Taenia asiaticus infection intestinal flora markers as well as screening method and application thereof
By metagenomic sequencing and random forest model construction of microbial DNA in feces of Asian Talisworm infected patients and healthy people, 28 specific species of bacteria were screened as markers, solving the problem of insufficient diagnostic sensitivity and specificity in the prior art, and achieving high accuracy early diagnosis.
Patent Information
- Application Number
- CN202510034380.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-09
- Publication Date
- 2025-04-15
AI Technical Summary
The prior art has problems of insufficient sensitivity and specificity in the early diagnosis of Asian tapeworm infection, especially immunological detection methods often cross-react with other tapeworms and flukes, resulting in a high false positive rate.
By metagenomic sequencing of microbial DNA in feces of healthy people and Asian Talisworm infected patients, the abundance of intestinal microorganisms was calculated, and a random forest model was constructed, and 28 specific species of flora were screened as markers of intestinal flora infection in Asian Talisworm infection.
The high sensitivity and specificity of early diagnosis were achieved, and the constructed predictive model was able to effectively distinguish between Asian stalactite infected patients and healthy people in a variety of samples, with an accuracy of more than 96%.
Smart Images
Figure CN119824086A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of disease diagnosis methods, and specifically relates to a group of intestinal flora markers of Asian tapeworm infection and a screening method and application thereof. Background Art
[0002] Infection with Asian tapeworms has no clinical symptoms in the early stages of infection, and usually only after the worms grow into adults do obvious symptoms appear, such as abdominal pain, nausea, diarrhea, or constipation. Currently, early diagnosis of Asian tapeworm infection is usually achieved through stool examination, immunological examination, ultrasound examination, and imaging examination. Immunological detection methods are superior to ultrasound examinations in early diagnosis, but they often cross-react with other tapeworms and flukes, resulting in a high false positive rate. There is an urgent need to develop new methods that are sensitive, specific, and have diagnostic value.
[0003] Molecular biological methods based on specific intestinal microbial markers have been proven to be sensitive, specific, and efficient detection methods in a variety of diseases. Unfortunately, however, there are currently no reports on highly specific and sensitive intestinal flora markers for Asian tapeworm infection. Summary of the invention
[0004] In view of this, the purpose of the present invention is to provide a group of intestinal flora markers for Asian tapeworm infection, which have the characteristics of good early diagnosis specificity and high sensitivity, laying a foundation for disease research.
[0005] The present invention provides a group of intestinal flora markers of Asian tapeworm infection, including the following species: Phascolarctobacterium succinatutens, GGB3293_SGB4348, Prevotellamassiliatimonensis, Prevotella copri clade_C, Bacteroidaceae bacterium ), Adlercreutzia equolifaciens, Prevotella strain SGB1653, Roseburia sp. strain AM23_20, Ruminococcus sp. strain NSJ_71, Coprococcus eutactus, GGB3277_SGB4327, Bifidobacterium longum, Roseburia sp. strain AF02_12, GGB3256_SGB4303, Clostridium strain SGB4751, Solobacterium strain SGB6833, Bifidobacteriumpseudocatenulatum, Flavonifractor plautii, Desulfovibrio piger), GGB1266_SGB1699, Prevotella pectinovora, Bifidobacteriumadolescentis, GGB38171_SGB72433, Clostridiales Family_XIII_bacterium strain WCA_MUC_591_APC_4B, Bacteroidales bacterium, Prevotella copri clade_A, Gemmiger strain SGB15295 and Prevotella strain SGB1680.
[0006] The present invention provides a method for screening intestinal flora markers of Asian tapeworm infection, comprising the following steps:
[0007] Microbial DNA was extracted from the feces of healthy individuals and patients infected with Asian tapeworm;
[0008] The microbial DNA in the feces of healthy individuals and patients infected with Asian tapeworms obtained above was used as material for metagenomic sequencing to calculate the abundance values of intestinal microorganisms at the species level;
[0009] Using the abundance values of the intestinal microorganisms at the species level as input data, constructing a random forest model;
[0010] The importance scores of the variable features in the random forest model were calculated, the importance scores of the variable features were sorted from high to low, and 28 specific species with higher importance scores were selected to obtain intestinal flora markers for Asian tapeworm infection.
[0011] Preferably, the method for calculating the abundance value of intestinal microorganisms at the species level includes counting the number of sequencing reads corresponding to each sequence obtained by sequencing to obtain the abundance value of the species.
[0012] Preferably, the criterion for selecting the 28 specific species with higher importance scores is to select the number of specific species with importance scores greater than 2.
[0013] The present invention provides the use of the intestinal flora marker of Asian tapeworm infection or the intestinal flora marker of Asian tapeworm infection obtained by the screening method in constructing a prediction model for Asian tapeworm infection.
[0014] Preferably, the method for constructing a prediction model for Asian tapeworm infection comprises the following steps:
[0015] The abundance values of the intestinal flora markers of Asian tapeworm infection are used as input data to construct a prediction model for Asian tapeworm infection.
[0016] Preferably, when constructing the prediction model for Asian tapeworm infection, the method further includes inputting the optimal mtry parameter;
[0017] The screening method of the optimal mtry parameter is to perform a grid search using a ten-fold cross validation method;
[0018] The optimal mtry parameter is 2.
[0019] The present invention provides a system for predicting Asian tapeworm infection, comprising the following components:
[0020] Data input module: used for inputting the abundance value of the intestinal flora marker of Asian tapeworm infection in the sample to be tested;
[0021] Prediction and analysis module: used to analyze the data input by the data input module using the Asian tapeworm infection prediction model constructed in the application and obtain analysis results;
[0022] And result display module: used to output the analysis results obtained by the prediction analysis model to the terminal for display.
[0023] The present invention provides a device for predicting Asian tapeworm infection, which is equipped with the Asian tapeworm infection prediction system.
[0024] The present invention provides a group of intestinal flora markers for Asian tapeworm infection, including 28 specific bacteria. The present invention obtains the abundance of bacteria by performing metagenomic sequencing on the bacterial components in the feces of healthy patients with Asian tapeworm infection, finds the flora of a specific species as a microbial marker through screening operations, and uses the data with class labels to optimize the model parameters to improve the accuracy and sensitivity of the model, predicts the probability of Asian tapeworm infection according to the output risk value, and indicates the balance state of intestinal microorganisms at the same time. Experiments have shown that the intestinal flora markers for Asian tapeworm infection provided by the present invention can be effectively used to construct a risk prediction model for Asian tapeworm infection, and have high prediction sensitivity and good specificity. It can effectively distinguish patients with Asian tapeworm infection and healthy people in multiple samples with an accuracy of more than 96%, laying a foundation for the study of diseases related to Asian tapeworm infection. BRIEF DESCRIPTION OF THE DRAWINGS
[0025] Figure 1 A flow chart of detecting Asian tapeworm disease based on Asian tapeworm disease microbial markers provided by the present invention;
[0026] Figure 2 The importance (Mean DecreaseAccuracy) graph of 28 features screened by the random forest model in Example 3;
[0027] Figures 3 to 10 The comparison results of the differences in microbial characteristics between the healthy group and the disease group in Example 3;
[0028] Fig.11 4, where the left figure is the ROC curve of the training set and the right figure is the ROC curve of the test set. DETAILED DESCRIPTION
[0029] The invention provides a group of intestinal flora markers of Asian tapeworm infection, including the following species: Phascolarctobacterium succinatutens (Taxonomy ID: 626940), GGB3293_SGB4348, Prevotellamassilia timonensi (Taxonomy ID: 1852370), Prevotella copri clade_C (Prevotella mycobacterium branch C), Bacteroidaceae bacterium (Bacteroidaceae bacterium, Taxonomy ID: 2212467 ), Adlercreutzia equolifaciens (Taxonomy ID: 446660), Prevotella (Taxonomy ID: 838) strain SGB1653, Roseburia sp. Taxonomy ID: 2049040 strain AM23_20, Ruminococcus sp. Taxonomy ID: 41978 strain NSJ_71, Coprococcus eutactus (Taxonomy ID: 33043), GGB3277_SGB4327, Bifidobacterium longum (Taxonomy ID: 216816), Roseburia sp. Taxonomy ID: 2049040 strain AF02_12, GGB3256_SGB4303, Clostridium The strains of SGB4751, Solobacterium, SGB6833, Bifidobacteriumpseudocatenulatum, Taxonomy ID: 28026, Flavonifractor plautii, Taxonomy ID: 292800, Desulfovibriopiger, Taxonomy ID: 901, GGB1266_SGB1699, Prevotellapectinovora, Taxonomy ID: 1602169, Bifidobacteriumadolescentis, Taxonomy ID: 1680, GGB38171_SGB72433, Clostridiales Family III, Taxonomy ID: 28026, Flavonifractor plautii, Taxonomy ID: 292800, Desulfovibriopiger, Taxonomy ID: 901, GGB1266_SGB1699, Prevotellapectinovora, Taxonomy ID: 1602169, Bifidobacteriumadolescentis, Taxonomy ID: 1680, GGB38171_SGB72433, Clostridiales Family XIII, bacteria ID: 2137877) strain WCA_MUC_591_APC_4B, Bacteroidales bacteriumTaxonomy ID: 2030927, Prevotella copri clade_A, Gemmiger strain SGB15295 and Prevotella Taxonomy ID: 838 strain SGB1680.
[0030] The present invention provides a method for screening intestinal flora markers of Asian tapeworm infection, which specifically comprises the following steps:
[0031] Microbial DNA was extracted from the feces of healthy individuals and patients infected with Asian tapeworm;
[0032] The microbial DNA in the feces of healthy individuals and patients infected with Asian tapeworms obtained above was used as material for metagenomic sequencing to calculate the abundance values of intestinal microorganisms at the species level;
[0033] Using the abundance values of the intestinal microorganisms at the species level as input data, constructing a random forest model;
[0034] The importance scores of the variable features in the random forest model were calculated, the importance scores of the variable features were sorted from high to low, and 28 specific species with higher importance scores were selected to obtain intestinal flora markers for Asian tapeworm infection.
[0035] In the present invention, the number of healthy individuals or patients infected with Asian tapeworm is preferably not less than 50, and can be 55 or 60. The feces of the healthy individuals or patients infected with Asian tapeworm is used as a test set sample. The present invention has no special restrictions on the method for extracting microbial DNA in feces, and an extraction method known in the art can be used. In an embodiment of the present invention, it is completed using a fecal DNA kit.
[0036] In the present invention, the metagenomic sequencing is entrusted to Anshengda Company. After obtaining the sequencing results, the sequencing results are preferably preprocessed to obtain the preprocessed sequencing results. The preprocessing method of the sequencing results preferably includes using KneadData (v0.10.0) software to remove low-quality data and possible host DNA data in the original data. The judgment criteria for the low-quality data are reads with a beginning and end quality less than 3, an average quality value less than 20, and a length less than 36 bases. The present invention also preferably performs microbial community abundance analysis and community classification on the pretreated sequencing results. The microbial community abundance analysis, a method for calculating the abundance value of intestinal microorganisms at the species level, preferably includes the number of sequencing reads corresponding to each sequence obtained by statistical sequencing, and obtains the abundance value of the species. The software for the microbial community abundance analysis is preferably MetaPhlAn4 (v4.0.3). The community classification preferably uses a database (mpa_vJan21_CHOCOPhlAnSGB_202103) to align each sequence to a bacterial database for classification annotation, so as to achieve intestinal microorganism classification.
[0037] In the present invention, the method for calculating the importance score of the variable feature in the random forest model is preferably to use the important() function for calculation. The standard for selecting 28 specific species with higher importance scores is preferably to select the number of specific species with importance scores greater than 2.
[0038] The present invention provides the use of the intestinal flora marker of Asian tapeworm infection or the intestinal flora marker of Asian tapeworm infection obtained by the screening method in constructing a prediction model for Asian tapeworm infection.
[0039] In the present invention, the method for constructing a prediction model for Asian tapeworm infection preferably includes the following steps: using the abundance value of the intestinal flora marker of Asian tapeworm infection as input data to construct a prediction model for Asian tapeworm infection. When constructing the prediction model for Asian tapeworm infection, the input of the optimal mtry parameter is also included; the screening method for the optimal mtry parameter is to perform a grid search using a ten-fold cross-validation method; the optimal mtry parameter is 2.
[0040] In the present invention, after constructing the prediction model for Asian tapeworm infection, it is preferred to further include model verification. The method for model verification includes inputting the abundance values of 28 specific species in the sequencing data of healthy individuals and patients infected with Asian tapeworms as input data into the constructed prediction model for Asian tapeworm infection, and judging the reliability of the prediction result of the prediction model for Asian tapeworm infection according to the consistency of the output value risk value compared with the actual situation: when the consistency reaches more than 70%, it means that the prediction result of the prediction model for Asian tapeworm infection is reliable; otherwise, the prediction result of the prediction model for Asian tapeworm infection is unreliable. The number of samples of healthy individuals or patients infected with Asian tapeworms is not less than 20, and can be 25 cases and 30 cases. When the output value risk value is greater than or equal to 0.5, it is judged that the source of the sample is a patient with Asian tapeworm disease, and when the output value risk value is less than 0.5, it is judged that the source of the sample is a healthy individual. The verification consistency of the present invention is 100%. At the same time, the Asian tapeworm infection prediction model of the present invention also performs sensitivity and specificity tests, specifically drawing ROC curves, and calculating the area under the curve (ROC-AUC) and 95% confidence interval curves of the average subject curve of the test set. The results show that the accuracy of the Asian tapeworm infection prediction model constructed by the present invention is 96% (AUC = 0.992), indicating that the Asian tapeworm infection prediction model constructed by the present invention has high prediction sensitivity and good specificity, and can effectively distinguish positive results (Asian tapeworm disease patients) and negative results (healthy individuals) in multiple samples, laying a foundation for the diagnosis and research of diseases related to Asian tapeworm infection.
[0041] The present invention provides a system for predicting Asian tapeworm infection, comprising the following components:
[0042] Data input module: used for inputting the abundance value of the intestinal flora marker of Asian tapeworm infection in the sample to be tested;
[0043] Prediction and analysis module: used to analyze the data input by the data input module using the Asian tapeworm infection prediction model constructed in the application and obtain analysis results;
[0044] And result display module: used to output the analysis results obtained by the prediction analysis model to the terminal for display.
[0045] The present invention provides a device for predicting Asian tapeworm infection, which is equipped with the Asian tapeworm infection prediction system.
[0046] The following is a detailed description of a group of intestinal flora markers for Asian tapeworm infection and their screening methods and applications provided by the present invention in conjunction with the examples, but they should not be construed as limiting the scope of protection of the present invention.
[0047] Example 1
[0048] The method for obtaining sample sequencing data is shown in Figure 1 , specifically including the following steps:
[0049] (1) Fecal samples from 79 patients with Asian tapeworms and 87 healthy controls were collected from the hospital. DNA was extracted from the fecal samples of patients with Asian tapeworms and healthy controls using a stool DNA extraction kit, and metagenomic sequencing was commissioned to Anshengda Company to obtain metagenomic sequencing data of intestinal flora;
[0050] (2) KneadData (v0.10.0) software was used to remove low-quality data and possible host DNA data from the original data to obtain high-quality data; low-quality data refers to reads with a beginning and end quality less than 3, an average quality value less than 20, and a length less than 36 bases.
[0051] (3) The MetaPhlAn4 (v4.0.3) software was used to analyze the microbial community abundance of the above high-quality data, and the database (mpa_vJan21_CHOCOPhlAnSGB_202103) was used to classify the community to obtain the bacterial abundance values of all samples.
[0052] Example 2
[0053] Model algorithm selection and parameter optimization
[0054] Use machine learning to select model algorithms.
[0055] Random forest uses a classifier of multiple decision trees, and the category of its output is determined by the mode of the category output by the individual trees. The cases of the sample data of the present invention have been clinically diagnosed and have classified labels, so they will be explored and selected in the supervised machine learning classification model. The relative abundance values of bacteria in all samples are used as input data, and the diagnostic results of the samples are used as output classification labels. The algorithm is constructed according to the following steps:
[0056] (1) The bacterial abundance values and sample information data of all samples in Example 1 were randomly divided into a 70% training set (n=116) and a 30% test set (n=50);
[0057] (2) Construct a machine learning classifier. The abundance values of all 440 bacterial species in the training set were used as input data for training to obtain a preliminary random forest model, which was used to judge the importance of subsequent variable features.
[0058] Example 3
[0059] Selection of specific bacterial species and optimization of parameters
[0060] (1) Calculate the Mean DecreaseAccuracy score of the training set variable features obtained by the random forest model (see Figure 2 ), and sort them from high to low according to the scores.
[0061] (2) Select bacteria with importance scores greater than 2.
[0062] (3) The abundance values of 28 specific species of the input sample are used to construct a random forest model. The microbial markers include the following 28 specific species combinations:
[0063] "Phascolarctobacterium_succinatutens", "GGB3293_SGB4348", "Prevotellamassilia_timonensis", "Prevotella_copri_clade_C", "Bacteroidaceae_bacterium", "Adlercreutzia_equolifaciens", "Prevote lla_SGB1653","Roseburia_sp_AM23_20","Ruminococcus_sp_NSJ_71","Coprococcus_eutactus","GGB32 77_SGB4327","Bifidobacterium_longum","Roseburia_sp_AF02_12","GGB3256_SGB4303","Clostridium_ SGB4751","Solobacterium_SGB6833","Bifidobacterium_pseudocatenulatum","Flavonifractor_plautii","Desulfovibrio_piger","GGB1266_SGB1699","Prevotella_pectinovora","Bifidobacterium_adol escentis", "GGB38171_SGB72433", "Clostridiales_Family_XIII_bacterium_WCA_MUC_591_APC_4B", "Bacteroidales_bacterium", "Prevotella_copri_clade_A", "Gemmiger_SGB15295" and "Prevotella_SGB1680".
[0064] Compared with the healthy control group, the relative abundance of 22 bacteria in the 28 specific species combinations increased, and 6 bacteria decreased (see Figures 3 to 10 ).
[0065] Example 4
[0066] Construction and testing of the optimal random forest model
[0067] (1) Input the 28 specific species obtained in Example 3 to obtain the optimal random forest model. Store the model for subsequent risk prediction of Taeniasis Asiatica for measurement data;
[0068] (2) Model evaluation: The test set was used to input the above model and perform sensitivity and specificity tests; the ROC curve was drawn and the area under the curve (ROC-AUC) and 95% confidence interval curve ( Fig.11 ):
[0069] (3) Risk value output: The trained algorithm model is input with the test data obtained from the experiment to obtain a probability between 0 (control) and 1 (suffering from Asian tapeworm disease). Finally, the probability value of 1 (suffering from the disease) is confirmed as the risk value. A probability value less than 0.5 is considered healthy, and a probability value greater than 0.5 is considered a high-risk group for Asian tapeworm disease. Ultrasound examination is recommended for diagnosis. If a person suffers from Asian tapeworm disease, a basic judgment of the intestinal fungus can also be given to the clinician for reference. The actual disease conditions and risk values of the 50 test subjects are shown in Table 1.
[0070] Table 1 Actual disease status and risk values of 50 test set subjects
[0071]
[0072]
[0073] As shown in Table 1, the microbial markers of Asian tapeworm disease of the present invention can be effectively used to construct a risk prediction model for Asian tapeworm disease, and have high prediction sensitivity and good specificity, and can effectively distinguish positive results (Asian tapeworm disease patients) and negative results (healthy people) in multiple samples, providing effective data for diagnosis. In actual situations, the accuracy of most prediction models is generally 50% to 90%, and it is difficult to reach 100%. However, the present invention can achieve an accuracy of 96% (AUC=0.992) by selecting appropriate and corresponding input variables and cooperating with the optimized model, laying a foundation for disease research.
[0074] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.
Claims
1. A group of intestinal flora markers of Asian tapeworm infection, characterized in that: The following strains are included: Phascolarctobacterium succinatutens, GGB3293_SGB4348, Prevotellamassiliatimonensi, Prevotella copri clade_C, Bacteroidaceae bacterium, Adlercreutzia equolifaciens, Prevotella strain SGB1653, Roseburia sp. strain AM23_20, Ruminococcus sp. strain NSJ_71, Coprococcus eutactus, GGB3277_SGB4327, Bifidobacterium longum, Roseburia sp. strain AF02_12, GGB3256_SGB4303, Clostridium strain SGB4751, Solobacterium strain SGB6833, Bifidobacteriumpseudocatenulatum, Flavonifractorplautii, Desulfovibriopiger, GGB1266_SGB1699, Prevotella pectinovora, Bifidobacterium adolescentis, GGB38171_SGB72433, Clostridiales Family_XIII_bacterium strain WCA_MUC_591_APC_4B, Bacteroidales bacterium, Prevotella copri clade_A), Gemmiger strain SGB15295 and Prevotella strain SGB1680.
2. A method for screening intestinal flora markers of Asian tapeworm infection according to claim 1, characterized in that: The following steps are involved: Microbial DNA was extracted from feces of healthy individuals and patients infected with Asian tapeworm; The microbial DNA in the feces of healthy individuals and patients infected with Asian tapeworms obtained above was used as material for metagenomic sequencing to calculate the abundance values of intestinal microorganisms at the species level; Using the abundance values of the intestinal microorganisms at the species level as input data, constructing a random forest model; The importance scores of the variable features in the random forest model were calculated, the importance scores of the variable features were sorted from high to low, and 28 specific species with higher importance scores were selected to obtain intestinal flora markers for Asian tapeworm infection.
3. The screening method according to claim 2, characterized in that: The method for calculating the abundance value of intestinal microorganisms at the species level includes counting the number of sequencing reads corresponding to each sequence obtained by sequencing to obtain the abundance value of the species.
4. The screening method according to claim 2, characterized in that: The criterion for selecting the 28 specific species with higher importance scores is to select the number of specific species with importance scores greater than 2.
5. Use of the intestinal flora marker for Asian tapeworm infection according to claim 1 or the intestinal flora marker for Asian tapeworm infection obtained by the screening method according to any one of claims 2 to 4 in constructing a prediction model for Asian tapeworm infection.
6. The use according to claim 5, characterized in that: The method for constructing a prediction model for Asian tapeworm infection comprises the following steps: The abundance value of the intestinal flora marker of Asian tapeworm infection described in claim 1 is used as input data to construct a prediction model for Asian tapeworm infection.
7. The use according to claim 6, characterized in that: When constructing the Asian tapeworm infection prediction model, the optimal mtry parameter is also input.
8. The use according to claim 7, characterized in that: The screening method of the optimal mtry parameter is to perform a grid search using a ten-fold cross validation method; The optimal mtry parameter is 2.
9. A system for predicting Asian tapeworm infection, characterized in that: It includes the following components: Data input module: used for inputting the abundance value of the intestinal flora marker of Asian tapeworm infection as described in claim 1 in the sample to be tested; Prediction and analysis module: used to analyze the data input by the data input module using the Asian tapeworm infection prediction model constructed in the application of any one of claims 6 to 8, and obtain analysis results; And result display module: used to output the analysis results obtained by the prediction analysis model to the terminal for display.
10. A device for predicting Asian tapeworm infection, characterized in that: Equipped with the Asian tapeworm infection prediction system according to claim 9.