CN113945724B - Suspense depression risk prediction device and kit and application thereof - Google Patents

Suspense depression risk prediction device and kit and application thereof Download PDF

Info

Publication number
CN113945724B
CN113945724B CN202111273591.6A CN202111273591A CN113945724B CN 113945724 B CN113945724 B CN 113945724B CN 202111273591 A CN202111273591 A CN 202111273591A CN 113945724 B CN113945724 B CN 113945724B
Authority
CN
China
Prior art keywords
depression
subliminal
sample
training
sub
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202111273591.6A
Other languages
Chinese (zh)
Other versions
CN113945724A (en
Inventor
王颖
肖书
陈观茂
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Jinan University
Original Assignee
Jinan University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Jinan University filed Critical Jinan University
Priority to CN202111273591.6A priority Critical patent/CN113945724B/en
Publication of CN113945724A publication Critical patent/CN113945724A/en
Application granted granted Critical
Publication of CN113945724B publication Critical patent/CN113945724B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
    • G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
    • G01N33/68—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving proteins, peptides or amino acids
    • G01N33/6863—Cytokines, i.e. immune system proteins modifying a biological response such as cell growth proliferation or differentiation, e.g. TNF, CNF, GM-CSF, lymphotoxin, MIF or their receptors
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
    • G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
    • G01N33/68—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving proteins, peptides or amino acids
    • G01N33/6863—Cytokines, i.e. immune system proteins modifying a biological response such as cell growth proliferation or differentiation, e.g. TNF, CNF, GM-CSF, lymphotoxin, MIF or their receptors
    • G01N33/6869—Interleukin
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N2800/00—Detection or diagnosis of diseases
    • G01N2800/30—Psychoses; Psychiatry
    • G01N2800/304—Mood disorders, e.g. bipolar, depression

Landscapes

  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Immunology (AREA)
  • Molecular Biology (AREA)
  • Engineering & Computer Science (AREA)
  • Cell Biology (AREA)
  • Chemical & Material Sciences (AREA)
  • Biomedical Technology (AREA)
  • Urology & Nephrology (AREA)
  • Hematology (AREA)
  • Microbiology (AREA)
  • Biotechnology (AREA)
  • Proteomics, Peptides & Aminoacids (AREA)
  • Food Science & Technology (AREA)
  • Medicinal Chemistry (AREA)
  • Physics & Mathematics (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Pathology (AREA)
  • Investigating Or Analysing Biological Materials (AREA)

Abstract

The invention discloses a subliminal depression risk prediction device, a kit and application thereof, and relates to the technical field of biology. The invention provides 8 detection markers capable of effectively predicting subliminal depression, which specifically comprise IL-1 beta, IL-4, IL-10, IL-17, IL-2R alpha, LIF, MIF and SDF-1 alpha, and based on the detection of the expression levels of the markers, the detection markers can effectively distinguish subliminal depression and healthy people, can realize rapid and large-scale screening of the people with subliminal depression, intervene in time on the people and avoid the people from developing into depression.

Description

Suspense depression risk prediction device and kit and application thereof
Technical Field
The invention relates to the technical field of biology, in particular to a subliminal depression risk prediction device, a kit and application thereof.
Background
Subliminal depression (SU), also called subclinical depression (subclinical depression) or minor depression (minor depression), means that two or more depression symptoms appear in an individual, the duration of the depression symptoms is more than two weeks, the diagnosis standard of depression in the fifth edition (DMS-v) of the american statistical diagnostic for mental disorders does not be met, but the mental sub-health state of individual social dysfunction and occupational function impairment is caused, and the state is between the state without depression symptoms and depression. The progression of depression is a continuous process, with subliminal depression as a precursor state of depression, which may continue to progress to depression or be restored by a timely prognosis. Numerous studies and epidemiological data have shown that the incidence of subliminal depression is about 10% to 28% in the population, significantly higher than depression and shows a trend of increasing year by year. Studies have found that the risk of developing depression in sub-threshold depressed populations is significantly higher than in the general population, with up to 26% of the probability of developing depression in 12 months in sub-threshold depressed populations without any intervention. Subliminal depression is an important risk factor for individuals to develop depression, and is therefore crucial for early detection, early diagnosis and early intervention of subliminal depression.
Subthreshold depressed individuals have less and atypical symptoms and less suicidal thoughts and behaviors than depressed patients, and therefore do not attract sufficient social attention. Secondly, the pathogenesis of subliminal depression is not clear at present, which is mainly judged by combining clinical symptoms with a depression symptom assessment scale, and a reliable biological index with clinical significance is lacked, so that higher missed diagnosis is caused. In addition, such conventional diagnostic methods are highly qualified and demanding on the examiner, and are time-consuming and labor-consuming, inconvenient for epidemiological investigations or quick and efficient for screening out subliminal depression subjects in a large population. Therefore, there is a great need to improve the screening efficiency of subliminal depression in an effort to find objective biomarkers for subliminal depression populations.
An increasing number of people are concerned about depression not only in neuropsychiatric diseases, but also in systemic diseases of multiple systems involving the central nervous system, the endocrine system and the immune system, and the systems are related to each other and affect each other. Numerous studies have demonstrated altered levels of serum cytokine expression in depression patients, and changes in cytokine levels may reflect changes in disease progression. Thus, more and more studies on depression have begun to switch from intracerebral to peripheral blood cell protein expression.
Chinese patent application 201110033299.7 discloses a gene expression diagnosis chip for depression with sub-syndrome, which is composed of a solid phase carrier, wherein 46 gene probes are arranged on the solid phase carrier. However, no method for rapidly and accurately screening out the subliminal depression population by detecting serum cytokines exists at present.
In view of this, the invention is particularly proposed.
Disclosure of Invention
The invention aims to provide a subliminal depression risk prediction device, a kit and application thereof.
The invention is realized by the following steps:
in a first aspect, the present invention provides the use of an agent for detecting the expression level of a serum cytokine in the manufacture of a kit for predicting the risk of subliminal depression, wherein the serum cytokine comprises: at least one of IL-1 β, IL-4, IL-10, IL-17, IL-2R α, LIF, MIF, and SDF-1 α. I is
In a second aspect, the present invention provides a method for training a prediction model of subliminal depression risk, comprising: and acquiring a training sample and a labeling result corresponding to the training sample. And inputting the serum cytokine expression level of the training sample into a pre-established subliminal depression risk prediction model to obtain a prediction result of the training sample. Wherein the sub-threshold depression risk prediction model is used for judging the risk of the sample suffering from sub-threshold depression according to the expression level of serum cytokines of the sample, and the serum cytokines comprise: at least one of IL-1 β, IL-4, IL-10, IL-17, IL-2R α, LIF, MIF, and SDF-1 α. Subsequently, a parameter update is performed on the constructed sub-threshold depression risk prediction model based on the annotation result and the prediction result.
In a third aspect, embodiments of the present invention provide a subliminal depression risk prediction apparatus, which includes an obtaining module and a prediction module. The acquisition module is used for acquiring the detection result of the serum cytokine expression level of the sample to be detected. The prediction module is used for inputting the detection result of the serum cytokine expression level of the sample to be detected into a pre-constructed subliminal depression risk prediction model to obtain a prediction result of the sample; wherein the serum cytokines comprise: at least one of IL-1 β, IL-4, IL-10, IL-17, IL-2R α, LIF, MIF, and SDF-1 α; the sub-threshold depression risk prediction model is obtained by training the training method of the sub-threshold depression risk prediction model according to the previous embodiment.
In a fourth aspect, an embodiment of the present invention provides an apparatus for constructing a prediction model of subliminal depression risk, which includes an obtaining module, a processing module, and a parameter updating module. And the acquisition module is used for acquiring the training samples and the labeling results corresponding to the training samples. The processing module is used for inputting the detection result of the serum cytokine expression level of the training sample into a pre-established subliminal depression risk prediction model to obtain the prediction result of the training sample; wherein, the model for predicting the risk of subliminal depression is used for judging the risk of subliminal depression of a sample according to the detection result of the serum cytokine expression level of the sample, and the serum cytokine comprises: at least one of IIL-1 beta, IL-4, IL-10, IL-17, IL-2R alpha, LIF, MIF and SDF-1 alpha. And the parameter updating module is used for updating parameters of the sub-threshold depression risk prediction model according to the marked result and the prediction result to obtain the trained sub-threshold depression risk prediction model.
In a fifth aspect, an embodiment of the present invention provides an electronic device, which includes a processor and a memory. The memory is for storing a program that, when executed by the processor, causes the processor to implement a method of training a sub-threshold depression risk prediction model as described in previous embodiments, or a sub-threshold depression risk prediction method as follows: inputting the obtained detection result of the serum cytokine expression level of the sample to be detected into a pre-constructed subliminal depression risk prediction model to obtain a prediction result of the sample; wherein the serum cytokines comprise: at least one of IIL-1 β, IL-4, IL-10, IL-17, IL-2R α, LIF, MIF and SDF-1 α; the sub-threshold depression risk prediction model is obtained by training the training method for the sub-threshold depression risk prediction in the previous embodiment.
In a sixth aspect, embodiments of the present invention provide a computer-readable storage medium, on which a computer program is stored, which when executed by a processor implements a training method for a sub-threshold depression risk prediction model as described in the previous embodiments, or the following sub-threshold depression risk prediction method: inputting the detection result of the serum cytokine expression level of the sample to be detected into a pre-constructed subliminal depression model to obtain a prediction result of the sample; wherein the serum cytokines comprise: at least one of IL-1 β, IL-4, IL-10, IL-17, IL-2R α, LIF, MIF, and SDF-1 α; the sub-threshold depression model is obtained by training the training method of the sub-threshold depression model in the previous embodiment.
In a seventh aspect, embodiments of the invention provide a kit for subthreshold depression risk prediction, comprising reagents for detecting serum cytokine expression levels; the serum cytokines include: at least one of IL-1 β, IL-4, IL-10, IL-17, IL-2R α, LIF, MIF, and SDF-1 α.
The invention has the following beneficial effects:
the invention provides 8 detection markers capable of effectively predicting subliminal depression, which can effectively distinguish subliminal depression and healthy people based on the detection of the expression levels of the markers, can realize the rapid screening of a large number of people with subliminal depression, intervenes in time on the people and avoids the people from developing into depression.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.
FIG. 1 is a graph comparing the difference in the expression levels of 8 cytokines in example 1, i.e., IL-1 β, IL-4, IL-10, IL-17, IL-2R α, LIF, MIF and SDF-1 α, in serum of subliminal depressed persons and healthy persons;
FIG. 2 is the results of classifying subliminal depressed people and healthy people by the predictive model of example 1; wherein, a is a classified refined 3D view under the optimal parameters (optimal C is 64, g is 0.14), and B is a visualization result of the classification map;
FIG. 3 is the results of classifying subliminal depressed people and healthy people by the predictive model of example 2; wherein A is a classified refined 3D view under the optimal parameters, and B is a visual result of the classified graph;
FIG. 4 is the results of classifying subliminal depressed people and healthy people by the predictive model of example 3; wherein A is a classified refined 3D view under the optimal parameters, and B is a visual result of the classified view;
FIG. 5 is the results of the classification of subliminal depressed and healthy populations by the predictive model of example 4; wherein A is a classified refined 3D view under the optimal parameters, and B is a visual result of the classified view;
FIG. 6 is the results of the predictive model of example 5 classifying a subliminal depressed population and a healthy population; wherein A is a classified refined 3D view under the optimal parameters, and B is a visual result of the classified view;
FIG. 7 is the results of classifying subliminal depressed people and healthy people using the predictive model of comparative example 1; wherein A is a classified refined 3D view under the optimal parameters, and B is a visual result of the classified view;
FIG. 8 is the results of classifying subliminal depressed people and healthy people using the predictive model of comparative example 2; wherein A is a classified refined 3D view under the optimal parameters, and B is a visual result of the classified view;
FIG. 9 is the results of classifying subliminal depressed people and healthy people using the predictive model of comparative example 3; wherein A is a classified refined 3D view under the optimal parameters, and B is a visual result of the classified graph;
FIG. 10 is the results of classifying subliminal depressed persons and healthy persons by the predictive model of comparative example 4; wherein A is a classification refined 3D view under the optimal parameters, and B is a visualization result of the classification map.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below. The examples, in which specific conditions are not specified, were conducted under conventional conditions or conditions recommended by the manufacturer. The reagents or instruments used are not indicated by the manufacturer, and are all conventional products available commercially.
The embodiment of the invention provides an application of a reagent for detecting the expression level of a serum cytokine in preparing a kit for predicting the risk of subliminal depression, wherein the serum cytokine comprises: at least one of IL-1 β, IL-4, IL-10, IL-17, IL-2R α, LIF, MIF, and SDF-1 α.
The six serum cytokines are characteristic serum cytokines for predicting the subliminal depression population, and based on the expression levels of the cytokines, whether a sample is the subliminal depression population can be effectively distinguished, a quick and effective screening mode is provided for the subliminal depression population, and the population is prevented from further developing into depression.
The present invention does not specifically limit the kind of the reagent, and in an alternative embodiment, the reagent is selected from at least one of a primer pair, a probe, and a chip. It is understood that in the case of target determination, the reagents may be implemented based on existing primer and probe design criteria and will not be described in detail herein. Any reagent capable of obtaining the expression level of the serum cytokines can be applied to the preparation of the kit for the people with the subliminal depression, and the reagent and the kit belong to the protection scope of the application.
In alternative embodiments, any one or more combinations of 8 cytokines may be selected, preferably any 5, 6 or 7 combinations, more preferably 8 combinations, and the greater the number of cytokines selected, the more accurate the prediction will be. More preferably, the combination of reagents that detect 8 cytokines enables prediction of subthreshold risk to be performed more accurately and efficiently.
The embodiment of the invention also provides a training method of the subthreshold depression risk prediction model, which comprises the following steps: acquiring a training sample and a labeling result corresponding to the training sample;
inputting the serum cytokine expression level of the training sample into a pre-established subliminal depression risk prediction model to obtain a prediction result of the training sample; wherein the sub-threshold depression risk prediction model is used for judging the risk of the sample suffering from sub-threshold depression according to the expression level of serum cytokines of the sample, and the serum cytokines comprise: at least one of IL-1 β, IL-4, IL-10, IL-17, IL-2R α, LIF, MIF, and SDF-1 α. At least one of;
and updating parameters of the constructed sub-threshold depression risk prediction model based on the annotation result and the prediction result. Preferably, the parameters are updated to perform parameter optimization based on cross-validation.
The training samples comprise diseased samples and healthy samples, and the sample size of each of the diseased samples and the healthy samples is more than or equal to 10, preferably more than or equal to 30.
The embodiment of the invention also provides a subliminal depression risk prediction device which comprises an acquisition module and a prediction module.
Specifically, the acquisition module is used for acquiring the detection result of the serum cytokine expression level of the sample to be detected. The prediction module is used for inputting the detection result of the serum cytokine expression level of the sample to be detected into a pre-constructed subliminal depression risk prediction model to obtain a prediction result of the sample; wherein the serum cytokines comprise: at least one of IL-1 β, IL-4, IL-10, IL-17, IL-2R α, LIF, MIF, and SDF-1 α; the sub-threshold depression risk prediction model is obtained by training the training method of the sub-threshold depression risk prediction model in any embodiment.
Preferably, the serum cytokine includes at least one of IL-1 β, IL-4, IL-10, IL-17, IL-2R α, LIF, MIF, and SDF-1 α. At least one of (a); more preferably, the serum cytokines include IL-1 β, IL-4, IL-10, IL-17, IL-2R α, LIF, MIF and SDF-1 α.
The embodiment of the invention also provides a training device of the subliminal depression risk prediction model, which comprises an acquisition module, a construction module and a parameter updating module.
Specifically, the obtaining module is configured to obtain a training sample and a labeling result corresponding to the training sample. The construction module is used for inputting the detection result of the serum cytokine expression level of the training sample into a pre-established subliminal depression risk prediction model to obtain the prediction result of the training sample; wherein, the model for predicting the risk of subliminal depression is used for judging the risk of subliminal depression of a sample according to the detection result of the serum cytokine expression level of the sample, and the serum cytokine comprises: at least one of IL-1 β, IL-4, IL-10, IL-17, IL-2R α, LIF, MIF, and SDF-1 α. And the parameter updating module is used for updating parameters of the sub-threshold depression risk prediction model according to the marked result and the prediction result to obtain the trained sub-threshold depression risk prediction model.
An embodiment of the present invention further provides an electronic device, where the electronic device includes: a processor and a memory. Wherein the memory is for storing a program that, when executed by the processor, causes the processor to implement a method of training a sub-threshold depression risk prediction model as described in any of the preceding embodiments, or a method of sub-threshold depression risk prediction as follows: inputting the obtained detection result of the expression level of the serum cytokine of the sample to be detected into a pre-trained subliminal depression risk prediction model to obtain a prediction result of the sample; wherein the serum cytokines comprise: at least one of IL-1 β, IL-4, IL-10, IL-17, IL-2R α, LIF, MIF, and SDF-1 α; the model for predicting the risk of subliminal depression is constructed by the training method for predicting the risk of subliminal depression in any embodiment.
The Memory may be, but is not limited to, a Random Access Memory (RAM), a Read Only Memory (ROM), a Programmable Read Only Memory (PROM), an Erasable Read Only Memory (EPROM), an electrically Erasable Read Only Memory (EEPROM), and the like.
The processor may be an integrated circuit chip having signal processing capabilities. The Processor may be a general-purpose Processor including a Central Processing Unit (CPU), a Network Processor (NP), etc.; but also Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components.
In practical applications, the electronic device may be a server, a cloud platform, a mobile phone, a tablet computer, a notebook computer, an ultra-mobile personal computer (UMPC), a handheld computer, a netbook, a Personal Digital Assistant (PDA), a wearable electronic device, a virtual reality device, and the like, and therefore, the embodiment of the present application does not limit the type of the electronic device.
Embodiments of the present invention further provide a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements a training method for a sub-threshold depression risk prediction model according to any of the foregoing embodiments, or the following sub-threshold depression risk prediction method: inputting the detection result of the serum cytokine expression level of the sample to be detected into a pre-trained subliminal depression model to obtain a prediction result of the sample; wherein the serum cytokines comprise: at least one of IL-1 β, IL-4, IL-10, IL-17, IL-2R α, LIF, MIF, and SDF-1 α; the sub-threshold depression model is constructed by the training method of the sub-threshold depression model according to any embodiment.
It should be noted that, in the electronic device and the computer-readable storage medium, the preferred combination of cytokines is as described in any of the foregoing embodiments, and is not described again.
The computer readable medium includes: u disk, removable hard disk, read only memory, random access memory, magnetic or optical disk, etc. for storing program codes.
The embodiment of the invention provides a kit for predicting the risk of subliminal depression, which comprises: reagents for detecting the expression level of serum cytokines; the serum cytokines include: IL-1 β, IL-4, IL-10, IL-17, IL-2R α, LIF, MIF and SDF-1 α.
It should be noted that the selection of the reagents in the kit is the same as that described in the foregoing application examples, and is not repeated herein.
The features and properties of the present invention are described in further detail below with reference to examples.
Example 1
A method of training a sub-threshold depression risk prediction model comprising the following steps.
Step 1:
collecting peripheral blood whole blood samples of a plurality of testers and placing the samples in procoagulant tubes; the multiple testers were divided into a first tester and a second tester, the first tester being 98 sub-threshold depressed populations (diseased samples) and the second tester being 51 healthy volunteers (healthy samples).
In the embodiment of the invention, the whole blood samples of the peripheral blood of a plurality of testers can be directly extracted through peripheral venous blood collection needles, in order to establish the relationship between the serum cytokine characteristic information and the risk of the subthreshold depression population, the subthreshold depression population is taken as a first tester, and the testers without subthreshold depression, namely healthy volunteers, are taken as a second tester in the plurality of testers for data sample collection.
Step 2:
pretreating and storing the collected whole blood sample: standing whole blood at room temperature for 40 min, centrifuging at 1000g/min for 15min, sucking supernatant after centrifugation to a new batch of EP tubes, centrifuging at 4 deg.C and 3000rpm/min for 10min, collecting supernatant, subpackaging and storing at 60 ul/tube specification, and standing at-80 deg.C in refrigerator.
The serum cytokine concentration of the sample was determined: the determination kit is a Bio-Plex liquid phase chip protein multiplex detection kit which is purchased from BIO-RAD company. The quantitative detection of the cell factor is carried out by sequentially preparing a gradient standard substance, an incubation sample, an incubation detection antibody, an incubation Streptavidin-PE (SA-PE) and a termination reaction according to the specification of a Bio-Plex liquid chip kit. Operating a Bio-Plex 200 suspension chip system instrument, sequentially operating instrument preheating (arm up), starting (start up) and correcting (calibre) programs under the prompt of Bio-Plex Manager 6.1 software, editing experiment protocol according to information provided by the kit, operating the experiment in the experiment protocol, reading a plate and obtaining an experiment result.
Analyzing the detection result of the cell factor by adopting a non-parametric test statistical analysis method: in the embodiment of the invention, the characteristic indexes of the serum cytokines are IL-1 beta, IL-4, IL-10, IL-17, IL-2R alpha, LIF, MIF and SDF-1 alpha, and the characteristic indexes represent multiple functions of innate immunity, adaptive immunity, cell growth, tissue injury repair and the like of a tester.
It should be noted that the characteristic serum cytokines can be understood as: serum cytokines with significantly different expression levels were present between healthy volunteers and sub-threshold depressed populations.
And step 3:
training a subthreshold depression risk prediction model:
in the embodiment of the present invention, a support vector machine model that can predict sub-threshold depressed people is established by using a more common libsvm software package, which specifically may be: the first and second testers 6 characteristic serum cytokine data and the first and second tester label data were input. The first tester tag may be assumed to be 1 and the second tester tag may be assumed to be-1; normalizing the first and second tester characteristic serum cytokine data to [01] range data; selecting a Gaussian radial basis function kernel with high flexibility and wide application in a libsvm software package as a support vector machine kernel, and according to a formula:
Figure BDA0003329509650000111
Figure BDA0003329509650000112
c，γ:c＝2m；γ＝2n。
wherein W is the weight of each feature value, and determines the importance of the contribution of the feature to the classification; c is a hyper-parameter between the edge space and the prediction error, also called a penalty parameter; gamma is the gaussian kernel size, reflecting the height and shape of the kernel function, x is each eigenvalue of the input,
Figure BDA0003329509650000113
is the difference variable of the soft margin.
And continuously improving the parameters c and gamma, and continuously adjusting the vector machine model function to improve the prediction capability of the model until a trained risk prediction model of the subliminal depression population is obtained.
And 4, step 4:
and training the prediction model of the subliminal depression population according to the serum characteristic cytokine indexes corresponding to each tester until the trained predication model of the subliminal depression population is obtained.
In the embodiment of the invention, a common steady k-fold cross validation analysis method is adopted to optimize the risk prediction model of the subliminal depression population, and the method specifically comprises the following steps: for the plurality of testers, the total number of the first tester and the second tester can be assumed to be n, k can be assumed to be 10, i.e. 10 folds, that is, the number of people included in each fold is n/10, and the number of people in each fold is random; and (3) randomly keeping the serum characteristic cell factors of the testers in the fold 1 as a test data set during each training, and adjusting the kernel parameters c and gamma of the support vector machine according to the Gaussian radial basis function every time by randomly iterating 1000 times by using the serum characteristic cell factors of the testers in the fold 9 as a training data set, so as to obtain the classification accuracy of the first tester and the second tester after each parameter change, thereby obtaining the good classification accuracy of the first tester and the second tester and further obtaining the trained subliminal depression risk prediction model.
The results of comparing the concentrations of cytokines IL-1 β, IL-4, IL-10, IL-17, IL-2 Ra, LIF, MIF and SDF-1 α in peripheral blood of sub-threshold depressed population and healthy volunteer groups using non-parametric test statistical analysis are shown in FIG. 1.
The concentrations of IL-1. beta., IL-4, IL-10, IL-17, IL-2R. alpha., LIF, MIF and SDF-1. alpha. were determined, and the results are shown in FIG. 1.
As can be seen from FIG. 1, IL-1 β, IL-4, IL-10, IL-17, IL-2R α, LIF, MIF and SDF-1 α showed significant differences in their expression levels in the serum of subliminal depressed people and healthy people.
Example 2
A method of constructing and training a prediction model of risk of subliminal depression is provided, which is substantially the same as in example 1 except that the characteristic serum cytokines for constructing and training the prediction model of risk of subliminal depression are 7, specifically IL-1 β, IL-4, IL-10, IL-17, IL-2 ra, LIF, MIF; when a model is adopted to predict a sample to be tested, the corresponding serum characteristic cytokine indexes are also the indexes of IL-1 beta, IL-4, IL-10, IL-17, IL-2R alpha, LIF and MIF.
Example 3
A method of constructing and training a predictive model of risk of subliminal depression is provided, substantially the same as in example 1, except that the characteristic serum cytokines used to construct and train the predictive model of risk of subliminal depression are 7, specifically IL-1 β, IL-4, IL-10, IL-17, IL-2 ra, LIF and SDF-1 α; when a model is adopted to predict a sample to be tested, the corresponding serum characteristic cytokine indexes are also the indexes of IL-1 beta, IL-4, IL-10, IL-17, IL-2R alpha, LIF and SDF-1 alpha.
Example 4
A method of constructing and training a predictive model of subliminal depression risk is provided, substantially the same as in example 1, except that the characteristic serum cytokines used to construct and train the predictive model of subliminal depression risk are 6, specifically IL-1 β, IL-4, IL-10, IL-17, IL-2 ra and SDF-1 α; when the model is adopted to predict a sample to be tested, the corresponding serum characteristic cytokine indexes are also the indexes of IL-1 beta, IL-4, IL-10, IL-17, IL-2R alpha and SDF-1 alpha.
Example 5
A method of constructing and training a prediction model of risk of subliminal depression is provided, which is substantially the same as in example 1 except that the characteristic serum cytokines for constructing and training the prediction model of risk of subliminal depression are 6, specifically IL-1 β, IL-4, IL-10, IL-17, LIF and MIF; when a model is adopted to predict a sample to be tested, the corresponding serum characteristic cytokine indexes are also the indexes of IL-1 beta, IL-4, IL-10, IL-17, LIF and MIF.
Comparative example 1
A method is provided for constructing and training a model for predicting risk of subliminal depression, substantially the same as in example 1, except that SCF is used in place of the cytokine SDF-1 α, and the characteristic serum cytokines used for constructing and training the model for predicting risk of subliminal depression are specifically IL-1 β, IL-4, IL-10, IL-17, IL-2 ra, LIF, MIF and SCF; when a model is adopted to predict a sample to be tested, the corresponding serum characteristic cytokine indexes are also the indexes of IL-1 beta, IL-4, IL-10, IL-17, IL-2R alpha, LIF, MIF and SCF.
Comparative example 2
A method of constructing and training a prediction model of risk of subliminal depression is provided, which is substantially the same as in example 1 except that IL-6 is used in place of the cytokine MIF, and the characteristic serum cytokines used for constructing and training the prediction model of risk of subliminal depression are specifically IL-1 β, IL-4, IL-10, IL-17, IL-2 ra, LIF, SDF-1 α and IL-6; when a model is adopted to predict a sample to be tested, the corresponding serum characteristic cytokine indexes are also the indexes of IL-1 beta, IL-4, IL-10, IL-17, IL-2R alpha, LIF, SDF-1 alpha and IL-6.
Comparative example 3
The method for constructing and training the subliminal depression risk prediction model is approximately the same as that in example 1, except that IL-2 and IL-6 are adopted to replace cytokines LIF and MIF respectively, and the characteristic serum cytokines used for constructing and training the subliminal depression risk prediction model are specifically IL-1 beta, IL-4, IL-10, IL-17, IL-2 Ra, SDF-1 alpha, IL-2 and IL-6; when a model is adopted to predict a sample to be tested, the corresponding serum characteristic cytokine indexes are also the indexes of IL-1 beta, IL-4, IL-10, IL-17, IL-2R alpha, SDF-1 alpha, IL-2 and IL-6.
Comparative example 4
A method of constructing and training a model for predicting risk of subliminal depression is provided, which is substantially the same as in example 1 except that IL-6 and IL-9 are used in place of the cytokines MIF and SDF-1 α, respectively, and the characteristic serum cytokines used for constructing and training the model for predicting risk of subliminal depression are specifically IL-1 β, IL-4, IL-10, IL-17, IL-2 ra, LIF, IL-6 and IL-9; when a model is adopted to predict a sample to be tested, the corresponding serum characteristic cytokine indexes are also the indexes of IL-1 beta, IL-4, IL-10, IL-17, IL-2R alpha, LIF, IL-6 and IL-9.
Test example 1
Known samples (a first tester and a second tester) are predicted by using the sub-threshold depression risk prediction models provided in examples 1-5 and comparative examples 1-4, and the support vector machine results are visualized to distinguish sub-threshold depression groups from healthy groups.
The predicted results of example 1 are shown in fig. 2. As can be seen from FIG. 2, the prediction model for the risk of subliminal depression provided by the invention has higher sensitivity and specificity for the detection of subliminal depression population, and the accuracy rate reaches 82.6%.
The predicted results of example 2 are shown in fig. 3. As can be seen from fig. 3, the accuracy of detection for the sub-threshold depressed population in the sub-threshold depression risk prediction model of example 2 was 78.5%.
The predicted results of example 3 are shown in fig. 4. As can be seen from fig. 4, the accuracy of detection of the sub-threshold depressed population in the sub-threshold depression risk prediction model of example 3 was 75.8%.
The predicted results of example 4 are shown in fig. 5. As can be seen from fig. 5, the accuracy of detection for the sub-threshold depressed population in the sub-threshold depression risk prediction model of example 4 was 75.2%.
The predicted results of example 5 are shown in fig. 6. As can be seen from fig. 6, the accuracy of detection for the sub-threshold depressed population in the sub-threshold depression risk prediction model of example 5 was 73.2%.
The predicted results of comparative example 1 are shown in fig. 7. As can be seen from fig. 7, the accuracy of detection for the sub-threshold depressed population in the sub-threshold depression risk prediction model of example 6 was 79.9%.
The predicted results of comparative example 2 are shown in fig. 8. As can be seen from fig. 8, the accuracy of the detection of the sub-threshold depression population of the prediction model of the sub-threshold depression risk of comparative example 1 was 75.2%.
The predicted result of comparative example 3 is shown in fig. 9. As can be seen from fig. 9, the accuracy of the detection of the sub-threshold depression population of the prediction model of the sub-threshold depression risk of comparative example 2 reaches 72.5%.
The predicted result of comparative example 4 is shown in fig. 10. As can be seen from fig. 10, the accuracy of the detection of the sub-threshold depression population by the sub-threshold depression risk prediction model of comparative example 3 reaches 71.8%.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (7)

1. The application of a reagent for detecting the expression level of serum cytokines in the preparation of a kit for predicting the risk of subliminal depression is characterized in that the serum cytokines are composed of IL-1 beta, IL-4, IL-10, IL-17, IL-2R alpha, LIF, MIF and SDF-1 alpha.
2. The use of claim 1, wherein the reagent is selected from at least one of a primer pair, a probe, and a chip.
3. A method of training a sub-threshold depression risk prediction model, comprising:
obtaining the expression level of serum cytokines of a training sample and a corresponding labeling result of the training sample, wherein the labeling result is diseased or healthy;
inputting the serum cytokine expression level of the training sample into a pre-established subliminal depression risk prediction model to obtain a prediction result of the training sample; the subliminal depression risk prediction model is a support vector machine model and is used for judging the risk of the sample suffering from subliminal depression according to the expression level of a sample serum cytokine, wherein the serum cytokine consists of IL-1 beta, IL-4, IL-10, IL-17, IL-2 Ra, LIF, MIF and SDF-1 alpha;
and updating parameters of the constructed subliminal depression risk prediction model based on the annotation result and the prediction result.
4. A subliminal depression risk prediction device, characterized in that it comprises:
the acquisition module is used for acquiring the detection result of the serum cytokine expression level of the sample to be detected;
the prediction module is used for inputting the detection result of the serum cytokine expression level of the sample to be detected into a pre-constructed subliminal depression risk prediction model to obtain a prediction result of the sample; wherein the serum cytokine is composed of IL-1 beta, IL-4, IL-10, IL-17, IL-2R alpha, LIF, MIF and SDF-1 alpha; the model for predicting the risk of subliminal depression is obtained by training the training method of the model for predicting the risk of subliminal depression according to claim 3.
5. Training device for a sub-threshold depression risk prediction model, characterized in that it comprises:
the acquisition module is used for acquiring the serum cytokine expression level of a training sample and a labeling result corresponding to the training sample; the noted result is diseased or healthy;
the processing module is used for inputting the detection result of the serum cytokine expression level of the training sample into a pre-established subliminal depression risk prediction model to obtain the prediction result of the training sample; the subliminal depression risk prediction model is a support vector machine model and is used for judging the risk of the sample suffering from subliminal depression according to the detection result of the expression level of the sample serum cytokines, wherein the serum cytokines consist of IL-1 beta, IL-4, IL-10, IL-17, IL-2R alpha, LIF, MIF and SDF-1 alpha;
and the parameter updating module is used for updating parameters of the sub-threshold depression risk prediction model according to the marked result and the prediction result to obtain the trained sub-threshold depression risk prediction model.
6. An electronic device, characterized in that the electronic device comprises: a processor and a memory; the memory for storing a program that, when executed by the processor, causes the processor to implement a method of training a sub-threshold depression risk prediction model according to claim 3, or a method of sub-threshold depression risk prediction as follows:
inputting the detection result of the serum cytokine expression level of the sample to be detected into a pre-trained subliminal depression risk prediction model to obtain a prediction result of the sample; wherein the serum cytokine is composed of IL-1 beta, IL-4, IL-10, IL-17, IL-2R alpha, LIF, MIF and SDF-1 alpha; the sub-threshold depression risk prediction model is obtained by training the training method for sub-threshold depression risk prediction according to claim 3.
7. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out a method of training a sub-threshold depression risk prediction model according to claim 3, or a method of sub-threshold depression risk prediction as follows:
inputting the detection result of the serum cytokine expression level of the sample to be detected into a pre-trained subliminal depression model to obtain a prediction result of the sample; wherein the serum cytokine is composed of IL-1 beta, IL-4, IL-10, IL-17, IL-2R alpha, LIF, MIF and SDF-1 alpha; the sub-threshold depression model is constructed by the training method of the sub-threshold depression model according to claim 3.
CN202111273591.6A 2021-10-29 2021-10-29 Suspense depression risk prediction device and kit and application thereof Active CN113945724B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111273591.6A CN113945724B (en) 2021-10-29 2021-10-29 Suspense depression risk prediction device and kit and application thereof

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111273591.6A CN113945724B (en) 2021-10-29 2021-10-29 Suspense depression risk prediction device and kit and application thereof

Publications (2)

Publication Number Publication Date
CN113945724A CN113945724A (en) 2022-01-18
CN113945724B true CN113945724B (en) 2022-07-05

Family

ID=79337266

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111273591.6A Active CN113945724B (en) 2021-10-29 2021-10-29 Suspense depression risk prediction device and kit and application thereof

Country Status (1)

Country Link
CN (1) CN113945724B (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103513043A (en) * 2013-10-15 2014-01-15 华中师范大学 Protein chip capable of quickly providing early-stage depression pre-warning
CN107220654A (en) * 2016-03-22 2017-09-29 李刚 A kind of Stem Cell Activity detecting system based on svm algorithms
CN112881547A (en) * 2021-01-12 2021-06-01 中国科学院大学宁波华美医院 Screening method of early liver cancer diagnosis marker for liver cirrhosis and hepatitis population
CN113444798A (en) * 2021-06-30 2021-09-28 北京泱深生物信息技术有限公司 Renal cancer survival risk biomarker group, diagnosis product and application

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110702917B (en) * 2019-09-05 2023-08-15 首都医科大学附属北京安定医院 Application of serum amyloid P in preparation of related products for diagnosis and treatment of depression
CN115198010A (en) * 2021-04-26 2022-10-18 武汉儿童医院 New application of PDK4, NRG1 and EPHB2 detection agent and depression detection reagent

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103513043A (en) * 2013-10-15 2014-01-15 华中师范大学 Protein chip capable of quickly providing early-stage depression pre-warning
CN107220654A (en) * 2016-03-22 2017-09-29 李刚 A kind of Stem Cell Activity detecting system based on svm algorithms
CN112881547A (en) * 2021-01-12 2021-06-01 中国科学院大学宁波华美医院 Screening method of early liver cancer diagnosis marker for liver cirrhosis and hepatitis population
CN113444798A (en) * 2021-06-30 2021-09-28 北京泱深生物信息技术有限公司 Renal cancer survival risk biomarker group, diagnosis product and application

Also Published As

Publication number Publication date
CN113945724A (en) 2022-01-18

Similar Documents

Publication Publication Date Title
US20210041440A1 (en) Methods and apparatus for identifying disease status using biomarkers
US11769596B2 (en) Plasma based protein profiling for early stage lung cancer diagnosis
CN105005680B (en) Use categorizing system and its method of kit identification and diagnosis pulmonary disease
Boyd Defining laboratory reference values and decision limits: populations, intervals, and interpretations
JP7361187B2 (en) Automated validation of medical data
WO2002044715A1 (en) Methods for efficiently minig broad data sets for biological markers
JP2007534086A (en) Pattern recognition method for diagnosis of systemic autoimmune disease
Sivajohan et al. Clinical use of artificial intelligence in endometriosis: A scoping review
Milas et al. Red cell distribution width to platelet ratio for liver fibrosis: a systematic review and meta-analysis of diagnostic accuracy
CN107169264B (en) complex disease diagnosis system
Rahnenführer et al. Statistical analysis of high-dimensional biomedical data: a gentle introduction to analytical goals, common approaches and challenges
CN113945724B (en) Suspense depression risk prediction device and kit and application thereof
Wang et al. Optimal threshold selection methods under tree or umbrella ordering
Kingsmore et al. Recent advances in the use of machine learning and artificial intelligence to improve diagnosis, predict flares, and enrich clinical trials in lupus
Zhu et al. A new method for classifying different phenotypes of kidney transplantation
Eckerle et al. Estimating the probability of bacterial infection using a novel biomarker among pediatric patients in the emergency department
CN116741384B (en) Bedside care-based severe acute pancreatitis clinical data management method
Zhang et al. An investigation of how normalisation and local modelling techniques confound machine learning performance in a mental health study
Mohammed et al. Two-Stage Classification of Breast Tumor Biomarkers for Iraqi Women
Lalitha et al. Deep Learning-Based Analysis of Liquid Biopsies for Uterus Cancer Biomarker Discovery
US20110111432A1 (en) Diagnostic method for peanut allergy
Jegorović et al. Comparison of the ELISA method with the nephelometric method for determination of serum amyloid A in patients with COVID 19
Long Bias
Ghalwash Improving the Computational Complexity of the COOL Screening Tool
ATAS et al. Detection of Thrombocytopenia, Anemia and Leukocytosis by Using Ensemble Learning

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant