Prediction system and device for predicting congenital heart disease combined with delayed intelligence development based on gene and storage medium
By constructing a machine learning model based on the expression data of LAMP3, VSIG4 and/or LCN8 genes, the accuracy and reliability of congenital heart disease diagnosis in the prior art are solved, and efficient prediction and diagnosis of congenital heart disease combined with backward intellectual development are achieved.
Patent Information
- Application Number
- CN202411127034.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-16
- Publication Date
- 2024-12-03
AI Technical Summary
The existing diagnosis model of congenital heart disease has problems with accuracy and reliability, especially during fetal diagnosis, the fetal position, the thickness of the pregnant woman's abdominal wall, amniotic fluid volume and the professional technical level of ultrasonic doctors, resulting in early prediction and intervention. difficulty.
Based on the expression data of the target gene of LAMP3, VSIG4 and/or LCN8, the samples to be tested are predicted by constructing a machine learning model to determine whether it is a sample of congenital heart disease combined with backward intellectual development.
It has achieved a more accurate and efficient diagnosis of congenital heart disease combined with backward intellectual development, and has high diagnostic efficacy, convenience and effectiveness.
Smart Images

Figure CN119068974A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of computer technology, and in particular, relates to a prediction system, device and storage medium for predicting congenital heart disease combined with intellectual disability based on gene prediction. Background Art
[0002] Congenital heart disease is congenital heart disease, referred to as CHD. Congenital heart disease is a congenital malformation caused by abnormal development of the heart and great blood vessels during the fetal period. It is the most common heart disease in children. According to hemodynamics combined with pathophysiological changes, congenital heart disease can be divided into cyanotic or non-cyanotic types, and can also be divided into three categories according to the presence or absence of shunt: no shunt (such as pulmonary artery stenosis, aortic coarctation), left-to-right shunt (such as atrial septal defect, ventricular septal defect, patent ductus arteriosus) and right-to-left shunt (such as tetralogy of Fallot, transposition of great vessels). Epidemiological statistics show that the number of patients with congenital heart disease has increased year by year, bringing a heavy burden to families and society. The drawbacks of the existing diagnosis model of congenital heart disease are also the main reason for this burden.
[0003] Congenital heart disease combined with intellectual disability is one of the common congenital heart diseases. Due to incomplete heart development, it may cause cardiac ejection dysfunction, thus causing intellectual disability. The current treatments for this disease mainly include traditional cardiac surgery and palliative surgery. At present, the prenatal diagnosis of congenital heart disease at home and abroad is mainly based on echocardiography. However, the diagnosis of congenital heart disease by echocardiography is affected by the pregnancy week, fetal position, abdominal wall thickness of pregnant women, amniotic fluid volume and the professional and technical level of ultrasound doctors, which affect the final test results. It can be seen that the development of a new method that can be used for the early prediction of congenital heart disease combined with intellectual disability is of great clinical significance for the early intervention treatment of patients. Summary of the invention
[0004] In view of this, the purpose of the present invention is to provide a prediction system, device and storage medium for congenital heart disease combined with intellectual retardation based on gene prediction. In addition, the present invention also provides a prediction method for congenital heart disease combined with intellectual retardation based on gene prediction, the method comprising: obtaining the expression data of the target gene of the sample to be tested, the target gene is LAMP3, VSIG4 and / or LCN8; inputting the expression data of the target gene into the constructed prediction model, the prediction model predicts whether the sample to be tested is a congenital heart disease combined with intellectual retardation sample based on the expression data of the target gene; based on the output result of the prediction model, predict and output whether the sample to be tested is a congenital heart disease combined with intellectual retardation sample result. It has been verified that the prediction method has a high diagnostic efficacy for the prediction of congenital heart disease combined with intellectual retardation.
[0005] The above-mentioned object of the present invention is achieved through the following technical solutions:
[0006] The first aspect of the present invention provides a prediction method for congenital heart disease combined with intellectual disability based on gene prediction, the method comprising:
[0007] Acquiring data, acquiring expression data of target genes of the sample to be tested, wherein the target genes are LAMP3, VSIG4 and / or LCN8;
[0008] Analyze the data, input the expression data of the target gene into the constructed prediction model, and the prediction model predicts whether the sample to be tested is a sample of congenital heart disease combined with intellectual disability based on the expression data of the target gene;
[0009] Output the prediction result, and predict and output whether the sample to be tested is a sample of congenital heart disease combined with intellectual developmental delay based on the output result of the prediction model.
[0010] Furthermore, the method for constructing the prediction model includes:
[0011] Acquiring target gene expression data, wherein the target gene is LAMP3, VSIG4 and / or LCN8, and the target gene expression data is from a population with congenital heart disease and intellectual disability and a population with congenital heart disease and intellectual disability;
[0012] Construct a prediction model and verify its effectiveness. Divide the target gene expression data into a training set and a test set. Extract the target gene expression data in the training set and input it into a machine learning model to construct a prediction model. Verify it through the test set to evaluate the effectiveness of the prediction model, and obtain a constructed prediction model.
[0013] Furthermore, the machine learning model includes a linear regression model, a logistic regression model, a decision tree model, a random forest model, a support vector machine model, a K nearest neighbor model, a neural network model and / or a naive Bayes model.
[0014] Further, whether the sample to be tested is a sample of congenital heart disease combined with intellectual disability is determined based on the following criteria:
[0015] If the expression level of any one or more of the target genes LAMP3 and LCN8 is lower than the threshold, and / or the expression level of VSIG4 is higher than the threshold, then the sample to be tested is a sample of congenital heart disease combined with intellectual disability;
[0016] If the expression level of any one or more of the target genes LAMP3 and LCN8 is higher than the threshold, and / or the expression level of VSIG4 is lower than the threshold, the sample to be tested is a sample without congenital heart disease and intellectual disability.
[0017] Further, the expression data of the target gene includes mRNA expression data of the target gene or protein expression data of the target gene;
[0018] The mRNA expression data is obtained by high-throughput sequencing technology, RT-PCR, qRT-PCR or in situ hybridization technology;
[0019] The protein expression data are obtained by mass spectrometry, immunoblotting, immunohistochemistry or immunoprecipitation.
[0020] The second aspect of the present invention provides a prediction system for congenital heart disease combined with intellectual disability based on gene prediction, the system comprising:
[0021] A data acquisition unit is used to acquire expression data of a target gene of a sample to be tested, wherein the target gene is LAMP3, VSIG4 and / or LCN8;
[0022] A data analysis unit, used for inputting the expression data of the target gene into a constructed prediction model, wherein the prediction model predicts whether the sample to be tested is a sample of congenital heart disease combined with intellectual disability based on the expression data of the target gene;
[0023] The prediction result output unit is used to predict and output whether the sample to be tested is a sample of congenital heart disease combined with intellectual development retardation based on the output result of the prediction model.
[0024] Furthermore, the method for constructing the prediction model includes:
[0025] Acquiring target gene expression data, wherein the target gene is LAMP3, VSIG4 and / or LCN8, and the target gene expression data is from a population with congenital heart disease and intellectual disability and a population with congenital heart disease and intellectual disability;
[0026] Constructing a prediction model and validating its effectiveness, dividing the target gene expression data into a training set and a test set, extracting the target gene expression data in the training set and inputting it into a machine learning model to construct a prediction model, and validating it through the test set to evaluate the effectiveness of the prediction model, thereby obtaining a constructed prediction model;
[0027] The machine learning model includes a linear regression model, a logistic regression model, a decision tree model, a random forest model, a support vector machine model, a K nearest neighbor model, a neural network model and / or a naive Bayes model.
[0028] Further, whether the sample to be tested is a sample of congenital heart disease combined with intellectual disability is determined based on the following criteria:
[0029] If the expression level of any one or more of the target genes LAMP3 and LCN8 is lower than the threshold, and / or the expression level of VSIG4 is higher than the threshold, then the sample to be tested is a sample of congenital heart disease combined with intellectual disability;
[0030] If the expression level of any one or more of the target genes LAMP3 and LCN8 is higher than the threshold, and / or the expression level of VSIG4 is lower than the threshold, the sample to be tested is a sample without congenital heart disease and intellectual disability.
[0031] The third aspect of the present invention provides an electronic device for disease prediction based on genetic prediction of congenital heart disease combined with intellectual disability, the device comprising a memory and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by one or more processors to implement the prediction method instructions based on genetic prediction of congenital heart disease combined with intellectual disability described in the first aspect of the present invention.
[0032] The fourth aspect of the present invention provides a non-temporary computer-readable storage medium. When the instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the prediction method for congenital heart disease combined with intellectual disability based on genetic prediction as described in the first aspect of the present invention.
[0033] In addition, the present invention also provides the use of a reagent for detecting the expression level of LAMP3, VSIG4 and / or LCN8 in the preparation of a diagnostic product for diagnosing congenital heart disease combined with intellectual disability.
[0034] Compared with the prior art, the present invention has the following beneficial effects:
[0035] The present invention provides a new and highly accurate prediction method for congenital heart disease combined with intellectual disability based on gene prediction, the method comprising: obtaining expression data of target genes of a sample to be tested, the target genes being LAMP3, VSIG4 and / or LCN8; inputting the expression data of the target genes into a constructed prediction model, the prediction model predicting whether the sample to be tested is a sample of congenital heart disease combined with intellectual disability based on the expression data of the target genes; predicting and outputting whether the sample to be tested is a sample of congenital heart disease combined with intellectual disability based on the output result of the prediction model, the above method is used to achieve a more accurate prediction of the disease of congenital heart disease combined with intellectual disability, and has high convenience and effectiveness. BRIEF DESCRIPTION OF THE DRAWINGS
[0036] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.
[0037] Figure 1 A schematic flow chart of a prediction method for congenital heart disease combined with intellectual disability based on gene prediction provided by an embodiment of the present invention;
[0038] Figure 2 A schematic diagram of a prediction system for congenital heart disease combined with intellectual disability based on gene prediction provided by an embodiment of the present invention;
[0039] Figure 3 A schematic diagram of an electronic device for disease prediction based on gene prediction of congenital heart disease combined with intellectual disability provided by an embodiment of the present invention;
[0040] Figure 4 This is an example diagram of the FASTQ data format for read segments;
[0041] Figure 5 is the sequencing quality histogram;
[0042] Figure 6 is the average quality score of the sequence;
[0043] Figure 7 This is a correlation analysis diagram between mRNA samples;
[0044] Figure 8 The volcano plot and heat map (top100) corresponding to the differentially expressed genes;
[0045] Fig. 9 The enrichment results of differentially expressed genes (GO enrichment on the left and KEGG enrichment on the right);
[0046] Fig.10 Figure 2 is the differential expression result of LAMP3, VSIG4 and LCN8, where A is a sample from a congenital heart disease patient with normal nutrition and normal intellectual development, and B is a sample from a congenital heart disease patient with normal nutrition and intellectual retardation;
[0047] Fig.11 The results of the diagnostic efficacy validation of LAMP3, VSIG4, and LCN8 for predicting congenital heart disease combined with intellectual disability were presented;
[0048] Fig.12 The results of the diagnostic efficacy validation of any two of LAMP3, VSIG4, and LCN8 combined for predicting congenital heart disease with intellectual disability;
[0049] Fig.13 This is the result of validating the diagnostic efficacy of the combination of LAMP3, VSIG4 and LCN8 in predicting congenital heart disease combined with intellectual disability. DETAILED DESCRIPTION
[0050] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.
[0051] In some of the processes described in the specification and claims of the present invention and the above-mentioned drawings, multiple operations appearing in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this document or may be executed in parallel. The serial numbers of the operations, such as S101, S102, etc., are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel.
[0052] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.
[0053] Figure 1 The present invention provides a method for predicting congenital heart disease combined with intellectual disability based on gene prediction. Specifically, the method comprises the following steps:
[0054] S101: Acquire expression data of target genes of the sample to be tested, wherein the target genes are LAMP3, VSIG4 and / or LCN8;
[0055] In one embodiment, the sample to be tested is derived from a subject.
[0056] In one embodiment, the subject includes mammals and non-mammals. Examples of mammals include, but are not limited to, any member of the class Mammalia: humans, non-human primates such as chimpanzees and other apes and monkeys; farm animals such as cattle, horses, sheep, goats, pigs; domestic animals such as rabbits, dogs and cats; laboratory animals, including rodents such as rats, mice and guinea pigs, etc. Examples of non-mammals include, but are not limited to, birds, fish or other non-mammals, etc.
[0057] In one embodiment, the subject is a human.
[0058] In one embodiment, the sample to be tested refers to a composition obtained from or derived from a target subject, which contains cell entities and / or other molecular entities to be characterized and / or identified, for example, based on physical, biochemical, chemical and / or physiological characteristics. The sample can be obtained from the subject's blood and other fluid samples and tissue samples of biological origin, such as biopsy tissue samples or tissue cultures or cells derived therefrom. The source of the tissue sample can be solid tissue, such as from fresh, frozen and / or preserved organ or tissue samples, biopsy tissues or aspirates; blood or any blood component; body fluids; cells from any time of individual pregnancy or development; or plasma.
[0059] In one embodiment, the sample to be tested includes but is not limited to: tissue, blood, serum, plasma, blood-derived cells, lymph, synovial fluid, cerebrospinal fluid, pleural fluid, peritoneal fluid, bladder washing fluid, secretions (e.g., breast secretions), oral wash fluid, swabs (e.g., oral swabs), touch preparations, fine needle aspirations, cell extracts, and combinations thereof.
[0060] In one embodiment, the sample to be tested is a blood sample from a subject.
[0061] In one embodiment, the gene expression data includes gene mRNA expression data or gene protein expression data.
[0062] In one embodiment, the mRNA expression data is obtained by high-throughput sequencing technology, RT-PCR, qRT-PCR or in situ hybridization technology. Any method or technology that can be used to detect the mRNA expression level corresponding to a gene can be used in the present invention.
[0063] In one embodiment, the protein expression data is obtained by mass spectrometry, immunoblotting, immunohistochemistry or immunoprecipitation. Any method or technique that can be used to detect the protein expression corresponding to a gene can be used in the present invention.
[0064] In one embodiment, the gene expression data is the mRNA expression data of the gene, and the mRNA expression data of the gene can be obtained by high-throughput sequencing technology. In a specific embodiment of the present invention, the high-throughput sequencing technology is the Illumina high-throughput sequencing platform.
[0065] In one embodiment, the target genes include LAMP3, VSIG4, LCN8, a combination of LAMP3 and VSIG4, a combination of LAMP3 and LCN8, and a combination of LAMP3, VSIG4 and LCN8.
[0066] S102: inputting the expression data of the target gene into a constructed prediction model, wherein the prediction model predicts whether the sample to be tested is a sample of congenital heart disease combined with intellectual disability based on the expression data of the target gene;
[0067] In one embodiment, the target genes include LAMP3, VSIG4, LCN8, a combination of LAMP3 and VSIG4, a combination of LAMP3 and LCN8, and a combination of LAMP3, VSIG4 and LCN8.
[0068] In one embodiment, the method for constructing the prediction model includes:
[0069] Acquiring target gene expression data, wherein the target gene is LAMP3, VSIG4 and / or LCN8, and the target gene expression data is from a population with congenital heart disease and intellectual disability and a population with congenital heart disease and intellectual disability;
[0070] Construct a prediction model and verify its effectiveness. Divide the target gene expression data into a training set and a test set. Extract the target gene expression data in the training set and input it into a machine learning model to construct a prediction model. Verify it through the test set to evaluate the effectiveness of the prediction model, and obtain a constructed prediction model.
[0071] In one embodiment, the machine learning model includes a linear regression model, a logistic regression model, a decision tree model, a random forest model, a support vector machine model, a K nearest neighbor model, a neural network model and / or a naive Bayes model.
[0072] In one embodiment, whether the sample to be tested is a sample of congenital heart disease combined with intellectual disability is determined based on the following criteria:
[0073] If the expression level of any one or more of the target genes LAMP3 and LCN8 is lower than the threshold, and / or the expression level of VSIG4 is higher than the threshold, then the sample to be tested is a sample of congenital heart disease combined with intellectual disability;
[0074] If the expression level of any one or more of the target genes LAMP3 and LCN8 is higher than the threshold, and / or the expression level of VSIG4 is lower than the threshold, the sample to be tested is a sample without congenital heart disease and intellectual disability.
[0075] In one embodiment, the inventors collected blood samples from patients with congenital heart disease combined with intellectual disability and patients with congenital heart disease combined with normal intellectual development for sequencing analysis to verify the expression of the above target genes and the diagnostic efficacy of the disease of congenital heart disease combined with intellectual disability. The clinical information of the patients with congenital heart disease combined with intellectual disability and patients with congenital heart disease combined with normal intellectual development is shown in Table 1 below.
[0076] Table 1 Clinical information of patients with congenital heart disease and intellectual disability and patients with congenital heart disease and normal intellectual development
[0077]
[0078]
[0079] In one embodiment, based on the sequencing by synthesis (SBS) technology, the Illumina high-throughput sequencing platform is used to sequence the cDNA library, which can produce a large number of high-quality reads. These reads or bases produced by the sequencing platform are called raw data. To facilitate the analysis of sequencing data, the raw image data obtained by Illumina sequencing is converted into sequence data, i.e., FASTQ format, through Base Calling to obtain the most original sequencing data file. The FASTQ format file can record the bases of the measured reads and their quality scores. Figure 4 As shown, the FASTQ format is stored in units of sequencing reads, and each read occupies 4 lines, where the first and third lines consist of file identification marks (sequence identifiers) and read names (IDs) (the first line starts with "@" and the third line starts with "+"; the ID in the third line can be omitted, but the "+" cannot be omitted), the second line is the base sequence, and the fourth line is the sequencing quality score of the base at the corresponding position.
[0080] An Illumina sequencer has two flow cells in one run. A flow cell contains eight lanes. One lane contains two columns. Each column contains 60 tiles. Each tile is seeded with a different cluster. The detailed information in the generated sequencing file identification mark (Sequence Identifiers) is shown in Table 2.
[0081] Table 2 Interpretation of sequencing file identification marks
[0082]
[0083]
[0084] The quality score of the read is represented by different characters, where the ASCII value of each character minus 33 is the corresponding sequencing quality value. Generally, the base quality ranges from 0-40, that is, the corresponding ASCII code is from "!" (0+33) to "I" (40+33). If the sequencing error rate is represented by E and the base quality value of Illunima HiSeq / Miseq is represented by Q, then the following relationship exists: Formula 1: Q = -10log 10 (E) The error rate of sequencing reads increases as sequencing progresses (see Table 3), which is caused by the consumption of chemical reagents during the sequencing process and is a common feature of the Illumina high-throughput sequencing platform.
[0085] Table 3. Concise correspondence between sequencing error rate and sequencing quality value
[0086]
[0087] In one embodiment, the present application performs data volume statistics on the sequence of the original data, and the results are shown in Table 4. In one embodiment, the present application performs quality control on the sequencing data. Since the original sequencing data may contain sequencing adapter sequences, low-quality reads and other types of redundant sequences, this will seriously affect the quality of subsequent assembly. In order to ensure the accuracy of subsequent bioinformatics analysis, the original sequencing data is first filtered to obtain high-quality sequencing data (clean data). Cutadapt can find and delete redundant sequences such as adapters, primers, polyA tails and other types in reads. Cutadapt (version, 3.5, default parameters) is used to clean the sequencing data. In one embodiment, the present application performs statistics on the data after quality control, and MultiQC supports the integration of quality control results, and performs data volume statistics on the sequence after data quality control. In one embodiment, the present application performs data quality assessment after quality control. Illumina sequencing belongs to the second generation sequencing technology, and a single run can generate billions of reads. Such a large amount of data cannot show the quality of each read one by one; using statistical methods, all sequencing reads are quality assessed, which can intuitively reflect the sequencing quality and library construction quality of the sample from a macro perspective. We perform sequencing-related quality assessment on the sequencing data of each sample. The average quality value of each base position in the reads is as follows: Figure 5 The statistics of the average sequence quality score and the number of reads are shown in Figure 6As shown. In one embodiment, the present application performs gene quantification, HISAT2 (version, 2.1.0) aligns the genome, samtools (version, 1.9) sorts and indexes bam files, QualiMap (version, 2.2.2) quality controls bam files, and featureCounts (version, 1.6.4) performs gene quantification. The reference genome version is GRCh38.primary_assembly.genome.fa.
[0088] Table 4 Sequencing data statistics
[0089]
[0090]
[0091] The correlation of gene expression levels between samples is an important indicator for testing the reliability of the experiment and whether the sample selection is reasonable. Biological repetition is necessary for biological experiments. There are two main purposes for setting up biological repetition: one is to prove that the biological experimental operations involved are not accidental, but repeatable; the other is to ensure that the subsequent differential gene analysis obtains more reliable results. The correlation of gene expression levels between samples is an important indicator for testing the reliability of the experiment and whether the sample selection is reasonable. The closer the correlation coefficient is to 1, the higher the similarity of the expression patterns between samples. If there are biological repetitions in the sample, the correlation coefficient between biological repetitions is usually required to be higher. The Encode plan recommends that the square of the Pearson correlation coefficient (R2) be greater than 0.92 (under ideal sampling and experimental conditions). In a specific embodiment, we require that R2 between biological replicate samples be at least greater than 0.8. The correlation coefficients within and between groups are calculated based on the count values of all genes / transcripts in the sample, and drawn into a heat map, which can intuitively display the sample differences between groups and the sample repetitions within the group. The sample correlation heat map involved in this embodiment is shown in the figure. Figure 7 shown.
[0092] In one embodiment, the present application performs differential expression analysis. In the transcriptome, determining whether the expression level of a gene in different samples is different is one of the core contents of the analysis. After obtaining the gene expression level, the expression data can be statistically analyzed to screen genes with significant differences between different samples. The differential analysis is mainly divided into two steps: (1) First, the original read count is normalized, mainly to correct the sequencing depth. (2) The hypothesis test probability (P-value) is calculated through a statistical model. The differential expression analysis uses DEseq2, and the screening parameters are pvalue<0.05 and |log2foldchange|>1. Using the above standards, 362 differentially expressed genes were screened between patients with congenital heart disease and intellectual development retardation and patients with congenital heart disease and normal intellectual development, of which 145 genes were upregulated and 217 genes were downregulated. The volcano plot and heat map are shown in the figure. Figure 8 shown.
[0093] In one embodiment, the present application further performs functional enrichment analysis on the differentially expressed genes between the above-mentioned patients with congenital heart disease and intellectual disability and patients with congenital heart disease and normal intellectual development, and uses the David database (https: / / david.ncifcrf.gov / tools.jsp) to perform GO functional enrichment and KEGG functional enrichment analysis on the differentially expressed genes. The screening criteria are: pvalue<0.05, and the enrichment results are as follows: Fig. 9 And as shown in Table 5.
[0094] Table 5 KEGG enrichment results
[0095]
[0096] In one embodiment, the present application screened and obtained differentially expressed genes between patients with congenital heart disease combined with intellectual disability and patients with congenital heart disease combined with normal intellectual development, including LAMP3, VSIG4, and LCN8. It was verified that LAMP3, VSIG4, and LCN8 were significantly differentially expressed between patients with congenital heart disease combined with intellectual disability and patients with congenital heart disease combined with normal intellectual development (see Fig.10 ).
[0097] In one embodiment, the present application further verifies the diagnostic efficacy of the screened differentially expressed genes LAMP3, VSIG4, and LCN8 in predicting congenital heart disease combined with intellectual retardation, and uses the R package "pROC" (version 1.15.0) to perform receiver operating characteristic (ROC) analysis, and calculates the area under the receiver operating characteristic curve (AUC) to evaluate the accuracy of the above-mentioned single differentially expressed gene, any two differentially expressed genes combined, and three differentially expressed genes combined in predicting congenital heart disease combined with intellectual retardation, wherein the range of the AUC value is 0-1. Among them, when judging the diagnostic efficacy of a single differentially expressed gene in predicting congenital heart disease combined with intellectual retardation, the expression amount of a single differentially expressed gene is directly used for analysis, and the level corresponding to the point with the largest Youden index is selected as its cutoff value. When judging the diagnostic efficacy of any two differentially expressed genes combined or three differentially expressed genes combined in predicting congenital heart disease combined with intellectual developmental delay, Logistic regression analysis was performed on the expression levels of each differentially expressed gene, and the probability of each patient having congenital heart disease combined with intellectual developmental delay was calculated through the fitted regression curve, and different probability division thresholds were determined. According to the determined probability division thresholds, the accuracy, specificity and sensitivity of any two differentially expressed genes combined or three differentially expressed genes combined in predicting congenital heart disease combined with intellectual developmental delay were calculated.
[0098] Verification results are as follows Figure 11-13As shown in the figure, the results showed that single genes LAMP3, VSIG4, and LCN8 all had high diagnostic efficacy for predicting congenital heart disease combined with intellectual disability. Among them, the AUC value corresponding to LAMP3 was 0.836, the sensitivity was 60%, the specificity was 92.9%, and the cutoff value was 36.000, the AUC value corresponding to VSIG4 was 0.814, the sensitivity was 80%, the specificity was 78.6%, and the cutoff value was 24.000, and the AUC value corresponding to LCN8 was 0.800, the sensitivity was 100%, the specificity was 71.4%, and the cutoff value was 7.000. The combination of any two genes among LAMP3, VSIG4 and LCN8 also has a high diagnostic efficacy for predicting congenital heart disease with intellectual developmental retardation. Among them, the AUC value corresponding to LAMP3+VSIG4 is 0.829, the sensitivity is 80%, the specificity is 78.6%, and the cutoff value is 0.280, the AUC value corresponding to LAMP3+LCN8 is 0.943, the sensitivity is 100%, the specificity is 85.7%, and the cutoff value is 0.270, and the AUC value corresponding to VSIG4+LCN8 is 0.914, the sensitivity is 80%, the specificity is 92.9%, and the cutoff value is 0.349. The combination of LAMP3, VSIG4 and LCN8 also has a high diagnostic efficacy for predicting congenital heart disease with intellectual developmental retardation. The AUC value corresponding to LAMP3+VSIG4+LCN8 is 0.914, the sensitivity is 100%, the specificity is 78.6%, and the cutoff value is 0.184. The above verification results prove that genes LAMP3, VSIG4 and / or LCN8 have high diagnostic efficacy for predicting congenital heart disease combined with intellectual disability, which also proves the effectiveness of the prediction method provided in the present application (with high diagnostic efficacy, accuracy, sensitivity and specificity), and can be used for the effective prediction of congenital heart disease combined with intellectual disability.
[0099] In one embodiment, as described above, the present application collects blood samples of patients with congenital heart disease combined with intellectual retardation and patients with congenital heart disease combined with normal intellectual development for sequencing analysis, and obtains the target gene expression data corresponding to the two clinical characteristics of patients with congenital heart disease combined with intellectual retardation and patients with congenital heart disease combined with normal intellectual development. The clinical information of the patients with congenital heart disease combined with intellectual retardation and patients with congenital heart disease combined with normal intellectual development is shown in Table 1. The target gene expression data is extracted and input into the machine learning model to construct a prediction model, and a constructed prediction model is obtained.
[0100] In one embodiment, the machine learning model includes a linear regression model, a logistic regression model, a random forest model, a Lasso regression model, a Ridge regression model, a linear discriminant analysis model, a nearest neighbor model, a decision tree model, a perceptron model, a neural network model, a support vector machine model, a naive Bayes model, an AdaBoost model, a GBDT model, an XGBoost model, a LightGBM model and / or a CatBoost model.
[0101] In one embodiment, the present invention further verifies the diagnostic efficacy of the constructed prediction model, that is, another data set (validation set) containing the target gene expression data corresponding to patients with congenital heart disease and intellectual retardation and patients with congenital heart disease and normal intellectual development is taken, and the efficacy of the constructed prediction model is verified in the data set. The results show that the above-constructed prediction model also has good diagnostic efficacy in the validation set.
[0102] In one embodiment, when the construction method of the above-mentioned prediction model is determined, the prediction model contains the threshold value, that is, when the prediction model is determined, the threshold value is also determined. Based on the determined threshold value, the classification result of whether the sample to be tested is a sample of congenital heart disease combined with intellectual retardation can be predicted. The specific judgment result based on the prediction model is: if the expression level of any one or more of the target genes LAMP3 and LCN8 is lower than the threshold value, and / or the expression level of VSIG4 is higher than the threshold value, then the sample to be tested is a sample of congenital heart disease combined with intellectual retardation; if the expression level of any one or more of the target genes LAMP3 and LCN8 is higher than the threshold value, and / or the expression level of VSIG4 is lower than the threshold value, then the sample to be tested is a sample of non-congenital heart disease combined with intellectual retardation.
[0103] S103: Predicting and outputting whether the sample to be tested is a sample of congenital heart disease combined with intellectual disability based on the output result of the prediction model;
[0104] In one embodiment, when the construction method of the above-mentioned prediction model is determined, the prediction model contains the threshold value, that is, when the prediction model is determined, the threshold value is also determined. Based on the determined threshold value, the classification result of whether the sample to be tested is a sample of congenital heart disease combined with intellectual retardation can be predicted. The specific judgment result based on the prediction model is: if the expression level of any one or more of the target genes LAMP3 and LCN8 is lower than the threshold value, and / or the expression level of VSIG4 is higher than the threshold value, then the sample to be tested is a sample of congenital heart disease combined with intellectual retardation; if the expression level of any one or more of the target genes LAMP3 and LCN8 is higher than the threshold value, and / or the expression level of VSIG4 is lower than the threshold value, then the sample to be tested is a sample of non-congenital heart disease combined with intellectual retardation.
[0105] Figure 2 A prediction system for congenital heart disease combined with intellectual disability based on gene prediction is provided in an embodiment of the present invention. Specifically, the system comprises:
[0106] A data acquisition unit is used to acquire expression data of a target gene of a sample to be tested, wherein the target gene is LAMP3, VSIG4 and / or LCN8;
[0107] A data analysis unit, used for inputting the expression data of the target gene into a constructed prediction model, wherein the prediction model predicts whether the sample to be tested is a sample of congenital heart disease combined with intellectual disability based on the expression data of the target gene;
[0108] The prediction result output unit is used to predict and output whether the sample to be tested is a sample of congenital heart disease combined with intellectual development retardation based on the output result of the prediction model.
[0109] In one embodiment, the method for constructing the prediction model includes:
[0110] Acquiring target gene expression data, wherein the target gene is LAMP3, VSIG4 and / or LCN8, and the target gene expression data is from a population with congenital heart disease and intellectual disability and a population with congenital heart disease and intellectual disability;
[0111] Constructing a prediction model and validating its effectiveness, dividing the target gene expression data into a training set and a test set, extracting the target gene expression data in the training set and inputting it into a machine learning model to construct a prediction model, and validating it through the test set to evaluate the effectiveness of the prediction model, thereby obtaining a constructed prediction model;
[0112] The machine learning model includes a linear regression model, a logistic regression model, a decision tree model, a random forest model, a support vector machine model, a K nearest neighbor model, a neural network model and / or a naive Bayes model.
[0113] In one embodiment, whether the sample to be tested is a sample of congenital heart disease combined with intellectual disability is determined based on the following criteria:
[0114] If the expression level of any one or more of the target genes LAMP3 and LCN8 is lower than the threshold, and / or the expression level of VSIG4 is higher than the threshold, then the sample to be tested is a sample of congenital heart disease combined with intellectual disability;
[0115] If the expression level of any one or more of the target genes LAMP3 and LCN8 is higher than the threshold, and / or the expression level of VSIG4 is lower than the threshold, the sample to be tested is a sample without congenital heart disease and intellectual disability.
[0116] Figure 3 It is a schematic diagram of an electronic device for disease prediction based on gene prediction of congenital heart disease combined with intellectual disability provided by an embodiment of the present invention. Specifically, the computer device includes a memory and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by one or more processors to implement the prediction method instructions based on gene prediction of congenital heart disease combined with intellectual disability as described above.
[0117] An embodiment of the present invention also provides a non-temporary computer-readable storage medium. Specifically, when the instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the prediction method for congenital heart disease combined with intellectual disability based on gene prediction as described above.
[0118] The validation results of this validation example show that assigning intrinsic weights to indications can moderately improve the performance of the method relative to the default setting.
[0119] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0120] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.
[0121] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0122] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.
[0123] A person of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, and the storage medium may include: a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, etc.
[0124] A person skilled in the art can understand that all or part of the steps in the above-mentioned embodiment method can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. The above-mentioned storage medium can be a read-only memory, a disk or an optical disk, etc.
[0125] The above is a detailed introduction to a computer device provided by the present invention. For a person skilled in the art, according to the concept of the embodiments of the present invention, there may be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as limiting the present invention.
Claims
1. A prediction method based on gene prediction for congenital heart disease combined with intellectual disability. It is characterized in that The method comprises: Acquiring data, acquiring expression data of target genes of the sample to be tested, wherein the target genes are LAMP3, VSIG4 and / or LCN8; Analyze the data, input the expression data of the target gene into the constructed prediction model, and the prediction model predicts whether the sample to be tested is a sample of congenital heart disease combined with intellectual disability based on the expression data of the target gene; Output the prediction result, and predict and output whether the sample to be tested is a sample of congenital heart disease combined with intellectual developmental delay based on the output result of the prediction model.
2. The method for predicting congenital heart disease combined with intellectual disability based on gene prediction according to claim 1, It is characterized in that The method for constructing the prediction model includes: Acquiring target gene expression data, wherein the target gene is LAMP3, VSIG4 and / or LCN8, and the target gene expression data is from a population with congenital heart disease and intellectual disability and a population with congenital heart disease and intellectual disability; Construct a prediction model and verify its effectiveness. Divide the target gene expression data into a training set and a test set. Extract the target gene expression data in the training set and input it into a machine learning model to construct a prediction model. Verify it through the test set to evaluate the effectiveness of the prediction model, and obtain a constructed prediction model.
3. The prediction method for congenital heart disease combined with intellectual disability based on gene prediction according to claim 2, It is characterized in that The machine learning model includes a linear regression model, a logistic regression model, a decision tree model, a random forest model, a support vector machine model, a K nearest neighbor model, a neural network model and / or a naive Bayes model.
4. The method for predicting congenital heart disease combined with intellectual disability based on gene prediction according to claim 1, It is characterized in that Whether the sample to be tested is a sample of congenital heart disease combined with intellectual disability is determined based on the following criteria: If the expression level of any one or more of the target genes LAMP3 and LCN8 is lower than the threshold, and / or the expression level of VSIG4 is higher than the threshold, then the sample to be tested is a sample of congenital heart disease combined with intellectual disability; If the expression level of any one or more of the target genes LAMP3 and LCN8 is higher than the threshold, and / or the expression level of VSIG4 is lower than the threshold, the sample to be tested is a sample without congenital heart disease and intellectual disability.
5. The method for predicting congenital heart disease combined with intellectual disability based on gene prediction according to claim 1, It is characterized in that The target gene expression data includes the target gene mRNA expression data or the target gene protein expression data; The mRNA expression data is obtained by high-throughput sequencing technology, RT-PCR, qRT-PCR or in situ hybridization technology; The protein expression data are obtained by mass spectrometry, immunoblotting, immunohistochemistry or immunoprecipitation.
6. A prediction system based on gene prediction for congenital heart disease combined with intellectual disability. It is characterized in that The system comprises: A data acquisition unit is used to acquire expression data of a target gene of a sample to be tested, wherein the target gene is LAMP3, VSIG4 and / or LCN8; A data analysis unit, used for inputting the expression data of the target gene into a constructed prediction model, wherein the prediction model predicts whether the sample to be tested is a sample of congenital heart disease combined with intellectual disability based on the expression data of the target gene; The prediction result output unit is used to predict and output whether the sample to be tested is a sample of congenital heart disease combined with intellectual development retardation based on the output result of the prediction model.
7. The prediction system for congenital heart disease combined with intellectual disability based on gene prediction according to claim 6, It is characterized in that The method for constructing the prediction model includes: Acquiring target gene expression data, wherein the target gene is LAMP3, VSIG4 and / or LCN8, and the target gene expression data is from a population with congenital heart disease and intellectual disability and a population with congenital heart disease and intellectual disability; Constructing a prediction model and validating its effectiveness, dividing the target gene expression data into a training set and a test set, extracting the target gene expression data in the training set and inputting it into a machine learning model to construct a prediction model, and validating it through the test set to evaluate the effectiveness of the prediction model, thereby obtaining a constructed prediction model; The machine learning model includes a linear regression model, a logistic regression model, a decision tree model, a random forest model, a support vector machine model, a K nearest neighbor model, a neural network model and / or a naive Bayes model.
8. The prediction system for congenital heart disease combined with intellectual disability based on gene prediction according to claim 6, It is characterized in that Whether the sample to be tested is a sample of congenital heart disease combined with intellectual disability is determined based on the following criteria: If the expression level of any one or more of the target genes LAMP3 and LCN8 is lower than the threshold, and / or the expression level of VSIG4 is higher than the threshold, then the sample to be tested is a sample of congenital heart disease combined with intellectual disability; If the expression level of any one or more of the target genes LAMP3 and LCN8 is higher than the threshold, and / or the expression level of VSIG4 is lower than the threshold, the sample to be tested is a sample without congenital heart disease and intellectual disability.
9. An electronic device for predicting congenital heart disease combined with intellectual disability based on gene prediction. It is characterized in that The device includes a memory and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by one or more processors to implement the instructions of the prediction method for congenital heart disease combined with intellectual disability based on gene prediction as described in any one of claims 1-5.
10. A non-transitory computer-readable storage medium, It is characterized in that When the instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the prediction method for congenital heart disease combined with intellectual disability based on gene prediction according to any one of claims 1 to 5.
