392 species, including 2,831 zoonotic and 7,182 plant sequence data.

[0007] 2) Encapsulation of a non-redundant pathogenic fungal library using the DADA2 analysis process

[0008] In order to better process the Illumina sequencing data of fungi, eliminate the amplification and sequencing errors caused by the complexity of the 18S rRNA and ITS region variation and length, and the interference of possible chimera sequences on species annotation, the present invention uses the DADA2 analysis process (Callahan et al., 2016) to accurately generate a noise-free sequence feature table (ASV table) to replace the traditional OTU method. At the same time, according to the characteristics of the target sequence data, the error rate model and denoising threshold are adjusted through machine learning, and the parameters of the standard DADA2 process are optimized to improve the analysis accuracy and processing efficiency. The specific process of DADA2 is as follows: 1) Install and load DADA2 in the R environment; 2) Use the plotQualityProfile function to visualize the quality of the original FASTQ file, evaluate the sequence quality, and determine the cutting and filtering strategy; 3) Set filtering and cutting parameters according to the quality map to remove low-quality sequences; 4) Accurately correct errors by learning the sequencing error model of each data set; 5) Use the learned error model to denoise the sequence and identify actual biological sequence variations; 6) Splice the double-end sequences into a complete sequence; 7) Construct the denoised and spliced ​​sequences into a feature table, where each column corresponds to a sample and each row corresponds to a noise-free feature sequence (ASV); 8) Remove chimeras, use the non-redundant pathogenic fungi database in the first step to classify the sequences in the ASV table, classify the sequences into the corresponding biological taxonomic groups, and output the final ASV table and classification information table. By using DADA2 encapsulation for the 18S V4 region and ITS region with the highest annotation efficiency, an ASV table containing only potential pathogenic fungal species can be directly obtained. This can accurately detect and identify the abundance and community composition characteristics of pathogenic fungi in samples at one time, and establish a new method for detecting pathogenic bacteria in environmental-plant-animal ecosystems.

[0009] 3) Optimization of DNA extraction technology

[0010] The present invention can be used for the detection of pathogenic fungi in various types of samples such as plants, animals and soil. Currently, there are corresponding kits on the market that can be used for rapid extraction of DNA, but the cost is relatively high. Since fungi have thicker cell walls, in order to obtain higher quality DNA samples, it is usually necessary to fully grind the samples first and then use enzymatic cell wall breaking, liquid nitrogen freeze-thaw or ultrasonic lysis methods, which is time-consuming, labor-intensive and prone to contamination. The present invention optimizes conventional DNA extraction techniques, selects appropriate samples (such as blood, saliva, leaves, bacterial cultures, etc.) from plants, animals, microorganisms or other organisms, and the samples are usually immediately stored at low temperatures (-80°C) or added to a preservation solution to prevent DNA degradation. In the present invention, sample DNA extraction includes the following steps:

[0011] (1) Mix the soil sample to be tested with a fungal lysis buffer (preferably: 2% cetyltrimethylammonium bromide, 1.4 M NaCl, 100 mM Tris-HCl, 20 mM EDTA), heat to 65 °C and treat for 30 minutes to release cellular NDA;

[0012] (2) Transfer the mixture to a centrifuge tube, centrifuge at high speed (12000 rpm) for 10 minutes to separate cell debris and other impurities, and remove the supernatant for later use;

[0013] (3) Add DNA magnetic beads to the aforementioned supernatant. After it fully adsorbs DNA, use a magnet to fix it in a container and drain the remaining liquid;

[0014] (4) After washing the surface of the magnetic beads with 70% ethanol multiple times to slowly wash away other salts and impurities, elute the DNA adsorbed on the magnetic beads with PCR buffer. The lysis and digestion of the sample to be tested can be completed under the conditions of 30 min and 65 °C. Finally, elute and measure the absorbance A260 / A280 to check whether it is between 1.8 and 2 to test the purity of DNA for subsequent sequencing analysis.

[0015] 4) Use the results of pathogenic fungal types and abundances for risk prediction

[0016] Through this process, an ASV table containing only pathogenic fungal information can be obtained. Some pathogenic fungi with higher risks are marked. In R language, subsequent processing of the ASV table can be carried out, such as visualization of data such as pathogenic fungal abundances and diversities. In order to make the data more intuitively and reliably applied, the calculation of a risk index (RI) is established.

[0017] RI = log 10 (∑A sum ) + 10 × HR sum

[0018] where RI represents the risk index, A sum represents the absolute abundance of pathogenic fungi in the sample, that is, the sequence reads of each ASV in the ASV table. HR sum represents the relative abundance of high-risk pathogenic fungi, that is, the proportion of marked high-risk ASVs among all pathogenic bacteria. The higher the RI, the higher the abundances of various pathogenic fungi and the proportion of high-risk pathogenic fungi in the sample, and the greater the risk. Through the test and analysis of 100 pairs of existing healthy and diseased samples, it is determined that RI ≤ 5.5 indicates a low pathogenic risk, 5.5 < RI ≤ 7 indicates a general pathogenic risk, and RI ≥ 7 indicates a high pathogenic risk, and a corresponding ASV abundance stacked bar chart is generated for easy understanding by users.

[0019] Advantages of the present invention

[0020] 1. Traditional fungal detection uses isolation and culture methods, which are time-consuming, cumbersome, and low in sensitivity. It takes several weeks and it is easy to overlook some trace amounts of pathogenic fungi. The present invention does not require isolation and culture, and can significantly improve detection efficiency by directly sequencing and analyzing DNA.

[0021] 2. PCR amplification can detect pathogens relatively efficiently, but is limited by the types of pathogens to be detected and is susceptible to trace contamination, resulting in false positives, and usually requires more than three repeated verifications. The present invention is based on high-throughput sequencing analysis and can obtain a large amount of pathogenic fungal information at one time.

[0022] 3. Existing pathogenic fungi detection processes such as FUNGuild can play an annotation role, but the number of species that can be detected is limited to 2825. The present invention can identify 6392 plant and zoonotic pathogenic fungi at one time, including many newly discovered species in recent years, and annotate the species that appear frequently and are proven to be more harmful.

[0023] 4. The present invention is based on a self-built non-redundant pathogenic fungi database. Currently, there are a large number of studies on species composition characteristics based on amplicon sequencing, but there are fewer studies on pathogenic fungi. In addition, this database can be continuously improved and upgraded based on the research and discovery of new pathogenic fungi, so the goal of rapid and highly specific detection of pathogenic fungi can be achieved, providing conditions for subsequent response and research.

[0024] 5. The present invention can obtain an ASV table containing only pathogenic fungi information, and can also combine the pathogenic fungi type and abundance results for risk prediction, issue early warnings for high-risk fungal complex pollution environments, better formulate response measures, prevent diseases before they occur, and help maintain biodiversity and ecosystem health. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 A flowchart of the method of the present invention is shown in FIG.

[0026] Figure 2 Schematic diagram of the pathogenic fungus detection process

[0027] Figure 3 Schematic diagram of 10013 sequences of 6392 species in the database obtained in the embodiment

[0028] Figure 4 Comparison of the abundance of pathogenic fungi at different levels in diseased and healthy rhizosphere soils

[0029] Figure 5 Comparison of composite fungal contamination risk index in different samples DETAILED DESCRIPTION

[0030] The present invention is described in detail below by way of examples. Experimental methods for which specific conditions are not specified in the text are generally carried out under conventional conditions or under conditions recommended by the instrument manufacturer. Unless otherwise defined, all professional and scientific terms used in the text have the same meanings as those familiar to those skilled in the art.

[0031] Implementation example A total of 16 rhizosphere soil samples of Codonopsis pilosula, 8 root rot diseased soils and 8 healthy soils were collected from the field in Longxi and Minxian, Gansu Province. This example demonstrates the implementation of a new method for pathogenic bacteria detection, including the following specific steps:

[0032] Combination Figure 1 and Figure 2 , a method for detecting complex contamination of pathogenic fungi based on 18S rRNA or ITS amplicon sequencing, comprising the following steps:

[0033] S1. Construct a non-redundant pathogenic fungus database. The present invention integrates relatively complete pathogenic fungal species information through literature and network retrieval methods, and constructs a non-redundant pathogenic fungus database for high-throughput sequencing analysis by comparing the sequence resources of the latest UNITE, Silva, NT, 18S, and ITS databases, thereby realizing the analysis of pathogenic fungal types and abundance based on 18S rRNA or ITS gene sequencing, providing technical support for ensuring ecological stability and human health.

[0034] In this embodiment, 91 web pages and literature resources containing fungal pathogenicity information were analyzed and integrated through network retrieval, and 8057 Latin names of fungal species were obtained after removing duplicates, as well as sequencing source files of 34 different types of samples containing potential pathogenic fungi. While summarizing the sequences of different new pathogenic fungi, the fungal 18S rRNA and ITS sequences in the UNITE, Silva, NT, 18S, and ITS databases were merged to remove redundancy. Subsequently, the Latin names of pathogenic fungi were matched with the sequence information library to obtain the corresponding subspecies and strain information. The database finally obtained contained 10,013 sequences of 6,392 species, including 2,831 zoonotic and 7,182 plant sequence data. Figure 3 shown.

[0035] S2. Efficient extraction of sample DNA; the present invention can be used for the detection of pathogenic fungi in various types of samples such as plants, animals and soil. Currently, there are corresponding kits on the market that can quickly extract DNA, but the cost is relatively high. Since fungi have thicker cell walls, in order to obtain higher quality DNA samples, it is usually necessary to fully grind the samples first and then use enzymatic cell wall breaking, liquid nitrogen freeze-thaw or ultrasonic lysis methods, which is time-consuming, labor-intensive and prone to pollution. The present invention optimizes the extraction technology, takes an appropriate amount of sample and lysis solution and mixes them evenly, and can complete the lysis and digestion of the sample to be tested under the conditions of 30 minutes and 60°C, release the DNA in the sample, and then combine it with the magnetic bead method, add an adsorbent to use a magnetic bead rod to adsorb the DNA, and then purify it with a washing solution, and finally elute it to measure the absorbance A260 / A280 to see if it is between 1.8 and 2 to test the purity of the DNA for subsequent sequencing analysis.

[0036] S3, perform 18S rRNA or ITS gene sequencing of the sample to obtain sequencing data;

[0037] S4. Obtain the ASV table of pathogenic fungi using the sequencing data in conjunction with the DADA2 amplicon analysis process; In order to better process the Illumina sequencing data of fungi, eliminate amplification and sequencing errors caused by the complexity of 18S rRNA and ITS regional variations and lengths, and possible interference of chimera sequences on species annotation, the present invention adjusts the error rate model and denoising threshold according to the characteristics of the target sequence data, optimizes the standard DADA2 process, and can achieve data quality assessment, filtering and denoising, merging read pairs, removing chimeras, and comparing annotations in one step. The 18S V7 region and ITS region with the highest annotation efficiency are encapsulated, and a noise-free sequence feature table (ASV table) can be accurately generated to replace the traditional OTU method. The comparison annotation of representative sequences with the pathogenic fungi database can be directly obtained, and the abundance and community composition characteristics of pathogenic fungi in the sample can be accurately detected and identified at one time, and a new method for detecting pathogenic bacteria in the environment-plant-animal ecosystem is established.

[0038] S5. Visualize the community composition and diversity information in the ASV table to predict the risk of fungal pathogenicity.

[0039] Through this process, an ASV table containing only pathogenic fungi information can be obtained. Some pathogenic fungi with higher risks are marked. Subsequent processing of the ASV table can be performed in R language, such as visualization of pathogenic fungi abundance, diversity and other data. In order to make the data more intuitive and reliable, the calculation of the risk index (RI) was established.

[0040] RI=log 10 (∑A sum )+10×HR sum

[0041] Where RI represents the risk index, A sum Represents the absolute abundance of pathogenic fungi in the sample, that is, the sequence reads of each ASV in the ASV table. sum Indicates the relative abundance of high-risk pathogenic fungi, that is, the proportion of labeled high-risk ASVs in all pathogens. The higher the RI, the higher the abundance of various pathogenic fungi and the proportion of high-risk pathogenic fungi in the sample, and the greater the risk.

[0042] The final pathogenic fungal ASV species table is shown in Table 1. The table uses the sample name as the column name, and different pathogenic fungal species correspond to different rows. The data in the table are absolute abundances, which are divided by the absolute abundance data. Figure 4 Comparison of risk index of different samples Figure 5 .

[0043] Table 1 Species of pathogenic fungi ASV in samples

[0044]

[0045] As can be seen from Table 1, this invention can obtain an ASV table containing only pathogenic fungal species, and can display the host type of pathogenic fungi, and also mark common high-risk species with *. Figure 4 It can be seen that the pathogenic fungi in different types of soil samples are different at the phylum level and genus level. At the phylum level, there are more Ascomycota and Morterellomycota in the diseased samples, and at the genus level, the abundance of Mortierella and Fusarium in the diseased samples is higher. Figure 5 It can be seen that the risk index of diseased samples is significantly higher than that of healthy samples (p<0.05). Therefore, it is concluded that the present invention can effectively diagnose complex fungal contamination and obtain complete and accurate information on pathogenic fungal species.

Claims

1. A method for detecting complex contamination of pathogenic fungi based on 18S rRNA and ITS amplicon sequencing, characterized in that It includes the following steps: S1. Build a non-redundant pathogenic fungi database; S2. Efficient extraction of sample DNA; S3, perform 18S rRNA and ITS gene sequencing of the sample to obtain sequencing data; S4, using the sequencing data and the DADA2 analysis process to encapsulate the data and obtain the ASV table of pathogenic fungi; S5. Visualize the community composition and diversity information in the ASV table to predict the risk of complex contamination by pathogenic fungi.

2. The method according to claim 1, characterized in that S1 specifically includes the following steps: S1-1. Obtain data and information on pathogenic fungi through Internet and literature retrieval; S1-2, merge and remove redundancy of fungal 18S rRNA or ITS sequences in UNITE, Sliva, NT, 18S, and ITS databases; S1-3. Match the Latin names of pathogenic fungi with the sequence information database to obtain the corresponding subspecies and strain information, and complete the construction of a non-redundant pathogenic fungi database.

3. The method according to claim 1, characterized in that In S2, the DNA of the sample is extracted by adsorbing the DNA using fungal lysis buffer combined with magnetic beads and then purified by washing solution.

4. The method according to claim 3, characterized in that The specific steps for sample DNA extraction are as follows: (1) Mix the soil sample to be tested with fungal lysis buffer and heat to 65°C for 30 minutes to release cellular NDA; (2) Transfer the mixture to a centrifuge tube and centrifuge at high speed for 10 minutes to separate cell debris and other impurities, and remove the supernatant for later use; (3) Add DNA magnetic beads to the supernatant, and after they fully absorb DNA, fix them in the container using a magnet and drain the remaining liquid; (4) After washing the beads with 70% ethanol several times to slowly remove other salts and impurities on the surface of the beads, the DNA adsorbed on the beads was eluted with PCR buffer.

5. The method according to claim 1, characterized in that In S3, 18S rRNA or ITS gene sequencing of the samples was completed using the Illumina PE250 sequencing platform.

6. The method according to claim 1, characterized in that In S4, the DADA2 analysis process was used to evaluate the quality of sequencing data, filter and denoise, merge read pairs, remove chimeras, and compare annotations to obtain the ASV table of pathogenic fungi.

7. The method according to claim 1, characterized in that In S4, the DADA2 analysis process specifically includes: 1) Install and load DADA2 in the R environment; 2) Use the plotQualityProfile function to visualize the quality of the original FASTQ file and decide on the cutting and filtering strategy based on the sequence quality; 3) Set filtering and cutting parameters according to the quality profile to remove low-quality sequences; 4) Accurately correct errors by learning a sequencing error model for each dataset; 5) Use the learned error model to denoise the sequence and identify actual biological sequence variations; 6) Splice the double-ended sequences into a complete sequence; 7) Construct a feature table from the denoised and concatenated sequences, where each column corresponds to a sample and each row corresponds to a noise-free feature sequence ASV; 8) Remove chimeras, classify the sequences in the ASV table using a non-redundant pathogenic fungi database, assign the sequences to the corresponding biological taxonomic groups, and output the final ASV table and classification information table.

8. The method according to claim 1, characterized in that In S5, the pathogenic fungi information in the ASV table was statistically analyzed using R language to calculate the RI value: RI=log 10 (∑A sum )+10×HR sum Among them, A sum Represents the absolute abundance of pathogenic fungi in the sample, that is, the sequence reads of each ASV in the ASV table; HR sum It represents the relative abundance of high-risk pathogenic fungi, that is, the proportion of marked high-risk ASVs in all pathogens; By testing and analyzing healthy and diseased samples, it is determined that RI ≤ 5.5 indicates a low pathogenic risk, 5.5 < RI ≤ 7 indicates a general pathogenic risk, and RI ≥ 7 indicates a high pathogenic risk, and a corresponding ASV abundance stacked bar chart is generated for easy understanding by users.

9. A system for detecting complex contamination of pathogenic fungi based on 18S rRNA and ITS amplicon sequencing, characterized in that: The system includes at least one processor and a memory connected to the processor, wherein: The memory is used to store computer programs or instructions; The processor is used to obtain and execute the computer programs or instructions in the memory, so that the system executes the method according to any one of claims 1-8.

Images