Method for preparing protein marker for detecting atopic dermatitis

Abstract

To provide a method for detecting atopic dermatitis in a low-invasive or non-invasive manner.SOLUTION: The present invention collects protein markers for detecting atopic dermatitis from skin surface lipids collected from subjects. The present invention detects atopic dermatitis using the protein markers.SELECTED DRAWING: None

Description

Technical Field

[0001] The present invention relates to a method for preparing a protein marker for detecting atopic dermatitis and a method for detecting atopic dermatitis using the protein marker. marker.

Background Art

[0002] Atopic dermatitis (hereinafter also referred to as "AD") is an eczematous skin disease that mainly develops in people with an atopic predisposition. Typical symptoms of AD include chronic and recurrent itching, rashes, erythema, etc. that occur symmetrically on both sides, as well as parakeratosis, reduced barrier function, dry skin, etc. Most cases of AD develop in infants and tend to improve with growth , but in recent years, adult-type and refractory AD have also been increasing.

[0003] Newborns or infants with genetic factors that are prone to allergies and atopy are prone to infantile eczema, atopic dermatitis, food allergies, and furthermore, bronchial asthma, allergic rhinitis, etc., and are reported to develop various allergic diseases with age (allergy march). Thus, allergic diseases are likely to develop another allergic disease once one disease has developed, and their treatment often takes a long time. Therefore, it is necessary to suppress the onset of allergic diseases in infants. Regarding the evaluation of the pathological condition of AD, various evaluation methods have been conventionally used in clinical practice. As evaluation items by dermatologists, Severity Scoring of Atopic

[0004] Dermatitis (SCORAD) and Eczema Area and Sev c Dermatitis(SCORAD)やEczema Area and Sev c Dermatitis(SCORAD)やEczema Area and Sev ​The Quality Index (EASI) is often used. However, these evaluation methods are Because it is highly subjective, it is used in conjunction with more objective biomarkers. This is often the case. Serological assessment items for AD include total IgE levels in the blood, peripheral blood eosinophil count, Serum lactate dehydrogenase (LDH) level, serum thymus a nd activation-regulated chemokine (TARC) value These are commonly used. However, serological evaluation is invasive because it involves blood sampling. However, it cannot always be said that a diagnosis can be made with sufficient accuracy.

[0005] Because the skin is in contact with the outside world, it is a tissue from which biological samples can be collected with minimal invasiveness. Next, various types of skin samples such as skin tissue collected by biopsy and stratum corneum removed by tape will be used. Nucleic acid or protein markers have been isolated. Non-patent documents 1-6 and Patent Document 1 contain: By applying a weakly adhesive tape to the skin, interleuk can be non-invasively administered from the skin surface. ILs, TNF-α, INF-γ, human β-defensin (hBD2), etc. It is recorded that pteromarkers were collected and the collected markers were used to examine skin diseases and conditions. It is described in Patent Document 2, which describes RNA, etc., derived from the skin cells of the subject from lipids on the skin surface. It is described that nucleic acids are isolated and used as samples for biological analysis. [Prior art documents] [Patent Documents]

[0006] [Patent Document 1] International Public Gazette No. 2014 / 144289 [Patent Document 2] International Public Gazette No. 2018 / 008319

Non-Patent Literature

[0007]

Non-Patent Literature 1

Non-Patent Literature 2

Non-Patent Literature 3

Non-Patent Literature 4

Non-Patent Literature 5

Non-Patent Literature 6

Summary of the Invention

Problems to be Solved by the Invention

[0008] The present invention relates to a method for preparing a protein marker for detecting atopic dermatitis from a subject in a minimally invasive or non-invasive manner, and a method for detecting atopic dermatitis using the protein marker. ーを調製する方法、及び当該タンパク質マーカーを用いたアトピー性皮膚炎の検出方法に 関する。

Means for Solving the Problems

[0009] In one aspect, the present invention provides a method for preparing a protein marker for detecting atopic dermatitis, which includes recovering at least one protein selected from the group consisting of the proteins shown in Tables 1-1 to 1 -13 from the skin surface lipids collected from a subject. ることを含む、アトピー性皮膚炎検出用タンパク質マーカーの調製方法を提供する。 別の一態様において、本発明は、被験体から採取した皮膚表上脂質から、下記表1-1 Detecting at least one protein selected from the group consisting of the proteins shown in ~1-13, a method for detecting atopic dermatitis in a subject is provided. In yet another aspect, the present invention provides a protein marker for detecting atopic dermatitis, comprising at least one protein selected from the group consisting of the proteins shown in Table 2-1 to 2-5 below.

Effects of the Invention

[0010] According to the present invention, a protein marker for detecting atopic dermatitis can be recovered from a subject by a simple, minimally invasive or non-invasive method, or atopic dermatitis can be detected using the marker. Therefore, the present invention enables the diagnosis of atopic dermatitis in various subjects including infants for whom invasive biological sample collection is not easy. Further, the method of the present invention can contribute to the early diagnosis and treatment of atopic dermatitis in infants and adults.

Modes for Carrying Out the Invention

[0011] All patent documents, non-patent documents, and other publications cited in this specification are hereby incorporated by reference in their entirety.

[0012] The names of the proteins disclosed in this specification follow the Gene Name or Protein Name described in UniProt ([https: / / www.uniprot .org / ]).

[0013] In this specification, "infant" refers to the age from 0 years old until entering elementary school, specifically, a person from 0 years old to 5 years old. In this specification, "adult" in a broad sense corresponds to "infant". It refers to people who do not engage in sexual activity, but preferably to those who have completed puberty, specifically those aged 16 or older. People are generally preferred, and those 20 years of age or older are even more preferred.

[0014] In this specification, "skin surface lipids (SS) "L)" refers to the lipid-soluble fraction present on the surface of the skin, and is sometimes called sebum. SSL mainly consists of secretions secreted from exocrine glands such as sebaceous glands in the skin, and is found on the skin surface. It exists on the surface of the skin in the form of a thin layer that covers the surface.

[0015] In this specification, unless otherwise specified, "skin" refers to the epidermis, dermis, hair follicles, etc. of the body surface. It is a general term for areas that include tissues such as sweat glands, sebaceous glands, and other glands.

[0016] In this invention, "detection" of atopic dermatitis refers to the detection of SSL in a subject from whom SSL was collected. This refers to clarifying the presence or absence of atopic dermatitis, and involves examination, measurement, determination, and evaluation. Alternatively, it can be rephrased as evaluation support. Furthermore, in this specification, "judgment" or The term "evaluation" does not include judgments or assessments made by physicians.

[0017] In this specification, "feature" is synonymous with "explanatory variable" in machine learning. In the detailed document, the selection of protein markers for detecting atopic dermatitis is used in machine learning. The proteins used are sometimes referred to as "characteristic proteins."

[0018] The inventors have found that SSL contains a protein useful for detecting atopic dermatitis (AD). We found that these proteins are used as protein markers for detecting AD. It can be used. Therefore, it is convenient to collect SSL from the skin surface of the subject. Furthermore, a minimally invasive or non-invasive method for detecting AD in a subject's biological sample and its The protein markers contained within can be recovered.

[0019] Therefore, in one embodiment, the present invention provides a protein marker for AD detection. In one embodiment, the present invention provides a method for preparing a protein marker for AD detection. This method involves extracting a target AD detection protein marker from SSL collected from the subject. This includes recovery. In another embodiment, the present invention provides an AD detection method. The method involves detecting the AD detection protein marker from SSL collected from the subject. Includes.

[0020] Method for preparing a protein marker for AD detection and method for detecting AD according to the present invention (hereinafter referred to as "summary") In the method of the present invention (also known as the method of the present invention), the subject is any organism that has SSL on its skin. Good. Examples of subjects include mammals, including humans and non-human mammals, and are preferred. The subjects are human. The subjects are not limited by sex or age and may include infants to adults. Preferably, the subject is a mammal that requires or desires the detection of AD (preferably) The subject is a human. For example, the subject is a mammal (preferably a human) suspected of developing AD. )

[0021] In one embodiment, the method of the present invention further includes collecting SSL from a subject. It is acceptable. The skin areas from which SSL can be collected include the head, face, neck, trunk, limbs, and other parts of the body. Examples include skin in any part of the body, preferably areas with symptoms such as AD-like eczema or dryness. Skin is one example.

[0022] For the collection of SSL from the subject's skin, a method used for the recovery or removal of SSL from the skin is employed. Any means available can be employed. Preferably, the SSL absorbent material described later, S SL adhesive material or an instrument for scraping SSL off the skin can be used. As for L absorbent materials or SSL adhesive materials, any material that has affinity for SSL is particularly suitable. The following are examples of materials, but are not limited to polypropylene, pulp, etc. The method of collecting SSL from the skin. A more detailed example of the order is SS applied to sheet-like materials such as oil-blotting paper and oil-blotting film. Methods for absorbing L, methods for adhering SSL to glass plates, tape, etc., spatula, screw Methods include using rapists to scrape off and recover the SSL. To improve adhesion, SSL absorbent material pre-soaked with a highly lipid-soluble solvent may be used. On the other hand, if the SSL absorbent material contains a highly water-soluble solvent or moisture, the adsorption of SSL will be reduced. Because it is inhibited, a highly water-soluble solvent or one with a low water content is preferable. SSL absorption It is preferable to use the material in a dry state.

[0023] The collected SSL may be immediately used in the protein extraction process described below, but the It may be stored until it is used in the protein extraction process. If stored, SSL can be prepared after collection. It is preferable to store it immediately under low temperature conditions. The storage temperature conditions for SSL are below 0°C. If present, preferably -20±20℃ to -80±20℃, more preferably -20±10℃ ℃ to -80±10℃, more preferably -20±20℃ to -40±20℃, even more preferably The temperature range is -20±10℃ to -40±10℃, more preferably -20±10℃, even more preferably The temperature is -20±5℃. The storage period for SSL is not particularly limited, but is preferably 12 Less than a month, for example, 6 hours or more but less than 12 months, more preferably 6 months or less, for example, less than a day Up to 6 months, more preferably 3 months or less, for example, 3 days to 3 months.

[0024] For protein extraction from collected SSL, the extraction of proteins from biological samples or sperm extraction is performed. Methods commonly used in production, for example, water, phosphate-buffered saline A solution containing Triton X-100, Tween 20, etc. as a surfactant. Liquid extraction method, or M-PER buffer (Thermo Fisher Scientific), MPEX PTS Reagent(GL science) , QIAzol Lysis Reagent (Qiagen), EasyPep TM M ini MS Sample Prep Kit(ThermoFisher Scie Extraction methods using commercially available protein extraction reagents and kits such as (ntific) can be used. Cut.

[0025] The extracted SSL-derived protein contains one type of protein marker for AD detection, which will be described later. The above may be included. The SSL-derived protein may be immediately used for AD detection as described below. However, the protein may be stored under normal storage conditions until it is used for AD detection.

[0026] As shown in the examples described below, 418 types of SSL-derived proteins are shown in Tables 1-1 to 1-13. Quality refers to proteins whose abundance in SSL is significantly different in AD patients compared to healthy individuals. Furthermore, machine learning using the abundance of these proteins in SSL as features has been performed. The built prediction model makes it possible to predict AD. Therefore, Tables 1-1 to 1-13 show The SSL-derived protein shown can be used as a protein marker for AD detection. Of the proteins shown in Tables 1-1 to 1-13, 147 types of proteins are shown in Tables 2-1 to 2-5. The protein, as shown in the examples described below, is a novel protein that has not been previously reported to be associated with AD. It is a standard protein marker for detecting AD. Furthermore, 65 types of markers are shown in Tables 3-1 to 3-2. Proteins include the 200 types of proteins shown in Tables 4-1 to 4-6 below, and Tables 5-1 to 5- This protein is common to the 283 types of proteins shown in 9, and is used as an AD detection protein. It can be used as a car.

[0027] [Table 1-1]

[0028] [Table 1-2]

[0029] [Table 1-3]

[0030] [Table 1-4]

[0031] [Table 1-5]

[0032] [Table 1-6]

[0033] Table 1-7

[0034] Table 1-8

[0035] Table 1-9

[0036] Table 1-10

[0037] Table 1-11

[0038] Table 1-12

[0039] Table 1-13

[0040] Table 2-1

[0041] Table 2-2

[0042] Table 2-3

[0043] [Table 2-4]

[0044] [Table 2-5]

[0045] [Table 3-1]

[0046] [Table 3-2]

[0047] More specifically, the SSL-derived proteins shown in Tables 1-1 to 1-13 are shown in Tables 4-1 to 4 below. -6 and the proteins shown in Tables 5-1 to 5-9, and the proteins shown in Tables 4-1 to 4-6 This refers to the following tables 7-1 to 7-4, Table 8, Tables 11-1 to 11-4, Tables 12-1 to 12-4 and Table The proteins listed in Table 13, and those listed in Tables 5-1 to 5-9, are listed in Tables 9-1 to 9 below. -7, Tables 10-1 to 10-2, Tables 14-1 to 14-7, Tables 15-1 to 15-4 and Table 16 Contains the following protein.

[0048] As shown in the examples described later, the SSLs of healthy children and children with AD are extracted, and healthy children or This refers to the quantitative value of the protein obtained in more than 75% of subjects in either group of children affected by Alzheimer's disease. The results showed that the prevalence was 1.5 times higher in children with AD compared to healthy children (p ≤ 0). 05) 116 types of proteins that increased (Table 7-1 to 7-4), and 0.75 times or less (p Twelve proteins (Table 8) were identified that decreased to ≤0.05. Similarly, in healthy adults... Extracted from SSLs of adult AD patients 2, representing 75% or more of either the healthy group or the AD patient group. The quantitative values ​​of proteins obtained from the subjects were analyzed. As a result, compared to healthy individuals, In AD patients, the relative abundance of 205 proteins increased by more than 1.5 times (p ≤ 0.05). Tables 9-1 to 9-7), and 37 types of proteins that decreased to 0.75 times or less (p ≤ 0.05) The quality (Tables 10-1 to 10-2) was identified.

[0049] Therefore, in one embodiment, the AD detection method of the present invention is used in the SSL of the subject. Based on the amount of the protein marker for AD detection (e.g., marker concentration in SSL), AD is detected. Including detection.

[0050] For example, Tables 7-1 to 7-4, Table 8, Tables 9-1 to 9-7 and Table 1 in the subject SSL Based on the concentration of one or more protein markers shown in 0-1 to 10-2, the SSL is derived from Whether the subject is AD or not (in other words, whether the SSL originates from a subject who is AD) It is possible to detect (whether or not). In the AD detection method of the present invention, Tables 7-1 to 7-4 The proteins shown in Tables 8, 9-1 to 9-7, and 10-1 to 10-2 are any one of the following: or a combination of two or more of these, used as a protein marker for AD detection. This is possible. For example, one or more markers (target markers) in the SSL of the subject. By measuring the concentration of the marker and comparing the measured concentration with that of a healthy control group, the subject can determine the following: It is possible to detect whether or not it is AD.

[0051] The group consists of proteins whose target markers are shown in Tables 7-1 to 7-4 and Tables 9-1 to 9-7. If at least one of the selected types is found, the concentration of the target marker in the subject is similar to that of the healthy group. If the value is higher compared to the target mark in the subject, the subject may be detected as having AD. For example, the target mark in the subject If the concentration of Kerr is statistically significantly higher than that of the healthy control group, the subject may be detected as having Alzheimer's disease (AD). For example, the concentration of the target marker in the subject is preferably 11 compared to the healthy control group. If the percentage is 0% or more, more preferably 120% or more, and even more preferably 150% or more, then the The subject may be detected as having AD. Two or more AD detection protein markers can be used as target markers. When using -, a certain percentage of the target marker, for example 50% or more, preferably 70% or more More preferably, 90% or more, and even more preferably 100%, whether or not the above criteria are met. Based on this, AD can be detected in the subject.

[0052] The target marker is selected from the group consisting of proteins shown in Tables 8 and 10-1 to 10-2. If there is at least one such type, the concentration of the target marker in the subject is compared to that of the healthy control group. If the concentration is low, the subject may be detected as having AD. For example, the concentration of the target marker in the subject If the degree is statistically significantly lower compared to the healthy control group, the subject may be detected as having AD. For example, the concentration of the target marker in the subject is preferably 90% or less compared to the healthy control group. More preferably 80% or less, and even more preferably 75% or less, the subject is tested for AD. It may be possible when using two or more AD detection protein markers as target markers. This includes a certain percentage of the target marker, for example, 50% or more, preferably 70% or more. Based on whether 90% or more, and more preferably 100%, of the above criteria are met, It can detect AD in the test subject.

[0053] The concentration of the protein marker for AD detection in SSL was determined by ELISA, immunohistochemistry, fluorescence, and electrophoresis. Conventional protein detection or quantification methods such as pneumatophoresis, chromatography, and mass spectrometry are used. It can be measured using this method. Of these, mass spectrometry such as LC-MS / MS is preferred. For concentration measurement, the above SSL-derived protein is used as a sample, and the target is measured according to the usual procedure. The detection or quantification of one or more protein markers is performed. The concentration of the target marker is based on the absolute amount of the target marker in SSL, This could also be the relative concentration of other standard substances in L or the total protein.

[0054] The healthy control group may be a population that does not suffer from Alzheimer's disease. The healthy control group may be organisms of the same species as the subject. Any group of subjects is acceptable. If necessary, select a group to make up the healthy control group according to the characteristics of the subjects. You may choose to use a healthy group. For example, if the subjects are infants, you may use a healthy group of infants as the control group. This may involve using a healthy adult population as the control group when the subject is an adult. This can be done. The concentration of the AD detection protein marker in the healthy control group is determined by the procedure described above. It can be measured. Preferably, the concentration of the marker in the healthy control group is measured in advance. There are. More preferably, in the healthy control group, Tables 7-1 to 7-4, Table 8, Tables 9-1 to 9-7 and The concentrations of all markers shown in Tables 10-1 to 10-2 have been measured in advance.

[0055] Alternatively, select from the group of proteins shown in Tables 7-1 to 7-4 and Tables 9-1 to 9-7. A group consisting of at least one selected protein and the proteins shown in Tables 8 and 10-1 to 10-2. It may be used as a target marker in combination with at least one more selected type. The detection criteria are the same as above.

[0056] In one embodiment of the AD detection method of the present invention, when the subject is an infant, the target marker is A small number of protein markers for AD detection can be selected from those shown in Tables 7-1 to 7-4 and Table 8. At least one type is preferred, and if the subject is an adult, the target markers are Tables 9-1 to 9-7 and Table 1 At least one selected from the AD detection protein markers shown in 0-1 to 10-2 Seeds are preferable.

[0057] Another preferred example of an AD detection protein marker for infants is shown in Table 11-1 below. The following 127 types of proteins are listed in Tables 11-1 to 11-4. The protein was extracted from the silicic sacs of healthy children and children with Alzheimer's disease, and quantitative values ​​were obtained in over 75% of all subjects. Of the proteins obtained, those present in children with AD were 1.5 times or more in proportion compared to healthy children (p This refers to proteins that have increased to ≤0.05 or decreased to 0.75 times or less (p ≤ 0.05). Another preferred example of an adult AD detection protein marker is shown in Table 14-1~1 below. The 220 proteins shown in 4-7 are cited as preferred examples. (Table 14-1~14-) The proteins shown in 7 were extracted from SSLs of healthy adults and adult AD patients, and were used in all subjects. Of the proteins for which quantitative values ​​were obtained in 75% or more of cases, the relative abundance in AD patients compared to healthy individuals was The number of factors increased by 1.5 times or more (p ≤ 0.05) or decreased to 0.75 times or less (p ≤ 0.05). It is protein.

[0058] Therefore, in another embodiment of the AD detection method of the present invention, when the subject is an infant The target marker is selected from the AD detection protein markers shown in Tables 11-1 to 11-4. At least one of the selected types is preferred, and if the subject is an adult, the target marker is as shown in Table 14-1~ At least one protein marker selected from the AD detection protein markers shown in 14-7 is preferred. Alternatively, if the subjects include both infants and adults, refer to Tables 11-1 to 11-4. At least one protein selected from the group consisting of the proteins shown in Tables 14-1 to 14-7, and A target marker is created by combining at least one protein selected from the group consisting of the proteins shown. It may be used as such.

[0059] In a further embodiment, the AD detection method of the present invention is the AD in the SSL of the subject. Predictions constructed using the amount of the detection protein marker (e.g., marker concentration in SSL) This includes detecting AD based on a measurement model. For example, the protein marker for AD detection Using the amount of - as the explanatory variable and whether or not it is AD as the dependent variable, a machine learning algorithm is used to perform AD We will construct a discriminant formula (predictive model) to distinguish between patients and healthy individuals. Using this discriminant formula, we will... It can detect the amount (concentration) of the marker, whether it is an absolute value or a relative value. It may also be normalized.

[0060] In one embodiment, quantitative values ​​of target markers derived from SSLs of AD patients and healthy SSLs of an AD patient are compared. Using quantitative values ​​of the target markers derived from the source as a training sample, a discriminant formula is used to distinguish between AD patients and healthy individuals. A predictive model is constructed, and based on this discriminant formula, patients with atopic dermatitis and healthy individuals are distinguished. The cutoff value (reference value) is determined. Then, the target marker is obtained from the SSL sample taken from the subject. The quantity is measured, the obtained measurement value is substituted into the discriminant formula, and the result obtained from the discriminant formula is used as a reference. By comparing it with the illumination value, the presence or absence of AD in the subject can be detected.

[0061] The variables used to construct a discriminant include explanatory variables and dependent variables. Examples of explanatory variables include... , the expression level of the AD detection protein marker selected by the following method (e.g., SSL medium concentration) The degree can be used. As the dependent variable, for example, whether the sample is derived from a healthy person or A You can use the question, "Is it from patient D (presence or absence of AD)?"

[0062] Features include statistically significant differences between the two groups being classified, for example, differences in expression levels between the two groups. Proteins that show significant fluctuations (differentially expressed proteins) are designated as characteristic proteins, and their expression levels Bell can be used. Also, feature proteins are publicly known, such as machine learning algorithms. It can be extracted using the following algorithm. For example, the random forest shown below Using the expression levels of proteins with high variable importance in the dataset, or using the R language's "Borut" Feature proteins can be extracted using packages such as the "a" package.

[0063] As for algorithms used in constructing discriminant formulas, public algorithms such as those used in machine learning are available. Knowledge can be utilized. An example of a machine learning algorithm is Random Forex. Random forest (SVM), a linear kernel support vector machine (SVM) Support Vector Machine (SVM rbf) kernel (linear), rbf kernel Linear network, Generalized Linear Model d linear model), regularized linear discriminant analysis (Regularized li (near discriminant analysis), regularized logistic regression Examples include (Regularized logistic regression). The constructed prediction model is then populated with validation data to calculate the predicted value, and this predicted value is then compared to the actual measured value. The model that best fits, for example, the model with the highest accuracy, is optimal. It can be selected as a predictive model. Furthermore, the detection rate (Recalculated) can be calculated from the predicted and actual values. l) Calculate the accuracy and the F-value, which is their harmonic mean, and then calculate the F The model with the largest value can be selected as the optimal predictive model.

[0064] When using the Random Forest algorithm in constructing a discriminant, the predictive model As an indicator of accuracy, the out-of-bounds (OOB) error rate (the rate at which estimations are incorrect for unknown data) is used. ) can be calculated (Breiman L. Machine Learning (2001) 45;5-32). Random In the forest, we use a method called bootstrap to extract duplicates from all samples. Please forgive me, but we randomly select about two-thirds of the sample size and classify them using a method called a decision tree. Create a container. At this time, any samples that were not extracted will be marked as Out of Bug (OOB). It is called upon to make an out-of-bounds (OOB) prediction of the target variable using a single decision tree and compare it with the correct label. This allows us to calculate the error rate (OOB error rate in decision trees). te). Repeat the same process 500 times and check the out-of-bounds error in 500 decision trees. The average value of the rate is used as the OOB error for the random forest model. It can be expressed as "or rate".

[0065] Note that the number of decision trees (ntree value) used to construct a random forest model is the default value. The standard number of strings is 500, but this can be changed to any number as needed. Furthermore, The number of variables (mtry value) used to create the discriminant expression for a sample in a single decision tree is, The FORT is the square root of the number of explanatory variables, but it can range from one to all explanatory variables as needed. The value of mtry can be changed to any of the following values ​​up to a certain number. You can use the "ret" package. Run the methods of the "caret" package Specify a dam forest and try 8 different mtry values, for example, when Accuracy is the highest The mtry value can be selected as the optimal mtry value. The number of trials can be changed to any number of trials as needed.

[0066] When using the Random Forest algorithm in constructing a discriminant, the model structure The importance of the explanatory variables used in the construction can be quantified (variable importance). For example, the mean decrease in the Gini coefficient can be used. It is possible.

[0067] The method for determining the cutoff value (reference value) is not particularly restricted and can be determined according to a known method. This can be done. For example, an ROC (Receiver Operator) created using a discriminant can be created. It can be obtained from the characteristic curve. In the ROC curve, the vertical axis represents the probability (sensitivity) of a positive result in a positive patient, and the horizontal axis represents the probability of a negative result. The probability of getting a negative result in a sexually transmitted infection (specificity) is subtracted from 1 (false positive rate), which is the probable result. The ROC curve shows "true positive (sensitivity)" and "false positive (1-specificity)". Regarding this, the value that maximizes "true positive (sensitivity)" - "false positive (1 - specificity)" (Youden The index can be used as the cutoff value (reference value).

[0068] When building a predictive model using data from a large number of proteins, if necessary, You may compress the data using PCA (Periodic Computer Analysis) before building a predictive model. Then, dimensionality reduction is performed by principal component analysis of the protein quantification values, and the main principal components are predicted using a model. These can be used as explanatory variables for constructing the formula.

[0069] As shown in the examples described below, Table 11-1~1 shows differences in expression between healthy children and children with AD. Proteins 1-4 were used as feature proteins, and their quantitative data (Log2(Abundant) The nce+1) value is used as the explanatory variable, and healthy children and children with AD are used as the dependent variable, and a machine learning algorithm is used. Random forest (Breiman L. Machine Learning (2001) 45;5-32) is used as the rhythm. We attempted to build a detection model. The constructed prediction model makes it possible to predict infant AD. This was demonstrated. Furthermore, as shown in the examples described later, it was observed in healthy adults and adult AD patients. For the proteins in Tables 14-1 to 14-7 that showed fluctuations, a similarly constructed predictive model was also used. Dell demonstrated that it is possible to predict adult AD. Therefore, the AD detection method of the present invention In one embodiment, the subjects are infants, and the target markers are as shown in Tables 11-1 to 11-4. These are the 127 proteins shown. In another embodiment of the AD detection method of the present invention, The subjects were adults, and the target markers were 220 types of proteins shown in Tables 14-1 to 14-7. It's about quality.

[0070] As shown in the examples described later, healthy children and children with AD were used as subjects, and SSL was collected from the subjects. The quantitative data of the derived protein (Log2(Abundance+1) value) was used as the explanatory variable. The target variables were healthy children and children with AD, and the machine learning algorithm used was Random Forest. We attempted to extract feature proteins and construct a prediction model using this method. During the model construction process, calculations were performed. Based on the resulting Gini coefficients, the top 140 proteins in terms of variable importance (Table 12-1~12) -4) was selected as a feature protein, and a predictive model was constructed using it. The predictive model demonstrated that it is possible to predict infant and toddler AD. Furthermore, the examples described below show that As shown, the subjects were healthy individuals (adults) and AD patients (adults), and SSL-derived samples were taken from the subjects. Using quantitative protein data (Log2(Abundance+1) value), similarly, We attempted to extract characteristic proteins and construct a predictive model. Based on the Gini coefficient, we considered the importance of the variables. We selected 110 types of proteins (Table 15-1 to 15-4) as characteristic proteins, A predictive model was constructed using this. The constructed predictive model makes it possible to predict adult AD. It was shown that... Therefore, in one embodiment of the AD detection method of the present invention, the subject is The subjects are infants and young children, and the target markers are the 140 proteins shown in Tables 12-1 to 12-4. In another embodiment of the AD detection method of the present invention, the subject is an adult, and the target marker - refers to the 110 types of proteins shown in Tables 15-1 to 15-4.

[0071] As shown in the examples described below, healthy children and children with AD were used as subjects, and SSL from the subjects The quantitative data of the protein (Log2(Abundance+1) value) is used as the explanatory variable. The target variables are healthy children and children with AD, and the machine learning algorithm used is the Boruta method (Kurs Using a et al. Fundamental Informaticae (2010) 101;271-286), feature proteins Extraction was performed (maximum number of trials: 1000, p-value: less than 0.01). 35 types of proteins were extracted (Table). 13) was extracted as a feature protein. Quantitative data of these proteins were extracted as features. The predictive model constructed using the random forest method can predict infant and toddler AD. This was demonstrated. Furthermore, as shown in the examples described later, healthy individuals (adults) and AD patients (adults) Subjects were selected, and quantitative data of SSL-derived proteins from the subjects (Log2(Abund Feature protein extraction was performed similarly using the nce+1) value as an explanatory variable. 24 types The protein (Table 16) was extracted as a characteristic protein. Using this, the same process was performed. It has been shown that a predictive model built using DamForest can predict adult Alzheimer's disease. Therefore, in another embodiment of the AD detection method of the present invention, the subject is an infant The protein markers for AD detection are the 35 proteins shown in Table 13. Another embodiment of Akira's AD detection method involves an adult subject and an AD detection tamper. The protein markers are the 24 proteins shown in Table 16.

[0072] Tables 7-1 to 7-7 show the following protein markers for AD detection extracted from the above-mentioned AD detection protein markers through differential expression analysis. -4, 130 types of proteins (A) included in Table 8 and any of Tables 11-1 to 11-4, Table 12-1~12- selected as feature proteins by random forest analysis. The 140 types of proteins shown in 4 (B), and the characteristic proteins identified by the Boruta method The union (A∪B∪C) of the 35 proteins (C) selected in Table 13 is shown in Table 4-1. These are the 200 types of proteins shown in Tables 4-4-6. The proteins are as shown in Tables 4-1 to 4-6. At least one protein selected from the group is preferred in the present invention for infants with AD. It is used as a detection marker. The amount of at least one protein is measured in the subject and in the healthy control group. By comparing with the above, infant AD can be detected. Alternatively, at least A predictive model is constructed using one type of protein as a feature protein, and based on this predictive model... This allows for the detection of AD in infants and young children.

[0073] [Table 4-1]

[0074] [Table 4-2]

[0075] [Table 4-3]

[0076] [Table 4-4]

[0077] [Table 4-5]

[0078] [Table 4-6]

[0079] Of the proteins shown in Tables 4-1 to 4-6 above, POF1B (Protein POF1B), MNDA (My eloid cell nuclear differentiation antigen), SERPINB4 (Serpin B4), CLEC3B (Tet ranectin), PLEC (Plectin), LGALS7 (Galectin-7), H2AC4 (Histone H2A type 1-B / E ), SERPINB3 (Serpin B3), AMBP (Protein AMBP), PFN1 (Profilin-1), DSC3 (Desmo collin-3), IGHG1(Immunoglobulin heavy constant gamma 1), ORM1(Alpha-1-acid g lycoprotein 1), RECQL (ATP-dependent DNA helicase Q1), RPL26 (60S ribosomal pr otein L26), KLK13 (Kallikrein-13), RPL22 (60S ribosomal protein L22), APOA2 ( Apolipoprotein A-II), SERPINB5 (Serpin B5), LCN15 (Lipocalin-15), IGHG3 (Immu noglobulin heavy constant gamma 3), CAP1 (Adenylyl cyclase-associated protein 1) The 23 proteins consisting of ) and SPRR2F (Small proline-rich protein 2F) are as described above. These are proteins common to proteins (A), (B), and (C). At least one protein selected from the group consisting of proteins is used in this invention. It is used as a preferred marker for detecting AD in infants. The at least one protein By comparing the amount of [the substance] between the test subject and the healthy control group, infant and child Alzheimer's disease (AD) can be detected. Alternatively, a predictive model can be constructed using at least one of these proteins as a feature protein. Based on this predictive model, infant AD can be detected.

[0080] In a preferred embodiment of the present invention, from SSL collected from infant subjects, the 2 At least one, preferably two or more, selected from the group consisting of three types of proteins, Preferably 5 or more types, more preferably 10 or more types, and even more preferably all proteins. To quantify. Furthermore, in this invention, a small number of proteins selected from the group consisting of the 23 proteins mentioned above are used. In addition to one type of protein, there are 200 types of proteins shown in Tables 4-1 to 4-6 below. One or more proteins selected from the group consisting of proteins (excluding the 23 types of proteins mentioned above) Quantitative analysis may be performed. For example, at least selected from the group consisting of the 23 proteins. In addition to one type of protein, there are 127 other types of proteins shown in Tables 11-1 to 11-4 (previous At least one protein selected from the group consisting of the following 23 proteins (excluding those listed in Table 12-1~) A group consisting of 140 types of proteins (excluding the 23 types of proteins mentioned above) as shown in 12-4. At least one selected from the above, and / or 35 proteins shown in Table 13 (the above) At least one protein selected from the group consisting of (excluding 23 proteins) may be quantified. When selecting a protein from Tables 11-1 to 11-4, the significance of the expression change should be considered. You may also preferentially select those with higher values ​​(e.g., smaller p-values). See also Table 12. When selecting proteins from -1 to 12-4, choose the protein with the highest variable importance first. Proteins that are selected preferentially, or those ranked from the top 50, preferably 30th, in terms of variable importance. You may select from the group. The amount of at least one type of protein as described above is compared between the subject and the healthy control group. By comparing with the above, infant AD can be detected. Alternatively, the above A predictive model is constructed using at least one protein as a feature protein, and the predictive model Infant AD can be detected based on this information.

[0081] Table 9-1 to 9: Of the above-mentioned protein markers for detecting AD, those extracted by differential expression analysis. -7, Tables 10-1 to 10-2 and 14-1 to 14-7 show 242 types of proteins (D ) and Table 15-1~15- selected as feature proteins by random forest The 110 types of proteins shown in 4 (E), and the characteristic proteins identified by the Boruta method The union (D∪E∪F) of the 24 proteins (F) selected in Table 16 is shown in Table 5-1. These are the 283 types of proteins shown in Tables 5-9. From the proteins shown in Tables 5-1 to 5-9... At least one protein selected from the group is preferred in the present invention for adult AD testing. It is used as a protein marker for testing. The amount of at least one of these proteins is tested. By comparing the patient with a healthy control group, adult Alzheimer's disease (AD) can be detected. Alternatively, the small At the very least, construct a predictive model using one type of protein as a feature protein, and the predictive model Adult AD can be detected based on this.

[0082] [Table 5-1]

[0083] [Table 5-2]

[0084] [Table 5-3]

[0085] [Table 5-4]

[0086] [Table 5-5]

[0087] [Table 5-6]

[0088] [Table 5-7]

[0089] [Table 5-8]

[0090] [Table 5-9]

[0091] Of the proteins shown in Tables 5-1 to 5-9 above, SERPINB1 (Leukocyte elastase inh ibitor), TTR (Transthyretin), DHX36 (ATP-dependent DNA / RNA helicase DHX36), I TIH4 (Inter-alpha-trypsin inhibitor heavy chain H4), GC (Vitamin D-binding prot ein), ALB (Serum albumin), SERPING1 (Plasma protease C1 inhibitor), DDX55 (AT P-dependent RNA helicase DDX55), IGHV1-46 (Immunoglobulin heavy variable 1-46) , EZR (Ezrin), VTN (Vitronectin), AHSG (Alpha-2-HS-glycoprotein), HPX (Hemope xin), PPIA (Peptidyl-prolyl cis-trans isomerase A), KNG1 (Kininogen-1), FN1 ( Fibronectin), PLG (Plasminogen), PRDX6 (Peroxiredoxin-6) and FLG2 (Filaggrin-2) The 19 proteins consisting of ) are common to the above proteins (D), (E), and (F). It is a protein. At least one protein selected from these 19 types of proteins. The substance is used in the present invention as a more preferred marker for detecting adult AD. Adult Alzheimer's disease can be detected by comparing the amount of at least one type of protein between the test subject and the healthy control group. This can be done. Alternatively, the at least one protein can be used as a characteristic protein. A predictive model can be constructed, and adult AD can be detected based on this predictive model.

[0092] In a preferred embodiment of the present invention, from SSL collected from an adult subject, the 19 At least one, preferably two or more, selected from the group consisting of proteins of different species. More preferably 5 or more types, more preferably 10 or more types, and even more preferably all proteins Quantitatively. Furthermore, in this invention, at least one selected from the group consisting of the 19 proteins mentioned above. In addition to one type of protein, there are 283 other proteins as shown in Tables 5-1 to 5-9 below. At least one protein selected from the group consisting of (excluding the 19 proteins mentioned above) The quality may be quantified. For example, at least one selected from the group consisting of the 19 proteins. In addition to one type of protein, there are 220 other proteins shown in Tables 14-1 to 14-7. At least one selected from the group consisting of (excluding the 19 proteins mentioned above), Table 15- From the 110 types of proteins shown in 1-15-4 (excluding the 19 types of proteins mentioned above) At least one selected from the group and / or one of the 24 proteins shown in Table 16 ( Even if at least one selected from the group consisting of (excluding the 19 proteins mentioned above) is quantified, Good. In this case, when selecting a protein from Tables 14-1 to 14-7, consider the possibility of expression changes. You may also preferentially select those with higher intention (e.g., smaller p-values). When selecting proteins from 15-1 to 15-4, choose the protein with the higher variable importance. Prioritize selecting from these, or select the top 50, preferably 30, tampers based on their variable importance. The protein may be selected from the group. The amount of at least one of these proteins may be compared between the subject and the healthy control group. By comparing them, adult AD can be detected. Alternatively, at least one of the following types A predictive model was constructed using protein as a feature protein, and based on this predictive model, adult A D can be detected.

[0093] Exemplary embodiments of the present invention include the following substances, manufacturing methods, uses, methods, etc., as described in this specification. Disclosed in this document. However, the present invention is not limited to these embodiments.

[0094] [1] Proteins shown in Tables 1-1 to 1-13 above, obtained from skin surface lipids collected from the subject. Atopic A method for preparing protein markers for detecting genital dermatitis. [2] Proteins shown in Tables 1-1 to 1-13 above, obtained from skin surface lipids collected from the subjects. The method involves detecting at least one protein selected from the group consisting of the following in the subject. A method for detecting atopic dermatitis. [3] Preferably, at least one of the proteins At least one protein selected from the group of proteins shown in Tables 2-1 to 2-5. Is it of a certain quality? At least one protein selected from the group of proteins shown in Tables 3-1 to 3-2. It is of a certain quality. The method described in [1] or [2]. [4] The subject is preferably an infant or young child. The at least one of the aforementioned proteins Preferably, at least one protein selected from the group consisting of the proteins shown in Tables 4-1 to 4-6. It is also a type of protein, More preferably, selected from the group consisting of proteins shown in Tables 7-1 to 7-4 and Table 8. It is at least one type of protein, More preferably, selected from the group consisting of proteins shown in Tables 11-1 to 11-4. It is at least one type of protein, or one of the proteins shown in Tables 12-1 to 12-4. It is at least one protein selected from the group shown, or one of the proteins shown in Table 13. It is at least one protein selected from the group consisting of chromosomes, More preferably, POF1B, MNDA, SERPINB4, CLEC3B, PLEC, LGALS7, H2AC4, SERPIN B3, AMBP, PFN1, DSC3, IGHG1, ORM1, RECQL, RPL26, KLK13, RPL22, APOA2, SERPINB5, At least one protein selected from the group consisting of LCN15, IGGH3, CAP1, and SPRR2F Including, More preferably, POF1B, MNDA, SERPINB4, CLEC3B, PLEC, LGALS7, H2AC4, SERPIN B3, AMBP, PFN1, DSC3, IGHG1, ORM1, RECQL, RPL26, KLK13, RPL22, APOA2, SERPINB5, At least one protein selected from the group consisting of LCN15, IGGH3, CAP1, and SPRR2F The group consisting of proteins shown in Tables 11-1 to 11-4, Tables 12-1 to 12-4, and Table 13. A combination with at least one other protein that is more likely to be selected. The method described in [1] or [2]. [5] The subject is preferably an adult, The at least one of the aforementioned proteins Preferably, at least one protein selected from the group consisting of the proteins shown in Tables 5-1 to 5-9. It is also a type of protein, More preferably, from the proteins shown in Tables 9-1 to 9-7 and Tables 10-1 to 10-2 It is at least one protein selected from the following group, More preferably, selected from the group consisting of proteins shown in Tables 14-1 to 14-7. It is at least one type of protein, or one of the proteins shown in Tables 15-1 to 15-4. It is at least one protein selected from the group shown, or one of the proteins shown in Table 16. It is at least one protein selected from the group consisting of chromosomes, More preferably, SERPINB1, TTR, DHX36, ITIH4, GC, ALB, SERPING1, DDX55, IGH selected from the group consisting of V1-46, EZR, VTN, AHSG, HPX, PPIA, KNG1, FN1, PLG, PRDX6 and FLG2 and contains at least one protein selected therefrom, more preferably, SERPINB1, TTR, DHX36, ITIH4, GC, ALB, SERPING1, DDX55, IGH selected from the group consisting of V1-46, EZR, VTN, AHSG, HPX, PPIA, KNG1, FN1, PLG, PRDX6 and FLG2 and at least one other protein selected from the group consisting of the proteins shown in Tables 14-1 to 14-7, Tables 15-1 to 15-4 and Table 16, is a combination with, the method according to [1] or [2]. 〔6〕Preferably, the subject is an infant, the at least one protein is preferably at least one protein selected from the group consisting of the proteins shown in Tables 7-1 to 7-4, and the method preferably includes detecting the subject as having atopic dermatitis when the concentration of the at least one protein is increased compared to the healthy group. the method according to [2]. 〔7〕Preferably, the subject is an infant, the at least one protein is preferably at least one protein selected from the group consisting of the proteins shown in Table 8, and the method preferably includes detecting the subject as having atopic dermatitis when the concentration of the at least one protein is decreased compared to the healthy group. the method according to [2]. 〔8〕Preferably, the subject is an adult, the at least one protein is preferably at least one protein selected from the group consisting of the proteins shown in Tables 9-1 to 9-7, and the method preferably includes detecting the subject as having atopic dermatitis when the concentration of the at least one protein is decreased compared to the healthy group. the method according to [2]. 〔8〕Preferably, the subject is an adult, the at least one protein is preferably at least one protein selected from the group consisting of the proteins shown in Tables 9-1 to 9-7, At least one protein selected from the following group, wherein when the concentration of the at least one protein is preferably increased compared to a healthy group the method includes detecting the subject as having atopic dermatitis. The method according to item [2]. 〔9〕The subject is preferably an adult, The at least one protein is preferably at least one protein selected from the group consisting of the proteins shown in Tables 10-1 to 10-2 wherein when the concentration of the at least one protein is preferably decreased compared to a healthy group the method includes detecting the subject as having atopic dermatitis. The method according to item [2]. 〔10〕The method preferably includes detecting AD based on a prediction model constructed with the concentration of the at least one protein as an explanatory variable and the presence or absence of AD as an objective variable, more preferably, detecting AD based on a cut-off value for discriminating patients with atopic dermatitis and healthy subjects, the cut-off value being calculated from a discriminant formula constructed with the concentration of the at least one protein derived from patients with atopic dermatitis and the concentration of the protein derived from healthy subjects as teacher samples, substituting the concentration of the at least one protein obtained from the skin surface lipids of the subject into the discriminant formula, and comparing the obtained result with the cut-off value to evaluate the presence or absence of atopic dermatitis in the subject. The method according to any one of items [2] to [5]. 〔11〕Preferably, detecting skin surface lipids derived from a subject having or suspected of having atopic dermatitis. The method according to any one of items [2] to

[10] . 〔11〕Preferably, detecting skin surface lipids derived from a subject having or suspected of having atopic dermatitis. The method according to any one of items [2] to

[10] . The method according to any one of items [2] to

[10] . The method according to any one of items [2] to

[10] . The method according to any one of items [2] to

[10] . The method according to any one of items [2] to

[10] . The method according to any one of items [2] to

[10] . The method according to any one of items [2] to [5]. 〔11〕Preferably, detecting skin surface lipids derived from a subject having or suspected of having atopic dermatitis. The method according to any one of items [2] to

[10] . The method according to any one of items [2] to

[10] .

[12] The subject is preferably an infant, The at least one of the aforementioned proteins is preferably one of the proteins shown in Tables 7-1 to 7-4. It is at least one protein selected from the following group, Preferably, the above method increases the concentration of at least one protein compared to the healthy group. If the above-mentioned lipids on the skin surface are present in a person with atopic dermatitis or suspected of having atopic dermatitis. Including detection as originating from the subject, The method described in

[11] .

[13] The subject is preferably an infant, The at least one of the aforementioned proteins is preferably from the group consisting of the proteins shown in Table 8. At least one selected protein, Preferably, the above method results in a decrease in the concentration of at least one protein compared to the healthy group. If the above-mentioned lipids on the skin surface are present in a person with atopic dermatitis or suspected of having atopic dermatitis. Including detection as originating from the subject, The method described in

[11] .

[14] The subject is preferably an adult, The at least one of the aforementioned proteins is preferably one of the proteins shown in Tables 9-1 to 9-7. It is at least one protein selected from the following group, Preferably, the above method increases the concentration of at least one protein compared to the healthy group. If the above-mentioned lipids on the skin surface are present in a person with atopic dermatitis or suspected of having atopic dermatitis. Including detection as originating from the subject, The method described in

[11] .

[15] The subject is preferably an adult, The at least one of the aforementioned proteins is preferably one of the proteins shown in Tables 10-1 to 10-2. It is at least one protein selected from the group consisting of qualities, The method preferably detects the skin surface lipid as being derived from a subject having or suspected of developing atopic dermatitis when the concentration of at least one of the proteins is decreased as compared with the healthy group. When it is like this, including detecting it as being derived from a subject. The method according to

[11] . Preferably, [1] further includes collecting the skin surface lipid from the subject. The method according to any one of [1] to

[15] .

[0095] 〔17〕A protein marker for detecting atopic dermatitis, comprising at least one selected from the group consisting of the proteins shown in Tables 1-1 to 1-13. Preferably, the at least one protein is at least one protein selected from the group consisting of the proteins shown in Tables 2-1 to 2-5, or the at least one protein is at least one protein selected from the group consisting of the proteins shown in Tables 3-1 to 3-2. The marker according to

[17] . 〔19〕Preferably, the marker is a marker for detecting infantile atopic dermatitis, and the at least one protein is preferably at least one protein selected from the group consisting of the proteins shown in Tables 7-1 to 7-4 and Table 8, more preferably at least one protein selected from the group consisting of the proteins shown in Tables 11-1 to 11-4, even more preferably at least one protein selected from the group consisting of the proteins shown in Tables 4-1 to 4-6, even more preferably at least one protein selected from the group consisting of the proteins shown in Tables 4-1 to 4-6, even more preferably at least one protein selected from the group consisting of the proteins shown in Tables 4-1 to 4-6, even more preferably at least one protein selected from the group consisting of the proteins shown in Tables 4-1 to 4-6, even more preferably POF1B, MNDA, SERPINB4, CLEC3B, PLEC, LGALS7, H2AC4, SERPIN B3, AMBP, PFN1, DSC3, IGHG1, ORM1, RECQL, RPL26, KLK13, RPL22, APOA2, SERPINB5, At least one protein selected from the group consisting of LCN15, IGGH3, CAP1, and SPRR2F be, The marker described in

[17] .

[20] The subject is preferably the marker is preferably for detecting adult atopic dermatitis It is a marker, The at least one of the aforementioned proteins Preferably, the proteins consist of those shown in Tables 9-1 to 9-7 and Tables 10-1 to 10-2. It is at least one protein selected from the group, More preferably, the proteins are selected from the group consisting of the proteins shown in Tables 14-1 to 14-7. It is at least one type of protein, More preferably, a small number of proteins selected from the group consisting of the proteins shown in Tables 5-1 to 5-9 are used. At the very least, it is a type of protein. More preferably, SERPINB1, TTR, DHX36, ITIH4, GC, ALB, SERPING1, DDX55, IGH Select from the group consisting of V1-46, EZR, VTN, AHSG, HPX, PPIA, KNG1, FN1, PLG, PRDX6, and FLG2. At least one protein is selected. The marker described in

[17] .

[0096]

[21] At least one selected from the group consisting of proteins shown in Tables 1-1 to 1-13 above. Use of a single protein as a marker for detecting atopic dermatitis.

[22] At least one selected from the group consisting of proteins shown in Tables 1-1 to 1-13 above. Use of a single protein in the production of a protein marker for detecting atopic dermatitis.

[23] Preferably, the at least one protein is At least one protein selected from the group of proteins shown in Tables 2-1 to 2-5. Is it of a certain quality? At least one protein selected from the group of proteins shown in Tables 3-1 to 3-2. It is of a certain quality. Use as described in

[21] or

[22] .

[24] The marker is preferably a marker for detecting atopic dermatitis in infants and young children. The at least one of the aforementioned proteins Preferably, the proteins are selected from the group consisting of those shown in Tables 7-1 to 7-4 and Table 8. It is at least one type of protein, More preferably, the proteins are selected from the group consisting of the proteins shown in Tables 11-1 to 11-4. It is at least one type of protein, More preferably, a small number of proteins selected from the group consisting of the proteins shown in Tables 4-1 to 4-6. At the very least, it is a type of protein. More preferably, POF1B, MNDA, SERPINB4, CLEC3B, PLEC, LGALS7, H2AC4, SERPIN B3, AMBP, PFN1, DSC3, IGHG1, ORM1, RECQL, RPL26, KLK13, RPL22, APOA2, SERPINB5, At least one protein selected from the group consisting of LCN15, IGGH3, CAP1, and SPRR2F be, Use as described in

[21] or

[22] .

[25] The subject is preferably the marker is preferably for detecting adult atopic dermatitis It is a marker, The at least one of the aforementioned proteins Preferably, the proteins consist of those shown in Tables 9-1 to 9-7 and Tables 10-1 to 10-2. It is at least one protein selected from the group, More preferably, the proteins are selected from the group consisting of the proteins shown in Tables 14-1 to 14-7. It is at least one type of protein, More preferably, a small number of proteins selected from the group consisting of the proteins shown in Tables 5-1 to 5-9 are used. At the very least, it is a type of protein. More preferably, SERPINB1, TTR, DHX36, ITIH4, GC, ALB, SERPING1, DDX55, IGH Select from the group consisting of V1-46, EZR, VTN, AHSG, HPX, PPIA, KNG1, FN1, PLG, PRDX6, and FLG2. At least one protein is selected. Use as described in

[21] or

[22] . [Examples]

[0097] The present invention will be described in more detail below based on examples, but the present invention is not limited thereto. No.

[0098] Example 1: Expression variation of atopic dermatitis-related tannins using infant SSL-derived protein Protein identification 1) Subjects and SSL collection 23 healthy children (6 months to 5 years old, male and female) (healthy group), and children with atopic dermatitis (AD group) Sixteen children (6 months to 5 years old, male and female) with AD were included as subjects. Recruitment of children with AD So, we gathered children with Alzheimer's disease who met the UKWP criteria based on parental assessment, and then... Patients were selected based on parental consent obtained through a referral program. Regarding the selected children with Alzheimer's disease, A dermatologist will examine the entire skin and conduct a medical interview, based on the atopic dermatitis treatment guidelines. A diagnosis of AD was made. Among the children diagnosed with AD, the atopic dermatitis treatment group Based on the severity assessment criteria described in Idline, mild to severe AD-like eczema or dryness on the face. Infants and young children exhibiting symptoms were selected as subjects. The entire face of each subject was examined (excluding the affected area for children with Alzheimer's disease). (Mu) Use an oil-absorbing film (5 x 8 cm, made of polypropylene, 3M) to remove sebum. It was recovered. The oil-absorbing film was transferred to a glass vial and used for protein extraction. It was then stored at -80°C for about a month.

[0099] 2) Protein preparation Cut the oil-absorbing film from 1) above to an appropriate size, QIAzol Lysis Using Reagent (Qiagen), protein precipitation was performed according to the provided protocol. The following was obtained. From the obtained protein precipitate, MPEX PTS Reagent (GL s Using (science), after dissolving the protein in the solubilizer according to the attached protocol, Lipin digestion was performed to obtain a peptide solution. The obtained peptide solution was dried under reduced pressure (35°C). Afterwards, water containing 0.1% formic acid and 2% acetonitrile. It was dissolved in a solution. Pierce TM Quantitative Fluoromet ric Peptide Assay(ThermoFisher Scientifi c) Follow the protocol and use a microplate reader (Corona Electric The peptide concentration in the solution was measured using ). If the required amount of peptide was not obtained, AD patients A peptide solution from one patient was excluded from the following analysis: LC-MS / MS For the analysis, the peptide concentration applied to the MS instrument was kept constant, and the protein quantification value was obtained. The result was calculated.

[0100] 3) LC-MS / MS analysis and data analysis The sample peptide solution obtained in step 2) above was subjected to LC-MS / MS analysis under the conditions shown in Table 6 below. It was used for that purpose.

[0101] [Table 6]

[0102] For the analysis of spectral data obtained by LC-MS / MS analysis, Proteome Discoverer ver.2.2(ThermoFisher Scientif We used (ic). For protein identification, we used the reference databases Swiss Prot and Tax. Set onomy to Homo sapiens, and Mascot database s The search was performed using earch (Matrix Science). The search involved Enzyme Let e ​​be Trypsin, Missed cleavage be 2, and Dynamic mo Oxidation (M), Acetyl (N-term), Acetyl (Protein N-term), Static Modificati Set ons to Carbamidomethyl(C). False disco The search targeted peptides satisfying the very rate (FDR) p < 0.01. Identification Label-free quantitative analysis (LF) of the proteins was performed using precursor ions. Q (Label-Free Quantification) was performed. Peptide-derived Protein quantification values ​​are calculated based on the peak intensity of precursor ions, and the peak intensity is detected. Values ​​below the limit were treated as missing values. The calculation of protein abundance ratios used Summed Abu The gradation-based method was used. To calculate the p-value indicating the significance of the difference between groups, An ANOVA (Individual Based, t-test) was used.

[0103] 4) Results Of the identified proteins, the False Discovery Rate (FDR) Proteins with a value of 0.1 or higher were excluded from the analysis. (Either the healthy control group or the AD group) For over 75% of the subjects, non-missing protein quantification values ​​were calculated for 533 types of proteins. Proteins were extracted for analysis. The abundance ratio was 1.5 times or more in the AD group compared to the healthy control group (p 116 types of proteins that increased to ≤0.05 (Tables 7-1 to 7-4), and 0.75 times or more Twelve proteins that decreased to below (p ≤ 0.05) were identified (Table 8).

[0104] [Table 7-1]

[0105] [Table 7-2]

[0106] [Table 7-3]

[0107] [Table 7-4]

[0108] [Table 8]

[0109] Example 2: Atopic dermatitis-related differential expression protein using adult SSL-derived protein Identification of the substance 1) Subjects and SSL collection Eighteen healthy individuals (20-59 years old, male) (healthy group), and patients with atopic dermatitis (AD patients). The study included 26 male participants (20-59 years old) in the AD group. Consent was obtained through a medical consent form. The subjects in the AD group were assessed by a dermatologist regarding the severity of their condition. The patient has been diagnosed with mild or moderate atopic dermatitis and has mild to moderate AD-like eczema on their face. We selected individuals who exhibited symptoms such as dryness. From each subject's entire face (including the affected area for AD patients) Sebum was collected using an oil-absorbing film (5 x 8 cm, made of polypropylene, 3M). Transfer the oil-absorbing film to a vial and keep at -80°C until ready for protein extraction. It was stored for about a month.

[0110] 2) Protein preparation Replace MPEX PTS Reagent (GL science) with EasyPep TM Mini MS Sample Prep Kit(ThermoFisher S Aside from obtaining the peptide solution using a scientific (Scientific) device according to the attached protocol, The peptide concentration was measured using the same procedure as in Example 1.

[0111] 3) LC-MS / MS analysis and data analysis Protein analysis and data analysis were performed under the same conditions and procedures as in Example 1.

[0112] 4) Results Of the identified proteins, the False Discovery Rate (FDR) Proteins with a value of 0.1 or higher were excluded from the analysis. In either the healthy control group or the AD group... Protein quantification values ​​were calculated for 1075 types of proteins in over 75% of subjects, with no missing values. Proteins were extracted for analysis. Note that many missing values ​​were observed in the protein quantification data. One patient with disease D was excluded from the analysis. The prevalence of disease was 1.5 times or more in the AD group compared to the healthy control group (p 205 types of proteins that increased to ≤0.05 (Table 9-1 to 9-7), and 0.75 times or more We identified 37 proteins that decreased to below (p ≤ 0.05) (Tables 10-1 to 10-2). .

[0113] [Table 9-1]

[0114] [Table 9-2]

[0115] [Table 9-3]

[0116] [Table 9-4]

[0117] [Table 9-5]

[0118] [Table 9-6]

[0119] [Table 9-7]

[0120] [Table 10-1]

[0121] [Table 10-2]

[0122] Example 3: Construction of a discriminant model for detecting atopic dermatitis in infants and young children. (Data used) In order to approximate the quantitative protein data obtained in Example 1 to a normal distribution, The peak intensity before the rise is used as the protein quantification value, and each protein quantification value represents the total detected protein. Log2(Abundance) is the base-2 logarithm of the value obtained by dividing by the sum of the quantitative values. The value of (+1) was calculated. The obtained Log2(Abundance+1) value was used in the machine learning model. It was used to construct the model. Using the same method as in Example 1, more than 75% (more than 29 people) of all subjects were found to be deficient. For 475 types of proteins for which quantitative values ​​other than numerical values ​​were calculated, we extracted them and analyzed them. did.

[0123] 3-1 Construction of a discrimination model using differentially expressed proteins 1) Selection of characteristic proteins Among the 475 types of proteins listed above, those that showed a statistically significant difference in prevalence in children with AD compared to healthy children were... We identified 127 proteins whose expression levels were altered (Tables 11-1 to 11-4). The protein in question was selected as a feature protein, and its quantitative data was used as the feature. 2) Model construction The Log2(Abundance+1) values ​​of the above 127 proteins were used as explanatory variables. The dependent variables used were healthy children and children with AD (presence or absence of AD). The R language's "caret" program... In the package, specify the Random Forest algorithm as the method, and perform a single decision. The optimal number of variables (mtry value) used in constructing the ruler was tuned. Using the mtry value determined by this, the random forest algorithm is executed, O The OB error rate was calculated. As a result, 127 types of proteins were used as feature cells. The model used as a protein had an error rate of 18.42%.

[0124] [Table 11-1]

[0125] [Table 11-2]

[0126] [Table 11-3]

[0127] [Table 11-4]

[0128] 3-2 Construction of a discrimination model using proteins with high variable importance in random forests 1) Selection of characteristic proteins The Log2(Abundance+1) values ​​of the above 475 types of proteins were used as explanatory variables. The dependent variables used were healthy children and children with AD (presence or absence of AD). The R language was used with "caret". In the package, specify the Random Forest algorithm as a method, and in one go The optimal number of variables (mtry value) used in constructing the decision tree was tuned. Using the mtry value determined by the algorithm, the random forest algorithm is executed. The top 140 proteins in terms of variable importance based on the Gini coefficient were calculated (Tables 12-1 to 12-4). ). These 140 types of proteins, and all 475 types used in the selection of characteristic proteins. Proteins were used as feature proteins, and their quantitative data were used as features. 2) Model construction The above 140 types of proteins or the Log2(Abundance) of all 475 types of proteins The e+1) value was used as the explanatory variable, and healthy children and children with AD (presence or absence of AD) were used as the dependent variables. The R language's "caret" package uses the Random Forest algorithm as a method. Specify the optimal value for the number of variables (mtry value) used to construct a decision tree in one go. The tuning was performed. Using the mtry value determined by the tuning, a random forest was created. The algorithm was executed to calculate the estimated out-of-bounds error rate (OOB error rate). As a result, the error rate when using all 475 proteins as feature proteins was 28.95%. In contrast, when the top 140 proteins in terms of variable importance were used as feature proteins... The error rate for matching was 7.89%.

[0129] [Table 12-1]

[0130] [Table 12-2]

[0131] [Table 12-3]

[0132] [Table 12-4]

[0133] 3-3 Construction of a discrimination model using feature proteins extracted by the Boruta method 1) Selection of characteristic proteins The Log2(Abundance+1) values ​​of the above 475 types of proteins were used as explanatory variables. Using healthy children and children with AD (presence or absence of AD) as the dependent variables, the R language "Boruta" was used. The package algorithm was executed. The maximum number of trials was 1000, and the p-value was 0.01. Thirty-five proteins with a value less than 100% were extracted (Table 13) and selected as characteristic proteins. Quantitative data of those proteins were used as features. 2) Model construction The Log2(Abundance+1) values ​​of the above 35 proteins were used as explanatory variables. The dependent variables used were children with normal development and children with AD (presence or absence of AD). The R language "caret" package was used. In the cage, specify the Random Forest algorithm as the method and make a single decision. The optimal number of variables (mtry value) used in building the tree was tuned. Therefore, using the determined mtry value, the random forest algorithm is executed, OO The B error rate was calculated. As a result, 35 types of proteins were used as feature proteins. The model used as a quality indicator had an error rate of 10.53%.

[0134] [Table 13]

[0135] Example 4: Construction of a discriminant model for detecting adult atopic dermatitis (Data used) In order to approximate the quantitative protein data obtained in Example 2 to a normal distribution, The peak intensity before the rise is used as the protein quantification value, and each protein quantification value represents the total detected protein. Log2(Abundance) is the base-2 logarithm of the value obtained by dividing by the sum of the quantitative values. The value of (+1) was calculated. The obtained Log2(Abundance+1) value was used in the machine learning model. This was used to construct the following. In the same manner as in Example 2, all subjects (however, the quantitative data of protein) (Excluding 3 individuals whose data did not follow a normal distribution) More than 75% (more than 31 individuals) had non-missing protein values. 985 types of proteins for which quantitative values ​​were calculated were extracted and used for analysis.

[0136] 4-1 Construction of a discrimination model using differentially expressed proteins 1) Selection of characteristic proteins Among the 985 types of proteins listed above, those whose expression was significantly altered in AD patients compared to healthy individuals were... We identified 220 types of proteins (Tables 14-1 to 14-7) and used them as characteristic proteins. A specific type was selected, and its quantitative data was used as a feature. 2) Model construction The Log2(Abundance+1) values ​​of the above 220 proteins were used as explanatory variables. The dependent variables used were healthy individuals and AD patients (presence or absence of AD). The R language's "caret" package was used. In the cage, specify the Random Forest algorithm as the method and make a single decision. The optimal number of variables (mtry value) used in building the tree was tuned. Therefore, using the determined mtry value, the random forest algorithm is executed, OO The B error rate was calculated. As a result, 220 types of proteins were used as feature proteins. The model used as a quality indicator had an error rate of 24.39%.

[0137] [Table 14-1]

[0138] [Table 14-2]

[0139] [Table 14-3]

[0140] [Table 14-4]

[0141] [Table 14-5]

[0142] [Table 14-6]

[0143] [Table 14-7]

[0144] 4-2 Construction of a discrimination model using proteins with high variable importance in random forests 1) Selection of characteristic proteins The Log2(Abundance+1) values ​​of the above 985 types of proteins were used as explanatory variables. The dependent variables used were healthy individuals and AD patients (presence or absence of AD). The R language's "caret" program... In the package, specify the Random Forest algorithm as the method, and perform a single decision. The optimal number of variables (mtry value) used in constructing the ruler was tuned. Using the mtry value determined by [the method], the random forest algorithm is executed, and [the process] The top 110 proteins in terms of variable importance based on two coefficients were calculated (Tables 15-1 to 15-4). These 110 types of proteins, and all 985 types used in the selection of characteristic proteins, Protein was used as the feature protein, and its quantitative data was used as the feature. 2) Model construction The above 110 types of proteins or the Log2(Abundance) of all 985 types of proteins The e+1) value was used as the explanatory variable, and healthy individuals and AD patients (presence or absence of AD) were used as the dependent variable. The R language's "caret" package provides a method for implementing the Random Forest algorithm. Specify the optimal number of variables (mtry value) to use in constructing a decision tree in one go. The tuning was performed. Using the mtry value determined by the tuning, the random forest was... The algorithm was executed to calculate the estimated out-of-bounds error rate (OOB error rate). As a result, the error rate when using all 985 proteins as feature proteins was 29.27%. In contrast, when the top 110 proteins in terms of variable importance were used as feature proteins... The error rate was 12.20%.

[0145] [Table 15-1]

[0146] [Table 15-2]

[0147] [Table 15-3]

[0148] [Table 15-4]

[0149] 4-3 Construction of a discriminant model using features extracted by the Boruta method 1) Selection of features The Log2(Abundance+1) values ​​of the above 985 types of proteins were used as explanatory variables. Using healthy individuals and AD patients (presence or absence of AD) as the dependent variables, the R language "Boruta" program was used. The package algorithm was executed. The maximum number of trials was set to 1000, and the p-value was 0.01. Twenty-four proteins were extracted (Table 16) and selected as characteristic proteins. Quantitative data of these proteins were used as features. 2) Model construction The Log2(Abundance+1) values ​​of the above 24 proteins were used as explanatory variables. The dependent variables used were the normal population and AD patients (presence or absence of AD). The R language's "caret" package was used. In the process, specify the Random Forest algorithm as the method, and use a single decision tree. The optimal number of variables (mtry value) used in the construction was tuned. Using the determined mtry value, the random forest algorithm is executed, and OOB is performed. The error rate was calculated. As a result, 24 types of proteins were used as feature proteins. The model used in this way had an error rate of 19.51%.

[0150] [Table 16]

[0151] A total of 418 types of proteins were obtained from the analysis of the above examples (see Tables 1-1 to 1-13 above). Regarding this, there is a paper that reports an association with AD using text mining (Elsevier). The number was investigated. The search found four or fewer reports related to AD, and the association with AD was described. 147 proteins were confirmed not to be affected (see Tables 2-1 to 2-5 above). These 147 proteins are novel markers for detecting Alzheimer's disease (AD).

Claims

1. A method for preparing a protein marker for detecting atopic dermatitis, comprising: The method includes recovering at least one protein selected from the group consisting of proteins shown in Table 1 below from lipids on the skin surface collected from the entire face of a subject, The subject is an infant. method. 【Table 1】

2. A method for measuring a protein expression level for detecting atopic dermatitis in a subject, comprising: The method includes measuring the expression level of at least one protein selected from the group consisting of proteins shown in Table 2 below from lipids on the skin surface collected from the entire face of a subject; The subject is an infant. method. 【Table 2】

3. The method of claim 2, wherein the detection includes detecting atopic dermatitis based on a constructed predictive model, with the expression level of the at least one protein as an explanatory variable and the presence or absence of atopic dermatitis as a target variable.

4. The method of claim 3, wherein the detection comprises detecting atopic dermatitis based on a cutoff value for distinguishing between atopic dermatitis patients and healthy subjects, the cutoff value being calculated from a discriminant for distinguishing between atopic dermatitis patients and healthy subjects, the cutoff value being constructed using the expression level of the at least one type of protein derived from atopic dermatitis patients and the expression level of the protein derived from healthy subjects as teacher samples, and comprising detecting to evaluate the presence or absence of atopic dermatitis in the subject by substituting the expression level of the at least one type of protein obtained from skin surface lipids collected from the entire face of the subject into the discriminant, and comparing the obtained result with the cutoff value.

Images