A method for pathogen identification and analysis based on metagenomic sequencing data

Abstract

The present invention relates to a pathogenic microorganism identification and analysis method based on metagenomic sequencing data, which comprehensively considers multiple evaluation factors, such as: the number of species sequences, the proportion of unique sequence numbers, the proportion of k-mer, relative abundance, clinical pathogenicity level, negative control detection degree and detection of the same batch, etc., determines the disease treatment evaluation index and calculates the pathogenicity score, and then classifies and screens the pathogenic microorganism database, background microorganism database and special concern database according to the size relationship between the pathogenicity score and the pathogenicity threshold, and finally outputs the pathogenic bacteria. The present invention can comprehensively and quantitatively evaluate the pathogenicity risk of each strain in the sample by comprehensively considering the multi-factor disease treatment evaluation index, more accurately judge whether a strain has a high pathogenicity potential, and help improve the accuracy of identification of pathogenic microorganisms and the reliability of identification results.

Description

Technical Field

[0001] The present invention relates to the technical field of pathogenic microorganism identification, and in particular to a method for pathogenic microorganism identification and analysis based on metagenomic sequencing data. Background Art

[0002] Pathogens are microorganisms that can cause infection in humans or animals, leading to disease. They include bacteria, fungi, viruses, parasites, and non-classical microorganisms. Diseases caused by pathogens are highly contagious and have the potential to cause a global pandemic. Therefore, detection of pathogens requires accuracy, speed, efficiency, and convenience.

[0003] Metagenomic sequencing (mNGS) is currently one of the most effective technologies for clinically identifying pathogenic microorganisms. Compared to traditional microbial detection methods, metagenomic sequencing can directly and unbiasedly sequence all nucleic acids in a sample, thereby detecting all microorganisms in the sample. This technology can quickly and accurately detect disease-causing pathogens and provides important value for etiological diagnosis and the identification of pathogens of emerging infectious diseases.

[0004] Currently, metagenomic sequencing technology produces complex results due to the influence of factors such as the environment, experimental background, and the human microbiome. Given this situation, manual analysis and screening of sequencing results are required. However, manual screening has certain limitations. First, manual screening takes a considerable amount of time, which may lead to delayed results. Second, due to the subjectivity of manual operations, there is a possibility of missing results. In addition, the accuracy of the interpretation of the results is also highly dependent on the professional level and experience of the interpreter. Therefore, to overcome these limitations, it is necessary to further explore automated and intelligent analysis tools to improve the data processing efficiency and accuracy of metagenomic sequencing technology. Summary of the Invention

[0005] To address these challenges, the present invention provides a method for pathogen identification and analysis based on metagenomic sequencing data. This method comprehensively evaluates multiple pathogenicity indicators to determine the pathogenicity score for each species in a sample. By setting appropriate thresholds, the present invention can identify pathogenic microorganisms in a sample and automatically analyze the pathogenicity risk of each species. This significantly improves the speed and accuracy of the interpretation process and provides a valuable reference for the future development of intelligent pathogen identification technologies.

[0006] The specific technical solutions of the present invention are:

[0007] A method for identifying and analyzing pathogenic microorganisms based on metagenomic sequencing data comprises the following steps:

[0008] Step S1: performing nucleic acid extraction, library construction, and sequencing on the sample data to obtain original pathogen metagenomic sequencing data, and performing data processing on the metagenomic sequencing data to obtain microbial sequencing data;

[0009] Step S2: Split the microbial sequencing data into k-mer sequence features and perform species identification to obtain a list of bacterial species in the sample;

[0010] Step S3: Setting the pathogenic bacteria evaluation index of the species and the corresponding weight;

[0011] Step S4: according to the corresponding indicators in the pathogenic bacteria evaluation index of the species, set the specific indicators in each indicator and calculate the corresponding weights;

[0012] Step S5: Calculate the corresponding indicator weight according to the pathogenic bacteria evaluation index, and finally calculate the pathogenicity score Score;

[0013] Step S6: Classify and screen the species according to their pathogenicity scores and pathogenicity levels, and output the pathogenic pathogens.

[0014] In the process of evaluating pathogens, the method of the present invention takes into account that the pathogenicity risk of a species is affected by the combined effects of multiple factors (species sequencing data, pathogenicity of the species, negative control detection data, and detection data from the same batch). Therefore, the sequencing data of the species, the pathogenicity of the species, the negative control detection data, and detection data from the same batch are used as evaluation indicators for pathogen judgment. The pathogenicity risks under these four indicators are quantified respectively, and then the pathogenicity risks are comprehensively quantified, thereby achieving a comprehensive, holistic, and effective evaluation based on the pathogenicity of pathogenic microorganisms.

[0015] Furthermore, the processing of the raw data in step S1 includes: splitting the primer barcode, filtering low-quality data, and removing human sequences to obtain a data set containing only pathogenic microorganism sequence information.

[0016] Furthermore, the pathogen evaluation index in step S3 includes species sequencing data, pathogenicity of the species, negative control detection data and detection data of the same batch.

[0017] Furthermore, the process of step S4 includes:

[0018] Step S41: Obtain the number of species sequences, the number of unique sequences, the k-mer ratio, and the relative abundance based on the species identification results, determine the evaluation indicators of the species sequencing data, including the number of sequences and relative abundance, and calculate the effective weight W1 of the sequencing data;

[0019] Step S42: Based on clinical concerns, pathogenicity of pathogens, and literature reports, a species pathogenicity level database is constructed, and the pathogenicity level database is divided into: level I, level II, and level III, and the pathogenicity weight W2 of the species is obtained:

[0020] Step S43: Calculate the number of sequences of the species in the negative control of this experiment and the historical negative control detection rate to obtain the negative control detection weight W3 of the species;

[0021] Step S44: Calculate the detection status of samples in the same batch, including the maximum, minimum, average, and median of the number of sequences detected in the same batch of samples, and calculate the same batch detection weight W4 of the species;

[0022] Furthermore, in step S5, the total weight is the sum of all weights to obtain the pathogenicity score Score, which is calculated as follows:

[0023]

[0024] Furthermore, in step S6, the weights of all species in sample T1 are calculated and sorted, and the pathogenic microorganism database, background microorganism database, and special concern database are classified and screened according to the relationship between the pathogenicity score and the pathogenicity threshold, and finally the pathogens are output.

[0025] Compared with the prior art, the present invention has the following advantages:

[0026] The pathogen identification and analysis method of the present invention comprehensively considers multiple evaluation factors (such as the number of species sequences, the proportion of unique sequences, the proportion of k-mers, relative abundance, clinical pathogenicity level, negative control detection rate, and detection of the same batch), determines the treatment evaluation index and calculates the pathogenicity score. Then, based on the relationship between the pathogenicity score and the pathogenicity threshold, it performs classification screening on the pathogenic microorganism database, the background microorganism database, and the special concern database, and finally outputs the pathogenic bacteria. This method can comprehensively and quantitatively assess the pathogenicity risk of each bacterial species in the sample, more accurately determine whether a bacterial species has a high pathogenic potential, and help improve the accuracy of pathogenic microorganism identification and the reliability of identification results. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 Flowchart for detecting pathogenic bacteria in an embodiment of the present invention.

[0028] Figure 2 Calculate the pathogenicity score of the sample pathogen in the embodiment of the present invention.

[0029] Figure 3 The flowchart of determining whether a bacterium is pathogenic is shown in an embodiment of the present invention. DETAILED DESCRIPTION

[0030] The present invention will be further described below with reference to the embodiments.

[0031] Example 1: A method for identifying pathogenic microorganisms based on metagenomic sequencing data

[0032] like Figure 1 As shown, the process of detecting pathogens includes steps S1-S6:

[0033] Step S1: Obtain metagenomic sequencing data of sample T1 and process the sequencing data to obtain microbial sequencing data C1;

[0034] Step S2: Split the microbial sequencing data set C1 into k-mer sequence features and perform species identification results. Obtain the number of species sequences, the number of unique species sequences, the k-mer ratio, the sample-negative control rate, the historical negative control detection rate, and the ratio of clinical samples in the same batch, and obtain the list of bacterial species in the sample (Species_Identification_result);

[0035] In step S2, the k-mer sequence feature length selected in this example is 35 bp, and the species reference database is obtained from the NCBI database;

[0036] The strain identification list obtained in step S2 detected 87 pathogenic microorganisms, and some of the identification results are shown in the following table:

[0037] Table 1 List of some identification results of strains

[0038]

[0039] Further, if Figure 2 As shown, sub-indicators are set for each evaluation indicator, and the corresponding weights of the evaluation indicators are calculated, and then the pathogenicity scores of the sample pathogens are calculated, which specifically involves steps S3-S5.

[0040] Step S3: Setting the pathogenic bacteria evaluation index of the species and the corresponding weight;

[0041] Preferably, four indicators are set as pathogenic bacteria evaluation indicators in step S3, including: species sequencing data, species pathogenicity, negative control detection data, and same-batch detection data. Among them, the weight of species valid data is set to 0.450, the weight of species corresponding clinical pathogenicity is 0.350, the weight of negative control detection analysis is 0.10, and the weight of same-batch detection analysis is 0.10. The sum of the weights of the four pathogenic bacteria evaluation indicators is 1.

[0042] Step S4: Set sub-indicators for each evaluation indicator and calculate the corresponding weight of the evaluation indicator, which specifically includes the following steps:

[0043] Step S41: Obtain the number of species sequences, the number of species unique sequences, the k-mer ratio, and the relative abundance based on the species identification results, determine the evaluation indicators of the species sequencing data, including the number of sequences and relative abundance, and calculate the effective weight W1 of the sequencing data;

[0044] Step S42: Construct a species pathogenicity level database, obtain clinical attention through discussion popularity on social media platforms and the number of search results, obtain the pathogenicity of pathogens through morbidity and mortality, disease burden research, medical insurance and health resource utilization data, and construct a species pathogenicity level database based on the reported status of literature impact factor and citation number. The pathogenicity level database is divided into: Level I (high clinical attention, high pathogenicity and many literature reports), Level II (high clinical attention, low pathogenicity and many literature reports), Level III (low clinical attention, low pathogenicity and few literature reports), and obtain the species pathogenicity weight W2:

[0045] Step S43: Calculate the number of sequences of the species in the negative control of this experiment and the historical negative control detection rate to obtain the negative control detection weight W3 of the species;

[0046] Step S44: Calculate the detection status of samples in the same batch, including the maximum, minimum, average, and median of the number of sequences detected in the same batch of samples, and calculate the same batch detection weight W4 of the species;

[0047] Preferably, the Unique sequence number in step S41 refers to the number of sequences of which the k-mer sequence feature of a sequencing data is completely aligned to the species; the Unique sequence number ratio refers to the ratio of the number of sequences uniquely aligned to the species to the number of sequences aligned to the species; and the k-mer ratio refers to the ratio of the number of k-mer features aligned to the species to the number of k-mer features aligned to the species;

[0048] Among them, the calculation formulas for the proportion of unique sequences and the proportion of k-mer are as follows:

[0049]

[0050]

[0051] Preferably, the number of species sequences of the parameter effective weight in step S41 is 0.450-0.550, the relative abundance is 0.450-0.550, and the sum of the weights is 1.

[0052] Preferably, the effective weight of the sequence number in step S41 is jointly determined by three sub-indicators: the number of seed sequences, the proportion of unique sequences, and the proportion of k-mer features; among which, the weight of the number of seed sequences is 0.540, the weight of the proportion of unique sequences is 0.300, and the weight of the proportion of k-mer features is 0.160.

[0053] Preferably, the seed sequence number weight sets three thresholds: when 0 < seed sequence number ≤ threshold 1, the seed sequence number is assigned 0.20; when threshold 1 < seed sequence number ≤ threshold 2, the seed sequence number is assigned 0.5; when threshold 2 < seed sequence number ≤ threshold 3, the seed sequence number is assigned 0.7; when threshold 2 < seed sequence number ≤ threshold 3, the seed sequence number is assigned 0.9.

[0054] Sequence number weight W reads The calculation formula is as follows:

[0055]

[0056] Preferably, the three thresholds of the number of seed sequences can be set to 0-10, 80-100 and 800-1000 respectively.

[0057] Preferably, the relative abundance W abundance Four thresholds can be set. When 0 < data parameter ≤ threshold 1, the data parameter is assigned 0.2; when threshold 1 < data parameter ≤ threshold 2, the data parameter is assigned 0.5; when threshold 2 < data parameter ≤ threshold 3, the number of sequences is assigned 0.7; when threshold 3 < data parameter ≤ threshold 4, the data parameter is assigned 0.85; when data parameter > threshold 4, the data parameter is assigned 0.95.

[0058] The four categories of relative abundance thresholds were set at 0.10%-0.11%, 1%-2%, 10%-11% and 20%-21%, respectively.

[0059] In step S41, the calculation formula of W1 is as follows:

[0060]

[0061] Preferably, the pathogenicity level database in step S42 includes common pathogenic microorganisms and common background colonizing microorganisms. The weight of level I in the pathogenicity level microorganisms is 0.89-0.91, the weight of level II is 0.69-0.71, and the weight of level III is 0.49-0.51. The weight of level I in the common background colonizing microorganisms is 0.49-0.51, the weight of level II is 0.29-0.31, and the weight of level III is 0.09-0.11.

[0062] In step S43, the number of sequences of the species in the negative control includes the sample-negative control rate in the negative control of the same batch of samples and the total negative control detection rate of the laboratory, where the weight W of the negative control of the same batch of samples ist is 0.745 - 0.755, and the weight Ws of the total detection rate of negative controls in the laboratory is 0.245 - 0.255. The formula for the negative control rate of the same batch of samples is as follows:

[0063]

[0064] When the number of detected sequence of species in the negative control is 0, the sample - negative control rate is 10000000;

[0065] Among them, when 0 < negative control - sample rate < 10, the weight is 0.25; when 10 < negative control - sample rate < 100, the weight is 0.6; when 101 < negative control - sample rate < 1000, the weight is 0.8; when 1001 < negative control - sample rate < 10000, the weight is 0.90; when 10001 < negative control - sample rate <1000000>0, the weight is 0.95; when negative control - sample rate = 10000000, the weight is 1.

[0066] In step S43, the formula for the total detection rate of negative controls in the laboratory is as follows:

[0067]

[0068] W3 = W t × sample - negative control rate + W s × (1 - total detection rate of negative controls).

[0069] In step S​​​​​​​​​​​​

[0074] According to the assumption that the importance of the number of seed sequences is slightly greater than that of the unique sequences and significantly greater than the proportion of the number of k-mer features, and the importance of the proportion of the number of unique sequences is slightly greater than the proportion of the number of k-mer features, an importance comparison matrix based on the number of sequences was constructed, where 1 indicates equal importance, 2 indicates slightly important, and 3 indicates significantly important. The weight vector of the three indicators is W = (0.54, 0.30, 0.16), that is, the weights of the number of seed sequences, the proportion of the number of unique sequences, and the proportion of the number of k-mer features are 0.540, 0.300, and 0.160, respectively. The results are shown in Table 2.

[0075] Table 2 Calculation results of importance comparison matrix and weight vector based on sequence number

[0076]

[0077] In step S41 , the four relative abundance thresholds are set to 0.1%, 1%, 10% and 20% respectively.

[0078] The pathogenicity level database and corresponding weights in step S42 are set as pathogenicity level I: 0.90, pathogenicity level II: 0.70, pathogenicity level III: 0.50, background level I: 0.50, background level II: 0.30, background level III: 0.10.

[0079] In step S43, the importance of the sample-negative control detection rate of negative control in this embodiment is significantly greater than the historical negative control detection results. An importance comparison matrix based on the negative control detection results is constructed, where 1 represents equal importance and 3 represents significantly important. The resulting weight vector of the two indicators is W = (0.75, 0.25), that is, the weights of the sample-negative control detection rate and the historical negative control detection are 0.75 and 0.25, respectively. The results are shown in Table 3.

[0080] Table 3 Importance comparison matrix and weight vector calculation results based on negative control detection results

[0081] Sample-negative control detection rate Historical negative control detection Weight Sample-negative control detection rate 1 3 0.75 Historical negative control detection 1 / 3 1 0.25

[0082] Step S5: Calculate the pathogenicity score based on the weight index of the species. The total weight is the sum of all weights to obtain the pathogenicity score Score. The calculation formula is as follows:

[0083]

[0084] The calculation results are shown in Table 4:

[0085] Table 4 Identification results

[0086]

[0087]

[0088] like Figure 3 As shown, referring to the species pathogenicity score, a pathogenicity threshold is set to comprehensively judge or screen out pathogenic bacteria, specifically involving step S6.

[0089] In step S6, the pathogenicity threshold is set and the process of determining pathogenic bacteria is as follows:

[0090] S61: If the weight of the species is less than the pathogenicity threshold, filter the species;

[0091] S62: If the weight of the species is greater than or equal to the pathogenicity threshold, the species is selected as the suspected pathogen list;

[0092] S621: If the species is in the pathogenic microorganism database, select it into the list of suspected pathogens;

[0093] S622: If the species is in the background colonizing microorganism database, select it and add it to the candidate pathogen list;

[0094] S623: The candidate pathogen list is sorted by weight, and the top three are selected to be included in the suspected pathogen list;

[0095] S63: Based on the pathogenic bacteria list, classify and rank them according to different genera. Among the same genus, the one with the highest pathogenicity score is selected as the pathogenic bacteria list, and all the pathogens of different genera are included in the pathogenic bacteria list.

[0096] In this example, the pathogenicity threshold is set to 0.70. Figure 3 The operation was performed to list species with a pathogenicity score greater than the pathogenicity threshold of 0.7 as suspected pathogens. The results were then screened through the pathogenic microorganism database and the background microorganism database, resulting in the identification of Mycoplasma pneumoniae, Streptococcus pneumoniae, herpes simplex virus type 1, human herpes virus type 4, and Haemophilus influenzae as suspected pathogens. Species with a pathogenicity score less than the pathogenicity threshold of 0.7 were then compared with the special concern database, which allowed the screening of Mycobacterium chelonae. Therefore, Mycoplasma pneumoniae, Streptococcus pneumoniae, herpes simplex virus type 1, human herpes virus type 4, Haemophilus influenzae, and Mycobacterium chelonae as pathogens were ultimately identified.

[0097] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any way. Any simple modification, change and equivalent transformation made to the above embodiment based on the technical essence of the present invention still fall within the scope of protection of the technical solution of the present invention.

Claims

1. A method for identifying and analyzing pathogenic microorganisms based on metagenomic sequencing data, characterized in that: The following steps are involved: Step S1: extracting nucleic acid, building a library and sequencing the sample data to obtain original pathogen metagenomic sequencing data, and processing the metagenomic sequencing data to obtain microbial sequencing data; Step S2: Split the microbial sequencing data into k-mer sequence features, perform species identification, obtain the number of species sequences, the number of unique sequences, the k-mer ratio, the sample-negative control rate, the historical negative control detection rate and the ratio of clinical samples in the same batch, and obtain a list of bacterial species in the sample; Step S3: setting the pathogenic bacteria evaluation index of the species to species sequencing data, species clinical pathogenicity, negative control detection data and same batch detection data; Step S4: according to the pathogenic bacteria evaluation index of the species described in step S3, set the specific index in each index and calculate the corresponding weight; Step S41: Obtain the number of species sequences, the number of unique sequences, the k-mer ratio and the relative abundance according to the species identification results, determine the evaluation index of the species sequencing data, and calculate the effective weight W1 of the species sequencing data; Step S42: construct a species pathogenicity level database based on clinical concerns, pathogenicity of pathogens, and literature reports, and divide the pathogenicity level database into: level I, level II, and level III. Determine and obtain the pathogenicity weight W2 of the species based on the level of the pathogenicity level database in which the species is located; Step S43: Calculate the number of sequences of species in negative control and the historical negative control detection rate, and obtain the negative control detection weight W3 of the species; Step S44: Calculate the detection status of samples in the same batch, including the maximum, minimum, average, and median of the number of sequences detected in the same batch of samples, and calculate the same batch detection weight W4 of the species; Step S5: Calculate the corresponding indicator weight according to the pathogenic bacteria evaluation index, and finally calculate the pathogenicity score Score; Step S6: Classify and screen the species according to their pathogenicity scores and pathogenicity levels, and output the pathogenic pathogens; The calculation formula of the effective weight W1 in step S41 is as follows: Among them, W in the calculation formula of effective weight W1 reads The calculation formula is as follows: Among them, W reads The calculation formula for the proportion of unique sequences and k-mer ratio in the calculation formula is as follows: The method for determining the pathogenicity weight W2 in step S42 is as follows: a. If the acquired species belongs to the pathogenicity level I microorganism, the pathogenicity weight W2 is 0.89-0.91, if it is level II, the pathogenicity weight W2 is 0.69-0.71, if it is level III, the pathogenicity weight W2 is 0.49-0.51, b. If the acquired species belongs to level I among the common background colonizing microorganisms, the pathogenicity weight W2 is 0.49-0.51, if it is level II, the pathogenicity weight W2 is 0.29-0.31, and if it is level III, the pathogenicity weight W2 is 0.09-0.11; The calculation formula of the negative control detection weight W3 in step S43 is as follows: W3=W t × sample - negative control rate + W s ×(1-total negative control detection rate) The calculation formula for the sample-negative control rate is as follows: The calculation formula for the total negative control detection rate is as follows: The calculation formula of the pathogenicity score Score in step S5 is as follows:

2. A method for identifying and analyzing pathogenic microorganisms based on metagenomic sequencing data according to claim 1, characterized in that: In step S3, the weight of the species sequencing data is set to 0.445-0.450, the weight of the species clinical pathogenicity is set to 0.245-0.250, the weight of the negative control detection data analysis is set to 0.145-0.150, and the weight of the same batch detection data analysis is set to 0.145-0.

150.

3. The method for pathogenic microorganism identification and analysis based on metagenomic sequencing data according to claim 1, characterized in that: The effective weight of the species sequence number in the calculation formula of the effective weight W1 in step S41 is 0.450-0.550, the effective weight of the relative abundance is 0.450-0.550, and the sum of the weights is 1.

4. The method for pathogenic microorganism identification and analysis based on metagenomic sequencing data according to claim 1, characterized in that: The effective weight W1 is calculated by the formula W reads In the calculation formula, the weight of the number of seed sequences is 0.540, the weight of the proportion of the number of unique sequences is 0.300, and the weight of the proportion of k-mers is 0.

160.

5. The method for pathogenic microorganism identification and analysis based on metagenomic sequencing data according to claim 1, characterized in that: In the calculation formula of the negative control detection weight W3 in step S43, W t It is 0.745-0.755, and Ws is 0.245-0.

255.

6. The method for pathogenic microorganism identification and analysis based on metagenomic sequencing data according to claim 1, characterized in that: The weight of species valid data in the pathogenicity score calculation formula in step S5 is set to 0.450.

Images