Method and system for determining allergic disease

Abstract

The present invention relates to an allergic disease determination method and the like comprising: a step for causing two or more types of probes selected from an allergen and a fragment thereof and a biological sample collected from a subject to contact with each other; a step for detecting a reaction between the probes and an antibody contained in the biological sample; a step for determining a type of the antibody reacted with the probes; a step for inputting data associated with the reaction of each of the at least two types of probes with the antibody and data associated with a type of the bound antibody into a prediction model and acquiring a prediction result; and a step for determining an allergic disease on the basis of the prediction result, wherein the prediction model is a nonlinear model or a model having two or more explanatory variables and two or more terms, the model being created using training data associated with the reaction of at least the probes with the antibody and training data associated with the type of the bound antibody.

Description

[Technical Field]

[0001] The present invention relates to a method for determining allergic diseases, and a kit, biochip, system, etc. used therefor. [Background technology]

[0002] The number of people suffering from allergic diseases is on the rise, and countermeasures are becoming a challenge. Allergic diseases are highly diverse, with numerous allergens causing them and varying symptoms and responses to treatment among patients. Research on allergies has made great strides over the past few decades, but the pathology is more complex than expected, and many aspects remain unknown. This diversity and complexity make it difficult to address allergic diseases. Accurate diagnosis is essential for addressing these diseases, but it is difficult to say that diagnostic methods for the diverse and complex allergic diseases have yet been fully established. The most common method for diagnosing allergic diseases is the measurement of allergen-specific IgE, but in recent years, attempts have been made to combine the results of IgG4 measurements with IgE (Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] WO2010 / 110454 Summary of the Invention [Problem to be solved by the invention]

[0004] There is a need for improved diagnostic methods for allergic diseases. [Means for solving the problem]

[0005] In one aspect, the present invention provides: [1] A method for determining an allergic disease, comprising: contacting two or more probes selected from allergens and fragments thereof with a biological sample collected from a subject; detecting a reaction between the probe and an antibody contained in a biological sample; a step of determining the type of antibody that reacted with the probe; A step of inputting data on the reaction of each of the at least two types of probes with antibodies and data on the type of antibody bound to the probe into a prediction model to obtain a prediction result; and a step of diagnosing an allergic disease based on the prediction result. wherein the prediction model is a nonlinear model or a model with two or more explanatory variables and two or more terms, created using at least training data on the reaction of the probe with an antibody and training data on the type of antibody bound. [2] The method according to [1], wherein the determination of the allergic disease is selected from determining whether the allergic disease is mild or severe and predicting the total ASCA value. [3] A step of contacting two or more probes selected from allergens and fragments thereof with a biological sample collected from a subject, and acquiring data on the reaction between the probes and antibodies contained in the biological sample and data on the type of antibody that reacted; inputting said data into a predictive model to obtain a predicted result; wherein the prediction model is a nonlinear model or a model with two or more explanatory variables and two or more terms, created using at least training data on the reaction of the probe with an antibody and training data on the type of antibody bound to the probe.

[0006] [4] The method according to any one of [1] to [3], wherein the type of antibody is selected from IgE and IgG4. [5] The method according to any one of [1] to [4], wherein the probe comprises overlapping fragments of an allergen. [6] The method according to any one of [1] to [5], wherein the probe is immobilized. [7] The method according to any one of [1] to [6], wherein the data relating to the reaction with the antibody is the intensity of the reaction with the antibody. [8] The method according to any one of [1] to [7], wherein additional training data is used to create the predictive model. [9] The method according to any one of [1] to [8], wherein the allergen is selected from a food allergen, an inhalant allergen, and a contact allergen.

[10] The method according to any one of [1] to [9], wherein the food allergen is selected from casein, lactalbumin, lactoglobulin, ovomucoid, ovalbumin, conalbumin, lysozyme, gliadin, amylase inhibitor, albumin, globulin, gluten, and tropomyosin.

[11] The method according to any one of [1] to

[10] , which is used for monitoring an allergic disease, wherein the biological samples collected from the subject include two or more types of biological samples collected at different times, and the step of determining an allergic disease includes comparing the prediction results obtained for the two or more types of biological samples with each other.

[12] The method according to any one of [1] to

[11] , wherein the allergic disease is food allergy.

[0007]

[13] A kit for use in the method according to any one of [1] to

[12] , which comprises two or more probes selected from allergens and fragments thereof.

[14] A biochip for use in the method according to any one of [1] to

[12] , in which two or more probes selected from allergens and fragments thereof are immobilized on a substrate.

[15] a data acquisition unit that acquires data on the reaction between the probes and antibodies contained in the biological sample and data on the type of the reacted antibodies, obtained by contacting two or more probes selected from allergens and their fragments with a biological sample collected from a subject; a data processing unit that inputs the data into a prediction model and obtains a prediction result; a determination unit that determines an allergic disease based on the prediction result; and Output section that outputs the judgment result wherein the prediction model is a nonlinear model or a model with two or more explanatory variables and two or more terms, created using at least training data on the reaction of the probe with an antibody and training data on the type of antibody bound.

[16] A program for causing a computer to execute the method according to any one of [1] to

[12] .

[17] A storage medium storing the program described in

[16] . [Effects of the Invention]

[0008] The present invention has, for example, one or more of the following effects depending on the embodiment. (1) It is possible to predict the severity of symptoms of allergic diseases. (2) It is possible to predict the response to provocation tests such as oral food challenge tests. (3) It has superior diagnostic ability compared to conventional specific IgE-based methods. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is a block diagram showing the functional configuration of the allergic disease determination system. [Figure 2] FIG. 2 is a flowchart showing an example of data processing in the system of the present invention. [Figure 3] FIG. 3 is a block diagram showing an example of the hardware configuration of a computer that realizes a stand-alone system of the present invention. [Figure 4] FIG. 4 is a schematic diagram of a partial peptide immobilized on beads. [Figure 5] Figure 5 shows the decision tree for prediction models 1-4. The prediction result "1" means severe and "0" means mild. [Figure 6] Figure 6 shows the decision tree for prediction models 1-5. The prediction result "1" means severe and "0" means mild. [Figure 7] Figure 7 shows the ROC curve created from the predicted values ​​for each subject obtained by prediction model 1-1. [Figure 8] FIG. 8 is a dot histogram created from the predicted values ​​for each subject obtained by prediction model 1-1. [Figure 9] Figure 9 shows the ROC curve created from the predicted values ​​for each subject obtained by prediction model 1-2. [Figure 10] FIG. 10 is a dot histogram created from the predicted values ​​for each subject obtained by prediction model 1-2. [Figure 11] Figure 11 shows the ROC curves created from the predicted values ​​for each subject obtained by prediction models 1-3. [Figure 12] FIG. 12 is a dot histogram created from the predicted values ​​for each subject obtained by prediction models 1-3. [Figure 13] FIG. 13 shows ROC curves created from the predicted values ​​for each subject obtained by prediction models 1-4. [Figure 14] FIG. 14 is a dot histogram created from the predicted values ​​for each subject obtained by prediction models 1-4. [Figure 15] Figure 15 shows ROC curves created from the predicted values ​​for each subject obtained by prediction models 1-5. [Figure 16] FIG. 16 is a dot histogram created from the predicted values ​​for each subject obtained by prediction models 1-5. [Figure 17] FIG. 17 shows ROC curves created from the predicted values ​​for each subject obtained by prediction models 1-6. [Figure 18] FIG. 18 is a dot histogram created from the predicted values ​​for each subject obtained using prediction models 1-6. [Figure 19] FIG. 19 shows ROC curves created from the predicted values ​​for each subject obtained by prediction models 1-7. [Figure 20] FIG. 20 is a dot histogram created from the predicted values ​​for each subject obtained by prediction models 1-7. [Figure 21] FIG. 21 shows ROC curves created from the predicted values ​​for each subject obtained by prediction models 1-8. [Figure 22] FIG. 22 is a dot histogram created from the predicted values ​​for each subject obtained by prediction models 1-8. [Figure 23] Figure 23 shows the decision tree for prediction model 2-3. The numbers in the prediction results represent the sufficiency rate of the ASCA value. [Figure 24] Figure 24 shows the decision tree for prediction model 2-3. The numbers in the prediction results represent the sufficiency rate of the ASCA value. [Figure 25] FIG. 25 is a scatter plot created from the predicted and observed values ​​for each subject obtained by prediction model 2-1. [Figure 26] FIG. 26 is a scatter plot created from the predicted and observed values ​​for each subject obtained by prediction model 2-2. [Figure 27] FIG. 27 is a scatter plot created from the predicted and observed values ​​for each subject obtained by prediction model 2-3. [Figure 28] FIG. 28 is a scatter plot created from the predicted and observed values ​​for each subject obtained by prediction model 2-4. [Figure 29] FIG. 29 is a scatter plot created from the predicted and observed values ​​for each subject obtained by prediction model 2-5. [Figure 30] FIG. 30 is a scatter plot created from the predicted and observed values ​​for each subject obtained by prediction model 2-6. DETAILED DESCRIPTION OF THE INVENTION

[0010] The following describes embodiments of the present invention. Note that the materials, configurations, etc. described below do not limit the present invention, and they can be combined with each other or modified in various ways within the scope of the present invention. Unless otherwise defined herein, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art. All patents, applications, and other publications (including online information) referenced herein are incorporated herein by reference in their entirety. This specification also incorporates the contents of the specification and drawings of Japanese Patent Application No. 2020-150142, filed on September 7, 2020, from which this application claims priority.

[0011] 1. How to determine if you have an allergic disease One aspect of the present invention relates to a method for determining an allergic disease (hereinafter, sometimes referred to as the "determination method of the present invention"). contacting two or more probes selected from allergens and fragments thereof with a biological sample collected from a subject; detecting a reaction between the probe and an antibody contained in a biological sample; a step of determining the type of antibody that reacted with the probe; A step of inputting data on the reaction of each of the at least two types of probes with antibodies and data on the type of antibody bound to the probe into a prediction model to obtain a prediction result; and a step of diagnosing an allergic disease based on the prediction result. The prediction model is a nonlinear model or a model with two or more explanatory variables and two or more terms, created using at least training data on the reaction of the probe with an antibody and training data on the type of antibody bound.

[0012] Allergic diseases are diseases in which the immune system overreacts to allergens, and include food allergies, atopic dermatitis, allergic rhinitis, allergic conjunctivitis, bronchial asthma, urticaria, anaphylactic shock, drug allergies, etc. In one embodiment, the allergic disease is a disease involving IgE. In another embodiment, the allergic disease belongs to type I allergy. In a specific embodiment, the allergic disease is food allergy.

[0013] In the determination method of the present invention, an allergen and / or a fragment thereof is used as a probe to detect an antibody (hereinafter, sometimes referred to as a target antibody) contained in a biological sample. Allergens used as probes include, but are not limited to, food allergens, inhalant allergens, contact allergens, etc.

[0014] Food allergens include, but are not limited to, allergens from milk and dairy products, eggs, fish, meat, mollusks, crustaceans, grains, nuts, fruits, vegetables, mushrooms, potatoes, etc. Fish allergens include allergens from salmon, mackerel, salmon roe, bass, flounder, cod, etc. Meat allergens include allergens from chicken, pork, gelatin, etc. Mollusk allergens include allergens from abalone and squid, etc. Crustacean allergens include allergens from shrimp and crab, etc. Grain allergens include allergens from wheat, rye, barley, oats, spelt, Kamut, buckwheat, peanuts, sesame, soybeans, lupin, etc. Nut allergens include allergens from almonds, cashews, walnuts, hazelnuts, walnuts, pecans, Brazil nuts, pistachios, macadamia nuts, etc. Fruit allergens include those from oranges, kiwifruit, bananas, peaches, apples, etc. Vegetable allergens include those from celery and mustard.

[0015] In some embodiments, the allergens include those ingredients listed as specific or quasi-specific ingredients in the Japanese Food Labeling Standards, such as shrimp, crab, wheat, buckwheat, eggs, milk, and peanuts, and quasi-specific ingredients such as almonds, abalone, squid, salmon roe, oranges, cashew nuts, kiwi fruit, beef, walnuts, sesame, salmon, mackerel, soybeans, chicken, bananas, pork, matsutake mushrooms, peaches, yams, apples, and gelatin. In another embodiment, the allergen comprises an allergen required to be declared under Regulation (EU) No. 1169 / 2011 of the European Parliament and of the Council, including gluten-containing grains (wheat, rye, barley, oats, spelt, kamut), shellfish, eggs, fish, peanuts, soybeans, milk and dairy products, tree nuts (almonds, hazelnuts, walnuts, cashews, pecans, Brazil nuts, pistachios, macadamia nuts), celery, mustard, sesame, lupin, and mollusks. In another embodiment, the allergen includes food allergens required for labeling under the U.S. Consumer Protection Act, including milk, eggs, fish (e.g., bass, flounder, cod), shellfish (e.g., crab, lobster, shrimp), tree nuts (e.g., almonds, walnuts, pecans), peanuts, wheat, and soybeans.

[0016] In another embodiment, food allergens include animal and plant allergens. Animal allergens include, for example, casein (αs1-casein, αs2-casein, β-casein, κ-casein, etc.), albumin (lactalbumin, ovalbumin, conalbumin, serum albumin, etc.), globulin (lactoglobulin, immunoglobulin, etc.), tropomyosin, and parvalbumin (β-parvalbumin, etc.). Plant allergens include, for example, coupin (7S globulin (vicilin), 11S globulin (legumin), 13S globulin, etc.), prolamin (α-amylase inhibitor, 2S albumin, nonspecific lipid transfer protein (nsLTP), etc.), PR-10 (Bet v 1 homolog), profilin, and oleosin. In some embodiments, the allergen is selected from casein, albumin, globulin, tropomyosin, parvalbumin, coupin, prolamin, PR-10 (Bet v 1 homolog), profilin, and oleosin.

[0017] In certain embodiments, cow's milk allergens include αs1-casein (Bos d 9), αs2-casein (Bos d 10), β-casein (Bos d 11), κ-casein (Bos d 12), α-lactalbumin (Bos d 4), serum albumin (Bos d 6), β-lactoglobulin (Bos d 5) and immunoglobulins (Bos d 7). In particular embodiments, egg allergens include ovomucoid (Gal d 1), ovalbumin (Gal d 2), ovotransferrin (conalbumin, Gal d 3), and lysozyme C (Gal d 4).

[0018] In certain embodiments, fish allergens include Atlantic salmon tropomyosin (Sal s 4), Atlantic salmon beta-parvalbumin (Sal s 1), beta-parvalbumin (Gad c 1), cichlid tropomyosin (Ore m 4), herring beta-parvalbumin (Clu h 1), carp beta-parvalbumin (Crp c 1), flounder beta-parvalbumin (Lep w 1), rainbow trout beta-parvalbumin (Onc m 1), sardine beta-parvalbumin (Sar sa 1), yellowfin tuna beta-parvalbumin (Thu a 1), swordfish beta-parvalbumin (Xip g 1).

[0019] In particular embodiments, mollusk allergens include abalone tropomyosin (Hall 1), Japanese pacific squid tropomyosin (Tod p 1), Pacific apple snail tropomyosin (Hel as 1), Pacific oyster tropomyosin (Cra g 1), and Sydney rock oyster tropomyosin (Sac g 1). In particular embodiments, shellfish allergens include Indian shrimp tropomyosin (Pen i 1), Marsh shrimp tropomyosin (Met e 1), American lobster tropomyosin (Hom a 1), black tiger prawn tropomyosin (Pen m 1), Brown shrimp tropomyosin (Pen a 1), Vannamei shrimp tropomyosin (Lit v 1), Hemisphere shrimp tropomyosin (Mel l 1), Northern shrimp tropomyosin (Pan b 1), Macrobrachium prawn tropomyosin (Mac r 1), European shrimp crab tropomyosin (Cra c 1), Japanese spiny lobster tropomyosin (Pan s 1), Taiwanese jacob tropomyosin (Por p 1), and Ishigaki nitropomyosin (Cha f 1).

[0020] In a particular embodiment, the cereal allergens are wheat gluten: wheat omega-5 gliadin (Tri a 19), wheat alpha / beta-gliadin (Tri a 21), wheat gamma-gliadin (Tri a 20), wheat high molecular weight glutenin (Tri a 26), wheat low molecular weight glutenin (Tri a 36), wheat alpha amylase inhibitor (monomer, Tri a 15 dimer, Tri a 28, tetramer, Tri a 29, Tria 30), wheat nsLTP (Tri a 14), barley alpha amylase inhibitor (Hor v 15), buckwheat 13S globulin (molecular weight 76 kDa), buckwheat 13S globulin beta-subunit (molecular weight 24 kDa, Fag e 1), buckwheat 7S globulin (vicilin, Fag e 3), buckwheat 2S albumin (Fag e 2), peanut 7S globulin (vicilin, Ara h 1), peanut 11S globulin (legumin, Ara h 3), peanut 2S albumin (Ara h 2, 6, 7), peanut nsLTP (Ara h 9, 16, 17), peanut Bet v 1 (Ara h 8), peanut profilin (Ara h 5), peanut oleosin (Ara h 10, 11, 14, 15), sesame 7S globulin (vicilin, Ses i 3), sesame 11S globulin (legumin, Ses i 6, 7), sesame 2S albumin (Ses i 1, 2), sesame oleosin (Ses i 4, 5), soybean 7S globulin (vicilin, Gly m 5), soybean 11S globulin (legumin, Gly m 6), soybean nsLTP (Gly m 1), soybean 2S albumin (Gly m 8), soybean Bet v 1 (Gly m 4), soybean profilin (Gly m 3), maize nsLTP (Zea m 14), pea 7S globulin (vicilin, Pis s 1), mung bean 7S globulin (vicilin, Vig r 2), mung bean Bet v 1 (Vig r 1), and maize profilin (Zea m 12).

[0021] In certain embodiments, the nut allergens are almond 11S globulin (legumin, Pru du 6), almond nsLTP1 (Pru du 3), almond profilin (Pru du 4), cashew nut 7S globulin (vicilin, Ana o 1), cashew nut 11S globulin (legumin, Ana o 2), cashew nut 2S albumin (Ana o 3), walnut 7S globulin (vicilin, Jug r 2), walnut 11S globulin (legumin, Jug r 4), walnut 2S albumin (Jug r 1), walnut nsLTP (Jug r 3), walnut PR-10 (Jug r 5), walnut profilin (Jug r 7), hazelnut 7S globulin (vicilin, Cor a 11), hazelnut 11S globulin (legumin, Cor a 9), hazelnut 2S albumin (Cor a 14), hazelnut nsLTP (Cor a 8), hazelnut Bet v 1 (Cor a 1), hazelnut profilin (Cor a 2), hazelnut oleosin (Cor a 12, 13), pecan 7S globulin (vicilin, Car i 2), pecan 11S globulin (legumin, Car i 4), pecan 2S albumin (Car i 1), Brazil nut 11S globulin (legumin, Ber e 2), Brazil nut 2S albumin (Ber e 1), pistachio nut 7S globulin (vicilin, Pis v 3), pistachio nut 11S globulin (legumin, Pis v 2, 5), pistachio nut 2S albumin (Pis v 1), and black walnut 11S globulin (legumin, Jung n 4).

[0022] In certain embodiments, the fruit allergens are orange nsLTP (Cit s 3), orange profilin (Cit s 2), green kiwi 2S albumin (Act d 13), golden kiwi nsLTP (Act c 10), green kiwi nsLTP (Act d 10), golden kiwi Bet v 1 (Act c 8), green kiwi Bet v 1 (Act d 8), green kiwi profilin (Act d 9), banana nsLTP (Mus a 3), banana profilin (Mus a 1), peach nsLTP (Pru p 3), peach Bet v 1 (Pru p 1), peach profilin (Pru p 4), apple nsLTP (Mal d 3), apple Bet v 1 (Mal d 1), apple profilin (Mal d 4), strawberry nsLTP (Fra a 3), apricot nsLTP (Pru ar 3), cherry nsLTP (Pru av 3), plum nsLTP (Pru d 3), pomegranate nsLTP (Pun g 1), red raspberry nsLTP (Rub I 3), grape nsLTP (Vit v 1), pear nsLTP (Pyr c 3), strawberry Bet v 1 (Fra a 1), apricot Bet v 1 (Pru ar 1), cherry Bet v 1 (Pru av 1), pear Bet v 1 (Pyr c 1), raspberry Bet v 1 (Rub i 1), strawberry profilin (Fra a 4), cherry profilin (Pru av 4), pear profilin (Pyr c 4), watermelon profilin (Citr l 2), muskmelon profilin (Cuc m 2), litchi profilin (Lit c 1), and pineapple profilin (Ana c 1).

[0023] In certain embodiments, fruit allergens include celery nsLTP (Api g 2, 6), celery Bet v 1 (Api g 1), celery profilin (Api g 4), yellow mustard 2S albumin (Sin a 1), yellow mustard profilin (Sin a 4), asparagus nsLTP (Aspa o 1), lettuce nsLTP (Lac s 1), cabbage nsLTP (Bra o 3), tomato nsLTP (Sola 1 13, 16, 17), carrot Bet v 1 (Dau c 1), tomato Bet v 1 (Sola 1 4), carrot profilin (Dau c 4), bell pepper profilin (Cap a 2), and tomato profilin (Sola 1 1).

[0024] Inhalant allergens include, but are not limited to, indoor allergens, pollen allergens, mold allergens, etc. Indoor allergens include, but are not limited to, allergens from dust, mites, tatami mats, buckwheat bramble, pet hair, clothing, bedding (cotton, silk, wool, feathers), etc. Pollen allergens include, but are not limited to, allergens from ragweed, knotweed, Japanese cedar, Japanese red pine, Japanese silver grass, typha, mugwort, olive, plane tree, European cornflower, European alder, Japanese white birch, hornbeam, European chestnut, European hazel, European bean, Japanese laurel, white oak, ash, Chinese laurel, Chinese hazel, saffron, Chinese laurel, Japanese mountain ash, Timothy grass, date palm, plantain, mesquite, Brassica rapa, Japanese laurel, lilac, etc. Mold allergens include, but are not limited to, allergens from Alternaria, Penicillium, Candida, Cladosporium, Aspergillus, etc.

[0025] In certain embodiments, the pollen allergens are ragweed nsLTPs (Amb a 6), mugwort nsLTPs (Art v 3), olive nsLTPs (Ole e 7), plane tree nsLTPs (Pla or 3), European oak nsLTPs (Par j 1), European alder Bet v 1 (Aln g 1), white birch Bet v 1 (Bet v 1), hornbeam Bet v 1 (Car b 1), chestnut Bet v 1 (Cas s 1), corylus avell Bet v 1 (Cor a 1), oak Bet v 1 (Fag s 1), common oak Bet v 1 (Ost c 1), white oak Bet v 1 (Qua a 1), redroot pigweed profilin (Ama r 2), ptaxa profilin (Amb a 8), Artemisia profilin (Art v 4), Birch profilin (Bet v 2), Chenopodium profilin (Che a 2), Corylus avellana profilin (Cor a 2), Saffron profilin (Cro s 2), Zoysiagrass profilin (Cyn d 12), Olive profilin (Ole e 2), Mountain ash profilin (Mer a 1), Common blueberry profilin (Par j 3), Timothy grass profilin (Phl p 12), Date palm profilin (Pho d 2), Plantain profilin (Pla l 2), Mesquite profilin (Pro j 2), Alder polcalcin (Aln g 4), Ragweed polcalcin (Amb a 9), Artemisia polcalcin (Art v 5), birch polcalcin (Bet v 4), Brassica rapa polcalcin (Bra r 5), schizoza polcalcin (Che a 3), silverleaf polcalcin (Cyn d 7), olive polcalcin (Ole e 3), European sedge polcalcin (Par j 4), timothy grass polcalcin (Phi p 7), Japanese sedge polcalcin (Sal k 7) and lilac polcalcin (Syr v 3).

[0026] Contact allergens include, but are not limited to, allergens from cosmetics, paints, clothing, latex (rubber), bedding, detergents, etc. In certain embodiments, latex allergens include latex nsLTP (Heb v 12).

[0027] In one embodiment, the reactivity of allergens with IgE and IgG4 differs depending on their fragments. Examples of such allergens include αs1-casein, αs2-casein, β-casein, κ-casein, β-lactoglobulin, and Brown shrimp tropomyosin.

[0028] In a preferred embodiment, the allergen used as a probe is a polypeptide. The allergen fragment used as a probe is not particularly limited as long as it is long enough for an antibody to bind to it, and may be 3 to 50 amino acids long, preferably 5 to 30 amino acids long, more preferably 10 to 20 amino acids long, and particularly preferably 12 to 18 amino acids long. The allergen fragment may be a contiguous fragment in terms of amino acid sequence, or, if the allergen contains a discontinuous structural epitope in the amino acid sequence, it may be a fragment containing the amino acid sequence of the structural epitope. When the fragments are contiguous, they may be overlapping fragments in which adjacent fragments partially overlap. SEQ ID NOs: 2 to 21 shown in Table 1 below are examples of a set of overlapping fragments of αs1-casein (SEQ ID NO: 1).

[0029] [Table 1]

[0030] The number of probes used in the determination method of the present invention may vary depending on the length of the amino acid sequence of the allergen to be detected and the number of allergens, but may be, for example, 2 to 10,000, 3 to 1,000, 9 to 512, or 18 to 144. The probes may be composed of multiple fragments of the same allergen, or one or more fragments from two or more different allergens. Furthermore, the probes may be composed of only full-length allergens, only allergen fragments, or both full-length allergens and allergen fragments.

[0031] The probe may be immobilized. The immobilization method is not particularly limited, and any known method can be used. Specific examples include a method in which the probe is directly adsorbed onto a measurement plate or chip, and a method in which the probe is bound to a carrier such as beads and the beads to which the probe is bound are immobilized onto a measurement plate or chip (WO2018 / 154814, WO2019 / 050017). Specific embodiments of immobilizing the probe onto the chip are described in detail in the biochip section below.

[0032] The subject in the determination method of the present invention may or may not have an allergic disease. In one embodiment, the subject is a subject suspected of having an allergic disease. In another embodiment, the subject is a subject diagnosed with an allergic disease. In some embodiments, the subject is a subject who is scheduled to undergo a test that may induce serious allergic symptoms such as anaphylaxis, such as a provocation test for allergic diseases (e.g., oral food challenge test, inhalation provocation test, nasal mucosal provocation test, eye reaction, etc.) or an intradermal test. In a specific embodiment, the subject is a subject who is scheduled to undergo an oral food challenge test.

[0033] The biological sample used in the determination method of the present invention is not particularly limited as long as it can contain the antibody to be detected, and examples thereof include blood, plasma, and serum. The biological samples may be two or more types collected at different times from the same subject. The results of each of these samples can be obtained and compared with each other or with a predetermined reference value, thereby monitoring allergic diseases. For example, if the determination item is whether the allergic disease is mild or severe, it can be further determined whether the disease remains mild or severe, whether it has progressed from mild to severe, or whether it has progressed from severe to mild. For example, by collecting samples before and after the start of treatment, or at two different time points after the start of treatment, it becomes possible to assess the effectiveness of treatment.

[0034] The types of antibodies that can be identified by the determination methods of the present invention are not limited to, and include, for example, IgA1, IgA2, IgD, IgE, IgG1, IgG2a, IgG2b, IgG3, IgG4, and IgM. In some embodiments, the types of antibodies that can be identified are IgE, IgG1, IgG2a, IgG2b, IgG3, and IgG4. In specific embodiments, the types of antibodies that can be identified are IgE and IgG4.

[0035] The step of contacting the probe with the biological sample can be carried out by any known method that enables specific binding between an antigen and an antibody, or by the method described in the Examples below. Specifically, for example, a liquid containing the biological sample can be loaded onto a probe immobilized on a substrate or carrier. The biological sample can be used as is or diluted.

[0036] The step of detecting the reaction between the probe and the target antibody can be carried out by any known method for detecting the reaction between an antigen and an antibody, or by the method described in the Examples below. Specifically, for example, the probe is contacted with a biological sample, and reacted for a predetermined time under conditions that allow binding of the probe and the antibody. The antibody that does not bind to the probe is then removed, and a detectably labeled antibody (secondary antibody) specific to the target antibody is applied. The unbound secondary antibody is then removed, and the label is then detected. Examples of labels include, but are not limited to, fluorescent substances, luminescent substances, enzymes (e.g., ALP, HRP), radioisotopes, etc., and methods for detecting the label include, but are not limited to, spectrophotometry, absorptiometry, fluorometry, colorimetry, autoradiography, etc. Detection can be qualitative or quantitative. Before contacting the probe with the biological sample, a blocking agent can be applied to suppress nonspecific adsorption of the antibody. In an embodiment using a biochip in which biotinylated probes are bound to a substrate or immobilization carrier modified with avidin or its derivatives, the blocking agent may contain biotin. The use of a blocking agent containing biotin can improve the signal-to-noise ratio.

[0037] The step of distinguishing the type of antibody reacting with the probe can be performed, for example, by using a secondary antibody specific to the desired type of antibody in the step of detecting the reaction between the probe and the antibody in the biological sample. The type of antibody may be distinguished one type at a time for each step of detecting the reaction between the probe and the antibody in the biological sample, or multiple types may be distinguished simultaneously. Simultaneous distinction of multiple types can be achieved, for example, by using different labels for each type of secondary antibody. Specifically, for example, a secondary antibody specific to type A antibodies and a secondary antibody specific to type B antibodies can be labeled with fluorescent labels of different wavelengths. Furthermore, by using the same probe in two or more reaction regions and reacting different secondary antibodies in each reaction region, multiple types of antibodies can be detected substantially simultaneously using the same label. The secondary antibody may be a monoclonal or polyclonal antibody. Using a monoclonal antibody can suppress nonspecific adsorption and increase the signal-to-noise ratio.

[0038] The step of determining the type of antibody that has reacted with the probe may be carried out simultaneously with the step of detecting the reaction between the probe and the target antibody, or may be carried out separately. The data relating to the reaction between the probe and the antibody may be quantitative or qualitative. In one embodiment, the data is quantitative data. The quantitative data may vary depending on the reaction detection method, and examples thereof include, but are not limited to, reaction intensity (absorbance, luminescence intensity, radiation intensity, etc.).

[0039] The prediction model used in the determination method of the present invention is a model for obtaining an index that serves as the basis for determining allergic diseases, and is a nonlinear model or a model with two or more explanatory variables and two or more terms, created using at least training data related to the reaction of the probe with an antibody and training data related to the type of antibody bound. The number of explanatory variables is not particularly limited as long as it is two or more. Non-limiting examples of the number of explanatory variables may vary depending on the type of prediction model, but may include, for example, 2 to 100 (i.e., 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, etc.), 2 to 50 , 2 to 30, 2 to 20, 2 to 15, 2 to 10, 3 to 100, 3 to 50, 3 to 30, 3 to 20, 3 to 15, 3 to 10, 4 to 100, 4 to 50, 4 to 30, 4 to 20, 4 to 15, 4 to 10, 5 to 100, 5 to 50, 5 to 20, 5 to 15, 5 to 10, etc. Note that for a nonlinear model, the number of explanatory variables may be less than two, for example, 1.

[0040] Examples of explanatory variables include variables related to the reaction between the probe and antibody (e.g., luminescence intensity, absorbance, fluorescence intensity, etc.), variables related to the type of bound antibody (e.g., isotypes such as IgE and IgG4), as well as the subject's age, sex, medical history, family history, underlying diseases, complications, genetic predisposition, total IgE level, IgE dilution rate (the dilution ratio when a sample from a subject with a high total IgE level is diluted to a range that enables detection of specific IgE), and test results for other allergens. In addition to the allergens listed above, other allergens include non-protein allergens such as formaldehyde, VOCs, metals, lacquer (urushiol), and iodine. Examples of allergen test results include test results for specific IgE antibodies using RAST, CAP, MAST, etc., and skin tests such as prick tests, scratch tests, and intradermal tests. For explanatory variables, raw values ​​may be used as is, or values ​​may be preprocessed (variable transformation) such as logarithmic transformation.

[0041] A prediction model may take the form of a prediction formula, a decision tree, or the like, depending on the statistical method used to create it. A prediction formula may include arithmetic operations, hinge functions, etc., and the number of terms is not particularly limited as long as it is two or more. Non-limiting examples of the number of terms in a prediction formula may vary depending on the type of prediction model, but may include, for example, 2 to 150 terms (i.e., 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40 terms, etc.), 2 to 100 terms, 2 to 50 terms, 2 to 30 terms, 2 to 20 terms, 2 The number of terms may be 1 to 15, 2 to 10, 4 to 100, 4 to 50, 4 to 30, 4 to 20, 4 to 15, 4 to 10, 6 to 150, 6 to 100, 6 to 50, 6 to 30, 6 to 20, 6 to 15, 6 to 10, 8 to 150, 8 to 100, 8 to 50, 8 to 30, 8 to 20, 8 to 15, 8 to 10, 10 to 150, 10 to 100, 10 to 50, 10 to 30, 10 to 20, 10 to 15, etc. Note that for nonlinear models, the number of terms may be less than two, for example, one term.

[0042] The training data is data from multiple subjects whose presence or absence of the allergic disease to be determined is known, and is used for creating and evaluating a prediction model. For example, when the determination method of the present invention determines whether an allergic reaction to allergen A is mild or severe, the training data includes data from subjects whose degree of allergic reaction to allergen A is known, and when the determination method of the present invention predicts an ASCA value for allergen A, the training data includes data from subjects whose ASCA value for allergen A is known. Using such training data as a base, various statistical methods are used to create a model that can better distinguish between subjects with mild symptoms and subjects with severe symptoms, or a model that can predict an ASCA value closer to the actual ASCA value.

[0043] Statistical methods used to create and evaluate predictive models include, but are not limited to, least squares, Ridge regression, LASSO regression, CART, and MARS. Nonlinear models are created using CART and MARS, while linear models are created using least squares, Ridge regression, and LASSO regression. The accuracy and generalizability of predictive models can be improved by cross-validation (CV). In cross-validation, for example, training data is divided into multiple sets, some of which are used as training data and the rest as validation data. The prediction results are then input into a predictive model to verify whether the results are biased. Specific cross-validation techniques include leave-one-out cross validation (LOOCV), 10-fold cross validation, and generalized cross validation. Obtaining similar results using multiple sets of training and validation data increases the likelihood of obtaining similar results (generalizability) when applied to samples used for actual assessment. Predictive models can be created manually or by machine learning. Prediction models can be evaluated using sensitivity, specificity, correlation coefficient (prediction rate), AUC in ROC curves, dot histograms, and the like.

[0044] In one embodiment, the prediction model has a correlation coefficient of 0.60 or higher. A higher correlation coefficient is preferable, and may be, for example, but not limited to, 0.60 or higher, 0.62 or higher, 0.64 or higher, 0.66 or higher, 0.68 or higher, 0.70 or higher, 0.72 or higher, 0.74 or higher, 0.76 or higher, 0.78 or higher, 0.80 or higher, 0.82 or higher, 0.84 or higher, 0.86 or higher, 0.88 or higher, 0.90 or higher, 0.92 or higher, 0.94 or higher, 0.96 or higher, 0.98 or higher, etc.

[0045] In one embodiment, the prediction model has an AUC of 0.900 or more when the vertical axis of the ROC curve is sensitivity ranging from 0 to 1 and the horizontal axis is (1-specificity) ranging from 0 to 1. The higher the AUC, the more preferable it is, and may be, for example, but not limited to, 0.900 or more, 0.905 or more, 0.910 or more, 0.915 or more, 0.920 or more, 0.925 or more, 0.930 or more, 0.935 or more, 0.940 or more, 0.945 or more, 0.950 or more, 0.955 or more, 0.960 or more, 0.965 or more, 0.970 or more, 0.975 or more, 0.980 or more, 0.985 or more, 0.990 or more, 0.995 or more, etc.

[0046] In addition to data on the reaction between the probe and the antibody and data on the type of antibody bound, additional training data related to the diagnosis of allergic diseases can be used to create a prediction model, including, but not limited to, the data described above for the explanatory variables.

[0047] The assessment of allergic diseases using the assessment method of the present invention includes, but is not limited to, determining whether an allergic disease is mild or severe (e.g., whether symptoms upon exposure to an allergen are mild or severe), predicting the grade (severity), predicting the total ASCA value, etc. The criteria for determining whether an allergic disease is mild or severe may vary depending on the circumstances in which the assessment results are used. For example, criteria for mildness include induced symptoms of Grade 1, an ASCA value below a predetermined value, and a level at which no medical treatment is required (a level at which follow-up observation or the like can be used). The criteria for mild or severe can be determined appropriately by a physician depending on the circumstances. The grade is an index of the severity of anaphylaxis and is determined based on the organ symptom with the highest symptom grade according to the following grading table (Yanagida et al., Journal of the Japanese Society of Pediatric Allergy, 2014;28:201). [Table 2]

[0048] The ASCA (Anaphylaxis Scoring Aichi) is a scoring system that quantitatively evaluates the severity of symptoms induced in an oral food challenge test. The induced symptoms are classified into five categories: respiratory, cutaneous, digestive, neurological, and circulatory. Subjective and objective symptoms are then ranked in order of severity to generate an organ score (1-60). The highest scores for each organ in a series of challenges are summed to arrive at a total score (maximum 240) (see, for example, Hino et al., Arerugi. 2013;62(8):968-79; Sakai et al., Asia Pac Allergy. 2017;7(4):234-242). If it is possible to determine whether symptoms upon exposure to an allergen will be mild or severe before exposure, this can help determine whether to perform tests involving allergen exposure (e.g., oral food challenge tests, inhalation provocation tests, nasal mucosal provocation tests, eye reaction provocation tests, and intradermal tests), and if so, whether to perform the tests on an outpatient basis or in an inpatient setting.

[0049] The diagnosis of an allergic disease based on the prediction result may vary depending on the prediction model and the determination item, but can be performed, for example, by comparing the prediction result with a reference value. In the case of a prediction model using a prediction formula, the prediction result (predicted value) obtained by the prediction formula can be compared with a preset reference value (e.g., a cutoff value) to determine whether it is positive or negative. If the reference value is set so that a higher predicted value than the reference value indicates a higher probability of severe disease and a lower predicted value indicates a higher probability of mild disease, the predicted value can be determined to be severe if it is higher than the reference value, and mild if it is lower. The cutoff value can be set using a known method such as the Youden index. Furthermore, if the prediction model is configured to directly output the determination item (e.g., determining whether it is severe or mild using a decision tree, predicting ASCA values ​​using a prediction formula or decision tree, etc.), the prediction result can also be used directly as the determination result.

[0050] 2. Methods for obtaining indicators for diagnosing allergic diseases Another aspect of the present invention is a method for detecting an allergic disease by contacting two or more probes selected from allergens and their fragments with a biological sample collected from a subject, and obtaining data on the reaction between the probes and antibodies contained in the biological sample and data on the type of the reacted antibodies; and inputting the data into a prediction model to obtain a prediction result as an index for determining an allergic disease. The present invention relates to a method for obtaining an index for diagnosing allergic diseases, comprising the steps of: (a) obtaining an index for diagnosing allergic diseases by using a method for diagnosing allergic diseases ... The index acquisition method of the present invention is similar to the determination method of the present invention, except that the step of determining an allergic disease based on the prediction result by the prediction model is not essential. The prediction result obtained by the index acquisition method of the present invention can be used as an index for determining an allergic disease. Specifically, for example, the presence or absence of a disease can be determined by comparing the prediction result with a reference value (such as a cutoff value) related to the disease condition to be determined. Furthermore, when the prediction model is configured to directly output the determination item (for example, determining whether the condition is severe or mild using a decision tree, predicting ASCA values ​​using a prediction formula or decision tree, etc.), the prediction result can also be used directly as the determination result.

[0051] 3. Allergy Disease Assessment Kit Another aspect of the present invention relates to a kit (hereinafter sometimes referred to as the "kit of the present invention") for use in the determination method or index acquisition method of the present invention, which comprises two or more probes selected from allergens and their fragments. The probe contained in the kit of the present invention is as described above for the determination method of the present invention. In the kit of the present invention, the probe may be immobilized or modified for immobilization. Modifications for immobilization include, for example, the addition of avidin or a derivative thereof, biotin or a derivative thereof, alkyne, azide, a phenolic OH group, a hydroxyl group, an amino group, a carboxyl group, a photoreactive group, a His tag, an antibody, etc. The kit of the present invention may further comprise a reagent for detecting the target antibody. Such reagents include, but are not limited to, a labeled secondary antibody that specifically binds to the target antibody, and, if necessary, a reagent for detecting the label (such as a luminescent reagent or a colorimetric reagent). The label attached to the secondary antibody is as described above for the determination method of the present invention. The kit of the present invention may further comprise a standard sample, instructions showing how to use the kit, such as an instruction manual or website information containing information on the method of use (e.g., a URL or two-dimensional code), or a medium on which information on the method of use is recorded, such as a flexible disk, CD, DVD, Blu-ray disk, memory card, USB memory, etc.

[0052] 4. Biochip for determining allergic diseases Another aspect of the present invention relates to a biochip (hereinafter sometimes referred to as the "biochip of the present invention") for use in the determination method or index acquisition method of the present invention, in which two or more probes selected from allergens and their fragments are immobilized on a substrate. The probe immobilized on the substrate of the biochip of the present invention is as described above in relation to the determination method of the present invention. The probe may be immobilized directly on the substrate or via a carrier (immobilization carrier). Hereinafter, the probe or the carrier on which the probe is immobilized may be referred to as the immobilized substance.

[0053] The substrate of the biochip is not particularly limited as long as it does not excessively adversely affect the test sample or biochemical reaction. Examples of suitable materials include resins and glass. Examples of suitable resins include, but are not limited to, thermosetting resins and thermoplastic resins. In particular, optically transparent resins such as polypropylene, polycarbonate, acrylic, polystyrene, polyethylene terephthalate, cycloolefin polymers, and cycloolefin copolymers can ensure good visible light transmittance. Examples of suitable polypropylenes include, but are not limited to, homopolypropylenes and random copolymers of polypropylene and polyethylene. Examples of suitable acrylics include, but are not limited to, polymethyl methacrylate, or copolymers of methyl methacrylate with other monomers such as methacrylic acid esters, acrylic acid esters, and styrene. Optically opaque resins can also be used. Examples of suitable optically opaque resins include, but are not limited to, resins such as polypropylene, polycarbonate, acrylic, polystyrene, polyethylene terephthalate, cycloolefin polymers, and cycloolefin copolymers to which a resin colorant (e.g., masterbatch) has been added. The light-opaque resin material is preferably one with high light-shielding properties and is preferably black in color. The thickness of the substrate is not particularly limited, but since it is desirable to have a certain degree of non-deformability during the manufacturing process, it is preferably 0.3 mm to 3.0 mm, more preferably 0.5 mm to 1.5 mm, and particularly preferably 0.7 mm to 1.0 mm.

[0054] In some embodiments, the resin material used for the substrate is hydrophobic. In some embodiments, the resin material is formed from a water-insoluble polymer. In some embodiments, the resin material does not contain dextran, polyethylene glycol, or derivatives thereof. The substrate may be hydrophilized. Hydrophilization prevents nonspecific adsorption to the substrate and firmly immobilizes the probe (or the probe-bound carrier) on the substrate. The hydrophilization may be performed on one or more portions of the substrate surface, or on the entire substrate surface. When a portion of the substrate surface is hydrophilized, for example, the entire upper surface of the substrate or one or more portions of the upper surface of the substrate may be hydrophilized. The hydrophilization treatment is not particularly limited, and examples include surface coating with hydrophilic inorganic materials such as silica or surfactants, chemical hydrophilization treatments such as plasma treatment, UV-ozone treatment, corona treatment, and flame treatment, and imparting hydrophilicity by physically forming fine nanostructures (WO2011 / 024947). When the substrate is made of resin, chemical hydrophilization treatments such as plasma treatment, UV-ozone treatment, corona treatment, and flame treatment are more preferred, as they can cleave the chemical bonds of molecules on the resin surface and generate polar functional groups depending on the type of resin, such as OH (hydroxyl group), CO (carbonyl group), COOH (carboxyl group), methoxy group, peroxide group, polar ether group, etc., and UV-ozone treatment is particularly preferred.

[0055] In one embodiment, the substrate has a surface including a hydrophilic reaction area, preferably a resin surface including a hydrophilic reaction area. The reaction area refers to a region on the substrate where an immobilized substance is immobilized and reacted with a target substance. The reaction area may be a portion or the entire substrate surface. When the reaction area is a portion of the substrate surface, the substrate surface other than the reaction area may be hydrophilic or hydrophobic. When the reaction area is a portion of the substrate surface, the reaction area may have various shapes, for example, a circle, an ellipse, a polygon, or a combination thereof. The reaction area may be continuous or discontinuous. For example, the reaction area may form multiple spots that do not contact each other. The reaction area may be surrounded by a boundary capable of retaining a liquid therein. Furthermore, the reaction area may be covered with polar functional groups generated by bonding with oxygen at the cleaved carbon-related bonds of the resin forming the substrate surface. Such functional groups include, but are not limited to, hydroxyl groups, carbonyl groups, carboxyl groups, methoxy groups, peroxide groups, polar ether groups, and the like.

[0056] A coating layer for preventing nonspecific adsorption and immobilizing probes can be provided on the substrate. Immobilizing a substance on the substrate via the coating layer prevents nonspecific adsorption and enhances detection sensitivity. The coating layer is not particularly limited as long as it can promote immobilization of the substance to be immobilized and / or suppress nonspecific adsorption, and can be composed of, for example, various polymers. Among polymers, water-soluble polymers are preferred. The use of water-soluble polymers makes it possible to avoid the use of non-aqueous solvents other than water and alcohol, which may denature the substance to be immobilized. Therefore, in the present invention, it is preferable to use a water-soluble polymer to prevent the denaturation of the substance to be immobilized. Furthermore, water-soluble polymers have the advantage of being excellent in suppressing nonspecific adsorption. Here, "water-soluble" refers, for example, to the solubility of the polymer in water (the number of grams dissolved in 100 g of water) being 5 or more. The number-average molecular weight of the polymer is not particularly limited and is typically about 3.5 to 5 million. By setting the molecular weight of the polymer to about 500 to several hundred thousand, it is possible to maintain an appropriate number of crosslinks between polymers, and to proceed with the reaction between the substance to be immobilized and the substance to be reacted with the substance to be immobilized (such as a photocrosslinking agent).

[0057] Examples of the water-soluble polymer include amphoteric polymers such as phosphorylcholine-containing polymers and nonionic polymers. Examples of the amphoteric polymer include polymers containing 2-methacryloyloxyphosphorylcholine (MPC) as the main component (for example, "LIPIDURE" manufactured by NOF Corporation). (R) " (LIPIDURE (R) -CR2001, LIPIDURE (R)Examples of nonionic polymers include polyalkylene glycols such as polyethylene glycol (PEG) and polypropylene glycol; vinyl alcohol, methyl vinyl ether, vinyl pyrrolidone, vinyl oxazolidone, vinyl methyl oxazolidone, 2-vinyl pyridine, 4-vinyl pyridine, N-vinyl succinimide, N-vinyl formamide, N-vinyl-N-methyl formamide, N-vinyl acetamide, N-vinyl-N-methyl acetamide, 2-hydroxyethyl methacrylate, polyethylene glycol methacrylate, polyethylene glycol acrylate, acrylamide, methacrylamide, N,N-dimethyl acrylamide, N-isopropyl acrylamide, diacetone acrylamide, methylolacrylamide, acryloyl morpholine, acryloyl pyrrolidine, acryloyl piperidine, styrene, chloromethyl styrene, bromomethyl styrene, vinyl acetate, methyl methacrylate, butyl acrylate, methyl cyanoacrylate, ethyl cyanoacrylate, and n-propyl cyanoacrylate. Examples of suitable polymers include, but are not limited to, nonionic vinyl polymers containing, as constituent components, monomer units such as methyl methacrylate, isopropyl cyanoacrylate, n-butyl cyanoacrylate, isobutyl cyanoacrylate, tert-butyl cyanoacrylate, glycidyl methacrylate, ethyl vinyl ether, n-propyl vinyl ether, isopropyl vinyl ether, n-butyl vinyl ether, isobutyl vinyl ether, and tert-butyl vinyl ether, either alone or as a mixture; natural polymers such as gelatin, casein, collagen, gum arabic, xanthan gum, tragacanth gum, guar gum, pullulan, pectin, sodium alginate, hyaluronic acid, chitosan, chitin derivatives, carrageenan, starches (carboxymethyl starch, aldehyde starch), dextrin, and cyclodextrin; and water-soluble cellulose derivatives such as methyl cellulose, viscose, hydroxyethyl cellulose, hydroxyethyl methyl cellulose, carboxymethyl cellulose, and hydroxypropyl cellulose.In addition, there are commercially available photo-crosslinkable water-soluble polymers based on these polymers, such as "BIOSURFINE-AWP" manufactured by Toyo Gosei Co., Ltd., which is based on polyvinyl alcohol, and "LIPIDURE" manufactured by NOF Corporation, which is based on 2-methacryloyloxyphosphorylcholine (MPC). (R) Among these, polyethylene glycol-based polymers (for example, vinyl polymers of polyethylene glycol (meth)acrylate), polymers containing MPC as the main component (for example, LIPIDURE®) and the like can also be used. (R) -CR2001) is preferred.

[0058] The substrate may be subjected to a surface treatment to enhance adhesion between the coating layer and the substrate. The surface treatment is not particularly limited, but examples thereof include treatments that cut the chemical bonds of molecules on the resin surface of the substrate and generate hydrophilic functional groups such as OH (hydroxyl group), CO (carbonyl group), COOH (carboxyl group), methoxy group, peroxide group, polar ether group, etc. depending on the type of resin, such as plasma treatment, UV-ozone treatment, corona treatment, and flame treatment.

[0059] The immobilization carrier is not particularly limited and may be, for example, a particulate carrier such as an organic or inorganic particle. The particulate carrier may be a magnetic particle. Non-limiting examples of magnetic particles include beads of uniform size in which a magnetizable substance such as γFe2O3 or Fe3O4 is coated with a uniformly hydrophilic polymer (e.g., a water-soluble polymer such as glycidyl methacrylate). Commercially available products include Dynabeads manufactured by Thermo Fisher Scientific, FG Beads manufactured by Tamagawa Seiki Co., Ltd., Sera-Mag magnetic beads manufactured by GE Healthcare, and Magnosphere manufactured by JSR Life Sciences. The immobilization carrier may be subjected to various surface treatments useful for immobilizing probes. Non-limiting examples of surface treatments include modification with reactive groups such as avidin or its derivatives, biotin or its derivatives, alkyne, azide, epoxy, hydroxyl group, amino group, carboxyl group, succinimide group, tosyl group, protein A, and protein G.

[0060] The probe may also have a modification useful for immobilization to a substrate or immobilization carrier. Non-limiting examples of modifications include avidin or its derivatives, biotin or its derivatives, alkyne, azide, phenolic OH group, hydroxyl group, amino group, carboxyl group, His tag, antibody, etc. In a specific embodiment, the probe is bound to biotin via a spacer, the immobilization carrier is avidinated, and the probe and the immobilization carrier are bound via biotin-avidin interaction. The spacer is not particularly limited as long as it does not inhibit the interaction between the immobilized substance (e.g., peptide) and antibody or the interaction between the reactive group and the group that reacts with it, and is non-cleavable in the context of use of the biochip of the present invention, and examples thereof include carbon chains, PEG, etc. The length of the spacer is not limited and may be, for example, 5 to 150 Å, 6 to 100 Å, 7 to 80 Å, 8 to 60 Å, or 9 to 50 Å. If the spacer is a carbon chain, it may be, for example, a C1 to C20 linear alkyl, a C3 to C10 linear alkyl, or a C4 to C8 linear alkyl, particularly a C6 linear alkyl. If the spacer is PEG, it may be, for example, a 2 to 24-mer, a 2 to 12-mer, a 2 to 8-mer, particularly a 2 to 4-mer. Use of a spacer makes it less likely that the interaction between the substance to be immobilized and the antibody will be inhibited by steric hindrance, etc.

[0061] The biochip of the present invention can be produced by any known method or the method described in the Examples. For example, a biochip (biochip A) in which a probe-bound carrier is immobilized on a hydrophilized substrate can be produced as follows. Specifically, for example, a substrate is produced by a step of hydrophilizing the substrate (hereinafter referred to as "step 1-1"), and then at least a substance to be immobilized, a photocrosslinker having at least two photoreactive groups per molecule, and a thickener and / or surfactant are spotted on the substrate, followed by a step of irradiating the substrate with light (hereinafter referred to as "step 1-2"), in which the substance to be immobilized is immobilized on the substrate.

[0062] An exemplary method for producing biochip A will be described below. Process 1-1 The hydrophilization treatment of the substrate is not particularly limited as long as it is a treatment that hydrophilizes the surface of the substrate. Examples include a method in which silica, a surfactant, or the like dissolved or suspended in a liquid is coated on the substrate by spin coating, painting, spraying, immersion, or the like, followed by drying; a method in which the substrate is chemically hydrophilized by plasma treatment, UV-ozone treatment, corona treatment, flame treatment, or the like; a method in which a fine nanostructure is transferred to the substrate or the substrate is roughened with a chemical solution to physically form a nanostructure and impart hydrophilicity; and a method in which the substrate is coated with a hydrophilic polymer. Chemical hydrophilization treatments such as plasma treatment, UV-ozone treatment, corona treatment, and flame treatment are more preferred, as they can cleave the chemical bonds of molecules on the resin surface and generate functional groups with hydrophilic polarity, such as OH (hydroxyl group), CO (carbonyl group), COOH (carboxyl group), methoxy group, peroxide group, and polar ether group, depending on the type of resin. UV-ozone treatment is particularly preferred.

[0063] A step of forming a boundary on the substrate to form a reaction area can be included. The method for forming the reaction area is not particularly limited, and examples include a method in which a ring preformed from the rubber composition or the resin material is adhered to the substrate using an adhesive, or a method in which a ring is formed from a substrate made of the resin material using various resin molding methods such as injection molding and vacuum molding, or mechanical cutting. The adhesive is not particularly limited as long as it does not affect the test sample or the biochemical reaction, and can be appropriately selected from commercially available adhesives.

[0064] Process 1-2 Step 1-2 includes a step of spotting at least the substance to be immobilized, a photocrosslinker having at least two photoreactive groups in one molecule, and a thickener and / or a surfactant in a grid pattern on the substrate, and a step of irradiating with light.

[0065] A biochip (biochip B) having a coating layer and a reaction area surrounded by a boundary capable of retaining a liquid therein can be manufactured, for example, by manufacturing a substrate through a process selected from a process of providing a coating layer on a substrate for preventing nonspecific adsorption and immobilizing a biological material, and a process of forming a boundary on the substrate to provide a reaction area (hereinafter referred to as "process 2-1"). a step of spotting at least a substance to be immobilized, a photocrosslinker having at least two photoreactive groups in one molecule, and a thickener and / or a surfactant on the substrate, and irradiating the substrate with light (hereinafter referred to as "step 2-2"); It may include:

[0066] An exemplary method for producing biochip B will now be described. Process 2-1 Step 2-1 includes a step selected from the step of providing a coating layer on the substrate to prevent nonspecific adsorption and immobilize a biological material, and the step of forming a boundary on the substrate to provide a reaction area. The order of these steps can be arbitrarily changed. Furthermore, a step of surface-treating the substrate may be included before the step of providing the coating layer.

[0067] The surface treatment method used in the step of surface treating the substrate is not particularly limited as long as it improves the adhesion between the substrate and the coating layer, and examples of surface modification methods include known methods such as plasma treatment, UV ozone treatment, and corona treatment.

[0068] Formation of a coating layer for preventing nonspecific adsorption to a substrate and immobilizing a biological material can be performed by known methods such as spin coating, painting, spraying, or immersion in a coating liquid containing a water-soluble polymer. For example, the coating liquid can be prepared by dissolving a water-soluble polymer in a solvent. Examples of solvents that can be used include water, lower alcohols miscible with water in any ratio, and mixtures thereof. Preferred lower alcohols include methanol, ethanol, and isopropanol. Of these, a mixed solvent of ethanol and water is preferred.

[0069] The concentration of the polymer in the coating liquid is not particularly limited, but the polymer concentration can be, for example, 0.0001 to 10 parts by mass, preferably 0.001 to 1 part by mass, and the photocrosslinking agent concentration can be, for example, 1 to 20 parts by mass, preferably 2 to 10 parts by mass, relative to the polymer.

[0070] The coating liquid containing the polymer preferably contains a photocrosslinking agent having at least two photoreactive groups in one molecule. In the present invention, the term "photoreactive group" refers to a group that generates radicals upon irradiation with light. The photocrosslinking agent can form covalent bonds with amino groups, carboxyl groups, carbon atoms constituting organic compounds, etc., by generating radicals from the photoreactive groups upon irradiation with light. Thus, by applying a coating liquid containing the photocrosslinking agent to a substrate and then irradiating the substrate with light, the substrate and polymer can be bonded via the photocrosslinking agent, and a polymer layer having a nonspecific adsorption prevention effect can be formed on the substrate. In the present invention, it is also possible to bond the substrate and polymer using the groups contained in the polymer by introducing photoreactive groups and / or groups capable of forming covalent or coordinate bonds with the substrate surface into the polymer without using the photocrosslinking agent or together with the photocrosslinking agent.

[0071] The water-soluble polymer and photocrosslinking agent contained in the coating layer of the present invention are known per se, can be produced by known production methods, and are commercially available. There is no particular limitation on the film thickness of the coating layer, but it is preferably 1 nm to 10 μm, more preferably 2 nm to 1 μm, and particularly preferably 10 nm to 100 nm.

[0072] In the present invention, after coating the coating layer as described above, it is preferable to stabilize the coating layer by aging it under constant temperature and humidity conditions. The temperature is preferably 5°C to 40°C, and more preferably 20°C to 30°C. The humidity is preferably 40% to 80%, and more preferably 50% to 70%. The aging period is preferably 1 day to 1 month, and more preferably 3 days to 2 weeks.

[0073] The method for forming the boundary on the substrate is not particularly limited, and examples include a method in which a ring preformed from the rubber composition or the resin material is adhered to the substrate using an adhesive, or a method in which a ring is formed from a substrate made of the resin material by various resin molding methods such as injection molding and vacuum molding, or by mechanical cutting, etc. The adhesive is not particularly limited as long as it does not affect the test sample or biochemical reaction, and can be appropriately selected from commercially available adhesives.

[0074] Process 2-2 Step 2-2 includes the steps of spotting at least the substance to be immobilized, a photocrosslinker having at least two photoreactive groups in one molecule, and a thickener and / or a surfactant in a grid pattern on the substrate, irradiating with light, and removing unreacted components.

[0075] In Step 1-2 or 2-2, when the substance to be immobilized is immobilized on an immobilization carrier in advance, a known immobilization method can be used. Examples of immobilization methods include, but are not limited to, a method in which the magnetic microparticles as the immobilization carrier are reacted with a previously prepared substance to be immobilized in a solution such as an appropriately selected known buffer solution, and the magnetic microparticles on which the substance to be immobilized is immobilized are recovered. For example, the magnetic microparticles may be washed in advance with a known buffer solution, or the magnetic microparticles on which the substance to be immobilized is immobilized may be washed with a known buffer solution when being recovered.

[0076] The photocrosslinking agent having at least two photoreactive groups per molecule used in step 1-2 or 2-2 is not particularly limited. Examples of the photoreactive group possessed by the photocrosslinking agent include an azide group (-N3), an acetyl group, a benzoyl group, and a diazirine group. In particular, an azide group is preferred because, upon irradiation with light, a nitrogen molecule is released and a nitrogen radical is generated. This nitrogen radical can bond not only with functional groups such as amino groups and carboxyl groups, but also with carbon atoms constituting organic compounds, and can form covalent bonds with almost any organic substance. Examples of photocrosslinking agents having an azide group include diazidothibenzo ... The photocrosslinking agent is preferably water-soluble. The term "water-soluble" in relation to the photocrosslinking agent means that it can be provided in aqueous solution at a concentration of 0.5 mM or more, preferably 2 mM or more.

[0077] The substance to be immobilized or the immobilization carrier on which the substance to be immobilized is immobilized, and the photocrosslinker are preferably dispersed or dissolved in a solution. The solution is not particularly limited, and known buffer solutions can be used. Examples of buffer solutions include PBS buffer, HEPES buffer, Tris buffer, and MES buffer. Phosphate-buffered saline is preferred when the substance to be immobilized is used as is, and HEPES buffer is preferred when the substance to be immobilized is used after immobilization on the immobilization carrier in order to prevent aggregation of the immobilization carrier. The solution in which the substance to be immobilized and the photocrosslinker are dispersed or dissolved is sometimes referred to as a stamp solution.

[0078] The concentration of the substance to be immobilized is not particularly limited, but is preferably 0.05 mg / mL to 2 mg / mL, more preferably 0.1 mg / mL to 1 mg / mL, when the substance to be immobilized is not immobilized on an immobilization carrier. When the substance to be immobilized is immobilized on an immobilization carrier, the substance to be immobilized is reacted with the immobilization carrier at a concentration of preferably 0.5 mg / mL to 50 mg / mL, more preferably 1 mg / mL to 25 mg / mL, and the immobilization carrier on which the substance to be immobilized is used at a concentration of preferably 1 mg / mL to 10 mg / mL, more preferably 2.5 mg / mL to 7.5 mg / mL. The concentration of the photocrosslinking agent is not particularly limited, but is preferably 0.01 mg / mL to 1 g / mL, and more preferably 0.2 mg / mL to 0.2 g / mL.

[0079] The solution in which the substance to be immobilized or the immobilization carrier on which the substance to be immobilized is immobilized and the photocrosslinker are dispersed or dissolved preferably further contains a thickener and / or a surfactant. The inclusion of a thickener allows for adjustment of the spot size when the solution is spotted on a substrate. More specifically, the higher the concentration of the thickener in the solution, the smaller the spot size (contact area with the substrate) tends to be when the same volume of solution is spotted. Furthermore, the inclusion of a surfactant can prevent the substance to be immobilized from accumulating at the air-liquid interface and localizing the substance to the edge of the spot. The surfactant also contributes to improving affinity with the substrate; the higher the concentration of the surfactant in the solution, the larger the spot size (contact area with the substrate) tends to be when the same volume of solution is spotted. Therefore, by adjusting the concentrations of the thickener and surfactant, stamps of any desired size can be formed.

[0080] The thickener is not particularly limited as long as it does not affect the test sample or biochemical reaction, and commercially available thickeners can be used. Examples include cellulose and its derivatives, polysaccharides, vinyl compounds, vinylidene compounds, polyglycol compounds, polyvinyl alcohol compounds, and polyalkylene oxide compounds. Specific examples include gellan gum, xanthan gum, curdlan, pullulan, guar gum derivatives, locust bean gum, carrageenan, pectin, β-glucan, tamarind gum, psyllium seed gum, dextran, glycerin, carboxymethyl cellulose, hydroxyethyl cellulose, hydroxymethylpropyl cellulose, lanolin, methylcellulose, petrolatum, polyethylene glycol, polyvinyl alcohol, polyvinylpyrrolidone, carboxyl vinyl polymer, polyvinylpyrrolidone, polyvinyl alcohol, dextrin fatty acid esters, and inulin fatty acid esters. It is preferable to use one or more selected from hydroxyethyl cellulose, methylcellulose, polyvinylpyrrolidone, and polyvinyl alcohol, and polyvinyl alcohol is more preferable.

[0081] The concentration of the thickener is not particularly limited, but when polyvinyl alcohol is used, for example, the concentration is preferably 0.01 to 1 part by weight, more preferably 0.02 to 0.5 parts by weight, and particularly preferably 0.03 to 0.3 parts by weight, relative to 100 parts by weight of the solution.

[0082] The surfactant is not particularly limited as long as it does not affect the test sample or biochemical reaction. Commercially available nonionic surfactants can be used, such as polyoxyethylene (10) octylphenyl ether [Triton X-100], polyoxyethylene (8) octylphenyl ether [Triton X-114], polyoxyethylene sorbitan monolaurate [Tween 20], polyoxyethylene sorbitan monooleate [Tween 80], polyoxyethylene (23) lauryl ether [Brij 35], polyoxyethylene (20) lauryl ether [Brij 58], Pluronic F-68, polyethylene glycol, or a mixture of two or more of these. It is preferable to use one or more surfactants selected from Triton X-100, Tween 20, and Tween 80, and it is more preferable to use Tween 20.

[0083] The concentration of the surfactant is not particularly limited, but when Tween 20 is used, for example, the concentration is preferably 0.001 to 1 part by weight, more preferably 0.005 to 0.5 parts by weight, and particularly preferably 0.01 to 0.3 parts by weight, relative to 100 parts by weight of the solution.

[0084] The viscosity of the stamping liquid is not particularly limited as long as it allows formation of a spot of the desired shape, but may be, for example, 0.4 mPa·s to 40 mPa·s at 25°C, preferably 0.5 mPa·s to 10 mPa·s, more preferably 0.6 mPa·s to 4 mPa·s, particularly preferably 0.8 mPa·s to 2 mPa·s, and particularly preferably about 1.3 mPa·s. The viscosity of the stamping liquid at 25°C may be 0.6 mPa·s to 10 mPa·s, 0.8 mPa·s to 4 mPa·s, 1.0 mPa·s to 2 mPa·s, 1.12 mPa·s to 2 mPa·s, 1.13 mPa·s to 2 mPa·s, 1.14 mPa·s to 2 mPa·s, 1.15 mPa·s to 2 mPa·s, 1.16 mPa·s to 2 mPa·s, 1.17 mPa·s to 2 mPa·s, 1.18 mPa·s to 2 mPa·s, 1.19 mPa·s to 2 mPa·s, 1.2 mPa·s to 2 mPa·s, etc.

[0085] The method for spotting the immobilized substance or the immobilization carrier on which the substance is immobilized and the photocrosslinker onto the substrate is not particularly limited, and examples thereof include spotting using a non-contact dispenser, spotting using a micropipette, pin-based spotting, and piezoelectric spotting. Spotting using a non-contact dispenser is preferred because it allows a more accurate amount of dispersion to be applied to the substrate and reduces variation in the amount of immobilized substance immobilized at each spot compared to methods that require contact with the substrate, such as pin-based spotting. In a specific embodiment, a non-contact dispenser with a nozzle diameter of 100 to 2000 times, preferably 150 to 1000 times, more preferably 200 to 700 times, and particularly 500 times the average maximum diameter of the solid carrier can be used. The spot diameter can be, for example, 400 μm to 800 μm, preferably 500 μm to 700 μm.

[0086] When spots are formed using a non-contact dispenser, the coefficient of variation in the amount (volume) of the substance to be immobilized on each spot can be made less than 11.8%. Thus, one embodiment of the biochip of the present invention has a plurality of spots on which substances to be immobilized are immobilized, and the coefficient of variation in the volume of the substance to be immobilized on each spot is less than 11.8%, for example, 11.5% or less, 11.0% or less, 10.5% or less, 10.0% or less, 9.5% or less, 9.0% or less, 8.5% or less, 8.0% or less, 7.5% or less, 7.0% or less, 6.5% or less, 6.0% or less, 5.5% or less, 5.0% or less, 4.5% or less, 4.0% or less, 3.5% or less, or 3.0% or less. Because the volume of the immobilized substance per spot is roughly proportional to the volume of the solid content per spot, the coefficient of variation of the volume of the immobilized substance per spot can be substituted for the coefficient of variation of the volume of the solid content per spot. The volume of the solid content present in a spot can be measured, for example, using image analysis software (such as Keyence's Multi-Analysis Application) from 3D images taken with a laser microscope (particularly a shape analysis laser microscope such as the VK-X series manufactured by Keyence Corporation).

[0087] Immobilization of the substance to be immobilized or the immobilization carrier on which the substance to be immobilized is immobilized can be carried out by irradiating with light after coating, preferably after drying the coated solution. The light used may be any light that can generate radicals from the photoreactive group used, and in particular, when an azide group is used as the photoreactive group, ultraviolet light (for example, wavelength 10 to 400 nm) is preferred. The irradiation time can be, for example, 10 seconds to 120 minutes, preferably 30 seconds to 60 minutes, and more preferably 1 minute to 30 minutes. Since the substance to be immobilized is rapidly immobilized by irradiation, the irradiation time is approximately equal to the time required for immobilization. The dose of the irradiated light is not particularly limited, but is usually 1 cm 2 The light irradiation is about 1 mW to 100 mW per unit time. This light irradiation causes the photoreactive group contained in the photocrosslinking agent (or, if the polymer has a photoreactive group, the photoreactive group) to generate radicals, which can bond the polymer layer and the substance to be immobilized via the photocrosslinking agent when the substance to be immobilized is not immobilized on an immobilization carrier, or can bond the polymer layer and the substance to be immobilized and / or the immobilization carrier when the substance to be immobilized is immobilized on an immobilization carrier.

[0088] After immobilizing the desired substance as described above, the substrate can be washed by a known method to remove unreacted components, etc., but such washing is not essential; the product may be manufactured with the stamping liquid remaining. In this way, a biochip on which the desired substance to be immobilized is immobilized can be obtained. If washing is not performed, components of the stamping liquid, such as a thickener and / or surfactant, may adhere to the spots on the biochip in addition to the immobilized substance. Such stamping liquid residue can be removed by appropriate washing before using the biochip.

[0089] 5. Allergy Disease Assessment System Another aspect of the present invention is a data acquisition unit that acquires data on the reaction between the probes and antibodies contained in the biological sample and data on the type of the reacted antibody, obtained by contacting two or more probes selected from allergens and fragments thereof with a biological sample collected from a subject; a data processing unit that inputs the data into a prediction model and obtains a prediction result; a determination unit that determines an allergic disease based on the prediction result; and Output section that outputs the judgment result The present invention relates to an allergic disease determination system (hereinafter, sometimes referred to as the "system of the present invention") comprising: a predictive model that is a nonlinear model or a numerical model having two or more explanatory variables and two or more terms, created using at least training data on the reaction of the probe with an antibody and training data on the type of antibody bound to the probe.

[0090] The system of the present invention will be described below with reference to the drawings. Note that the allergic disease determination system described below is intended to exemplify the present invention and is not intended to limit the present invention. Furthermore, the components of the system of the present invention that are common to the determination method of the present invention, such as allergic diseases, probes, data related to reactions between probes and antibodies contained in biological samples, data related to antibody types, and prediction models, are as described above for the determination method of the present invention.

[0091] 1 is a block diagram showing the functional configuration of an allergic disease determination system 100. The allergic disease determination system 100 includes a data acquisition unit 101, a data processing unit 102, a determination unit 103, and an output unit 104. The data acquisition unit 101 acquires and stores data related to the reaction between the probe and the target antibody and data related to the type of antibody that reacted. Other data to be used in the prediction model can also be acquired and stored as needed. Data acquisition can be performed directly from a detector that detects the reaction between the probe and the target antibody, from an input device such as a keyboard or touch panel, by reading electronic data via a communication interface such as a storage medium interface or a network interface, or by a combination of these acquisition methods. The data is stored in any writable data storage device, such as a ROM, RAM, magnetic disk, or magneto-optical disk.

[0092] The data processing unit 102 accesses the data stored in the data acquisition unit 101, reads out information necessary for the prediction model, inputs it into the prediction model stored in the data processing unit 101, and obtains a prediction result. The prediction result is sent to the determination unit 103, where it is linked to the determination result, and the determination result is sent to the output unit 104. Linking with the determination result can be performed by various methods. For example, when a discontinuous item (e.g., a binary classification of severe or mild) is determined using a prediction model based on a prediction formula, the prediction result (predicted value) obtained by the prediction formula can be linked to the determination result by comparing it with a preset reference value (e.g., a cutoff value). On the other hand, if the prediction model is configured to directly output the determination item (e.g., output of a discontinuous item using a decision tree or output of a continuous item (e.g., ASCA value) using a prediction formula), the prediction result can also be used as the determination result directly. The output unit 104 outputs the transmitted determination result. The output format is not particularly limited, and can be, for example, display on various displays, output by a printer, transmission of electronic data, storage in a portable storage medium, etc. The output unit 104 can also be configured to store the determination result sent from the determination unit 103.

[0093] In addition to the above, the system of the present invention may have further functions and configurations useful for diagnosing allergic diseases. For example, the system of the present invention can output information related to the determination item together with the determination result (related information output function). For example, if the determination item is whether symptoms upon exposure to an allergen are mild or severe, the determination result may include not only information on whether the symptoms are mild or severe, but also related information on various provocation tests (such as oral food challenge tests). Such related information may include, for example, information that, if the symptoms are mild, provocation tests can be performed on an outpatient basis, and, for example, information that, if the symptoms are severe, provocation tests should or should not be performed after hospitalization.

[0094] The predictive model can be updated as needed by optimizing it with additional training data, etc. The predictive model can be updated manually by a user, or automatically by the system accessing a server or the like where the updated predictive model is stored and acquiring the updated predictive model (predictive model update function). The reference value used for linking with the judgment result may be changeable. For example, the system of the present invention may be configured so that the reference value can be selected from a plurality of candidates, or so that the user can input the reference value. Therefore, in the system of the present invention, for example, the judgment unit 103 may have a function of allowing the user to select a reference value from a plurality of candidates or a function of allowing the user to input a reference value (reference value change function). The system of the present invention may include a detection unit that detects the reaction between the probe and the target antibody. The detection unit may include a detector that detects the reaction between the probe and the target antibody, and a transmitter that transmits data regarding the reaction between the probe and the target antibody, data regarding the type of antibody, etc. to the data acquisition unit. The system of the present invention may also be equipped with a user interface that enables operation of the system and data input, a power source, a communication interface that enables external information to be acquired via the Internet, etc.

[0095] An example of data processing in the system of the present invention, in which the prediction result (predicted value) is compared with a reference value to link the prediction result and the determination result, is described with reference to the flowchart in FIG. 2. First, when the system is started, the determination unit 103 presents candidate reference values ​​stored in the storage unit to the user (S1). When the user selects a reference value, the selected reference value is stored in the storage unit of the determination unit 103. Next, the data acquisition unit 101 prompts the user to input data, and the input data is stored in the storage unit of the data acquisition unit 101 (S2). Next, the data processing unit 102 reads the data stored in the storage unit of the data acquisition unit 101 and the prediction model stored in the storage unit of the data processing unit 102 (S3), inputs the data into the prediction model (S4), and obtains a predicted value (S5). If only the predicted value is to be output (Yes in S6, No in S7), the obtained predicted value is sent to the output unit 104, and the predicted value is output (S10), and the processing ends. The obtained predicted value can be directly used as an index for allergic disease determination. In particular, when the predicted value is continuous (e.g., the result of calculation using a prediction formula), it is possible to assist sensory judgment, such as a sense of level. When the predicted value and the judgment result are output (Yes in S6, Yes in S7), the predicted value is sent to the judgment unit 103 and compared with a reference value selected by the user and stored in the storage unit of the judgment unit 103 (S8). The predicted value and the judgment result are sent to the output unit 104, and the predicted value and the judgment result are output (S9), and the processing ends. When only the judgment result is output (No in S6), the predicted value is sent to the judgment unit 103 and compared with a reference value selected by the user (S11), and only the judgment result is sent to the output unit 104, and the judgment result is output (S12), and the processing ends. The judgment result may include information related to the judgment item (e.g., information on whether symptoms upon exposure to an allergen are mild or severe), as well as information accompanying that information (e.g., information regarding the implementation of the provocation test as described above).

[0096] The system of the present invention can also be realized by a computer. FIG. 3 is a block diagram showing an example of the hardware configuration of a computer 200 that realizes a standalone system of the present invention. The computer 200 includes a monitor 210, a communication interface 220, a CPU 230, a RAM 240, and a storage 250. The storage 250 stores a determination program 251 for executing various processes of the system of the present invention, acquired data 252 used for determining allergic diseases, a prediction model 253, and reference values ​​254. The CPU 230 reads and executes the code of the determination program 251 stored in the storage 250. Thus, the CPU 230 realizes functions such as the data processing unit 102 and the determination unit 103 in the system of the present invention. The storage 250 realizes the storage functions of the data acquisition unit 101, the processing unit 102, and the determination unit 103 in the system of the present invention. The RAM 240 stores intermediate data and the like during program execution. The monitor 210 realizes functions such as the output unit 104 and a user interface. The communication interface 220 realizes the functions of the data acquisition unit 101, the output unit 104, etc. In addition, by reading the code of the determination program 251, a part or all of the processing related to the system of the present invention may be realized by an OS or other software running on a computer.

[0097] The present invention also relates to a program for implementing the determination method and / or index acquisition method of the present invention using a computer or the like, and a storage medium on which the program is stored (e.g., a flexible disk, a hard disk, a magnetic disk, a magneto-optical disk such as a CD-ROM, MO, DVD, or BD, a magnetic tape, or a semiconductor memory such as a flash memory). [Example]

[0098] The present invention will be described in more detail below with reference to examples, but the scope of the present invention is not limited to these examples. <Example 1: Biochip manufacturing> A polycarbonate sheet (Mitsubishi Gas Chemical Company, MR58U, 0.8 mm thick) was irradiated for 2 minutes at a distance of 50 mm using a UV-ozone irradiation device (Sen Special Light Sources, SSP16-110, UV lamp: SUV110GS-36L). Next, 0.5 parts by weight of polyethylene glycol monomethacrylate polymer (Sanyu Chemical Laboratory, polyethylene glycol molecular weight 350) and 0.025 parts by weight of 4,4'-diazidostilbene-2,2'-disulfonic acid (Tokyo Chemical Industry Co., Ltd., 98%, hereafter referred to as bisazide) were dissolved in 75% aqueous ethanol to obtain a coating solution for the nonspecific adsorption inhibitor. Using a spin coater (MIKASA, MS-A100), 15 μL of this coating solution was coated onto the UV-treated polycarbonate at 800 rpm for 5 seconds and 5,000 rpm for 10 seconds. After coating, a UV irradiation device (UVP, CL-1000) was used to irradiate the film with 120 mW / cm 2 The substrate was irradiated with light at 1000 kJ / s for 10 minutes. A silicone rubber O-ring (14φ) was attached to the coated substrate with adhesive to form a reaction area. The substrate was then aged at 25°C and 65% humidity for 4 days to obtain a substrate for a biochip.

[0099] 0.8 parts by weight of sodium chloride, 0.02 parts by weight of potassium chloride, 0.29 parts by weight of disodium hydrogen phosphate dodecahydrate, and 0.02 parts by weight of potassium dihydrogen phosphate were dissolved in ultrapure water to obtain a PBS solution. 0.59 parts by weight of 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid was dissolved in ultrapure water to obtain a HEPES solution. 0.1 parts by weight of polyvinyl alcohol (Wako Pure Chemical Industries, Ltd., 160-03055) and 0.05 parts by weight of Tween 20 (Sigma-Aldrich, P7949-100ML) were dissolved in the HEPES solution to obtain a solvent for the stamping liquid. The immobilized substances were αs1-casein, β-casein, β-lactoglobulin, and partial peptides of αs1-casein (20 overlapping fragments consisting of 15 amino acid residues (SEQ ID NOS: 2-21)) biotinylated at the termini using an aminohexyl (Ahx) group as a spacer. Each was reacted with Streptavidin beads (particle size 200 nm, Tamagawa Seiki Co., Ltd.) in PBS solution at 4°C for 1.5 hours, purified, and dispersed in a stamping solution solvent to obtain a stamping solution. Figure 4 shows a schematic diagram of the partial peptides immobilized on the beads. The structures of the biotinylated peptides are shown below. [ka]

[0100] Bisazide was dissolved in ultrapure water to prepare a 10 mg / mL bisazide solution. The 10 mg / mL bisazide solution was diluted 5-fold. The bisazide solution was dissolved in the stamp solution to make the total volume 4.8 parts by weight. Stamp solutions, each containing one type of partial peptide, were spotted onto a biochip substrate using a non-contact dispenser (BioDot, Non-contact Micro Dispensing System AD1520) by a grid-type multi-point dispense spotting method, creating a total of 81 spots in a 9 x 9 arrangement. After spotting, the beads were dried for 10 minutes at 0.09 MPa in a vacuum dryer. After drying, a UV irradiation device (UVP, CL-1000) was used at 4 mW / cm to immobilize the beads with the partial peptide attached to the substrate using a photocrosslinking agent. 2 The biochip was obtained by irradiating the sample with 1000 kJ / cm² for 10 minutes.

[0101] <Example 2: Measurement using a chip> Using the biochip produced in Example 1, the reaction between IgE antibodies and IgG4 antibodies in serum collected from subjects and the partial peptides immobilized on the biochip was measured. The composition of the subjects is as shown in Table 3 below. Subjects with Immunocap-specific IgE (casein) levels of less than 0.35 were considered non-patients, subjects with Immunocap-specific IgE (casein) levels of 0.35 or higher and an ASCA level of 2 or lower, and subjects with Immunocap-specific IgE (casein) levels of 0.35 or higher, a total milk load of 6 mL or higher, and an ASCA level of 11 or lower were considered mild patients, and all other subjects were considered severe patients. [Table 3]

[0102] The reaction area of ​​the biochip was blocked for 1 hour at room temperature with a protein-free blocking agent containing biotin (PVDF Blocking Reagent for Can Get Signal (Toyobo, NYPBR01) with 0.02 wt% biotin added), and then washed three times with TBS-T solution (137 mM sodium chloride, 2.68 mM potassium chloride, 25 mM trishydroxymethylaminomethane, pH 7.4, 0.1 wt% Tween 20). 130 μL of 8-fold diluted serum was added as a test sample to the reaction area of ​​the biochip, and the reaction was allowed to proceed at room temperature for 8 minutes while shaking. The test sample was removed by suction and washed with TBS-T solution. After washing, 130 μL of ALP-labeled anti-human IgE monoclonal antibody (Abcam, ab99805) or ALP-labeled anti-human IgG4 monoclonal antibody (Abcam, ab99822) diluted 2000-fold with Can Get Signal Immunoreaction Enhancer Solution 2 (Toyobo, NKB-301) was added and incubated at room temperature for 4 minutes with shaking. The antibody was removed by aspiration and the plate was washed with TBS-T. 130 μL of luminescence reagent (Dynalight Substrate with Rapid Glow Enhancer, Molecular Probes, 4475406) ​​was added and incubated for 1 minute. The luminescence intensity was measured by counting the number of pixels in the luminescent area using Dynacom's SpotSolver and NHI's ImageJ software, using the non-spot area as the background.

[0103] (1) Determining mild or severe symptoms Prediction models were created using the following method. In each prediction model, an and cn represent coefficients, xn represents explanatory variables, h() represents the hinge function, and n represents a natural number. Each coefficient and explanatory variable is independent for each prediction model and is not necessarily the same between different prediction models. For example, coefficient a1 in the following prediction model 1-1 is independent of coefficient a1 in prediction model 1-2 and can take different values. In addition, normalized values ​​of measured values ​​(e.g., probe luminescence intensity) were used as explanatory variables.

[0104] <Prediction Model 1-1> A prediction model was constructed using LASSO regression assuming a binomial distribution, with the luminescence intensity of each probe, the subject's age, total IgE, and ImmunoCap-specific IgE (casein, milk) as explanatory variables, and mild or severe as the objective variable. Non-patients and mildly ill patients in Table 3 were classified as mild, and severe patients as severe (the same applies to prediction models 1-2 to 1-8). For the luminescence intensity of each probe, an interaction term between two factors (the product of the two variables) was added as an explanatory variable. Leave-one-out cross validation (LOOCV) was performed, and prediction model 1-1, represented by the following prediction formula, was obtained at the point where the prediction error was minimized. a1+x1×a2+x2×a3+x22×x3×a4+x4×x2×a5+x3×x5×a6+x3×x6×a7+x21×x6×a8+x7×x8×a9+x7×x9×a10+x7×x10×a11+x11×x12×a12+x11×x 13 x a13 + x11 x x9 x a14 + x11 x x14 x a15 + x2 x x8 x a16 + x15 x x6 x a17 + x15 x x16 x a18 + x17 x x18 x a19 + x13 x x19 x a20 + In the above prediction formula, x1 is the immunocap-specific IgE (milk), x2 to x7, x11, x12, x15, x17, x21, and x22 are the luminescence intensities of different probes (IgE), and x8 to x10, x13, x14, x16, and x18 to x20 are the luminescence intensities of different probes (IgG4).

[0105] <Prediction Model 1-2> A prediction model was constructed in the same manner as prediction model 1-1, except that only the luminescence intensity of each probe was used as an explanatory variable, and prediction model 1-2, represented by the following prediction formula, was obtained. a1+x1×a2+x2×x3×a3+x4×x5×a4+x3×x6×a5+x7×x8×a6+x7×x9×a7+x10×x11×a8+x10×x12×a9+x13×x11×a10+x13×x14×a11+x13×x12× a12+x13×x15×a13+x13×x16×a14+x1×x17×a15+x18×x9×a16+x18×x19×a17+x5×x20×a18+x14×x15×a19+x21×x15×a20+x15×x16×a21 In the above prediction formula, x1 to x11, x13, and x18 are the luminescence intensities of the different probes (IgE), and x12, x14 to x17, and x19 to x21 are the luminescence intensities of the different probes (IgG4).

[0106] <Prediction Models 1-3> A prediction model was constructed in the same manner as prediction model 1-1, except that the subject's age, total IgE, and Immunocap-specific IgE (casein, milk) were used as explanatory variables, and prediction model 1-3, represented by the following prediction formula, was obtained. a1+x1×a2+x2×a3 In the above prediction formula, x1 is Immunocap-specific IgE (milk), and x2 is Immunocap-specific IgE (casein).

[0107] <Prediction Model 1-4> A classification tree was constructed using the CART method with the same explanatory and objective variables as in prediction model 1-1. First, a large tree was constructed, and then the average misclassification rate was calculated using 10-fold cross validation. This was then pruned at the minimum position to obtain prediction model 1-4, represented by the decision tree in Figure 5. In the figure, x1 is the immunocap-specific IgE (milk), and x2 to x4 are the luminescence intensities of the different probes (IgE). <Prediction Models 1-5> Using the same explanatory variables and objective variables as in prediction model 1-3 and the same procedure as in prediction model 1-4, prediction model 1-5, represented by the decision tree in Figure 6, was obtained. In the figure, x1 is Immunocap-specific IgE (milk).

[0108] <Prediction Models 1-6> Using the same explanatory variables and objective variables as in prediction model 1-1, a model was constructed using the MARS method, assuming a binomial distribution and taking into account the interaction between the two factors. The position where the R-squared value was maximized was determined using 10-fold cross validation, and prediction model 1-6, represented by the following prediction formula, was obtained. a1+h(c1-x1)×a2+h(x1-c1)×x2×a3+h(X3-c2)×h(c3-X4)×a4+h(c4-X4)×a5+h(c5-X5)×a6 In the above prediction formula, x1 is the luminescence intensity of Immunocap-specific IgE (milk), x2 is the luminescence intensity of the probe (IgG4), x3 is total IgE, Immunocap-specific IgE (casein), and x4 to x5 are the luminescence intensities of different probes (IgE).

[0109] <Prediction Model 1-7> Using the same explanatory variables and objective variables as in prediction model 1-2, prediction model 1-7, represented by the following prediction formula, was obtained using the same procedure as in prediction model 1-6. a1+h(x1-c1)×a2+h(c1-x1)×a3+h(x2-c2)×a4+h(c3-x1)×x2×a5+h(x1-c4)×a6 In the above prediction formula, x1 and x2 are the luminescence intensities of the different probes (IgE).

[0110] <Prediction Model 1-8> Using the same explanatory variables and objective variables as in prediction model 1-3, and following the same procedure as in prediction model 1-6, prediction model 1-8, represented by the following prediction formula, was obtained. a1+h(c1-x1)×a2+h(x2-c2)×a3+h(x2-c3)×h(x3-c4)×a4 In the above prediction formula, x1 is Immunocap-specific IgE (milk), x2 is Immunocap-specific IgE (casein), and x3 is total IgE.

[0111] ROC curves and dot histograms (Figures 7-10, 13-14, 17-20) were created from the predicted values ​​for each subject obtained using prediction models 1-1, 1-2, 1-4, 1-6, and 1-7. For comparison, ROC curves and dot histograms (Figures 11-12, 15-16, and 21-22) were also created from the predicted values ​​for each subject obtained using prediction models 1-3, 1-5, and 1-8. The AUC (Area Under Curve) for each prediction model is summarized in Table 4. [Table 4]

[0112] (2) Prediction of ASCA values <Prediction Model 2-1> A model was constructed using LASSO regression, assuming a normal distribution, with the luminescence intensity of each probe, the subject's age, total IgE, ImmunoCap-specific IgE (casein, milk), and IgE dilution rate as explanatory variables, and the subject's ASCA value as the objective variable. For the luminescence intensity of each probe, an interaction term between two elements (the product of the two variables) was added as an explanatory variable. Leave-one-out cross validation was performed, and prediction model 2-1, represented by the following prediction formula, was obtained at the point where the prediction error was minimized. a1+x1×a2+x2×a3+x3×a4+x4×a5+x5×x3×a6+x6×x3×a7+x6×x4×a8+x7×x8×a9+x7×x9×a 10+x7×x10×a11+x7×x11×a12+x12×x8×a13+x13×x14×a14+x8×x9×a15+x3×x9×a16+x3 ×x15×a17+x14×x16×a18+x14×x17×a19+x14×x18×a20+x14×x19×a21+x20×x21×a22+x22×x15×a23+x23×x15×a24+x23×x24×a25+x15×x25×a26+x15×x26×a27+x15×x24×a28 In the above prediction formula, x1 is Immunocap-specific IgE (milk), x2 is the IgE dilution rate, x3 to x8, x12 to x14, and x20 are the luminescence intensities of different probes (IgE), and x9 to x11, x15 to x19, and x21 to x26 are the luminescence intensities of different probes (IgG4).

[0113] <Prediction Model 2-2> A prediction model was constructed in the same manner as prediction model 2-1, except that the subject's age, total IgE, and Immunocap-specific IgE (casein, milk) were used as explanatory variables, and prediction model 2-2, represented by the following prediction formula, was obtained. a1+x1×a2+x2×a3+x3×a4+x4×a5 In the above prediction formula, x1 is the subject's age, x2 is Immunocap-specific IgE (milk), x3 is Immunocap-specific IgE (casein), and x4 is total IgE.

[0114] <Prediction Model 2-3> A regression tree was constructed using the CART method with the same explanatory variables and objective variables as in prediction model 2-1. First, a large tree was constructed, and then the average misclassification rate was calculated using 10-fold cross validation. This was then pruned at the minimum position to create prediction model 2-3, represented by the decision tree in Figure 23. In the figure, x1 is the luminescence intensity of the probe (IgE), and x2 is ImmunoCap-specific IgE (milk). <Prediction Model 2-4> Using the same explanatory variables and objective variables as in prediction model 2-2 and the same procedure as in prediction model 2-3, prediction model 2-4, represented by the decision tree in Figure 24, was obtained. In the figure, x1 is Immunocap-specific IgE (milk).

[0115] <Prediction Model 2-5> Using the same explanatory variables and objective variables as in prediction model 2-1, a model was constructed using the MARS method, assuming a normal distribution and taking into account the interaction between two factors. Generalized cross validation was used to determine the position where the R-squared value was maximized, and prediction model 2-5, represented by the following prediction formula, was obtained. a1+h(x1-c1)×a2+h(c2-x2)×a3+h(c2-x2)×x3×a4+h(x4-c3)×a5+h(x1-c1)×x5×a6+h(x1-c1) ×h(x6-c4)×a7+h(x1-c1)×h(c4-x6)×a8+h(x6-c5)×a9+h(x1-c1)×x7×a10+x8×h(c2-x2)×a11 +h(c6-x9)×h(c3-x4)×a12+h(x10-c7)×a13+h(c8-x11)×h(c5-x6)×a14+h(c9-x12)×h(c3-x4 )×a15+h(c5-x6)×h(x13-c10)×a16+h(c11-x14)×h(c3-x4)×a17+h(c12-x15)×h(c3-x4)×a18 In the above prediction formula, x1 is Immunocap-specific IgE (milk), x5 is total IgE, x2, x8, x9, x12, x14, and x15 are the luminescence intensities of different probes (IgE), and x3, x4, x6, x7, x10, x11, and x13 are the luminescence intensities of different probes (IgG4).

[0116] <Prediction Model 2-6> Using the same explanatory variables and objective variables as in prediction model 2-2 and following the same procedure as in prediction model 1-6, prediction model 2-6, represented by the following prediction formula, was obtained. a1xh(c1-x1)×a2+h(c2-x2)×h(x1-c1)×a3 In the above prediction formula, x1 is Immunocap-specific IgE (casein) and x2 is Immunocap-specific IgE (milk).

[0117] Scatter plots (Figures 25, 27, and 29) were created for the predicted values ​​(horizontal axis) and observed values ​​(vertical axis) for each subject obtained using prediction models 2-1, 2-3, and 2-6. For comparison, scatter plots (Figures 26, 28, and 30) were also created for the predicted values ​​(horizontal axis) and observed values ​​(vertical axis) for each subject obtained using prediction models 2-2, 2-4, and 2-6. The correlation coefficients for each prediction model are summarized in Table 5. [Table 5]

[0118] These results demonstrate that the inclusion of IgE and IgG4 reactivity to allergens or their fragments as parameters allows for more accurate differentiation between severe and mild cases and prediction of ASCA values. [Explanation of symbols]

[0119] 100: Allergy disease assessment system 101: Data acquisition section 102: Data processing unit 103: Judgment section 104: Output section 200: Computer 210: Monitor 220: Communication interface 230:CPU 240:RAM 250:Storage 251: Judgment program 252: Acquired data 253: Predictive Model 254: Reference value

Claims

1. A data acquisition unit that acquires data on the reaction between a probe containing two or more types selected from allergen fragments and an antibody contained in a biological sample collected from a subject, the data being obtained by contacting the probe with the antibody and the type of antibody that reacted; a data processing unit that inputs the data into a prediction model and obtains a prediction result; a determination unit that determines an allergic disease based on the prediction result; and an output unit that outputs the determination result. wherein the prediction model is a nonlinear model or a model having two or more explanatory variables and two or more terms, created using at least training data on the reaction of the probe with an antibody and training data on the type of antibody bound to the probe; The data relating to the reaction with the antibody includes the strength of the reaction with the antibody, An allergic disease determination system, wherein the statistical method used to create the nonlinear model is the CART method or the MARS method.

2. The system according to claim 1, wherein the determination of the allergic disease is selected from determining whether the allergic disease is mild or severe and predicting the total ASCA (Anaphylaxis Scoring Aichi) value.

3. The system according to claim 1 or 2, wherein the type of antibody is selected from IgE and IgG4.

4. The system of any one of claims 1 to 3, wherein the probes comprise overlapping fragments of the allergen.

5. The system according to any one of claims 1 to 4, wherein the probe is immobilized.

6. The system of any one of claims 1 to 5, wherein additional training data is used to create the predictive model.

7. The system according to any one of claims 1 to 6, wherein the allergen is selected from a food allergen, an inhalant allergen and a contact allergen.

8. The system according to any one of claims 1 to 7, wherein the food allergen is selected from casein, lactalbumin, lactoglobulin, ovomucoid, ovalbumin, conalbumin, lysozyme, gliadin, amylase inhibitor, albumin, globulin, gluten and tropomyosin.

9. The system according to any one of claims 1 to 8, wherein the allergic disease is a food allergy.

10. A process for acquiring data on the reaction between the probe containing two or more types selected from fragments of allergens and an antibody contained in a biological sample collected from a subject, the data being obtained by contacting the probe with the antibody contained in the biological sample and the type of antibody that reacted, and a process for inputting the data into a prediction model to obtain a prediction result; the prediction model is a nonlinear model or a model having two or more explanatory variables and two or more terms, created using at least training data on the reaction of the probe with an antibody and training data on the type of antibody bound to the probe; The data relating to the reaction with the antibody includes the strength of the reaction with the antibody, A method for obtaining an index for determining allergic disease, wherein the statistical method used to create the nonlinear model is the CART method or the MARS method.

11. The method described in claim 10, wherein the type of antibody is selected from IgE and IgG4.

12. A method according to claim 10 or 11, wherein the probe comprises overlapping fragments of the allergen.

13. A method described in any one of claims 10 to 12, wherein the probe is immobilized.

14. A method described in any one of claims 10 to 13, wherein additional training data is used to create the predictive model.

15. A method according to any one of claims 10 to 14, wherein the allergen is selected from a food allergen, an inhalant allergen and a contact allergen.

16. A method according to any one of claims 10 to 15, wherein the food allergen is selected from casein, lactalbumin, lactoglobulin, ovomucoid, ovalbumin, conalbumin, lysozyme, gliadin, amylase inhibitor, albumin, globulin, gluten and tropomyosin.

17. A method according to any one of claims 10 to 16, wherein the allergic disease is a food allergy.

18. A program for causing a computer to execute a method for determining an allergic disease, comprising: The method comprises: contacting a probe containing two or more allergen fragments with a biological sample collected from a subject; detecting a reaction between the probe and an antibody contained in a biological sample; a step of determining the type of antibody that reacted with the probe; A step of inputting data on the reaction of each of the at least two types of probes with antibodies and data on the type of antibody bound to the probe into a prediction model to obtain a prediction result; and a step of diagnosing an allergic disease based on the prediction result, wherein the prediction model is a nonlinear model or a model having two or more explanatory variables and two or more terms, created using at least training data on the reaction of the probe with an antibody and training data on the type of antibody bound, Data on antibody reactions, including antibody reaction strength, A program in which the statistical method used to create the nonlinear model is the CART method or the MARS method.

19. A storage medium storing the program according to claim 18.

Images