Non-alcoholic fatty liver disease diagnosis model construction method and diagnosis device

Abstract

The invention discloses a non-alcoholic fatty liver disease diagnosis model construction method and a non-alcoholic fatty liver disease diagnosis device. The method comprises the following steps: S101, obtaining a non-alcoholic fatty liver disease data set through a high-throughput gene expression database; s102, dividing the data set into an NAFLD group and a Contro l group, and carrying out gene difference calibration analysis on the NAFLD group and the Contro l group so as to obtain a differential expression gene DEGs; s103, taking the intersection of the DEGs and the inflammation and neutrophile granulocyte extracellular trap related genes INRGs, so as to obtain M inflammation and neutrophile granulocyte extracellular trap related differential expression genes INRDEGs; s104, constructing an SVM (Support Vector Machine) model by using the obtained M INRDEGs through an SVM algorithm so as to obtain a gene number N with the lowest result error rate and the highest accuracy rate and N corresponding INRDEGs; and S105, constructing an LASSO regression model through LASSO regression analysis on the basis of the SVM model containing N INRDEGs so as to obtain a non-alcoholic fatty liver disease diagnosis model, and realizing low-cost and small-trauma non-alcoholic fatty liver disease diagnosis through a device containing the model.

Description

Technical Field

[0001] The present invention belongs to the technical field of molecular biology, and in particular relates to a method for constructing a diagnostic model for non-alcoholic fatty liver disease and a diagnostic device. Background Art

[0002] Nonalcoholic fatty liver disease (NAFLD) has affected about 38% of the adult population worldwide and has become the main cause of chronic liver disease and liver transplantation in my country, placing a huge burden on the health of the Chinese people. However, the existing diagnostic and treatment methods are still insufficient. Existing diagnostic devices for nonalcoholic fatty liver disease are mainly based on liver biopsy and liver imaging methods. Liver biopsy is traumatic, while liver imaging relies on the doctor's subjective judgment, and the manpower, material resources and time costs of using these two devices are not suitable for large-scale screening. The above diagnostic devices are difficult to provide early diagnosis of nonalcoholic fatty liver disease. Therefore, it is crucial to develop a simple and easy model or marker for the early diagnosis of nonalcoholic fatty liver disease.

[0003] Existing diagnostic devices for non-alcoholic fatty liver disease are mainly based on liver biopsy and liver imaging methods. Liver biopsy is invasive, while liver imaging relies on the doctor's subjective judgment. In addition, the manpower, material resources and time costs of using these two devices are not suitable for large-scale screening. The above diagnostic devices are difficult to provide early diagnosis of non-alcoholic fatty liver disease. Summary of the invention

[0004] The present invention aims to solve one of the technical problems in the related art at least to a certain extent. To this end, the purpose of the present invention is to propose a method for constructing a diagnostic model for non-alcoholic fatty liver disease and a diagnostic device.

[0005] To achieve the above object, according to the first aspect of the present invention, the present invention provides a method for constructing a diagnostic model for non-alcoholic fatty liver disease, comprising the steps of:

[0006] S101, Obtaining non-alcoholic fatty liver disease dataset through high-throughput gene expression database;

[0007] S102, dividing the non-alcoholic fatty liver disease data set into two groups, NAFLD and Control, and performing gene difference calibration analysis on the NAFLD group and the Control group to obtain differentially expressed gene DEGs;

[0008] S103, taking the intersection of the obtained differentially expressed gene DEGs and the inflammation and neutrophil extracellular trap-related genes INRGs to obtain M inflammation and neutrophil extracellular trap-related differentially expressed genes INRDEGs;

[0009] S104, constructing an SVM model using the obtained M inflammation and neutrophil extracellular trap-related differentially expressed genes INRDEGs through a support vector machine (SVM) algorithm to obtain the number of genes N with the lowest error rate and the highest accuracy, and N corresponding inflammation and neutrophil extracellular trap-related differentially expressed genes INRDEGs, where M>=N;

[0010] S105. Construct a LASSO regression model by performing LASSO regression analysis on N inflammation- and neutrophil extracellular trap-related differentially expressed genes INRDEGs contained in the SVM model to obtain a diagnostic model for non-alcoholic fatty liver disease.

[0011] According to one embodiment of the present invention, the non-alcoholic fatty liver disease dataset includes multiple batches of datasets; after step S101, the following steps are further included:

[0012] S1011. Remove batch effects from multiple batches of data sets to obtain an integrated non-alcoholic fatty liver disease data set.

[0013] According to one embodiment of the present invention, in step S104, N=4, and the four corresponding inflammation- and neutrophil extracellular trap-related differentially expressed genes INRDEGs are IL6, TNF, SPP1, and CCL5, respectively.

[0014] According to one embodiment of the present invention, after step S103, the method further includes the following steps:

[0015] S1031, performing logistic regression based on the obtained multiple inflammation and neutrophil extracellular trap-related differentially expressed genes INRDEGs and constructing a logistic regression model;

[0016] S1032. Obtain whether each inflammation and neutrophil extracellular trap-related differentially expressed gene INRDEGs is statistically significant in the logistic regression model; and retain the inflammation and neutrophil extracellular trap-related differentially expressed gene INRDEGs that are statistically significant.

[0017] According to one embodiment of the present invention, after step S103, the method further includes the following steps:

[0018] S1033. Analyze the expression differences of the inflammation and neutrophil extracellular trap-related differentially expressed genes INRDEGs between different sample groups in the non-alcoholic fatty liver disease dataset to obtain the expression difference information of each inflammation and neutrophil extracellular trap-related differentially expressed gene INRDEGs between different sample groups; and retain the inflammation and neutrophil extracellular trap-related differentially expressed genes INRDEGs that have significant differences between different sample groups.

[0019] According to one embodiment of the present invention, in step S102, the threshold for obtaining the differentially expressed genes DEGs is |logFC|>0.3 and adj.p<0.05.

[0020] According to one embodiment of the present invention, the N inflammation and neutrophil extracellular trap-related differentially expressed genes INRDEGs are used as hub genes, and after step S105, the following steps are further included:

[0021] S106. Drawing a nomogram based on the hub gene to show the relationship between the hub genes in the non-alcoholic fatty liver disease data set to verify the utility of the expression amount of the hub gene for the non-alcoholic fatty liver disease diagnostic model;

[0022] S107, drawing a calibration curve through Calibration analysis, and drawing the fitting of the actual probability and the probability predicted by the model under different situations in the curve to evaluate the prediction effect of the non-alcoholic fatty liver disease diagnosis model on the actual result;

[0023] S108. Based on the diagnostic model risk score RiskScore in the non-alcoholic fatty liver disease dataset, the role of the non-alcoholic fatty liver disease diagnostic model in clinical utility is evaluated by decision curve analysis DCA.

[0024] According to one embodiment of the present invention, after step S105, the method further includes the following steps:

[0025] S109, using the pROC software package in the R programming language to draw a ROC curve based on the risk score RiskScore of the diagnostic model in the non-alcoholic fatty liver disease data set, so as to perform risk scoring through the ROC curve;

[0026] The calculation formula of risk score RiskScore is as follows:

[0027] IL6*(-0.81085028038064)+TNF*(-0.135302961906175)+SPP1*(0.263579706135327)+CCL5*(0.203623519951156).

[0028] According to one embodiment of the present invention, after step S108, the following steps are further included:

[0029] S110, dividing the non-alcoholic fatty liver disease (NAFLD) results into a HighRisk group and a LowRisk group according to the median value of the risk score expression;

[0030] S111. Display the differences in expression levels of the N hub genes in the HighRisk group and the LowRisk group in the results of non-alcoholic fatty liver disease (NAFLD) through group comparison diagrams to obtain the statistical significance of the expression levels of the N hub genes in the HighRisk group and the LowRisk group.

[0031] According to the second aspect of the present invention, an embodiment of the present invention provides a non-alcoholic fatty liver disease diagnostic device, which includes a computer device, and the computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that when the processor executes the computer program, the non-alcoholic fatty liver disease diagnostic model constructed according to any one of claims 1 to 9 is implemented.

[0032] Through the above technical scheme, the method for constructing a diagnostic model of non-alcoholic fatty liver disease and the diagnostic device of the present invention include the steps of S101, obtaining a non-alcoholic fatty liver disease data set through a high-throughput gene expression database; S102, dividing the non-alcoholic fatty liver disease data set into two groups, NAFLD and Control, and performing gene difference calibration analysis on the two groups, so as to obtain differentially expressed genes DEGs; S103, taking the intersection of the obtained differentially expressed genes DEGs and inflammation and neutrophil extracellular trap related genes INRGs, so as to obtain M inflammation and neutrophil extracellular trap related differentially expressed genes INRDEGs; S104, obtaining a plurality of differentially expressed genes INRDEGs, and obtaining a plurality of differentially expressed genes INRDEGs. 104. The obtained M inflammation and neutrophil extracellular trap-related differentially expressed genes INRDEGs are used to construct an SVM model through the support vector machine SVM algorithm to obtain the number of genes N with the lowest error rate and the highest accuracy, and N corresponding inflammation and neutrophil extracellular trap-related differentially expressed genes INRDEGs, where M>=N; S105. The N inflammation and neutrophil extracellular trap-related differentially expressed genes INRDEGs contained in the SVM model are used to construct a LASSO regression model through LASSO regression analysis to obtain a diagnostic model for non-alcoholic fatty liver disease. The device containing the model can achieve low-cost and less invasive liver disease diagnosis.

[0033] Other features and advantages of the present invention will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 is a flow chart of a method for constructing a non-alcoholic fatty liver disease diagnostic model provided by an embodiment of the present invention;

[0035] Figure 2 is a flow chart of another method for constructing a non-alcoholic fatty liver disease diagnosis model provided by an embodiment of the present invention;

[0036] Figure 3 is another flow chart of a method for constructing a non-alcoholic fatty liver disease diagnosis model provided by an embodiment of the present invention;

[0037] Figure 4 It is a distribution box plot and a low-dimensional feature distribution PCA plot of a data set before and after removing the batch effect provided by an embodiment of the present invention;

[0038] Figure 5A A volcano plot is drawn based on the differential analysis results of the gene expression values ​​in the integrated GEO data set provided by the embodiment of the present invention;

[0039] Figure 5BIt is a Venn diagram of the intersection of the differentially expressed genes DEGs and the inflammation and neutrophil extracellular trap-related genes INRGs provided in the embodiments of the present invention;

[0040] Figure 5C It is the result of analyzing the expression difference of inflammation and neutrophil extracellular trap-related differentially expressed genes INRDEGs between different sample groups in the integrated GEO dataset Combined Datasets provided in the embodiment of the present invention;

[0041] Figure 6 It is a bar chart of the results of the gene ontology GO and pathway KEGG enrichment analysis provided by the embodiment of the present invention;

[0042] Figure 7 It is a GSEA analysis diagram of biological related functions and signaling pathways provided by an embodiment of the present invention;

[0043] Figure 8 This is a diagram for analyzing the diagnostic value of 12 inflammation and neutrophil extracellular trap-related differentially expressed genes in non-alcoholic fatty liver disease NAFLD provided by an embodiment of the present invention;

[0044] Fig. 9 This is a value verification analysis diagram of the non-alcoholic fatty liver disease NAFLD diagnostic model provided by an embodiment of the present invention;

[0045] Fig.10 It is a schematic diagram of the structure of a non-alcoholic fatty liver disease NAFLD diagnostic device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0046] In order to facilitate the understanding of the present invention, the present invention will be described more fully below. Preferred embodiments of the present invention are given below. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present invention more thorough and comprehensive.

[0047] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0048] On the one hand, see Figure 1 The present invention provides a method for constructing a non-alcoholic fatty liver disease diagnosis model, comprising the steps of:

[0049] S101, Obtaining non-alcoholic fatty liver disease dataset through high-throughput gene expression database;

[0050] S102, dividing the non-alcoholic fatty liver disease data set into two groups, NAFLD and Control, and performing gene difference calibration analysis on the two groups to obtain differentially expressed gene DEGs;

[0051] S103, taking the intersection of the obtained differentially expressed gene DEGs and the inflammation and neutrophil extracellular trap-related genes INRGs to obtain M inflammation and neutrophil extracellular trap-related differentially expressed genes INRDEGs;

[0052] S104, constructing an SVM model using the obtained M inflammation and neutrophil extracellular trap-related differentially expressed genes INRDEGs through a support vector machine (SVM) algorithm to obtain the number of genes N with the lowest error rate and the highest accuracy, and N corresponding inflammation and neutrophil extracellular trap-related differentially expressed genes INRDEGs, where M>=N;

[0053] S105. Construct a LASSO regression model by performing LASSO regression analysis on N inflammation- and neutrophil extracellular trap-related differentially expressed genes INRDEGs contained in the SVM model to obtain a diagnostic model for non-alcoholic fatty liver disease.

[0054] See also Figure 2 Specifically, in step S101, non-alcoholic fatty liver disease (NAFLD) data sets GSE89632, GSE49541, and GSE48452 can be downloaded from the high-throughput gene expression database GEO (Gene Expression Omnibus); since these data sets may come from different researchers. The data sets obtained in this way will have what we call batch effects, such as different experimental times, different experimental batches, different processing methods, different sequencing platforms, etc. Therefore, in one embodiment of the present invention, the non-alcoholic fatty liver disease data set includes multiple batches of data sets; after step S101, the following steps are also included:

[0055] S1011. Remove batch effects from multiple batches of data sets to obtain an integrated non-alcoholic fatty liver disease data set.

[0056] In an embodiment of the present invention, the batch effect was removed from the above dataset using the R programming language sva software package to obtain an integrated GEO dataset (Combined Datasets). Subsequently, the expression value differences of the dataset before and after the batch effect were compared using a distribution box plot. Secondly, the low-dimensional feature distribution of the dataset before and after the batch effect was compared using a PCA (Principal Component Analysis) plot. The results of the distribution box plot and PCA plot show that (as shown in Figure 4 As shown in the figure, after removing the batch processing, the batch effect of samples in the non-alcoholic fatty liver disease (NAFLD) dataset was basically eliminated.

[0057] In step S102, the data of the integrated GEO dataset (Combined Datasets) are divided into NAFLD and Control. In order to analyze the differences in gene expression values ​​for NAFLD and Control in the integrated GEO dataset (Combined Datasets), the R package limma is used to perform differential analysis on the integrated GEO dataset (Combined Datasets) to obtain differentially expressed genes of the two groups of data. The results are as follows: The integrated GEO dataset (Combined Datasets) has a total of 415 differentially expressed gene DEGs that meet the threshold of |logFC|>0.3 and adj.p<0.05; under this threshold, there are 193 up-regulated expression genes (logFC>0.3 and adj.p<0.05) and 222 down-regulated expression genes (logFC<0.3 and adj.p<0.05). A volcano plot (such as Figure 5A ).

[0058] In step S103, in order to obtain the differentially expressed genes INRDEGs related to inflammation and neutrophil extracellular traps, the intersection of all the differentially expressed genes (DEGs) with |logFC|>0.3 and adj.p<0.05 and the genes INRGs related to inflammation and neutrophil extracellular traps was taken and a Venn diagram was drawn (such as Figure 5B ). A total of M = 12 inflammation and neutrophil extracellular trap-related differentially expressed genes INRDEGs were obtained, namely IL6, NFIL3, IL1RL1, PTX3, CXCL2, CD274, SOCS3, LDLR, CCL2, CCL5, TNF, and SPP1.

[0059] In step S104, an SVM model was constructed based on 12 inflammation and neutrophil extracellular trap-related differentially expressed genes INRDEGs and the SVM (Support Vector Machine) algorithm to obtain the number of genes with the lowest error rate and the highest accuracy. Figure 8 B and Figure 8 As shown in C), when the number of genes is N=4, the accuracy of the SVM model is the highest. The four inflammation and neutrophil extracellular trap-related differentially expressed genes INRDEGs are: IL6, TNF, SPP1, and CCL5.

[0060] In step S105, based on the four inflammation and neutrophil extracellular trap-related differentially expressed genes (INRDEGs) included in the SVM model, a LASSO (Least Absolute Shrinkage and Selection Operator) regression analysis is performed to construct a LASSO regression model, i.e., a non-alcoholic fatty liver disease (NAFLD) diagnostic model. By drawing a LASSO regression model diagram (e.g. Figure 8 ) and the LASSO variable trajectory diagram (as shown in D of Figure 8 The results showed that the four inflammation and neutrophil extracellular trap-related differentially expressed genes (INRDEGs), i.e., hub genes, included in the LASSO regression model were IL6, TNF, SPP1, and CCL5.

[0061] The embodiment of the present invention provides a method for constructing a diagnostic model for non-alcoholic fatty liver disease, by using the characteristics that the expression levels of TNF and IL-6 in early liver tissue specimens of patients with non-alcoholic fatty liver disease are reduced, and the expression levels of SPP1 and CCL5 in early liver tissue specimens of patients with non-alcoholic fatty liver disease are increased. The diagnostic model constructed with these four proteins has good and stable early diagnostic efficacy for non-alcoholic fatty liver disease. Therefore, the combination of TNF, IL-6, SPP1, and CCL5 can be used as a marker for the early diagnosis of non-alcoholic fatty liver disease, and a detection model can be constructed using them, and a device including the detection model is provided to achieve early diagnosis of non-alcoholic fatty liver disease.

[0062] Furthermore, in an example of the present invention, after step S103, the following steps are further included:

[0063] S1031, performing logistic regression based on the obtained multiple inflammation and neutrophil extracellular trap-related differentially expressed genes INRDEGs and constructing a logistic regression model;

[0064] S1032. Obtain whether the differentially expressed genes INRDEGs related to inflammation and neutrophil extracellular traps are statistically significant in the logistic regression model;

[0065] The statistically significant inflammation- and neutrophil extracellular trap-related differentially expressed genes INRDEGs were retained;

[0066] S1033. Analyze the expression differences of the inflammation and neutrophil extracellular trap-related differentially expressed genes INRDEGs between different sample groups in the non-alcoholic fatty liver disease dataset to obtain the expression difference information of each inflammation and neutrophil extracellular trap-related differentially expressed gene INRDEGs between different sample groups; retain each inflammation and neutrophil extracellular trap-related differentially expressed gene (INRDEGs) that has significant differences between different sample groups.

[0067] Specifically, in step S1031, in order to determine the diagnostic value of 12 inflammation and neutrophil extracellular traps-related differentially expressed genes (INRDEGs) in non-alcoholic fatty liver disease (NAFLD), firstly, a logistic regression was performed based on the 12 inflammation and neutrophil extracellular traps-related differentially expressed genes (INRDEGs) and a logistic regression model was constructed and displayed through a forest plot. And the results in step S1032 show that (such as Figure 8 As shown in A of the Figure 3, 12 inflammation- and neutrophil extracellular trap-related differentially expressed genes (INRDEGs) were all statistically significant in the logistic regression model (p value < 0.05).

[0068] Specifically, in step S1033, we then analyze the expression differences of inflammation and neutrophil extracellular trap-related differentially expressed genes (INRDEGs) between different sample groups in the integrated GEO dataset (Combined Datasets) and use the ggplot2 software package of the R programming language to draw a group comparison chart to display the analysis results, such as Figure 5C As shown in the results, the 12 inflammation and neutrophil extracellular trap-related differentially expressed genes (INRDEGs) were significantly different between different sample groups, and except for genes CCL5 and SPP1 which were highly expressed in the NAFLD group, the remaining genes were highly expressed in the Control group.

[0069] Furthermore, in an example of the present invention, after step S103, the steps may also be included: applying the 12 inflammation and neutrophil extracellular trap-related differentially expressed genes (INRDEGs) to perform gene ontology (GO) and pathway (KEGG) enrichment analysis, and the relationship between the biological process (BP), cellular component (CC), molecular function (MF) and biological pathway (KEGG) of the 2 inflammation and neutrophil extracellular trap-related differentially expressed genes (INRDEGs) and non-alcoholic fatty liver disease (NAFLD).

[0070] Specifically, gene ontology (GO) and pathway (KEGG) enrichment analysis was used to further explore the relationship between the biological process (BP), cellular component (CC), molecular function (MF) and biological pathway (KEGG) of 12 inflammation and neutrophil extracellular trap-related differentially expressed genes (INRDEGs) and non-alcoholic fatty liver disease (NAFLD). These 12 inflammation and neutrophil extracellular trap-related differentially expressed genes (INRDEGs) were used for gene ontology (GO) and pathway (KEGG) enrichment analysis. The results showed that 12 inflammation and neutrophil extracellular trap-related differentially expressed genes (INRDEGs) in non-alcoholic fatty liver disease (NAFLD) were mainly enriched in physical processes (BP) such as cellular response to lipopolysaccharide, cellular response to bacterial molecules, cellular response to biological stimulation, response to lipopolysaccharide and response to bacterial molecules; cellular components (CC) such as the outer side of the plasma membrane, circulating endosomes, low-density lipoprotein particles, endolysosomal membranes and endoplasmic reticulum lumen; molecular functions (MF) such as cytokine activity, receptor ligand activity, signal receptor activation activity, cytokine receptor binding and chemokine activity: at the same time, they were also enriched in biological pathways (KEGG) such as TNF signaling pathway, rheumatoid arthritis, interaction between viral proteins and cytokines and cytokine receptors, lipids and atherosclerosis, and cytokine-cytokine receptor interaction. The results of gene ontology (GO) and pathway (KEGG) enrichment analysis were displayed by bar graphs (such as Figure 6 ).

[0071] Furthermore, in an example of the present invention, after step S103, the method may further include the following steps: studying the expression of M inflammation- and neutrophil extracellular trap-related differentially expressed genes INRDEGs in the integrated GEO datasets (Combined Datasets) and the biological processes involved (the relationship between the affected cellular components and the molecular functions played) through gene set enrichment analysis (GSEA), and determining the effects of the expression levels of all genes in the integrated GEO datasets (Combined Datasets) on non-alcoholic fatty liver disease (NAFLD).

[0072] Specifically, in order to determine the impact of all gene expression levels in the integrated GEO datasets on non-alcoholic fatty liver disease (NAFLD), gene set enrichment analysis (GSEA) was used to study the relationship between the expression of all genes in the integrated GEO datasets and the biological processes involved, the cell components affected, and the molecular functions played. The results showed that all genes in the integrated GEO datasets were significantly enriched in biologically related functions and signaling pathways (such as Hypoxia Dn, Apoptosis By CDKN1A Via TP53, Emt Breast Tumor Dn, Circadian Rhythm Genes, etc. Figure 7 ).

[0073] See also Figure 3 In one embodiment of the present invention, the N inflammation and neutrophil extracellular trap-related differentially expressed genes INRDEGs are used as hub genes, and after step S105, the following steps are further included:

[0074] S106. Drawing a nomogram based on the hub gene to show the relationship between the hub genes in the non-alcoholic fatty liver disease data set to verify the utility of the expression amount of the hub gene for the non-alcoholic fatty liver disease diagnostic model;

[0075] S107, drawing a calibration curve through Calibration analysis, and drawing the fitting of the actual probability and the probability predicted by the model under different situations in the curve to evaluate the prediction effect of the non-alcoholic fatty liver disease diagnosis model on the actual result;

[0076] S108. Based on the diagnostic model risk score RiskScore in the non-alcoholic fatty liver disease dataset, the role of the non-alcoholic fatty liver disease diagnostic model in clinical utility is evaluated by decision curve analysis DCA.

[0077] Specifically, in step S106, in order to further verify the value of the diagnostic model for non-alcoholic fatty liver disease (NAFLD), a nomogram was drawn based on hub genes to show the relationship between hub genes in the integrated GEO datasets. The results showed that ( Fig. 9As shown in A of , the utility of the hub gene IL6 expression level for the non-alcoholic fatty liver disease (NAFLD) diagnostic model is significantly higher than that of other variables; the utility of the TNF and CCL5 expression levels for the non-alcoholic fatty liver disease (NAFLD) diagnostic model is significantly lower than that of other variables.

[0078] In step S107, in order to determine the accuracy and resolution of the non-alcoholic fatty liver disease (NAFLD) diagnostic model, a calibration curve is drawn through calibration analysis, and the prediction effect of the model on the actual result is evaluated according to the fitting of the actual probability and the probability predicted by the model under different conditions. The calibration curve of the non-alcoholic fatty liver disease (NAFLD) diagnostic model shows (as shown in Fig. 9 The calibration line shown by the dotted line deviates slightly from the diagonal line of the ideal model, but is nearly consistent with it.

[0079] In step S108, the role of the diagnostic model for non-alcoholic fatty liver disease (NAFLD) in clinical utility is evaluated by decision curve analysis (DCA) based on the diagnostic model risk score (RiskScore) in the integrated GEO dataset (Combined Datasets) and the results are presented. Fig. 9 As shown in C of , the line of the model is stably higher than All positive and All negative within a certain range, and the net profit of the model is large, and the effect of the model is better.

[0080] See also Figure 3 In one embodiment of the present invention, after step S105, the method further includes the following steps: S109, using the pROC software package in the R programming language to draw a ROC curve based on the risk score RiskScore of the diagnostic model in the non-alcoholic fatty liver disease data set, so as to perform risk scoring through the ROC curve;

[0081] The Risk Score calculation formula is as follows:

[0082] IL6*(-0.81085028038064)+TNF*(-0.135302961906175)+SPP1*(0.263579706135327)+CCL5*(0.203623519951156).

[0083] Specifically, in step S109, the R package pROC is used to draw the ROC curve based on the risk score (RiskScore) of the diagnostic model in the integrated GEO dataset (Combined Datasets). The ROC curve shows (such as Fig. 9 The expression level of risk score (RiskScore) in the combined GEO dataset (Combined Datasets) showed a certain accuracy (0.7 <AUC<0.9)。

[0084] See also Figure 3 In one embodiment of the present invention, after step S108, the following steps are further included:

[0085] S110, dividing the non-alcoholic fatty liver disease (NAFLD) results into a HighRisk group and a LowRisk group according to the median value of the risk score expression;

[0086] S111. Display the differences in expression levels of the N hub genes in the HighRisk group and the LowRisk group in the results of non-alcoholic fatty liver disease (NAFLD) through group comparison diagrams to obtain the statistical significance of the expression levels of the N hub genes in the HighRisk group and the LowRisk group.

[0087] Specifically, in step S110, NAFLD is divided into a HighRisk group and a LowRisk group according to the median value of the risk score (RiskScore) expression of the non-alcoholic fatty liver disease (NAFLD) diagnostic model.

[0088] In step S111, in order to explore the expression difference of hub genes in NAFLD, the difference analysis results of the expression levels of the four hub genes in the HighRisk group and the LowRisk group in NAFLD are displayed through group comparison diagrams. The difference results are shown (such as Fig. 9 E), the expression levels of the four hub genes in the HighRisk group and LowRisk group of NAFLD were statistically significant (p value < 0.05), and TNF, SPP1 and CCL5 were highly expressed in the high-risk group (HighRisk), and IL6 was highly expressed in the low-risk group (LowRisk).

[0089] See also Fig.10On the other hand, the present invention also provides a non-alcoholic fatty liver disease diagnosis device, the non-alcoholic fatty liver disease diagnosis device includes a computer device, the computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the non-alcoholic fatty liver disease diagnosis model constructed above when executing the computer program. The improvements of the present invention compared with the existing diagnosis device are mainly as follows:

[0090] 1. Less traumatic, low manpower, material and time costs.

[0091] 2. Early diagnosis of non-alcoholic fatty liver disease.

[0092] The computer program may be stored in a computer-readable storage medium, and when the computer program is executed by a processor, the steps of each of the above method embodiments may be implemented. The computer program includes computer program code, and the computer program code may be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0093] It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practices in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practices, computer-readable media does not include electrical carrier signals and telecommunication signals.

[0094] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0095] The steps in the method of the embodiment of the present invention can be adjusted in order, combined or deleted according to actual needs.

[0096] The modules or units in the system of the embodiment of the present invention may be combined, divided or deleted according to actual needs.

[0097] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic preset hardware, or in a combination of computer software and electronic preset hardware. Whether these functions are performed in preset hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0098] In the embodiments provided by the present invention, it should be understood that the disclosed apparatus / computer equipment and methods can be implemented in other ways. For example, the apparatus / computer equipment embodiments described above are merely schematic. For example, the division of the modules or units is merely a logical function division. There may be other division methods in actual implementation, such as multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces, indirect coupling or communication connection of the apparatus or unit, which may be electrical, mechanical or other forms.

[0099] The preferred embodiments of the present invention are described in detail above. However, the present invention is not limited to the specific details in the above embodiments. Within the technical concept of the present invention, a variety of simple modifications can be made to the technical solution of the present invention, and these simple modifications all belong to the protection scope of the present invention.

[0100] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present invention will not further describe various possible combinations.

[0101] In addition, various embodiments of the present invention may be arbitrarily combined, and as long as they do not violate the concept of the present invention, they should also be regarded as the contents disclosed by the present invention.

Claims

1. A method for constructing a diagnostic model for non-alcoholic fatty liver disease, characterized in that: Includes steps: S101, Obtaining non-alcoholic fatty liver disease dataset through high-throughput gene expression database; S102, dividing the non-alcoholic fatty liver disease data set into two groups: a NAFLD group and a Control group, and performing gene difference calibration analysis on the NAFLD group and the Control group to obtain differentially expressed gene DEGs; S103, taking the intersection of the obtained differentially expressed gene DEGs and the inflammation and neutrophil extracellular trap-related genes INRGs to obtain M inflammation and neutrophil extracellular trap-related differentially expressed genes INRDEGs; S104, constructing an SVM model using the obtained M inflammation and neutrophil extracellular trap-related differentially expressed genes INRDEGs through a support vector machine (SVM) algorithm to obtain the number of genes N with the lowest error rate and the highest accuracy, and N corresponding inflammation and neutrophil extracellular trap-related differentially expressed genes INRDEGs, where M>=N; S105. Construct a LASSO regression model by performing LASSO regression analysis on N inflammation- and neutrophil extracellular trap-related differentially expressed genes INRDEGs contained in the SVM model to obtain a diagnostic model for non-alcoholic fatty liver disease.

2. The method for constructing a diagnostic model for non-alcoholic fatty liver disease according to claim 1, characterized in that: The non-alcoholic fatty liver disease dataset includes multiple batches of datasets; after step S101, the step further includes: S1011. Remove batch effects from multiple batches of data sets to obtain an integrated non-alcoholic fatty liver disease data set.

3. The method for constructing a diagnostic model for non-alcoholic fatty liver disease according to claim 1, characterized in that: In step S104, N=4, and the four corresponding inflammation- and neutrophil extracellular trap-related differentially expressed genes INRDEGs are IL6, TNF, SPP1, and CCL5.

4. The method for constructing a diagnostic model for non-alcoholic fatty liver disease according to claim 1, characterized in that: After step S103, the method further includes the following steps: S1031, performing logistic regression based on the obtained multiple inflammation and neutrophil extracellular trap-related differentially expressed genes INRDEGs and constructing a logistic regression model; S1032. Obtain whether each inflammation and neutrophil extracellular trap-related differentially expressed gene INRDEGs is statistically significant in the logistic regression model; and retain the inflammation and neutrophil extracellular trap-related differentially expressed gene INRDEGs that are statistically significant.

5. The method for constructing a diagnostic model for non-alcoholic fatty liver disease according to claim 1, characterized in that: After step S103, the method further includes the following steps: S1033. Analyze the expression differences of the inflammation and neutrophil extracellular trap-related differentially expressed genes INRDEGs between different sample groups in the non-alcoholic fatty liver disease dataset to obtain the expression difference information of each inflammation and neutrophil extracellular trap-related differentially expressed gene INRDEGs between different sample groups; and retain the inflammation and neutrophil extracellular trap-related differentially expressed genes INRDEGs that have significant differences between different sample groups.

6. The method for constructing a diagnostic model for non-alcoholic fatty liver disease according to claim 1, characterized in that: In step S102, the threshold for obtaining the differentially expressed genes DEGs is |logFC|>0.3 and adj.p<0.

05.

7. The method for constructing a diagnostic model for non-alcoholic fatty liver disease according to any one of claims 1 to 5, characterized in that: The N inflammation and neutrophil extracellular trap-related differentially expressed genes INRDEGs are used as hub genes, and after step S105, the following steps are further included: S106. Drawing a nomogram based on the hub gene to show the relationship between the hub genes in the non-alcoholic fatty liver disease data set to verify the utility of the expression amount of the hub gene for the non-alcoholic fatty liver disease diagnostic model; S107, drawing a calibration curve through Calibration analysis, and drawing the fitting of the actual probability and the probability predicted by the model under different situations in the curve to evaluate the prediction effect of the non-alcoholic fatty liver disease diagnosis model on the actual result; S108. Based on the diagnostic model risk score RiskScore in the non-alcoholic fatty liver disease dataset, the role of the non-alcoholic fatty liver disease diagnostic model in clinical utility is evaluated by decision curve analysis DCA.

8. The method for constructing a diagnostic model for non-alcoholic fatty liver disease according to claim 7, characterized in that: After step S105, the method further includes the following steps: S109, drawing a ROC curve based on the risk score RiskScore of the diagnostic model in the non-alcoholic fatty liver disease data set, so as to perform risk scoring through the ROC curve; The calculation formula of risk score RiskScore is as follows: IL6*(-0.81085028038064)+TNF*(-0.135302961906175)+SPP1*(0.263579706135327)+CCL5*(0.203623519951156).

9. The method for constructing a diagnostic model for non-alcoholic fatty liver disease according to claim 8, characterized in that: After step S108, the method further includes the following steps: S110, dividing the non-alcoholic fatty liver disease (NAFLD) results into a HighRisk group and a LowRisk group according to the median value of the risk score expression; S111. Display the differences in expression levels of the N hub genes in the HighRisk group and the LowRisk group in the results of non-alcoholic fatty liver disease (NAFLD) through group comparison diagrams to obtain the statistical significance of the expression levels of the N hub genes in the HighRisk group and the LowRisk group.

10. A non-alcoholic fatty liver disease diagnostic device, comprising a computer device, the computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the non-alcoholic fatty liver disease diagnostic model constructed according to any one of claims 1 to 9 is implemented.

Images