A biomarker combination and its use in predicting parkinson's disease

Abstract

The application discloses a biomarker combination and application thereof in predicting Parkinson's disease, and the biomarker combination is composed of ALCAM, IGFALS, SELPLG, ITIH1, ELN, ITIH2, LBP, GPX3, LY6H, SERPINC1, WSCD1, SERPINA6, GPLD1, PON1, GNS, SPON2, CTSL, CLEC3B, OLFM1, CD55, ORM1, BSG and GAPDH. The 23-protein biomarker combination in the application can be used for risk estimation and detection of Parkinson's disease patients, has the advantages of high sensitivity and high specificity, and provides favorable technical support for early screening diagnosis, intervention treatment and the like of Parkinson's disease patients.

Description

Technical Field

[0001] This invention belongs to the fields of biomedical technology and diagnostics, specifically relating to a combination of biomarkers and their application in predicting Parkinson's disease. Background Technology

[0002] Parkinson's disease (PD), also known as "tremor paralysis," is the second most common neurodegenerative disease after Alzheimer's disease, affecting millions of people worldwide. Its incidence increases significantly with age; recent surveys indicate that the prevalence of PD in people over 65 years of age in my country is as high as 1.7%. Medical research suggests that the main clinical features of PD are the loss of α-synuclein aggregates and dopamine neurons. Symptoms of PD vary from person to person, but are mostly motor symptoms (such as bradykinesia, tremor, and rigidity) and non-motor symptoms (such as constipation, depression, loss of smell, and sleep disorders).

[0003] Although Parkinson disease (PD) was first reported by the British physician James Parkinson in 1817, the molecular events that trigger PD are not fully understood, meaning that current treatments do not slow or reverse disease progression. Currently, the most accurate clinical diagnostic tools for PD rely primarily on molecular imaging (such as magnetic resonance diffusion tensor imaging and magnetic susceptibility-weighted imaging) and relevant clinical scales (such as the UPDRS scale and the Hoehn-Yahr scale). However, non-motor symptoms associated with PD may appear more than ten years earlier than motor symptoms, potentially missing the optimal window for rehabilitation treatment. Therefore, the discovery of novel biomarkers for early diagnosis of PD is particularly important.

[0004] Biomarkers are biochemical indicators that can mark changes or potential changes in the structure or function of systems, organs, tissues, cells, and subcellular structures. The human body contains a variety of biomarkers, and with the continuous advancement of molecular biology techniques, the types of biomarkers are increasing, including biomarkers of basic physical conditions, imaging data (e.g., mammograms), specific molecules (e.g., prostate-specific antigen PSA), gene mutations (e.g., BRCA mutations), cellular markers (e.g., circulating tumor cells), and biomarkers of proteins or metabolites. Currently, a major limitation of PD treatment is the inability to make a timely diagnosis. Therefore, the urgent need for highly sensitive and specific biomarkers for early diagnosis and detection of disease progression is particularly important. Summary of the Invention

[0005] In order to solve the above-mentioned technical problems, the present invention provides a combination of biomarkers and their application in predicting Parkinson's disease.

[0006] The first aspect of the present invention provides the use of a combination of biomarkers in the preparation of products for predicting or diagnosing Parkinson's disease, wherein the combination of biomarkers comprises ALCAM, IGFALS, SELPLG, ITIH1, ELN, ITIH2, LBP, GPX3, LY6H, SERPINC1, WSCD1, SERPINA6, GPLD1, PON1, GNS, SPON2, CTSL, CLEC3B, OLFM1, CD55, ORM1, BSG, and GAPDH.

[0007] A second aspect of the present invention provides a reagent for detecting a combination of biomarkers, the combination of which comprises ALCAM, IGFALS, SELPLG, ITIH1, ELN, ITIH2, LBP, GPX3, LY6H, SERPINC1, WSCD1, SERPINA6, GPLD1, PON1, GNS, SPON2, CTSL, CLEC3B, OLFM1, CD55, ORM1, BSG, and GAPDH.

[0008] In a preferred embodiment, the reagent is used to detect the expression level of the biomarker combination.

[0009] In a preferred embodiment, the reagent is a reagent that specifically binds to the biomarker, or a biomolecular reagent that specifically hybridizes with the nucleic acid encoding the biomarker.

[0010] In a preferred embodiment, the reagent is a reagent used for genome, transcriptome and / or proteome sequencing.

[0011] In a preferred embodiment, the expression level is the protein expression level and / or the mRNA transcription level, and / or the biomolecular reagent is selected from one or more primers, probes, and antibodies.

[0012] A third aspect of the present invention provides a biomarker combination comprising ALCAM, IGFALS, SELPLG, ITIH1, ELN, ITIH2, LBP, GPX3, LY6H, SERPINC1, WSCD1, SERPINA6, GPLD1, PON1, GNS, SPON2, CTSL, CLEC3B, OLFM1, CD55, ORM1, BSG, and GAPDH.

[0013] A fourth aspect of the present invention provides a kit comprising the reagents as described in the second aspect of the present invention and / or a combination of biomarkers as described in the third aspect of the present invention.

[0014] The fifth aspect of the present invention provides a method for constructing a Parkinson's disease prediction model, the method comprising: inputting protein expression data corresponding to biomarker combinations in a sample into a Caret R package containing a logistic regression model for machine learning to obtain a Parkinson's disease prediction model;

[0015] The biomarker combination consists of ALCAM, IGFALS, SELPLG, ITIH1, ELN, ITIH2, LBP, GPX3, LY6H, SERPINC1, WSCD1, SERPINA6, GPLD1, PON1, GNS, SPON2, CTSL, CLEC3B, OLFM1, CD55, ORM1, BSG, and GAPDH.

[0016] Specifically, the code for the logistic regression model is as follows:

[0017] The parameters of the generalized linear regression model are as follows: a backward regression method is used to screen biomarkers, and the model is trained using the `train` function of the R package `Caret`, and predicted using the `predict` function. Preferably, the R package for the generalized linear regression model includes: `predict.model = train(formula, data = train_data, method = "glm", family = 'binomial')` (formula: model formula, the combination of input molecules; `train_data`: training set); prediction code: `predict(predict.model, test_data)` (predict.model: the prediction model obtained from the training set; `test_data`: test set or external validation set).

[0018] The determination of the sample is based on logistic regression analysis. If the probability of the expression level of the biomarker combination is greater than 0.5, it is determined to be a positive sample; otherwise, it is a negative sample.

[0019] The code for the Caret R package described in this invention is open source.

[0020] In a preferred embodiment, the sample includes, but is not limited to, exosomes from Parkinson's disease patients, such as saliva, blood (serum or plasma), urine, and cerebrospinal fluid.

[0021] Preferably, the sample to be tested is the cerebrospinal fluid of a Parkinson's disease patient.

[0022] In a preferred embodiment, before machine learning is performed, the protein expression data of the sample is collected by DDA method and peptide matching is performed by Firmiana software.

[0023] In a preferred embodiment, the sample includes Parkinson's disease patients and healthy individuals.

[0024] In a preferred embodiment, the protein expression data input to the logistic regression model satisfies the following: the protein expression level corresponding to the combination of biomarkers in the sample is 1.5 times or more than 1.5 times the corresponding protein expression level in healthy individuals, and the p-value of the t-test is less than 0.05.

[0025] In a preferred embodiment, the peptide matching utilizes the UniProt human protein database.

[0026] In a preferred embodiment, before performing machine learning, the samples are grouped to obtain a modeling group sample and a validation group sample.

[0027] The modeling group samples are used to construct a Parkinson's disease prediction model, and the validation group samples are used to validate the Parkinson's disease prediction model.

[0028] In a preferred implementation, the protein expression data input into the logistic regression model is protein frequency greater than or equal to 30%.

[0029] In a preferred embodiment, the step of using validation group samples for validation includes: calculating the area under the curve, sensitivity, and specificity of the specificity curve of the protein expression data of the biomarker combination in the sample; and judging the accuracy of the prediction model based on the area under the curve, sensitivity, and specificity.

[0030] In a preferred embodiment, the method further includes determining whether the sample has Parkinson's disease. If the probability is greater than or equal to 0.5, the sample is determined to have Parkinson's disease; if the probability is less than 0.5, the sample is determined not to have Parkinson's disease.

[0031] In a preferred embodiment of the present invention, the protein expression data is obtained by LC-MS technology and collected using the DDA (data-dependent acquisition) detection method.

[0032] Preferably, the data acquired by the DDA detection method is subjected to peptide matching using Firmiana software. More preferably, the peptide matching database is the UniProt human protein database.

[0033] More preferably, the protein expression data processed by Firmiana are used: protein quantification is performed using a label-free intensity-based absolute quantification (iBAQ) method, the FOT (Fraction of total, defined as the iBAQ of the protein divided by the total iBAQ of all identified proteins in the sample) is calculated for each protein, and the FOT of each protein is used as protein expression data to input into the logistic regression model.

[0034] The sixth aspect of the present invention provides a predictive model for Parkinson's disease, which is constructed by the method described in the fifth aspect of the present invention.

[0035] The seventh aspect of the present invention provides a prediction system for Parkinson's disease, the prediction system including an analysis and judgment module, the analysis and judgment module containing a prediction model as described in the sixth aspect of the present invention, for outputting a prediction result of whether a sample has Parkinson's disease;

[0036] The biomarker combination consists of ALCAM, IGFALS, SELPLG, ITIH1, ELN, ITIH2, LBP, GPX3, LY6H, SERPINC1, WSCD1, SERPINA6, GPLD1, PON1, GNS, SPON2, CTSL, CLEC3B, OLFM1, CD55, ORM1, BSG, and GAPDH.

[0037] In a preferred embodiment, the detection module detects the protein expression level corresponding to the biomarker combination in the sample to be tested, and transmits the expression level data to the analysis and judgment module.

[0038] The eighth aspect of the present invention provides a method for predicting Parkinson's disease, which involves predicting a sample using reagents as described in the second aspect of the present invention, combinations of biomarkers as described in the third aspect of the present invention, kits as described in the fourth aspect of the present invention, prediction models as described in the sixth aspect of the present invention, or prediction systems as described in the seventh aspect of the present invention.

[0039] The ninth aspect of the present invention provides the application of a reagent as described in the second aspect of the present invention, a combination of biomarkers as described in the third aspect of the present invention, a kit as described in the fourth aspect of the present invention, a prediction model as described in the sixth aspect of the present invention, or a prediction system as described in the seventh aspect of the present invention in the prediction of Parkinson's disease.

[0040] The tenth aspect of the present invention provides the use of the reagent as described in the second aspect of the present invention in the preparation of a kit for predicting or diagnosing Parkinson's disease; wherein the biomarker combination comprises ALCAM, IGFALS, SELPLG, ITIH1, ELN, ITIH2, LBP, GPX3, LY6H, SERPINC1, WSCD1, SERPINA6, GPLD1, PON1, GNS, SPON2, CTSL, CLEC3B, OLFM1, CD55, ORM1, BSG, and GAPDH.

[0041] The eleventh aspect of the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, can implement the steps of the method described in the eighth aspect of the present invention, or implement the functions of the prediction model described in the sixth aspect of the present invention or the prediction system described in the seventh aspect of the present invention.

[0042] The twelfth aspect of the present invention provides an electronic device including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to perform the steps of the method as described in the eighth aspect of the present invention, or to perform the functions of a prediction model as described in the sixth aspect of the present invention or a prediction system as described in the seventh aspect of the present invention.

[0043] Based on common knowledge in the field, the above-mentioned preferred conditions can be combined arbitrarily to obtain various preferred embodiments of the present invention.

[0044] The reagents and raw materials used in this invention are all commercially available.

[0045] The positive and progressive effects of this invention are as follows:

[0046] Experiments revealed significant changes in the expression levels of the 23 protein biomarkers provided in this invention in the cerebrospinal fluid of Parkinson's disease patients. Therefore, these protein biomarkers can be used for risk prediction and detection in Parkinson's disease patients, exhibiting high sensitivity and specificity, providing valuable technical support for early screening, diagnosis, and intervention. Developing corresponding auxiliary early diagnostic kits based on the cerebrospinal fluid protein biomarkers of Parkinson's disease patients has broad research value and provides significant convenience for early clinical diagnosis and intervention. Attached Figure Description

[0047] Figure 1 ROC curves for 23 protein molecular biomarkers combined in the training set.

[0048] Figure 2The confusion matrix results for 23 protein biomarkers conjugated in the training set include prediction accuracy (100.00% for positive and 100.00% for negative), sensitivity (100.00%), and specificity (100.00%).

[0049] Figure 3 ROC curves for 23 protein molecular biomarkers combined in the test set.

[0050] Figure 4 The results of the confusion matrix for 23 protein biomarkers combined in the test set include prediction accuracy (85.71% for positive and 92.00% for negative), sensitivity (75.00%), and specificity (85.71%).

[0051] Figure 5 ROC curves for 23 protein molecular biomarkers used in combination on an external validation set.

[0052] Figure 6 The confusion matrix results for 23 protein biomarkers conjugated in an external validation set include prediction accuracy (100.00% for positive and 100.00% for negative), sensitivity (100.00%), and specificity (100.00%).

[0053] Figure 7 A schematic diagram of the structure of a system for predicting the risk of Parkinson's disease.

[0054] Figure 8 This is a schematic diagram of the electronic device. Detailed Implementation

[0055] The present invention is further illustrated below by way of embodiments, but the invention is not limited to the scope of the embodiments described herein. Experimental methods in the following embodiments that do not specify specific conditions were performed according to conventional methods and conditions, or as selected according to the product instructions.

[0056] The study included 86 Parkinson's disease samples and 34 healthy control samples. The design and implementation of this study were approved and supervised through an ethical vote. Written informed consent was obtained from all patients.

[0057] Example 1

[0058] 1. Cerebrospinal fluid sample pretreatment

[0059] Add 100 μL of 50 mM ammonium bicarbonate to a 30 μL cerebrospinal fluid sample, vortex for 1 min, and incubate at 95 °C for 4 min to denature the proteins. After cooling to room temperature, add 2 μg of trypsin to the system and shake at 37 °C for 18 h. Then add 10 μL of ammonia to stop the enzymatic digestion. Desalt the digested peptide samples, dry them, and freeze them at -80 °C until mass spectrometry analysis.

[0060] 2. Mass spectrometry detection of cerebrospinal fluid samples

[0061] The peptide sample was detected using an Orbitrap Fusion-Lumos triple-resolution mass spectrometry system (ThermoFisher Scientific, Rockford, USA) tandem high-performance liquid chromatography system (EASY-nLC1200, ThermoFisher), and the mass spectrometry data of the corresponding whole protein were obtained. The specific procedures were as follows:

[0062] Nanofluid chromatography was used, with a self-made C18 column (150 μm ID × 8 cm, 1.9 μm / m²). (Packaging material). Column oven temperature: 60℃. The dry powdered peptide was reconstituted with loading buffer (0.1% formic acid aqueous solution), loaded, and separated by a chromatographic column. Elution was performed with a linear 6-30% mobile phase B (ACN and 0.1% formic acid) at 600 nL / min. Detection was achieved using a 150-min gradient HPLC combined with data-dependent acquisition (DDA) mass spectrometry. DDA mass spectrometry parameters were set as follows: ion mode: positive ion; primary mass spectrometry resolution: 120 K, maximum injection time: 80 ms, AGCTarget: 3e6, scan range: 300-1400 m / z; secondary mass spectrometry resolution: 7.5 K, maximum injection time: 20 ms, AGCTarget: 5e4, collision energy: 27%. Data acquisition was controlled using Xcalibur software on the HPLC-tandem mass spectrometry system.

[0063] 3. Data Analysis

[0064] All data were processed using Firmiana to search the UniProt human protein database (updated July 4, 2013, 32,015 entries). Firmiana is a workflow based on the Galaxy system, consisting of multiple functional modules including a user login interface, raw data, identification and quantification, data analysis, and knowledge mining. The mass difference of the precursor ion was 20 ppm, and the mass difference of the daughter ion was 50 mmu. A maximum of two missed cleavage sites were allowed. The search engine set cysteine ​​carbamoyl methylation as a fixed modification and methionine N-acetylation and oxidation as variable modifications. The precursor ion charge range was set to +2, +3, and +4. The false detection rate (FDR) was set to 1%.

[0065] The identified peptide quantification results were recorded as the average peak area of ​​chromatographic fragment ions from all reference libraries. Protein quantification was performed using a label-free, intensity-based absolute quantification (iBAQ) method. Peak area values ​​were calculated as a subset of the corresponding proteins. The total score (FOT) was used to represent the normalized abundance of a specific protein in the sample. FOT was defined as the protein's iBAQ divided by the total iBAQ of all identified proteins in the sample. Proteins with at least one unique peptide and a 1% FDR were selected for further analysis. The FOT of each protein was calculated, and the FOT of each protein was used as protein expression data in a generalized linear regression model.

[0066] 4. Establish a prediction model

[0067] Thirty-five positive cases and 17 negative cases were used as the training set. First, 829 widely distributed proteins were screened based on a frequency > 30%. Then, molecules showing significant differences in expression (FOT fold greater than 1.5 and t-test p-value less than 0.05) were selected between samples from Parkinson's disease patients and healthy samples, resulting in 116 proteins being chosen as candidate biomarkers.

[0068] Based on a regression classifier and logistic regression analysis, the FOT values ​​of each candidate biomarker were input into the Caret R package to build a predictive model (FOT values ​​for the training and test sets are shown in Tables 1-2, and FOT values ​​for the external validation set are shown in Tables 3-4). According to the Caret R package code, the following 23 protein molecular biomarkers were selected as the optimal biomarker combination for predicting Parkinson's disease: ALCAM, IGFALS, SELPLG, ITIH1, ELN, ITIH2, LBP, GPX3, LY6H, SERPINC1, WSCD1, SERPINA6, GPLD1, PON1, GNS, SPON2, CTSL, CLEC3B, OLFM1, CD55, ORM1, BSG, and GAPDH.

[0069] The parameters of the generalized linear regression model are as follows: a backward regression method is used to screen biomarkers, and the model is trained using the `train` function of the R package `Caret`, and predicted using the `predict` function. Preferably, the R package for the generalized linear regression model includes: `predict.model = train(formula, data = train_data, method = "glm", family = 'binomial')` (formula: model formula, the combination of input molecules; `train_data`: training set); prediction code: `predict(predict.model, test_data)` (predict.model: the prediction model obtained from the training set; `test_data`: test set or external validation set).

[0070] The determination of the sample is based on logistic regression analysis. If the probability of the expression level of the biomarker combination is greater than 0.5, it is determined to be a positive sample (positive case); otherwise, it is a negative sample (negative case).

[0071] Table 1. FOT values ​​of 23 protein biomarkers in the training and test sets.

[0072]

[0073]

[0074] Table 2. FOT values ​​of 23 protein biomarkers in the training and test sets.

[0075]

[0076]

[0077]

[0078] Table 3. FOT values ​​of 23 protein biomarkers in the external validation set.

[0079]

[0080]

[0081] Table 4. FOT values ​​of 23 protein biomarkers in the external validation set.

[0082]

[0083] Twenty-three protein molecular markers (ALCAM, IGFALS, SELPLG, ITIH1, ELN, ITIH2, LBP, GPX3, LY6H, SERPINC1, WSCD1, SERPINA6, GPLD1, PON1, GNS, SPON2, CTSL, CLEC3B, OLFM1, CD55, ORM1, BSG, and GAPDH) were screened from cerebrospinal fluid samples of Parkinson's disease patients at Huashan Hospital affiliated with Fudan University. ROC curves (Receiver Operating Curves) were plotted based on their relative expression levels, and a model for analyzing these markers was established. ROC curves (Receiver Operating Curves) were plotted for the relative protein expression levels of these 23 markers, and the AUC (Area Under the ROC Curve) was calculated. The training set included 35 positive cases and 17 negative cases, with an AUC of 1.000 (see [link to training set]). Figure 1 ).Depend on Figure 2 The diagnostic sensitivity was 100.00%, specificity was 100.00%, positive accuracy was 100.00%, and negative accuracy was 100.00%. The test set included 25 positive cases and 7 negative cases, with an AUC of 0.894 (see...). Figure 3 ).Depend on Figure 4 The diagnostic sensitivity was 75.00%, specificity 85.71%, positive accuracy 85.71%, and negative accuracy 92.00%. The external validation set included 26 positive cases and 10 negative cases, with an AUC of 1.000 (see...). Figure 5 ).Depend on Figure 6 It can be seen that the diagnostic sensitivity is 100.00%, the specificity is 100.00%, the positive accuracy is 100.00%, and the negative accuracy is 100.00%. For the analysis method, please refer to Karimollah Hajian-Tilaki, Receiver Operating Characteristic (ROC) Curve Analysis for Medical Diagnostic Test Evaluation, Caspian J Intern Med 2013; 4(2):627-635.

[0084] The results above indicate that the combined use of 23 protein molecular markers (ALCAM, IGFALS, SELPLG, ITIH1, ELN, ITIH2, LBP, GPX3, LY6H, SERPINC1, WSCD1, SERPINA6, GPLD1, PON1, GNS, SPON2, CTSL, CLEC3B, OLFM1, CD55, ORM1, BSG, and GAPDH) in the cerebrospinal fluid of Parkinson's disease patients can be used for early screening and diagnosis of Parkinson's disease.

[0085] To determine whether a sample is a patient with Parkinson's disease, the expression level (FOT) of the protein molecular biomarker collected by DDA as described above is input into the obtained prediction model to obtain the output result of whether it is Parkinson's disease.

[0086] Example 2: System for predicting the risk of Parkinson's disease

[0087] System 61 for predicting Parkinson's disease risk: data processing module 52 and judgment and output module 53, and also includes data collection module 51. Figure 7 ).

[0088] The data collection module 51 is used to collect the expression level data of the combination of biomarkers in the tissue sample of the patient with Parkinson's disease and transmit it to the data processing module.

[0089] The data processing module 52 is used to analyze the expression level data of the received or input biomarker combination according to the data analysis method described in Example 1 to obtain calculation results. The expression level data of the biomarker combination can be collected by the data collection module 51, or it can be obtained from other sources.

[0090] The judgment and output module 53 is used to judge whether the calculation result meets the preset judgment condition, that is, the probability of having Parkinson's disease is greater than or equal to the predicted probability of not having Parkinson's disease, so as to predict the risk of Parkinson's disease and output the prediction result; wherein, in the judgment and output module, when the expression level data meets the judgment condition that the probability of having Parkinson's disease is greater than or equal to the predicted probability of not having Parkinson's disease, the output prediction result is "has a risk of having Parkinson's disease"; when the expression level data does not meet the judgment condition that the probability of having Parkinson's disease is less than the predicted probability of not having Parkinson's disease, the output prediction result is "does not have a risk of having Parkinson's disease".

[0091] Example 3 Electronic device

[0092] This embodiment provides an electronic device, which can be represented in the form of a computing device (e.g., a server device), including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, it can implement the method for predicting the risk of Parkinson's disease in Embodiment 1 of the present invention.

[0093] Figure 8 The hardware structure diagram of this embodiment is shown. The electronic device 4 specifically includes:

[0094] At least one processor 91, at least one memory 92, and a bus 93 for connecting different system components (including processor 91 and memory 92), wherein:

[0095] Bus 93 includes a data bus, an address bus, and a control bus.

[0096] The memory 92 includes volatile memory, such as random access memory (RAM) 921 and / or cache memory 922, and may further include read-only memory (ROM) 923.

[0097] The memory 92 also includes a program / utility 925 having a set (at least one) of program modules 924, including but not limited to: an operating system, one or more application programs, other program modules, and program data, each or some combination of these examples may include an implementation of a network environment.

[0098] The processor 91 executes various functional applications and data processing, such as the data analysis method of Embodiment 1 of the present invention, by running computer programs stored in the memory 92.

[0099] Electronic device 9 can further communicate with one or more external devices 94 (e.g., keyboard, pointing device, etc.). This communication can be performed via input / output (I / O) interface 95. Furthermore, electronic device 9 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public network, such as the Internet) via network adapter 96. Network adapter 96 communicates with other modules of electronic device 9 via bus 93. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with electronic device 9, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (disk array) systems, tape drives, and data backup storage systems, etc.

[0100] It should be noted that although several units / modules or sub-units / modules of the electronic device have been mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to the embodiments of this application, the features and functions of two or more units / modules described above can be embodied in one unit / module. Conversely, the features and functions of one unit / module described above can be further divided and embodied by multiple units / modules.

[0101] Example 4: Computer-readable storage medium

[0102] This invention provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the method for predicting the risk of Parkinson's disease in Embodiment 1 of this invention.

[0103] The readable storage medium may be more specifically adopted, including but not limited to: portable disk, hard disk, random access memory, read-only memory, erasable programmable read-only memory, optical storage device, magnetic storage device, or any suitable combination thereof.

[0104] In a possible implementation, the present invention can also be implemented as a program product comprising program code, which, when the program product is run on a terminal device, is used to cause the terminal device to perform the steps of the method for predicting the risk of Parkinson's disease as described in Embodiment 1 of the present invention.

[0105] The program code for executing the present invention can be written in any combination of one or more programming languages. The program code can be executed entirely on the user device, partially on the user device, as a standalone software package, partially on the user device and partially on a remote device, or entirely on a remote device.

[0106] Finally, the above specific implementation methods are only used to illustrate the technical solution of the present invention, and are not intended to limit it.

[0107] Full names of biomarkers (refer to the GeneCards database)

[0108] ALCAM: Activated Leukocyte Cell Adhesion Molecule

[0109] IGFALS: Insulin Like Growth Factor Binding Protein Acid Labile SubunitSELPLG: Selectin P Ligand

[0110] ITIH1:Inter-Alpha-Trypsin Inhibitor Heavy Chain 1

[0111] ELN:Elastin

[0112] ITIH2:Inter-Alpha-Trypsin Inhibitor Heavy Chain 2

[0113] LBP:Lipopolysaccharide Binding Protein

[0114] GPX3:Glutathione Peroxidase 3

[0115] LY6H:Lymphocyte Antigen 6Family Member H

[0116] SERPINC1:Serpin Family C Member 1

[0117] WSCD1:WSC Domain Containing 1

[0118] SERPINA6:Serpin Family A Member 6

[0119] GPLD1:Glycosylphosphatidylinositol Specific Phospholipase D1

[0120] PON1:Paraoxonase 1

[0121] GNS:Glucosamine(N-Acetyl)-6-Sulfatase

[0122] SPON2:Spondin 2

[0123] CTSL:Cathepsin L

[0124] CLEC3B:C-Type Lectin Domain Family 3Member B OLFM1:Olfactomedin 1

[0125] CD55:CD55 Molecule(Cromer Blood Group)

[0126] ORM1:Orosomucoid 1

[0127] BSG:Basigin(Ok Blood Group)

[0128] GAPDH:Glyceraldehyde-3-Phosphate Dehydrogenase。

Claims

1. A combination of biomarkers, characterized in that, The biomarker combination consists of ALCAM, IGFALS, SELPLG, ITIH1, ELN, ITIH2, LBP, GPX3, LY6H, SERPINC1, WSCD1, SERPINA6, GPLD1, PON1, GNS, SPON2, CTSL, CLEC3B, OLFM1, CD55, ORM1, BSG, and GAPDH.

2. A reagent kit, characterized in that, The kit contains the combination of biomarkers as described in claim 1.

3. The use of the biomarker combination as described in claim 1 or the kit as described in claim 2 in the preparation of products for predicting or diagnosing Parkinson's disease.

4. A method for constructing a Parkinson's disease prediction model, characterized in that, The method includes: inputting protein expression data corresponding to biomarker combinations from patient samples into the Caret R package containing a logistic regression model for machine learning to obtain a Parkinson's disease prediction model; The biomarker combination consists of ALCAM, IGFALS, SELPLG, ITIH1, ELN, ITIH2, LBP, GPX3, LY6H, SERPINC1, WSCD1, SERPINA6, GPLD1, PON1, GNS, SPON2, CTSL, CLEC3B, OLFM1, CD55, ORM1, BSG, and GAPDH.

5. The method as described in claim 4, characterized in that, The samples are derived from plasma, urine, saliva, or cerebrospinal fluid; and / or, before machine learning, the protein expression data of the samples are collected via DDA and peptide matching is performed using Firmiana software; and / or, the samples include Parkinson's disease patients and healthy individuals; and / or, the protein expression data input into the logistic regression model satisfies the following: the protein expression level corresponding to the biomarker combination in the sample is 1.5 or greater than the corresponding protein expression level in healthy individuals, and the p-value of the t-test is less than 0.

05.

6. The method as described in claim 5, characterized in that, The peptide matching utilizes the UniProt human protein database; and / or, before machine learning, the samples are grouped to obtain training set samples, test set samples, and external validation set samples. The training set samples are used to construct the Parkinson's disease prediction model, and the test set samples and external validation set samples are used to validate the Parkinson's disease prediction model; and / or, the protein expression data input to the logistic regression model is a protein frequency greater than or equal to 30%. And / or, the step of validation using an external validation set sample includes: calculating the area under the curve, sensitivity, and specificity of the specificity curve for the protein expression data of the biomarker combination in the sample; The accuracy of the prediction model is then judged based on the offline area, sensitivity, and specificity.

7. A predictive system for Parkinson's disease, characterized in that, The prediction system includes an analysis and judgment module; the analysis and judgment module contains a Parkinson's disease prediction model constructed by the method described in any one of claims 4-6, used to output a prediction result of whether a sample has Parkinson's disease; The biomarker combination consists of ALCAM, IGFALS, SELPLG, ITIH1, ELN, ITIH2, LBP, GPX3, LY6H, SERPINC1, WSCD1, SERPINA6, GPLD1, PON1, GNS, SPON2, CTSL, CLEC3B, OLFM1, CD55, ORM1, BSG, and GAPDH.

8. The prediction system as described in claim 7, characterized in that, The prediction system also includes a detection module, which detects the protein expression levels corresponding to the biomarker combinations in the sample to be tested and transmits the expression level data to the analysis and judgment module.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it can realize the function of the Parkinson's disease prediction model constructed by the method of any one of claims 4-6 or the prediction system of claim 7 or 8.

10. An electronic device comprising a memory and a processor, the memory storing a computer program, characterized in that, The processor is used to execute the computer program to implement the functions of the Parkinson's disease prediction model constructed by the method of any one of claims 4-6 or the prediction system of claim 7 or 8.

Images