Marker combination for ventricular remodeling diagnosis, application, diagnosis model and system

Abstract

The invention belongs to the technical field of biological medicine, a series of differential miRNAs and intestinal strains are innovatively screened out in a patient queue by applying intestinal flora metagenomics and intestinal bacteria source plasma exosome miRNA sequencing, variable screening is carried out by adopting LASSO regression and gradual backward regression, and finally, two miRNAs including miR1403p and miR10a5p and two intestinal strains including AnaliterunusspG32012 are obtained through screening. According to the heart failure early-stage recognition method based on the Bifidobacteriumlonium, Bifidobacteriumlonium is used as a marker combination, meanwhile, a diagnosis model is constructed, high sensitivity and specificity are achieved for recognition of ventricular remodeling, the efficiency of early-stage recognition of heart failure can be improved, ventricular remodeling can be better found, recognized and intervened as soon as possible, heart failure can be better prevented and delayed, the heart failure early-stage recognition method can be applied to children, full-life-cycle screening and monitoring can be achieved, and the heart failure early-stage recognition method based on the Bifidobacteriumlonium can be used for early-stage recognition of heart failure. The method is expected to be applied to clinical examination and has clinical transformation value.

Description

Technical Field

[0001] The present invention belongs to the field of biomedicine technology, and specifically relates to a marker combination and application, a diagnostic model and a system for ventricular remodeling diagnosis. Background Art

[0002] Heart failure is a clinical syndrome in which the filling and / or ejection capacity is impaired and the body's blood supply needs cannot be met due to abnormal cardiac structure and / or function. It is the terminal state of various cardiovascular diseases in the late stage. It is reported that the 1-year mortality rate of patients with heart failure is 15% to 30%, while the 5-year mortality rate is as high as 50% to 75%. The pathophysiological mechanism of heart failure is relatively complex. It is currently recognized that changes in the neurohumoral mechanism characterized by sympathetic nerve excitation and activation of the renin-angiotensin system ultimately lead to ventricular remodeling, which is often accompanied by an increase in a series of biomarkers. Heart diseases with increased left heart pressure load, such as aortic valve stenosis and aortic arch stenosis, are most likely to lead to ventricular remodeling and subsequent heart failure. For such patients, surgery is the only way to relieve obstruction. However, ventricular remodeling often exists before surgical correction and develops progressively. Even after surgical correction, residual stenosis and obstruction occur in a large number of cases, further slowly aggravating the process of ventricular remodeling. Failure to detect and intervene in ventricular remodeling in a timely manner can lead to irreversible myocardial hypertrophy and interstitial fibrosis, significantly increase the risk of heart failure and surgery, and worsen long-term outcomes. In order to achieve the goal of life-long health, early detection, early diagnosis, and early intervention of ventricular remodeling are required for such patients. However, the optimal time for intervention is often difficult to determine. Currently available indicators mainly depend on gradient pressure differences and the presence or absence of symptoms. These indicators cannot indicate the extent of heart damage and functional deterioration in the early stages. Therefore, the field of ventricular remodeling and heart failure urgently needs to develop more sensitive and convenient diagnostic methods and diagnostic systems, and establish effective methods for preventing ventricular remodeling.

[0003] At present, the biomarkers widely used in clinical practice include B-type natriuretic peptide (BNP), N-terminal pro-B-type natriuretic peptide (NT-proBNP), myocardial injury markers such as cardiac troponin I and cardiac troponin T, and myocardial extracellular matrix remodeling markers such as galectin 3. However, these clinical indicators can only indicate patients with heart failure or myocardial injury, and cannot effectively indicate ventricular remodeling. In recent years, due to the development of metabolomics technology, researchers have found that the levels of lipids, ketone bodies, and amino acid-related metabolites in the blood circulation of patients with heart failure are increased, such as long-chain acylcarnitines, ceramides, ketone bodies, branched-chain amino acids, etc., and have shown application prospects in the early diagnosis, prognosis assessment and treatment of heart failure. In addition, metabolites related to intestinal flora such as trimethylamine oxide and short-chain fatty acids have also gradually attracted attention, and all have the potential to be used as molecular markers of chronic heart failure. However, due to the great individual differences in metabolites and susceptibility to various factors such as diet, disease, drug use, gender, age, etc., the diagnostic effectiveness needs to be improved.

[0004] In addition, conventional assessment methods for ventricular remodeling before heart failure rely on histopathology and imaging. Although histology is the "gold standard" for diagnosing ventricular remodeling, ventricular muscle samples are usually not available, the operation is risky, and the feasibility is low. The main imaging examination methods include echocardiography, CT, and MRI. The combined diagnosis of the three imaging methods has a good display of cardiac morphology, ventricular thickness, ventricular wall motion, and contractile function. However, at present, only echocardiography is used as a routine examination tool in follow-up. CT and MRI are expensive, take a long time to scan, and have low patient acceptance and popularity. Commonly used molecular markers of heart failure, such as brain natriuretic peptide (BNP) and N-terminal pro-B-type natriuretic peptide (NT-proBNP), have outstanding performance in predicting the diagnosis and prognosis of patients with heart failure, but they are greatly affected by factors such as age and renal function, and have certain limitations, which limits their application in children. Myocardial injury markers such as creatine kinase, creatine kinase isoenzymes, cardiac troponin I, and cardiac troponin T specifically reflect myocardial injury and may have potential in acute heart failure caused by ischemic heart disease, but are difficult to evaluate ventricular remodeling caused by chronic heart failure. In addition, in recent years, researchers have discovered more new markers involving metabolites, RNA, etc., but these molecular biological markers are not yet mature and are in the animal experiment or preclinical testing stage, and most of the populations or animal models selected in the studies are adults, so it is impossible to know the sensitivity and specificity in the pediatric population. The current first-line drugs can reduce deaths after heart failure, but it is difficult to cure or reverse heart failure. Therefore, early diagnosis and treatment of heart failure may greatly slow down or even reverse ventricular remodeling and improve the prognosis of children with heart failure. But neither BNP / NT-proBNP nor other myocardial injury markers can do this. Summary of the invention

[0005] The present invention is made to solve the above-mentioned problems, and aims to provide a marker combination and application, a diagnostic model and a system for ventricular remodeling diagnosis.

[0006] The present invention provides a marker combination for diagnosing ventricular remodeling, which has the following characteristics and comprises: miRNA and intestinal bacteria, wherein the miRNA is miR_140_3p and miR_10a_5p, and the intestinal bacteria are Anaeritruncus_sp_G3_2012 and Bifidobacterium_longum.

[0007] The present invention also provides a diagnostic model constructed using the above-mentioned marker combination for ventricular remodeling diagnosis, characterized in that: the combination formula of the diagnostic model is:

[0008] Panel = 1.0711702 + (-0.1128574 * Anaeritruncus_sp_G3_2012) + (0.1847674 * miR_140_3p) + (-0.1081418 * miR_10a_5p) + (-0.6567257 * Bifidobacterium_longum), in which Panel is the diagnostic model, miR_140_3p and miR_10a_5p are the expression levels of miR_140_3p and miR_10a_5p in plasma exosomes, and Anaeritruncus_sp_G3_2012 and Bifidobacterium_longum are the abundance of Anaeritruncus_sp_G3_2012 and Bifidobacterium_longum in feces.

[0009] The present invention also provides a ventricular remodeling diagnosis system, characterized in that it includes: a detection module, used to detect the expression levels of miR_140_3p and miR_10a_5p in plasma exosomes and the abundance of Anaeritruncus_sp_G3_2012 and Bifidobacterium_longum in feces; a calculation module, used to input the detection results of the detection module into the combination formula of the diagnostic model to obtain the calculation results; an analysis module, used to use the calculation results as diagnostic indicators for ventricular remodeling diagnosis, wherein the calculation module is provided with the above-mentioned diagnostic model.

[0010] The present invention also provides an application of the above-mentioned marker combination for ventricular remodeling diagnosis in the preparation of a kit, characterized in that the kit is used to detect the expression levels of miR_140_3p and miR_10a_5p in plasma exosomes, and to detect the abundance of Anaeritruncus_sp_G3_2012 and Bifidobacterium_longum in feces.

[0011] Functions and Effects of the Invention

[0012] According to the marker combination and application, diagnostic model and system for ventricular remodeling diagnosis involved in the present invention, the present invention innovatively uses intestinal flora metagenomics and intestinal bacteria-derived plasma exosome miRNA sequencing in a patient cohort to screen out a series of differential miRNAs and intestinal bacterial species, and uses LASSO regression and stepwise backward regression for variable screening, and finally screens out two miRNAs including miR_140_3p, miR_10a_5p and two intestinal bacterial species including Anaeritruncus_sp_G3_2012, Bifidobacterium_longum as marker combinations, and constructs a diagnostic model at the same time, which has high sensitivity and specificity for identifying ventricular remodeling, can improve the efficiency of early identification of heart failure, better discover, identify and intervene in ventricular remodeling as early as possible, prevent and delay the occurrence of heart failure, and can be applied to children. For preoperative patients, it is expected to establish the best time for surgery, improve surgical outcomes, reduce medical risks, and cut medical costs; for postoperative patients, it can effectively predict ventricular recovery status, guide cardiac rehabilitation, timely identify changes in ventricular reserve function caused by restenosis, improve cardiac function and exercise tolerance, and ultimately improve the patient's quality of life, thereby achieving full life cycle screening and monitoring, which is expected to be used in clinical testing and has clinical translational value. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 is a volcano plot of differential miRNAs in an embodiment of the present invention;

[0014] Figure 2 It is the result of LEfSe identifying characteristic bacterial species of the heart failure group and the control group in the embodiment of the present invention;

[0015] Figure 3 It is the characteristic variable screening result using LASSO regression in the embodiment of the present invention;

[0016] Figure 4 This is the result of using 10-fold cross validation to screen important variables in an embodiment of the present invention;

[0017] Figure 5 is the ROC curve of the diagnostic model in the embodiment of the present invention;

[0018] Figure 6 This is the result of internal validation of the diagnostic model using Bootstrap resampling in the embodiment of the present invention;

[0019] Figure 7 is the linear regression analysis result of Panel and EF in the embodiment of the present invention;

[0020] Figure 8 is a diagnostic model calibration curve in an embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to make the technical means, creative features, objectives and effects of the present invention easy to understand, the following embodiments and accompanying drawings specifically illustrate the marker combination and application, diagnostic model and system for ventricular remodeling diagnosis of the present invention.

[0022] <Example>

[0023] 1. Recruitment of clinical case cohort and collection of stool and blood

[0024] First, a clinical cohort was recruited as a prospective study cohort. The statistical method used a 1:1 parallel design. The two-sided α for type I error was 0.05, and the β for type II error was 0.2 (i.e., the test power was 80%). Considering 5% loss to follow-up, it was calculated that the sample size required 52 patients with left ventricular remodeling and heart failure (poLV) due to left ventricular infarction and 60 patients with normal physical examination. This standard and number were followed during recruitment.

[0025] The inclusion criteria for the left heart obstruction ventricular remodeling and heart failure group were: confirmed diagnosis of aortic valve and subvalvular stenosis, aortic arch stenosis by ultrasound and cardiac CT; pressure gradient at left heart obstruction >50mmHg; high compliance of the children and parents. Exclusion criteria were: the presence of any other extracardiac malformations and chromosomal abnormalities; combined digestive system diseases; patients unwilling to sign informed consent. The inclusion criteria for the normal physical examination group were: confirmed absence of congenital heart disease and any other heart malformations by ultrasound and cardiac CT; high compliance of the children and parents. Exclusion criteria were: the presence of any other extracardiac malformations and chromosomal abnormalities; combined digestive system diseases; patients unwilling to sign informed consent. Standard fecal bacteria, blood collection and operation procedures (SOP) were established, and the collection, processing, and storage of fecal samples were completed by senior laboratory physicians and laboratory technicians.

[0026] 2. Extraction of plasma exosomes

[0027] First, centrifuge the patient's blood at 4 degrees and 2000g for 20 minutes to separate the blood supernatant, i.e., plasma. Process the plasma through a 0.22-micron filter, and pre-treat the ultrafiltration tube with 10ml PBS solution, centrifuge at 4 degrees and 4000g for 10 minutes. Take out PBS from the inner tube of the ultrafiltration tube and the collection tube. Add the sample to the inner tube of the ultrafiltration tube, centrifuge at 4 degrees and 4000g for 30 minutes. Aspirate the liquid in the collection tube, continue to add the sample to the inner tube of the ultrafiltration tube, centrifuge at 4°C and 4000g for 30 minutes, until all samples are centrifuged, add 10ml PBS to the inner tube for buffer replacement, gently blow the sample, centrifuge at 4°C and 4000g for 30 minutes, and collect the exosomes in the tube.

[0028] 3. Screening of differential miRNAs and intestinal flora in chronic heart failure by miRNA omics sequencing and 16S rRNA sequencing

[0029] Methods: The ascending aortic constriction (AAC) model of neonatal mice was used to simulate chronic heart failure in children. C57BL / 6 neonatal mice were selected and anesthetized by freezing for 3-4 minutes. The second rib of the sternum was cut open, and a transverse incision was made to open the chest cavity and expose the ascending aorta. A 10-0 silk thread was passed through the ascending aorta between the heart and the brachiocephalic artery. A 28G pad needle was placed parallel to the ascending aorta. After the ascending aorta was ligated, the pad needle was withdrawn to form constriction, and then the chest cavity was closed layer by layer. The neonatal mice in the sham group underwent a similar procedure. The chest was opened and the ascending aorta was separated at the same time, but it was not ligated, and no ascending aortic constriction was formed. The rest was the same as the surgery group. After surgery, the body temperature of the neonatal mice was quickly restored on a heating pad until they woke up, and they were returned to the cage for routine feeding. Blood and feces were collected at 4 weeks. Ten mice with ascending aortic constriction and 10 mice in the sham group were selected, and peripheral blood was collected, and plasma exosomes and feces were separated.

[0030] Gut microbiota metagenomic sequencing Fecal SNAP was first frozen and DNA was extracted using the SDS method. All samples were sequenced on the Illumina platform in PE150 mode. Adapters were trimmed using Trimmomatic-0.39 and low-quality reads were filtered. Then, sequencing reads were aligned to the host genome reference (GRCM39 / MM39) using Bowtie2 (Version 2.4.1) to remove host sequences. Functional analysis was performed using Humann3. Sample reads were mapped to this database to quantify gene presence and abundance on a per-species basis. A translation search was then performed against the UNIREF-based protein sequence catalog to find all reads that failed to map at the nucleotide level. The result is an abundance profile of gene families (Uniref90s), stratified by each species contributing to these genes, which can then be summarized into higher-level gene groupings such as ECS or KOS. Linear discriminant analysis was performed using LEFSE (Version 1.0) software. The Wilcoxon rank sum test (two-tailed) was used for comparison between two groups, and the Kruskal-Wallis test was used for comparison between three groups. Multiple testing was adjusted using the Benjamin-Hockberg method. The differences in β diversity among the groups were tested using permutational multivariate analysis of variance (PERAMOVA).

[0031] RNA was extracted using the conventional TRIZOL method for miRNA sequencing. 1 μg RNA was used as input material for RNA sample preparation for each sample. UltraMRNA Library Preparation Kit for (NEB, USA) to generate sequencing libraries and add index codes to the attribute sequences of each sample. Library fragments were purified using the Ampure XP system (Beckman Coulter, Beverly, USA) and cDNA fragments with a length of 250 to 300 bp were screened. Index-coded samples were clustered on the CBOT cluster generation system using the Truseq PE Cluster Kit V3-CBOT-HS (Illumia) according to the manufacturer's instructions. After cluster generation, library preparations were sequenced on the Illumina Novaseq platform to generate 150 bp paired-end reads. Genes with adjusted P values ​​< 0.05 and fold changes > 0.5 were subjected to subsequent IPA analysis.

[0032] Figure 1 is a volcano plot of differential miRNAs in an embodiment of the present invention; Figure 2 This is the result of LEfSe identifying characteristic bacterial species in the heart failure group and the control group in the embodiment of the present invention.

[0033] like Figure 1 and Figure 2 As shown, in the miRNA differential analysis, a total of 34 miRNAs were differentially upregulated and 52 metabolites were differentially downregulated in the heart failure group compared with the normal group.

[0034] LEfSe analysis showed that there were 9 differential bacterial genera between the heart failure group and the control group. The characteristic genera in the heart failure group were Bacteroides and Anaerotruncus_sp_G3_2012, and the characteristic genera in the sham operation group were Parasutterella, Parasutterella_excrementihominis, Proteobacteria, Bifidobacterium_longum, Proteobacteria_bacterium_CAG_139, Lactobacillus_murinus, and Actinobacteria.

[0035] 4. Quantitative verification of differential miRNA and intestinal flora content in plasma exosomes in clinical cohorts by real-time fluorescence quantitative polymerase chain reaction (qRT-PCR)

[0036] Methods: In order to translate from animal models to clinical applications, four miRNAs (miR_140_3p, miR_10a_5p, miR_1306_3p, miR_184_3p) and six differential bacterial genera (Parasutterella_excrementihominis, Anaeritruncus_sp_G3_2012, Bifidobacterium_longum, Actinobacteria, Proteobacteria_bacterium_CAG_139, Bacteroidetes) were selected from the above differential miRNAs and bacterial genera for further study in children with heart failure based on the high differences in the results of the mouse model. Subsequently, the venous blood and fecal extracts of the two groups of patients before surgical treatment were subjected to qRT-PCR to determine the content of miRNAs and differential bacterial genera and compared. RNA extraction and purification: The Novozymes Cell / Tissue Total RNA Extraction Kit (refer to the kit manual for specific operations) (RC101) was used to extract RNA from cells and animal tissues. The RNA was reverse transcribed using the Novozymes reverse transcription kit (R323) to obtain cDNA products.

[0037] The validation results are as follows: qRT-PCR results showed that the 4 differential miRNAs and 6 bacterial genera were significantly different between the heart failure group and the control group (p<0.001). Among them, in the heart failure group, miR_140_3p, miR_1306_3p, miR_184_3p, Parasutterella_excrementihominis, Proteobacteria_bacterium_CAG_139, and Bacteroidetes were specifically highly expressed, while miR_10a_5p, Anaeritruncus_sp_G3_2012, Bifidobacterium_longum, and Actinobacteria were downregulated.

[0038] 5. Use LASSO regression (least absolute shrinkage and selection algorithm) and 10-fold cross validation to screen important variables

[0039] Methods: In order to screen variables, LASSO regression was used to analyze the above-mentioned 4 differential miRNAs and 6 bacterial genera. LASSO regression is a statistical method for linear regression. By introducing the L1 penalty term, the coefficients of the model are shrunk and selected. For the coefficient of a certain characteristic variable, if its absolute value is less than half of the penalty term, it will be compressed to 0. LASSO can be used to solve the problem of collinearity of variables, compress the regression coefficients of some non-important variables to zero, and retain variables with non-zero coefficients, so as to achieve the purpose of variable screening and improve the generalization ability and interpretability of the model.

[0040] Then, the λ with the smallest binomial deviance was obtained through 10-fold cross validation. When lambda.lse was 1, a model with good performance but the least number of independent variables was obtained. LASSO regression and 10-fold cross validation were completed using the R 4.2.3 software glmnet package.

[0041] Figure 3 It is the characteristic variable screening result using LASSO regression in the embodiment of the present invention; Figure 4 This is the result of using 10-fold cross validation to screen important variables in an embodiment of the present invention. Figure 4 The two dotted lines in the figure are two special lambda values: lambda.min and lambda.lse. lambda.min represents the lambda value with the smallest error, and lambda.lse represents the lambda value that obtains the most streamlined model within a variance range of lambda.min.

[0042] like Figure 3 and Figure 4 As shown in the figure, the LASSO regression results and 10-fold cross validation showed that when the optimal λ (λ = 0.02750813, i.e. lambda.lse), 6 variables with non-zero coefficients were screened out, including: miR_140_3p (0.62440164), miR_10a_5p (-0.49312928), Parasutterella_excrementihominis (-0.45379016), Anaeritruncus_sp_G3_2012 (-0.03054508), Bifidobacterium_longum (-0.71022003), Actinobacteria (-0.20785760), while miR_1306_3p, miR_184_3p,

[0043] Proteobacteria_bacterium_CAG_139, Bacteroidetes.

[0044] 6. The diagnostic model was established using the stepwise backward regression method, and the predictive efficacy of the diagnostic model was evaluated using AUC and calibration curve.

[0045] Methods: For the two miRNAs and four intestinal flora screened by LASSO regression, the Akaike information criterion (AIC) stepwise backward regression method was used to further screen variables and establish a model. The general process of stepwise backward elimination is to first put all variables into the model, then try to remove one of the independent variables from the model, observe whether the model explains the variation of the dependent variable significantly, and then remove the least explanatory independent variable, and iterate continuously until no independent variable meets the conditions for elimination.

[0046] After the diagnostic model was established, the predictive efficacy of the diagnostic model was evaluated using AUC, Bootstrap internal validation, and calibration curve. Stepwise backward regression, AUC, and calibration curve were all completed using R 4.2.3. All tests were two-sided, and P < 0.05 was considered statistically significant.

[0047] Results: After stepwise backward regression, a diagnostic prediction model (Panel) including 2 miRNAs and 2 intestinal flora was finally obtained: miR_140_3p (OR, 1.203, 95% CI, 1.076-1.400, P = 0.005), miR_10a_5p (OR, 0.898, 95% CI, 0.816-0.958, P = 0.007), Anaeritruncus_sp_G3_2012 (OR, 0.893, 95% CI, 0.812-0.964, P = 0.01), and Bifidobacterium_longum (OR, 0.519, 95% CI, 0.267-0.786, P = 0.012).

[0048] The combination of the diagnostic model is:

[0049] Panel=1.0711702+(-0.1128574*Anaeritruncus_sp_G3_2012+

[0050] (0.1847674*miR_140_3p)+(-0.1081418*miR_10a_5p)+(-0.6567257*Bifidobacterium_longum).

[0051] By detecting the expression of miR_140_3p and miR_10a_5p in plasma exosomes and the abundance of Anaeritruncus_sp_G3_2012 and Bifidobacterium_longum in feces, the calculation results were substituted into the combination formula of the diagnostic model and used as the diagnostic index of ventricular remodeling. Specifically, the diagnostic threshold for ventricular remodeling is -0.04883208, that is, it can be considered that when the calculation result is greater than -0.049, ventricular remodeling is diagnosed.

[0052] Based on the above diagnostic model, this embodiment provides a ventricular remodeling diagnostic system, including: a detection module, used to detect the expression levels of miR_140_3p and miR_10a_5p in plasma exosomes and the abundance of Anaeritruncus_sp_G3_2012 and Bifidobacterium_longum in feces; a calculation module, used to input the detection result of the detection module into the combination formula of the diagnostic model to obtain the calculation result; an analysis module, used to use the calculation result as a diagnostic indicator for ventricular remodeling diagnosis, that is, when the result calculated based on the combination formula of the above diagnostic model is greater than -0.049, it is diagnosed that ventricular remodeling exists.

[0053] In this embodiment, the applicant recruited a group of clinical cohorts as an internal validation cohort (a total of 78 people, including patients with ventricular remodeling and heart failure due to left heart obstruction and several patients with normal physical examinations), and used the above-mentioned diagnostic system for diagnosis. The diagnostic results of the diagnostic system were compared with the clinical diagnostic results, and the sensitivity of the diagnostic system of this embodiment was 93.5%, and the specificity was 85.1%, with high diagnostic efficiency. In addition, 75 children with ventricular remodeling were recruited from other medical centers as an external validation cohort for diagnostic verification, of which 69 children had the results obtained by the above-mentioned diagnostic system with a threshold value of >-0.049. Compared with the clinical diagnostic results, the diagnostic accuracy of this embodiment can reach 92%.

[0054] Figure 5 : is the ROC curve of the diagnosis model in the embodiment of the present invention.

[0055] like Figure 5 As shown, the AUC value of the diagnostic model was 0.898, p < 0.001, which has a good biological marker value and has stronger diagnostic ability than traditional heart failure markers such as NT-proBNP (AUC = 0.654, p = 0.005) and HS_cTNI (AUC = 0.585, p = 0.124).

[0056] Figure 6 This is the result of internal validation of the diagnostic model using Bootstrap resampling in the embodiment of the present invention.

[0057] like Figure 6 As shown, the diagnostic model was internally validated by resampling using the Bootstrap method (B=1000), and the AUC value and 95% confidence interval of the smooth ROC curve were: 0.893 (0.833-0.945).

[0058] Figure 7 is the linear regression analysis result of Panel and EF in the embodiment of the present invention; Figure 8 is a diagnostic model calibration curve in an embodiment of the present invention.

[0059] like Figure 7 and Figure 8 As shown in the figure, the Panel and EF values ​​were linearly regressed, and the regression equation was y = -20.619x + 13.151, the R square value was 0.2947, p < 0.001, and there was a negative correlation. The calibration curve also showed that the diagnostic model had good predictive ability.

[0060] Functions and Effects of the Embodiments

[0061] According to the marker combination and application, diagnostic model and system for ventricular remodeling diagnosis involved in this embodiment, the present invention innovatively uses intestinal flora metagenomics and intestinal bacteria-derived plasma exosome miRNA sequencing in patient cohorts to screen out a series of differential miRNAs and intestinal bacterial species, and uses LASSO regression and stepwise backward regression for variable screening, and finally screens out two miRNAs including miR_140_3p, miR_10a_5p and two intestinal bacterial species including Anaeritruncus_sp_G3_2012, Bifidobacterium_longum as marker combinations, and constructs a diagnostic model at the same time, which has high sensitivity and specificity for identifying ventricular remodeling, can improve the efficiency of early identification of heart failure, better discover, identify, and intervene in ventricular remodeling as early as possible, prevent and delay the occurrence of heart failure, and can be applied to children's groups, can achieve full life cycle screening and monitoring, is expected to be used in clinical testing, and has clinical transformation value.

[0062] Those skilled in the art should understand that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A marker combination for diagnosing ventricular remodeling, characterized in that: include: miRNA and intestinal bacteria species, the miRNA is miR_140_3p and miR_10a_5p, and the intestinal bacteria species are Anaeritruncus_sp_G3_2012 and Bifidobacterium_longum.

2. A diagnostic model constructed using the marker combination for ventricular remodeling diagnosis according to claim 1, characterized in that: The combination of the diagnostic model is: Panel=1.0711702+(-0.1128574*Anaeritruncus_sp_G3_2012)+(0.1847674*miR_140_3p)+(-0.1081418*miR_10a_5p)+(-0.6567257*Bifidobacterium_longum), In the combination formula, Panel is a diagnostic model, miR_140_3p and miR_10a_5p are the expression levels of miR_140_3p and miR_10a_5p in plasma exosomes, Anaeritruncus_sp_G3_2012 and Bifidobacterium_longum are the abundances of Anaeritruncus_sp_G3_2012 and Bifidobacterium_longum in feces.

3. A ventricular remodeling diagnostic system, characterized in that: include: The detection module was used to detect the expression of miR_140_3p and miR_10a_5p in plasma exosomes and the abundance of Anaeritruncus_sp_G3_2012 and Bifidobacterium_longum in feces; A calculation module, used for inputting the detection result of the detection module into the combination formula of the diagnosis model to obtain a calculation result; an analysis module, configured to use the calculation result as a diagnostic indicator for ventricular remodeling diagnosis, Wherein, the calculation module is provided with the diagnosis model as claimed in claim 2.

4. A use of the marker combination for ventricular remodeling diagnosis according to claim 1 in preparing a kit, characterized in that: The kit was used to detect the expression levels of miR_140_3p and miR_10a_5p in plasma exosomes, and to detect the abundance of Anaeritruncus_sp_G3_2012 and Bifidobacterium_longum in feces.

Images