An analysis model for judging skin characteristics and its construction method

Abstract

The present invention discloses an analysis model for judging skin characteristics and a construction method thereof, and relates to the field of gene detection technology. The present invention discloses a construction method for judging skin characteristics analysis model, comprising the following steps: S1, obtaining the correspondence between the skin features of a sample and the skin characteristics, and establishing a database; S2, obtaining the gene data of the sample and formatting the gene data; S3, selecting a machine learning model according to the formatted gene data of the sample and the database, and constructing a skin characteristics judgment model for judging specific skin characteristics based on the sample gene data. The skin characteristics analysis model provided by the present invention can accurately judge skin characteristics, and the method for detecting using the skin characteristics model is simple, and skin characteristics can be judged efficiently.

Description

Technical Field

[0001] The present invention relates to the technical field of gene detection, and in particular to an analysis model for judging skin characteristics and a construction method thereof. Background Art

[0002] The skin is the largest organ in the human body and the part where the human body is in direct contact with the environment. It can isolate the human body from most adverse factors. The skin is mainly divided into the epidermis, dermis and subcutaneous fat from the outside to the inside. The epidermis is the innermost layer, the basal layer. The cells here will continue to divide after generation, and then gradually move upward and deform to form the other layers of the epidermis, from the outside to the inside, they are the stratum corneum, granular layer and spinous cell layer. The dermis is mainly composed of two types of proteins, namely collagen and elastin. The dermis is the main part of the distribution of the vascular network of the skin tissue, which can transport nutrients to the skin tissue. Subcutaneous fat is located in the lower part of the dermis, mainly composed of loose connective tissue and fat lobules, and has the functions of heat preservation, resistance to external mechanical shock and energy storage.

[0003] Currently, skin characteristics can be reflected through skin gene testing. Skin gene testing refers to the analysis of various skin capabilities through high-throughput sequencing biotechnology. Commercial skin gene testing services have appeared on the market. Common skin gene testing projects include six major items: whitening ability, moisturizing ability, anti-wrinkle ability, glycation risk, antioxidant ability, anti-inflammatory and anti-allergic. Skin gene testing has many problems, such as high difficulty in testing, inaccurate test results, long testing time, high requirements for the operating environment, and the requirement for testers to have certain qualifications.

[0004] Chinese patent CN109706253A discloses a skin gene detection method, which includes the following steps: S1. Providing a test sample of a test subject, wherein the test sample includes a saliva sample; S2. Extracting a nucleic acid sample from the saliva sample, and performing SNP detection on 8 sites of 8 genes in the nucleic acid sample, wherein the 8 genes and sites are rs13312741 site of the TYR gene, rs1805007 site of the MC1R gene, rs11568737 site of the SLC45A2 gene, rs4880 site of the SOD2 gene, rs1799750 site of the MMP1 gene, rs222747 site of the TRPV1 gene, rs61816761 site of the FLG gene, and rs11103631 site of the AQP3 gene; S3. Analyzing the test results according to the data obtained in step S2 to determine the skin characteristics of the test subject. This gene detection method can detect the characteristics of the skin at the genetic level, which can reflect the true condition of the tester's skin and provide targeted long-term skin care. However, the accuracy of the detection needs to be further improved.

[0005] Chinese patent CN106498071A discloses a system for detecting and evaluating skin quality at the genetic level, which is mainly used for detecting and evaluating the quality of healthy human skin. It uses typing of 19 gene loci to calculate the average score and obtain skin performance evaluation. The 19 gene loci include: Rs179950, Rs3025058, Rs3918242, Rs7616661, Rs4880, Rs1800566, Rs1050450, Rs1001179, Rs17553719, Rs1834640, Rs2228478, Rs1800414, Rs25487, Rs13181, Rs7531806, Rs747650, Rs478304, Rs8032158, and Rs873549. Through detection and evaluation, the invention can help some cosmetics companies or beauty salons provide customers with personalized skin care products and maximize the effect of use, and can also expand the potential consumer groups of some health care product manufacturers.

[0006] However, there is no report in the prior art that uses skin gene data to construct a skin characteristic analysis model for skin characteristic analysis. Summary of the invention

[0007] The object of the present invention is to provide an analysis model for determining skin characteristics.

[0008] To achieve the above-mentioned object of the invention, the technical solution of the present invention is as follows:

[0009] In one aspect, the present invention provides a method for constructing a skin characteristic analysis model, comprising the following steps:

[0010] S1. Obtain the correspondence between the skin features and skin characteristics of the sample and establish a database;

[0011] S2, obtaining the genetic data of the sample and formatting the genetic data;

[0012] S3. According to the formatted sample gene data and the database, a machine learning model is selected to construct a skin characteristic determination model for determining specific skin characteristics based on the sample gene data.

[0013] Specifically, the skin characteristics described in step S1 include: anti-photoaging ability, anti-dryness ability, anti-wrinkle ability, anti-glycation ability, antioxidant ability, and anti-skin barrier damage ability.

[0014] Specifically, the detection sites for anti-photoaging ability in step S1 include: rs1800414, rs62543565, rs10733310, rs4911414, rs1015362;

[0015] The detection sites of anti-desiccation ability in step S1 include: rs61816761, rs11103631;

[0016] The detection sites of anti-wrinkle ability in step S1 include: rs1800796, rs1799750;

[0017] The detection sites of anti-glycation ability in step S1 include: rs1800625, rs2070600, rs1049346;

[0018] The detection sites of antioxidant capacity in step S1 include: rs1800566, rs713041;

[0019] The detection sites for the ability to resist skin barrier damage in step S1 include: rs612529, rs2303067;

[0020] The detection sites for oil control ability in step S1 include: rs174547 and rs743572.

[0021] Specifically, the corresponding relationship between the anti-aging ability and the skin characteristics in step S1 is:

[0022] The skin's UV resistance is divided into three levels from high to low. Taking rs1800414 as the standard, CC is high-end, CT is mid-range, and TT is low-end; taking rs62543565 as the standard, TT is high-end, TC is mid-range, and CC is low-end; taking rs10733310 as the standard, TT is high-end, TG is mid-range, and GG is low-end; taking rs4911414 as the standard, GG is high-end, GT is mid-range, and TT is low-end; taking rs1015362 as the standard, TT is high-end, TC is mid-range, and CC is low-end.

[0023] Skin characteristics: The above capabilities are low-end, corresponding to specific skin characteristics: anti-photoaging level 1, and so on.

[0024] Specifically, the corresponding relationship between the anti-drying ability and the skin characteristics in step S1 is:

[0025] The skin's ability to regulate and delay water loss through the skin is divided into three levels from high to low. Taking rs61816761 as the standard, GG is high-end, GA is mid-range, and AA is low-end. Taking rs11103631 as an example, AA is high-end, AG is mid-range, and GG is low-end.

[0026] Skin characteristics: The above capabilities are low-end, corresponding to specific skin characteristics: anti-dryness level 1, and so on.

[0027] Specifically, the corresponding relationship between the anti-wrinkle ability and the skin characteristics in step S1 is:

[0028] The ability of the skin to regulate collagen synthesis in the skin was divided into three levels from high to low. Using rs1800796 as the standard, CC was high, CG was medium, and GG was low. Using rs1799750 as the standard, DD was high, DI was medium, and II was low.

[0029] Skin characteristics: The above capabilities are low-end, corresponding to specific skin characteristics: anti-wrinkle level 1, and so on.

[0030] Specifically, the corresponding relationship between the anti-glycation ability and the skin characteristics in step S1 is:

[0031] The ability of the receptor for advanced glycation end products protein in the skin reduction period is divided into three levels from high to low. Taking rs1800625 as the standard, TT is high-end, TC is medium-end, and CC is low-end; taking rs2070600 as the standard, CC is high-end, CT is medium-end, and TT is low-end; taking rs1049346 as the standard, GG is high-end, GA is medium-end, and AA is low-end.

[0032] Skin characteristics: The above capabilities are low-end, corresponding to specific skin characteristics: anti-glycation level 1, and so on.

[0033] Specifically, the corresponding relationship between the antioxidant capacity and the skin characteristics in step S1 is:

[0034] The skin's free radical scavenging ability was divided into three levels from high to low. Taking rs1800566 as the standard, GG was high-grade, GA was mid-grade, and AA was low-grade; taking rs713041 as the standard, CC was high-grade, TC was mid-grade, and TT was low-grade.

[0035] Skin characteristics: The above capabilities are low-end, corresponding to specific skin characteristics: antioxidant level 1, and so on.

[0036] Specifically, the corresponding relationship between the ability to resist skin barrier damage and skin characteristics in step S1 is:

[0037] The skin's ability to regulate / repair the skin barrier is divided into three levels from high to low. Taking rs612529 as the standard, TT is high-grade, TC is mid-grade, and CC is low-grade. Taking rs2303067 as the standard, GG is high-grade, GA is mid-grade, and AA is low-grade.

[0038] Skin characteristics: The above capabilities are low-end, corresponding to specific skin characteristics: Level 1 resistance to skin barrier damage, and so on.

[0039] Specifically, the corresponding relationship between the oil control ability and the skin characteristics in step S1 is:

[0040] The skin's ability to inhibit oil secretion is divided into three levels from high to low. Taking rs174547 as the standard, TT is high-end, TC is medium-end, and CC is low-end; taking rs743572 as the standard, AA is high-end, AG is medium-end, and GG is low-end.

[0041] Skin characteristics: The above capabilities are low-end, corresponding to specific skin characteristics: oil control level 1, and so on.

[0042] Specifically, the correspondence between SNP sites and genes is as follows:

[0043]

[0044] Specifically, the OCA2 gene regulates the pH value and maturation of melanosomes, while the BNC2 gene and RALY / ASIP gene regulate the conversion and deposition of melanin into the stratum corneum of the skin.

[0045] Specifically, the FLG gene regulates natural moisturizing factor and skin barrier.

[0046] Specifically, the MMP-1 gene controls collagen degradation. IL-6 can promote keratinocyte proliferation and migration and increase fibroblast proliferation, controlling collagen synthesis.

[0047] Specifically, the AGER gene regulates the advanced glycation end product receptor protein. The GLO1 gene inhibits dicarbonyl-mediated glycation reactions and protects cells from damage by advanced glycation end products.

[0048] Specifically, the NQO1 gene and the GPX4 gene regulate redox processes.

[0049] Specifically, the SPINK5 gene regulates the skin barrier, and the VSTM1 gene regulates the body's allergic response.

[0050] Specifically, the FADS1 gene regulates the ability of skin lipid synthesis, and the CYP17A1 gene regulates the ability of lipid cell induction.

[0051] Specifically, the AQP3 gene encodes a water channel protein in the skin that controls the flow of water in and out of the skin.

[0052] Specifically, the method for obtaining gene data of a sample described in step S2 comprises the following steps:

[0053] (1) Lyse the sample using a lysis buffer;

[0054] (2) Perform qPCR detection on the sample after lysis in step (1) to obtain data.

[0055] More specifically, the lysis solution in step (1) is lysis solution a; the lysis solution a comprises sodium dodecyl sulfate, tris(hydroxymethyl)aminomethane and silver diethyldithiocarbamate.

[0056] Preferably, the concentrations of sodium dodecyl sulfate, tris(hydroxymethyl)aminomethane and silver diethyldithiocarbamate are 1.3%-2.1%wt, 0.1-0.2M, and 120-170mM, respectively.

[0057] In some embodiments of the present invention, the concentrations of sodium dodecyl sulfate, tris(hydroxymethyl)aminomethane and silver diethyldithiocarbamate may be 1.3%wt, 0.1M, and 120mM, respectively.

[0058] In other embodiments, the concentrations of sodium dodecyl sulfate, tris(hydroxymethyl)aminomethane and silver diethyldithiocarbamate may be 2.1%wt, 0.2M, and 170mM, respectively.

[0059] More specifically, the volume of the lysate a is 3-5 times the volume of the sample.

[0060] Preferably, the volume of the lysate a is 3 times the volume of the sample.

[0061] More specifically, the lysate a is heated to 40-45° C. in a water bath before use.

[0062] More specifically, the sample in step (1) includes but is not limited to: oral swab, saliva or biological tissue.

[0063] Preferably, the sample in step (1) is an oral swab.

[0064] More specifically, the qPCR reaction system in step (2) includes reagent A, primers and reagent B; the reagent A includes Taq DNA polymerase, Tris-HCl, KCl, MgSO 4 , dNTP and BSA; the reagent B includes chitosan, lysine and 2,5-dimercapto-1 3 4-thiadiazole.

[0065] Preferably, the concentrations of Taq DNA polymerase, Tris-HCl, KCl, MgSO4, dNTP and BSA can be: 1.7-2.3U, 70-80 mM, 70-90 mM, 1.5-2.2 mM, 0.05-0.4 mM, 0.1-2 mg / mL, respectively.

[0066] In some embodiments, the concentrations of Taq DNA polymerase, Tris-HCl, KCl, MgSO4, dNTP and BSA can be: 1.7 U, 70 mM, 70 mM, 1.5 mM, 0.05 mM, 0.1 mg / mL, respectively.

[0067] In other embodiments, the concentrations of Taq DNA polymerase, Tris-HCl, KCl, MgSO4, dNTP and BSA can be: 2.3U, 80 mM, 90 mM, 2.2 mM, 0.4 mM, 2 mg / mL, respectively.

[0068] More specifically, the dosage of the reagent A is 1.8%-2.5%wt of the mass of the lysed sample.

[0069] Preferably, the mass ratio of chitosan, lysine and 2,5-dimercapto-1,34-thiadiazole can be 1-2:3-4:0.12-0.17.

[0070] In some embodiments, the mass ratio of chitosan, lysine and 2,5-dimercapto-1,34-thiadiazole can be 1:4:0.12.

[0071] In some other embodiments, the mass ratio of chitosan, lysine and 2,5-dimercapto-1,34-thiadiazole can be 2:3:0.17.

[0072] More specifically, the dosage of the reagent B is 1.3%-1.7%wt of the mass of the lysed sample.

[0073] More specifically, the primer sequences are shown in SEQ ID NO: 1-SEQ ID NO: 72.

[0074] More specifically, the reaction procedure of the qPCR in step (2) is: pre-denaturation at 95°C for 2 min, 1 cycle; 95°C for 10 s, 60°C for 25 s, 10 cycles; 95°C for 10 s, 60°C for 20 s, 28 cycles.

[0075] More specifically, after the reaction procedure starts, the heating rate is 6-9°C / s, and the cooling rate is 6-8°C / s.

[0076] Specifically, in step S2, the gene data is formatted to convert the pathway signal of the qPCR instrument into a corresponding weight value.

[0077] To be more specific, the qPCR instrument has four channel signals, namely ROX, CY5, FAM, and VIC. Each site corresponds to two of the channels. When the instrument analysis is completed, each channel will output a Boolean value to represent negative and positive. Negative means not detected, and positive means detected.

[0078] According to some embodiments of the present invention: Taking the site rs1800566 as an example, the genotype of the site corresponds to the test results of the channels FAM and VIC. When FAM is positive and VIC is positive, the genotype is GA. When FAM is positive and VIC is negative, the genotype is GG. When FAM is negative and VIC is positive, the genotype is AA.

[0079] After obtaining the site genotype, referring to Table 5, when the number of dangerous bases is 0, the site weight value is the wild-type weight value, when the number of dangerous bases is 1, the weight value is the heterozygous mutation weight value, when the number of dangerous bases is 2, the weight value is the homozygous mutation weight value. The specific weight values ​​are shown in the following table.

[0080]

[0081] Specifically, the machine learning model described in step S3 is selected from the EDM model.

[0082] Specifically, step S3 of constructing a skin characteristic judgment model comprises: using the EDM model to calculate the GRS value of the project, and based on the normal distribution of the GRS value, using the mean and standard deviation to perform grade division.

[0083] Furthermore, the calculation formula of the GRS value is: ;

[0084] Where: OR i SNP i The weight of X i is the number of dangerous bases; the project level is calculated using the GRS, mean and sd values ​​according to the formula:

[0085] Level 1: ;

[0086] Level 2: ;

[0087] Level 3: ;

[0088] The calculation formula of threshold coefficient xs is: ;

[0089] Where N is the number of sites included in the project.

[0090] In yet another aspect, the present invention provides a skin characteristic analysis model constructed by the above construction method.

[0091] In another aspect, the present invention provides a method for analyzing skin characteristics using the above analysis model, comprising the following steps:

[0092] (1) Lyse the sample using a lysis buffer;

[0093] (2) performing qPCR detection on the sample after lysis in step (1) to obtain data;

[0094] (3) Formatting the data obtained in step (2);

[0095] (4) Input the formatted data in step (3) into the skin characteristic judgment model to obtain the skin characteristic conclusion.

[0096] On the other hand, the present invention provides application of the above-mentioned skin characteristic analysis model in skin characteristic detection.

[0097] The beneficial effects of the present invention are:

[0098] The skin characteristic analysis model provided by the present invention can accurately determine skin characteristics, and the method of using the skin characteristic model for detection is simple, and skin characteristics can be determined efficiently, saving the time of gene detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0099] Figure 1 The fitting results of increase and decrease sites for anti-photoaging.

[0100] Figure 2 This is the fitting result of the skin barrier increase and decrease sites.

[0101] Figure 3 This is the fitting result of oil control increase and decrease sites.

[0102] Figure 4 This is the fitting result of the moisturizing increase and decrease sites.

[0103] Figure 5 The results are for fitting the antioxidant increase and decrease sites.

[0104] Figure 6 This is the fitting result of the anti-glycation increase and decrease sites.

[0105] Figure 7 This is the fitting result of the collagen increase and decrease sites. DETAILED DESCRIPTION

[0106] In order to make the technical means, creative features, purpose and efficacy of the present invention easy to understand, the present invention is further explained in conjunction with specific embodiments below, but the following embodiments are only preferred embodiments of the present invention, not all. Based on the embodiments in the implementation mode, other embodiments obtained by those skilled in the art without creative work all belong to the protection scope of the present invention. In the following embodiments, unless otherwise specified, the operating methods used are conventional operating methods, the equipment used are conventional equipment, and the equipment materials used in each embodiment are the same.

[0107] Example 1 Construction of skin characteristics judgment model

[0108] Constructing an analytical model for determining skin characteristics:

[0109] S1. Obtain the correspondence between the skin features and skin characteristics of the sample and establish a database;

[0110] S2, obtaining the genetic data of the sample and formatting the genetic data;

[0111] S3. According to the formatted sample gene data and the database, a machine learning model is selected to construct a skin characteristic determination model for determining specific skin characteristics based on the sample gene data.

[0112] The corresponding relationship between the “skin features” and “skin characteristics” and the relevant detection target genes (i.e., genes related to specific skin characteristics) in step S1 is:

[0113] (1) Anti-photoaging ability:

[0114] The skin's UV resistance is divided into three levels from high to low. Taking rs1800414 as the standard, CC is high-end, CT is mid-range, and TT is low-end; taking rs62543565 as the standard, TT is high-end, TC is mid-range, and CC is low-end; taking rs10733310 as the standard, TT is high-end, TG is mid-range, and GG is low-end; taking rs4911414 as the standard, GG is high-end, GT is mid-range, and TT is low-end; taking rs1015362 as the standard, TT is high-end, TC is mid-range, and CC is low-end.

[0115] Skin characteristics: The above capabilities are low-end, corresponding to specific skin characteristics: anti-photoaging level 1, and so on.

[0116] (2) Anti-drying ability:

[0117] The skin's ability to regulate and delay water loss through the skin is divided into three levels from high to low. Taking rs61816761 as the standard, GG is high-end, GA is mid-range, and AA is low-end. Taking rs11103631 as an example, AA is high-end, AG is mid-range, and GG is low-end.

[0118] Skin characteristics: The above capabilities are low-end, corresponding to specific skin characteristics: anti-dryness level 1, and so on.

[0119] (3) Anti-wrinkle ability:

[0120] The ability of the skin to regulate collagen synthesis in the skin was divided into three levels from high to low. Using rs1800796 as the standard, CC was high, CG was medium, and GG was low. Using rs1799750 as the standard, DD was high, DI was medium, and II was low.

[0121] Skin characteristics: The above capabilities are low-end, corresponding to specific skin characteristics: anti-wrinkle level 1, and so on.

[0122] (4) Anti-glycation ability related:

[0123] The ability of the receptor for advanced glycation end products protein in the skin reduction period is divided into three levels from high to low. Taking rs1800625 as the standard, TT is high-end, TC is medium-end, and CC is low-end; taking rs2070600 as the standard, CC is high-end, CT is medium-end, and TT is low-end; taking rs1049346 as the standard, GG is high-end, GA is medium-end, and AA is low-end.

[0124] Skin characteristics: The above capabilities are low-end, corresponding to specific skin characteristics: anti-glycation level 1, and so on.

[0125] (5) Antioxidant capacity related:

[0126] The skin's free radical scavenging ability was divided into three levels from high to low. Taking rs1800566 as the standard, GG was high-grade, GA was mid-grade, and AA was low-grade; taking rs713041 as the standard, CC was high-grade, TC was mid-grade, and TT was low-grade.

[0127] Skin characteristics: The above capabilities are low-end, corresponding to specific skin characteristics: antioxidant level 1, and so on.

[0128] (6) Related to the ability to resist skin barrier damage:

[0129] The skin's ability to regulate / repair the skin barrier is divided into three levels from high to low. Taking rs612529 as the standard, TT is high-grade, TC is mid-grade, and CC is low-grade. Taking rs2303067 as the standard, GG is high-grade, GA is mid-grade, and AA is low-grade.

[0130] Skin characteristics: The above capabilities are low-end, corresponding to specific skin characteristics: Level 1 resistance to skin barrier damage, and so on.

[0131] (7) Oil control ability related:

[0132] The skin's ability to inhibit oil secretion is divided into three levels from high to low. Taking rs174547 as the standard, TT is high-end, TC is medium-end, and CC is low-end; taking rs743572 as the standard, AA is high-end, AG is medium-end, and GG is low-end.

[0133] Skin characteristics: The above capabilities are low-end, corresponding to specific skin characteristics: oil control level 1, and so on.

[0134] Step S2 includes:

[0135] 1. Preparation of Reagents

[0136] The formula of lysis buffer a is shown in Table 1.

[0137] Table 1 Formula of lysate a

[0138] Element concentration Sodium dodecyl sulfate (SDS) 1.3%wt Tris(hydroxymethyl)aminomethane 0.1M Silver diethyldithiocarbamate (DTC) 120mM

[0139] The formula of reagent A is shown in Table 2.

[0140] Table 2 Reagent A formula

[0141] Element concentration Taq DNA Polymerase 1.7U Tris-HCl 70 mM KCl 70mM MgSO 4 1.5mM dNTP 0.05mM BSA 0.1mg / mL

[0142] The formula of reagent B is shown in Table 3.

[0143] Table 3 Reagent B formula

[0144] Element Mass fraction Chitosan 1 Lysine 4 2,5-dimercapto-1,3,4-thiadiazole 0.12

[0145] The above reagents can be prepared into lyophilized preparations for storage and transportation. The preparation method of the lyophilized preparations is: freeze-drying the components of the reagents separately and then mixing them according to a proportion, or mixing the components of the reagents according to a proportion and dispersing them in a solvent and then freeze-drying them. When used, dilute them with ABS buffer to a corresponding concentration.

[0146] 2. Detection Method

[0147] The information of the detected genes is shown in Table 4.

[0148] Table 4

[0149]

[0150] The specific steps of the test are as follows:

[0151] (1) Sample lysis: Add the sample (oral swab) into lysis solution a and mix thoroughly to obtain a lysis sample; the lysis solution a is heated to 40-45° C. in a water bath before use; the amount of lysis solution a added is sufficient to cover the sample, and in this embodiment 1, its volume is preferably 3 times the volume of the sample.

[0152] (2) Take reagent A, lysed sample, primers, and reagent B in sequence and add them to the qPCR reaction tube, vortex mix for 5-10 seconds, let stand for 5-10 seconds, and then vortex mix for 5-10 seconds to fully mix; the amount of reagent A is 1.8%wt of the mass of the lysed sample; the amount of reagent B is 1.3%wt of the mass of the lysed sample; the primer sequences are shown in Table 6.

[0153] (3) Amplification: The obtained qPCR reaction tube is placed in a qPCR device and reacted according to the PCR program; the PCR program is: pre-denaturation at 95°C for 2 min, 1 cycle; 95°C for 10 s, 60°C for 25 s, 10 cycles; 95°C for 10 s, 60°C for 20 s, 28 cycles.

[0154] In the PCR process, the heating rate is 6-9°C / s after the program starts, and the cooling rate is 6-8°C / s. The qPCR collects fluorescence signals at 60°C, and the collected fluorescence amplification signals include FAM, VIC and CY5, and the internal reference is ROX.

[0155] (4) Collect test result data on the qPCR instrument and format the test data as needed.

[0156] The gene data is formatted to convert the pathway signals of the qPCR instrument into corresponding weight values.

[0157] The qPCR instrument has four channel signals, namely ROX, CY5, FAM, and VIC. Each site corresponds to two of the channels. After the instrument analysis is completed, each channel will output a Boolean value to indicate negative or positive. Negative means not detected, and positive means detected.

[0158] Taking the site rs1800566 as an example, the genotype of this site corresponds to the test results of channels FAM and VIC. When FAM is positive and VIC is positive, the genotype is GA. When FAM is positive and VIC is negative, the genotype is GG. When FAM is negative and VIC is positive, the genotype is AA.

[0159] After obtaining the site genotype, referring to Table 5, when the number of dangerous bases is 0, the site weight value is the wild-type weight value, when the number of dangerous bases is 1, the weight value is the heterozygous mutation weight value, and when the number of dangerous bases is 2, the weight value is the homozygous mutation weight value.

[0160] Table 5

[0161]

[0162] Table 6

[0163]

[0164] Table 6 (continued)

[0165]

[0166] Table 6 (continued)

[0167]

[0168] Comparative Example 1: The only difference from Example 1 is that the detection sites are added on the basis of Example 1. The number of detection sites in Comparative Example 1 is 25, and the sequence of the added detection sites is shown in Table 7 below. The rest is the same as Example 1.

[0169] Table 7 Detection sites and primer sequences added in Comparative Example 1

[0170]

[0171] Table 7 (continued)

[0172]

[0173] Comparative Example 2: The difference from Example 1 is that the detection sites are reduced on the basis of Example 1. The detection sites in Comparative Example 2 are 11, and the detection sites are rs62543565, rs10733310, rs1015362, rs4911414, rs612529, rs174547, rs11103631, rs713041, rs1800625, rs2070600 and rs1800796, the sequence is the same as Example 1, and the rest are the same as Example 1.

[0174] Table 8 Site weights increased in comparative example 1

[0175]

[0176] Table 9 Correspondence between the gear position and genotype of the increased sites in Comparative Example 1

[0177]

[0178] Step S3 is: according to the formatted sample gene data and the database, a machine learning model is selected to construct a skin characteristic judgment model based on the sample gene data.

[0179] 1. Screen the samples, extract the sample DNA for testing to obtain the corresponding genotype, and label the sample skin characteristics.

[0180] 2. Use the logistic regression model to perform association analysis on the relationship between sample genotype and skin characteristics, and obtain the OR value corresponding to the genotype.

[0181] The skin characteristics judgment model based on sample gene data is constructed as follows:

[0182] The GRS value of the project is calculated using the EDM model. Based on the normal distribution of the GRS value, the mean and standard deviation are used for grading. The calculation formula for the GRS value is: (where OR i SNP i The weight of (see Table 5); X i is the number of dangerous bases; where m refers to the number of sites included in the project); use GRS, mean and sd values ​​to calculate the project level according to the formula, where mean is the arithmetic mean of the GRS values ​​of all samples involved in the calculation, and sd is the standard deviation of the GRS values ​​of all samples involved in the calculation:

[0183] Level 1: ;

[0184] Level 2: ;

[0185] Level 3: ;

[0186] The calculation formula of the threshold coefficient xs is: (N is the number of sites included in the project).

[0187] The above gene detection method and the Moji MAX skin detector were used to test the oral swab samples of 75 volunteers. The 75 people were divided into three groups, 15 in group one with natural skin condition, 30 in group two with bad living habits, and 30 in group three with careful skin care.

[0188] 3. Analysis results:

[0189] The results of the genetic test are as follows (level 1-3, the larger the number, the stronger the ability):

[0190] Table 10

[0191]

[0192] Table 10 (continued)

[0193]

[0194] Table 10 (continued)

[0195]

[0196] The skin detector test results are as follows (0-100 points, the higher the score, the stronger the level):

[0197] Table 11

[0198]

[0199] Table 11 (continued)

[0200]

[0201] Table 11 (continued)

[0202]

[0203] The condition level is calculated based on the current status test (skin tester), with less than 50 points as level 1, 50 to 80 points as level 2, and more than 80 points as level 3.

[0204] Table 12

[0205]

[0206] Table 12 (continued)

[0207]

[0208] Table 12 (continued)

[0209]

[0210] The seven-dimensional genetic and current status test results of each volunteer tested were compared. The comparison results are as follows. The data in Table 13 are the consistency rates of the seven-dimensional results of the current status test and genetic test of 75 samples.

[0211] Table 13

[0212]

[0213] The experimental data of Comparative Example 1 and Comparative Example 2 are seven normal distribution graphs, such as Figure 1-Figure 7 As shown. The higher the curve, the denser the probability of this interval. The steeper the normal curve, the smaller the standard deviation, which means that most variable values ​​are closer to the mean, so most values ​​are closely clustered around the mean. Judging criteria: The peak is close to the center, and the curve is high and steep, which is the best.

[0214] Figure 1 (Anti-photoaging) The results show that compared with normal sites, the distribution is basically unchanged when the sites are increased, and the peak value of the curve becomes lower when the sites are reduced. Conclusion: Normal sites are the best. Figure 2 (Skin barrier) The results show that compared with normal sites, the curve deviates from the center when the sites are increased; and the peak of the curve becomes lower when the sites are reduced. Conclusion: normal sites are optimal. Figure 3 (Oil control) The results show that compared with normal sites, increasing the number of sites leads to uneven distribution and the curve deviates from the center; reducing the number of sites leads to a lower peak value of the curve. Conclusion: normal sites are optimal. Figure 4 (Moisturizing) The results show that compared with the normal site, the curve deviates from the center when the site is increased, and the peak of the curve becomes lower when the site is reduced. Conclusion: The normal site is the best. Figure 5 (Antioxidant) The results show that compared with the normal site, the curve deviates from the center when the site is increased, and the peak of the curve becomes lower when the site is reduced. Conclusion: The normal site is the best. Figure 6 (Anti-glycation) The results showed that compared with normal sites: when the sites were increased, the curve deviated from the center; when the sites were reduced, the curve deviated from the center. Conclusion: normal sites are optimal. Figure 7 (Collagen) The results show that compared with normal sites: when the sites are increased, the curve deviates from the center; when the sites are reduced, the peak of the curve becomes lower. Conclusion: normal sites are optimal.

[0215] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for constructing a skin characteristic analysis model, characterized in that: The following steps are involved: S1. Obtaining the genetic data of the sample and formatting the genetic data; S2. Select a machine learning model based on the formatted sample genetic data to build a skin characteristic determination model for determining specific skin characteristics based on the sample genetic data; The gene data in step S1 include: gene data of anti-photoaging ability, anti-drying ability, anti-wrinkle ability, anti-glycation ability, antioxidant ability, anti-skin barrier damage ability, and oil control ability; The detection sites for anti-photoaging ability consisted of the following sites: rs1800414, rs62543565, rs10733310, rs4911414, and rs1015362; The detection loci for desiccation resistance consisted of the following loci: rs61816761, rs11103631; The detection loci for anti-wrinkle ability consisted of the following loci: rs1800796, rs1799750; The detection sites for anti-glycation ability consist of the following sites: rs1800625, rs2070600, and rs1049346; The detection loci of antioxidant capacity consisted of the following loci: rs1800566, rs713041; The detection sites for the ability to resist skin barrier damage consist of the following sites: rs612529, rs2303067; The detection loci for oil control ability consisted of the following loci: rs174547, rs743572; The method for formatting the gene data in step S1 is to convert the pathway signal of the qPCR instrument into a corresponding weight value; When the number of dangerous bases is 0, the site weight value is the wild-type weight value; when the number of dangerous bases is 1, the weight value is the heterozygous mutation weight value; when the number of dangerous bases is 2, the weight value is the homozygous mutation weight value; Step S2 constructs the skin characteristic judgment model by using the EDM model to calculate the GRS value of the project, and based on the normal distribution of the GRS value, uses the mean and standard deviation to perform grade classification; The calculation formula of the GRS value is: ; Where: OR i SNP i The weight of X i is the number of dangerous bases, m is the number of sites included in the project; the project level is calculated using the GRS, mean and sd values ​​according to the formula: Level 1: ; Level 2: ; Level 3: ; The calculation formula of threshold coefficient xs is: ; In the formula, N is the number of sites included in the project; When the test item is light aging resistance, OR i The weights are rs1800414, rs62543565, rs10733310, rs4911414, and rs1015362, respectively, N = 5; When the test item is anti-drying ability, OR i The weights are rs61816761 and rs11103631, respectively, N = 2; When the test item is anti-wrinkle ability, OR i The weights are rs1800796 and rs1799750, respectively, N = 2; When the test item is anti-glycation ability, OR i The weights are rs1800625, rs2070600, and rs1049346, respectively, N = 3; When the test item is antioxidant capacity, OR i The weights are rs1800566 and rs713041, respectively, N = 2; When the test item is the ability to resist skin barrier damage, OR i The weights are rs612529 and rs2303067, respectively, N = 2; When the test item is oil control ability, OR i are the weights of rs174547 and rs743572, respectively, and N=2.

2. The construction method according to claim 1, characterized in that: The method for obtaining gene data of a sample described in step S1 comprises the following steps: (1) Lyse the sample using a lysis buffer; (2) Perform qPCR detection on the sample after lysis in step (1) to obtain data.

3. A skin characteristic analysis model constructed according to the construction method according to any one of claims 1-2.

4. A method for analyzing skin characteristics using the skin characteristics analysis model according to claim 3, characterized in that: The following steps are involved: (1) Lyse the sample using a lysis buffer; (2) performing qPCR detection on the sample after lysis in step (1) to obtain data; (3) Formatting the data obtained in step (2); (4) Input the formatted data in step (3) into the skin characteristic judgment model to obtain the skin characteristic conclusion.

Images