Method for screening Alzheimer's disease markers by bioinformatics technology and application of Alzheimer's disease markers

Abstract

The invention relates to a method for screening Alzheimer's disease markers by using a bioinformatics technology and application of the Alzheimer's disease markers. Serum microRNA samples of Alzheimer's disease normal subjects (NC), mild cognitive impairment (MCI) and Alzheimer's disease (AD) patients in an NCBI public database are analyzed, and a bioinformatics analysis method is adopted to systematically identify common miRNA which commonly presents differential expression at the early stage and the late stage of the disease, so that target mRNA of the common miRNA is predicted, and a regulatory network is constructed; and finally screening out specific miRNA closely associated with the key mRNA as a disease monitoring marker. The method breaks through the high-cost limitation of traditional dependence on high-throughput sequencing, realizes efficient and reliable screening and identification of disease molecular markers by deeply mining public data resources, provides a new molecular target and detection means for early diagnosis, illness monitoring and intervention treatment of Alzheimer's disease, and has a wide application prospect. The important clinical application value is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of bioinformatics, and particularly relates to a method for screening Alzheimer's disease markers by bioinformatics and application thereof. BACKGROUND

[0002] Alzheimer's Disease (AD) is a complex nervous system disease characterized by progressive cognitive impairment and neurodegeneration, and is the main cause of senile dementia. Its incidence increases significantly with age. According to reports of authoritative agencies such as the International Alzheimer's Association, the number of AD patients worldwide has reached the order of tens of millions, and is showing a rapid growth trend, which has brought a heavy burden to global public health and social economy.

[0003] At present, the clinical diagnosis and treatment of AD faces two core problems: early diagnosis difficulty and unknown disease mechanism. The current clinical diagnosis of AD mainly relies on neuropsychological assessment (such as MMSE scale), imaging examination (such as PET-CT, MRI) and cerebrospinal fluid biomarker detection (such as Aβ42 / Aβ40 ratio, tau protein). However, these methods have obvious limitations: neuropsychological assessment is highly subjective and difficult to effectively distinguish mild cognitive impairment (MCI) from normal aging; imaging examination is costly and has radiation exposure risk, and is not sensitive to early microscopic pathological changes; cerebrospinal fluid detection is an invasive operation with poor patient compliance, and is not suitable for large-scale population screening and early diagnosis. Therefore, the development of non-invasive, efficient and low-cost biomarkers for early diagnosis has become an urgent need in the field of AD research.

[0004] MicroRNA (miRNA) is a class of endogenous non-coding small RNA molecules that precisely regulate gene expression at the post-transcriptional level. Circulating serum miRNAs have the characteristics of high stability and easy access through peripheral blood, and are considered as a very promising source of non-invasive diagnostic markers for diseases. In recent years, high-throughput sequencing technology has been applied to discover AD-related differentially expressed miRNAs. However, this method itself is costly, and due to significant sample heterogeneity, batch effects and non-uniform data analysis processes among different research cohorts, the repeatability of research results is poor, and it is difficult to obtain high-confidence markers with universal applicability, which greatly limits its clinical transformation and application.

[0005] On the other hand, with the explosive growth of omics data, public databases such as GEO of NCBI have stored a large amount of AD-related serum miRNA expression profile data. These data are a "gold mine" that has not been fully tapped, containing important information to reveal key molecular events of AD. However, there is currently a lack of a systematic bioinformatics method that can integrate, deeply mine and cross-verify these heterogeneous public data to overcome the inherent defects of traditional high-throughput sequencing, thereby achieving the goal of screening AD molecular markers with high confidence at low cost and high efficiency.

[0006] Therefore, there is an urgent need in the art for a new method that can effectively overcome the high cost and low repeatability of the prior art, and can systematically and reliably screen and identify AD molecular markers with clinical value from public data resources. SUMMARY

[0007] The purpose of the application is to solve the problems of high cost, low efficiency and poor repeatability of AD marker screening in the prior art, and to provide a method for screening AD markers using bioinformatics technology. The method realizes efficient and reliable marker screening by deeply mining NCBI public database resources, integrating samples of multiple disease stages (NC, MCI, AD) for analysis, and coupling multi-level bioinformatics strategies such as differential expression screening, protein-protein interaction network analysis and hub gene identification, thereby providing new molecular targets for early diagnosis, disease monitoring and intervention treatment of AD.

[0008] To achieve the above-mentioned purpose, the application adopts the following technical solutions:

[0009] (1) Data acquisition: obtaining serum microRNA expression profile datasets containing normal subjects (NC) group, mild cognitive impairment (MCI) group and Alzheimer's disease (AD) patient group from NCBI public database;

[0010] (2) Differential expression miRNA screening: using GEO2R tool to standardize and analyze the differences of the data obtained in step (1), comparing the miRNA expression levels of MCI group and NC group, AD group and NC group respectively, and screening out the common miRNA set that presents significant differential expression in both comparison groups.

[0011] (3) Target mRNA prediction and PPI network construction: performing target mRNA prediction on the common miRNA obtained in step (2) through multiple miRNA target prediction databases, taking the intersection of the prediction results of each database as candidate target mRNA; introducing the candidate target mRNA into a protein-protein interaction (PPI) analysis database to construct a PPI regulatory network;

[0012] (4) Core module screening: using MCODE algorithm to mine the core module of the PPI network constructed in step (3), and screening the core module with the highest score;

[0013] (5) Hub mRNA screening: using Cytoscape to visualize the PPI network, and screening multiple hub mRNAs with high connectivity from the PPI network of step (3);

[0014] (6) miRNA-mRNA interaction network construction and marker determination: based on the targeting prediction relationship between the common miRNA of step (2) and the hub mRNA of step (5), constructing a miRNA-mRNA interaction network, and screening microRNAs having a targeting relationship with the hub gene from the network as candidate molecular markers for Alzheimer's disease monitoring.

[0015] The screening standard of differentially expressed miRNA when analyzing the sample data by GEO2R in the application is p - value≤0.05, |log2FC|>1.

[0016] The plurality of miRNA target gene prediction databases in the application include TargetScan, miRDB and miRWalk; wherein the screening conditions of each database are as follows: the miRWalk database adopts P-value≤0.05, the TargetScan database adopts Target Score≥80, and the miRDB database adopts context++score≤-0.1.

[0017] The parameter setting of the MCODE algorithm in the application is: Degree Cutoff=2, Node Score Cutoff=0.2, K-Core=2, Maximum Depth=100.

[0018] The microRNA having a targeting relationship with the hub gene in the application refers to a microRNA having a targeting relationship with at least two hub genes.

[0019] The application discloses a combination of candidate molecular markers for Alzheimer's disease screened by the method.

[0020] An Alzheimer's disease-related molecular marker is one or more miRNAs: hsa-miR-211-Sp, hsa-miR-765, hsa-miR-3649, hsa-miR-5581-5p, and hsa-miR-5698.

[0021] Beneficial effects: Compared with the prior art, the present application has the following advantages:

[0022] (1) High cost-effectiveness: The present application is completely based on public database resources, breaking through the high cost limit of traditional reliance on self-conducted high-throughput sequencing, greatly reducing research costs.

[0023] (2) Strong systematicness and reliability: By integrating differential expression analysis, common screening, target gene prediction, PPI network analysis, module mining and hub gene identification and other multi-level bioinformatics methods, a systematic screening process is formed, and the screened markers have higher biological credibility and clinical transformation potential.

[0024] (3) Focus on disease progression: By simultaneously analyzing samples at NC, MCI and AD stages, and screening "common miRNAs" that are differentially expressed in both mild and severe stages of the disease, the core molecular events driving the occurrence and development of the disease can be better captured, which has unique value for early diagnosis and disease monitoring.

[0025] (4) Clear clinical application prospect: The finally screened miRNA markers provide new molecular tools and targets for early screening of AD, risk prediction of MCI to AD conversion and disease progression monitoring. BRIEF DESCRIPTION OF DRAWINGS

[0026] Figure 1 The screening process and results of differentially expressed miRNAs are shown. Among them, Figure 1 A is the volcano plot of differentially expressed miRNAs in the mild cognitive impairment (MCI) group compared with the normal subjects (NC) group; Figure 1 B is the volcano plot of differentially expressed miRNAs in the Alzheimer's disease (AD) group compared with the NC group; Figure 1 C is a statistical chart of the number of differentially expressed miRNAs in MCI and AD; Figure 1 D is a diagram showing the screening of common differentially expressed miRNAs in MCI and AD by a Venn diagram; Figure 1 E is the set of common differentially expressed miRNAs.

[0027] Figure 2 The results of target gene prediction and database intersection analysis of common differentially expressed miRNAs are shown.

[0028] Figure 3 The protein-protein interaction (PPI) network diagram constructed based on the candidate target mRNA is shown.

[0029] Figure 4 The core functional module mined from the PPI network using the MCODE algorithm is shown.

[0030] Figure 5 The key hub mRNA screened from the PPI network is shown.

[0031] Figure 6 The regulatory interaction network diagram between the common differentially expressed miRNA and the hub gene is shown. DETAILED DESCRIPTION

[0032] The application will be described in detail below with reference to the accompanying drawings and specific embodiments. The following examples are only used to illustrate the application and should not be regarded as limiting the scope of the application.

[0033] I. Analysis method:

[0034] 1. Data acquisition: The serum microRNA expression profile dataset (GSE120584) related to Alzheimer's disease (AD) was screened out by using the gene expression database GEO (https: / / www.ncbi.nlm.nih.gov / gds / ), and the sample data included 30 normal subject (NC) group samples, 30 mild cognitive impairment (MCI) group samples and 30 Alzheimer's disease (AD) patient group samples.

[0035] 2. Differentially expressed miRNA screening: After standardizing the sample data obtained in step 1 by using the GEO2R tool, the screening criteria for differentially expressed miRNA were set as p-value≤0.05 and |log2FC|>1, and the comparisons between the MCI group and the NC group and between the AD group and the NC group were performed, respectively, to screen out the common miRNA that showed significant differential expression in the MCI stage and the AD stage.

[0036] 3. Target mRNA prediction and PPI network construction: The screened miRNA was subjected to target mRNA prediction by using three databases, TargetScan (https: / / www.targetscan.org / vert_72 / ), miRDB (http: / / mirdb.org / ) and miRWalk (http: / / mirwalk.umm.uni-heidelberg.de / ), wherein the screening conditions of each database were as follows: P-value≤0.05 for the miRWalk database, Target Score≥80 for the TargetScan database, and context++ score≤-0.1 for the miRDB database; the intersection of the prediction results of the above three databases was taken as the candidate target mRNA. Subsequently, the candidate target mRNA was subjected to protein-protein interaction analysis by using the STRING database (https: / / cn.string-db.org / ) to construct a PPI regulatory network.

[0037] 4. Core module screening: MCODE algorithm was used to screen the core module of PPI network constructed in step (3), and the parameters of MCODE algorithm were set as: Degree Cutoff = 2, Node Score Cutoff = 0.2, K-Core = 2, Maximum Depth = 100, and the top 5 core modules with the highest scores were screened out;

[0038] 5. Hub mRNA screening: Cytoscape was used to visualize the PPI network, and the top 10 hub mRNAs were screened out from the PPI network of step 3;

[0039] 6. miRNA-mRNA interaction network construction and marker determination: based on the target prediction relationship between the common miRNAs in step 2 and the 10 hub mRNAs in step 5, a miRNA-mRNA interaction network was constructed, and microRNAs that have target relationship with at least two hub genes were screened out from the network as candidate molecular markers for Alzheimer's disease monitoring.

[0040] II. Results analysis:

[0041] 1. Differential expression miRNA screening results: as shown in the attached Figure 1 figure, 8 significantly differentially expressed miRNAs (4 up-regulated and 4 down-regulated) were identified in the comparison between MCI group and NC group; 55 significantly differentially expressed miRNAs (34 up-regulated and 21 down-regulated) were identified in the comparison between AD group and NC group. By taking the intersection of the results of the two comparisons through the Venn diagram, 5 common miRNAs that are differentially expressed in MCI and AD stages were finally determined, namely hsa-miR-211-5p, hsa-miR-765, hsa-miR-3649, hsa-miR-5581-5p and hsa-miR-5698.

[0042] 2. Target gene prediction results: 5 miRNAs were subjected to target gene prediction by using 3 databases (TargetScan, miRWalk and miRDB), and the genes common to the 3 databases were selected as the regulated genes. Finally, hsa-miR-211-5p targeted 186 mRNAs, hsa-miR-765 targeted 262 mRNAs, hsa-miR-3649 targeted 18 mRNAs, hsa-miR-5581-5p targeted 69 mRNAs, and hsa-miR-5698 targeted 190 mRNAs.

[0043] 3. PPI network construction results: as shown in the attached Figure 3The predicted target genes were used for protein-protein interaction analysis by using the STRING database to construct a PPI regulatory network.

[0044] 4. Core module screening: as shown in the accompanying Figure 4 The MCODE algorithm was used to mine the core module of the PPI network constructed in step 3,

[0045] 5. Hub mRNA screening: the PPI network was visualized by using Cytoscape, and the top 10 hub mRNAs with the highest connectivity were screened from the PPI network of step 3, namely CXCL12, HNF4A, HGF, RUNX2, SIRT1, ACTB, PRKACA, PPARGC1A, CREB1 and NR3C1.

[0046] 6. miRNA-mRNA interaction network construction and marker determination: the interaction network diagram of miRNA and hub genes was constructed as shown in the accompanying Figure 6 The microRNAs that have a targeting relationship with at least two hub genes were screened from the network as candidate molecular markers for monitoring Alzheimer's disease, namely hsa-miR-211-5p, hsa-miR-765, hsa-miR-3649, hsa-miR-5581-5p and hsa-miR-5698.

[0047] Conclusion: The present application provides a systematic method for screening AD molecular markers based on bioinformatics, which is efficient and low-cost. Through this method, we successfully identified a group of serum miRNA markers that are continuously abnormally expressed in the AD process and closely related to key hub genes from public data. This marker combination lays a solid foundation for the development of early diagnosis of AD and the exploration of new therapeutic targets, and has great scientific value and clinical application prospect.

[0048] The above examples are only used to understand the method of the present application and its core idea. It should be noted that for those skilled in the art, without departing from the principles of the present application, several improvements and modifications can be made to the present application, and these improvements and modifications will also fall within the protection scope of the claims of the present application.

Claims

1. A method for screening Alzheimer's disease biomarkers using bioinformatics technology, characterized in that, Includes the following steps: (1) Data acquisition: Serum microRNA expression profile datasets containing normal subjects (NC), mild cognitive impairment (MCI) and Alzheimer's disease (AD) patients were obtained from the NCBI public database; (2) Screening of differentially expressed miRNAs: The data obtained in step (1) were standardized and differentially analyzed using the GEO2R tool. The miRNA expression levels of the MCI group and the NC group, and the AD group and the NC group were compared respectively. A set of common miRNAs that showed significant differential expression in both comparison groups was screened out. (3) Target mRNA prediction and PPI network construction: Target mRNA prediction is performed on the common miRNAs obtained in step (2) through multiple miRNA target prediction databases, and the intersection of the prediction results of each database is taken as the candidate target mRNA; The candidate target mRNA is imported into the protein-protein interaction (PPI) analysis database to construct the PPI regulatory network; (4) Core module selection: The MCODE algorithm is used to mine the core modules of the PPI network constructed in step (3) and select the core modules with the highest scores. (5) hub mRNA screening: Cytoscape was used to visualize the PPI network and multiple hub mRNAs with high connectivity were screened from the PPI network in step (3). (6) Construction of miRNA-mRNA interaction network and identification of biomarkers: Based on the target prediction relationship between the common miRNA in step (2) and the hub mRNA in step (5), a miRNA-mRNA interaction network is constructed, and microRNAs that have a target relationship with the hub gene are screened from the network as candidate molecular biomarkers for Alzheimer's disease monitoring.

2. The method according to claim 1, characterized in that, In step (2), when analyzing the sample data using GEO2R, ​​the screening criteria for differentially expressed miRNAs are p-value ≤ 0.05 and |log2FC| > 1.

3. The method according to claim 1, characterized in that, In step (3), the multiple miRNA target gene prediction databases include TargetScan, miRDB and miRWalk; the screening conditions for each database are as follows: the miRWalk database uses P-value ≤ 0.05, the TargetScan database uses Target Score ≥ 80, and the miRDB database uses context++score ≤ -0.

1.

4. The method according to claim 1, characterized in that, In step (4), the parameters of the MCODE algorithm are set as follows: Degree Cutoff = 2, Node Score Cutoff = 0.2, K-Core = 2, Maximum Depth = 100.

5. The method according to claim 1, characterized in that, In step (6), the microRNA that has a targeting relationship with the hub genes refers to a microRNA that has a targeting relationship with at least two of the hub genes.

6. A combination of candidate molecular biomarkers for Alzheimer's disease obtained by screening using the method described in any one of claims 1-5.

7. A molecular marker associated with Alzheimer's disease, comprising one or more of the following miRNAs: hsa-miR-211-5p, hsa-miR-765, hsa-miR-3649, hsa-miR-5581-5p, and hsa-miR-5698.

Images