Method and device for generating drug preparation suggestions based on classification attributes

Abstract

The present application relates to a method and device for generating drug preparation suggestions based on classification attributes. The method includes: obtaining multiple samples; determining the detection frequency corresponding to each genotype of the target gene of the multiple samples, and the target gene has a mapping relationship with the target drug; determining the reference frequency corresponding to each genotype of the target gene; judging whether the preset strategy is met based on the detection frequency and the reference frequency; when the preset strategy is met, classifying the multiple samples through a data classification algorithm to generate multiple classification attributes; generating drug preparation suggestions based on the target gene and the multiple classification attributes, and the drug preparation suggestions contain the proportion information of the target drug. The method and device for generating drug preparation suggestions based on classification attributes involved in the present application can effectively improve the safety of drug use, the rationality of drug supply, and provide strong technical support for personalized medicine.

Description

Technical Field

[0001] The present application relates to the field of computer information processing, and in particular, to a method and device for generating drug preparation suggestions based on classification attributes. Background Art

[0002] Genetic testing is a technology that identifies genetic information by analyzing an individual's DNA sequence. This information can be used to understand an individual's health status, disease risk, drug response, etc. Through genetic testing reports, the test results are linked to existing medical literature or databases to help doctors or individuals understand the relationship between genes and diseases and drug responses.

[0003] Drug allocation refers to the rational arrangement of drug procurement and storage to ensure adequate and reasonable drug supply. Traditional drug allocation is mainly based on clinical experience and medication history, but with the development of personalized medicine, drug allocation has begun to combine genetic testing technology to achieve more accurate medication management.

[0004] The above information disclosed in the Background section is only for enhancement of understanding of the background of the present application and therefore it may contain information that does not constitute the prior art that is already known to a person of ordinary skill in the art. Summary of the invention

[0005] In view of this, the present application provides a method and device for generating drug preparation suggestions based on classification attributes, which can effectively improve the safety of drug use and the rationality of drug supply, and provide strong technical support for personalized medicine.

[0006] Other features and advantages of the present application will become apparent from the following detailed description, or may be learned in part by the practice of the present application.

[0007] According to one aspect of the present application, a method for generating drug preparation suggestions based on classification attributes is proposed, the method comprising: obtaining multiple samples; determining the detection frequency corresponding to each genotype of a target gene of the multiple samples, the target gene having a mapping relationship with a target drug; determining the reference frequency corresponding to each genotype of the target gene; judging whether a preset strategy is met based on the detection frequency and the reference frequency; when the preset strategy is met, classifying the multiple samples through a data classification algorithm to generate multiple classification attributes; generating a drug preparation suggestion based on the target gene and the multiple classification attributes, the drug preparation suggestion including the proportion information of the target drug.

[0008] In an exemplary embodiment of the present application, the plurality of samples include a plurality of batches of samples.

[0009] Judging whether a preset strategy is met based on the detection frequency and the reference frequency includes: determining the batch detection frequency corresponding to each genotype of the target gene of each batch of samples in the multiple batches of samples; judging whether the preset strategy is met based on the batch detection frequency and the reference frequency.

[0010] In an exemplary embodiment of the present application, it also includes: if the batch detection frequency based on a batch of samples and the reference frequency do not meet the preset strategy, re-inspecting the batch of samples; and judging whether the preset strategy is met based on the re-inspected batch detection frequency and the reference frequency.

[0011] In an exemplary embodiment of the present application, the multiple samples are classified by a data classification algorithm to generate multiple classification attributes, including: classifying the sample information by a classification model to generate multiple classification attributes; and / or classifying the sample information by a clustering algorithm to generate multiple classification attributes.

[0012] In an exemplary embodiment of the present application, generating a drug preparation recommendation based on the target gene and the multiple classification attributes includes: determining the target classification attribute; extracting samples belonging to the target classification attribute from the multiple samples; determining the detection frequency corresponding to each genotype of the target gene of the sample of the target classification attribute and generating an attribute drug preparation recommendation for the target classification attribute based on the detection frequency; generating the drug preparation recommendation through the attribute drug preparation recommendations corresponding to the multiple classification attributes.

[0013] In an exemplary embodiment of the present application, the medication preparation suggestion includes a drug ratio, the classification attribute includes a regional attribute, and the medication preparation suggestion is generated through attribute medication preparation suggestions corresponding to the multiple classification attributes, including: generating a drug ratio for a target regional attribute, wherein the drug ratio includes the proportion of the target drug.

[0014] In an exemplary embodiment of the present application, generating a drug ratio for the target regional attribute includes: determining an alternative drug for the target drug for the target regional attribute and a proportion of the alternative drug.

[0015] In an exemplary embodiment of the present application, determining an alternative drug for the target drug with the target regional attributes includes: determining a plurality of alternative drugs, wherein the alternative drugs are other drugs under the same drug classification item as the target drug; determining an alternative gene based on the alternative drug, wherein the alternative gene has a mapping relationship with the alternative drug; determining a detection frequency corresponding to each genotype of the alternative gene of the sample of the target region; and screening out an alternative drug for the target drug with the target regional attributes from the plurality of alternative drugs based on the detection frequency corresponding to each genotype of the alternative gene of the sample with the target regional attributes.

[0016] In an exemplary embodiment of the present application, the target gene includes a first target gene and a second target gene, wherein the reference frequency of the first target gene is known, and the reference frequency of the second target gene is unknown, and the detection frequencies corresponding to the respective genotypes of the target genes of the multiple samples are determined, and the target genes have a mapping relationship with the target drugs, including: determining the detection frequencies corresponding to the respective genotypes of the first target gene and the second target gene of the multiple samples, the first target gene has a mapping relationship with the first target drug, and the second target gene has a mapping relationship with the second target drug, and determining the reference frequencies corresponding to the respective genotypes of the target genes includes: determining the reference frequencies corresponding to the respective genotypes of the first target gene; based on the detection frequencies, the reference frequencies The frequency determines whether the preset strategy is met, including: judging whether the preset strategy is met based on the detection frequency of the first target gene and the reference frequency of the first target gene; when the preset strategy is met, classifying the multiple samples through a data classification algorithm to generate multiple classification attributes, including: when the detection frequency of the first target gene and the reference frequency of the first target gene meet the preset strategy, classifying the multiple samples through a data classification algorithm to generate multiple classification attributes, generating a drug preparation suggestion based on the target gene and the multiple classification attributes, the drug preparation suggestion includes the proportion information of the target drug, including: generating a drug preparation suggestion based on the second target gene and the multiple classification attributes, the drug preparation suggestion includes the proportion information of the second target drug.

[0017] In an exemplary embodiment of the present application, the drug ratio is calculated according to the following formula: g ( f i ) = i 1×( α f i ) + i 2×( βe γfi )+ i 3×[ δln(1+ f i )]+ i 4×( -af i 2 + bf i + c ),in, g ( f i ) is the proportion of drugs, f i is the detection frequency, i 1. i 2. i 3. i 4 is the weight coefficient, α , β , c , d , a , b , c as a parameter.

[0018] In an exemplary embodiment of the present application, it also includes: obtaining a classification attribute of a user; obtaining a target genotype of the user; and preparing medicines for the user based on the classification attribute and the target genotype in combination with the medicine preparation suggestion.

[0019] In an exemplary embodiment of the present application, it also includes: determining a regional classification attribute; determining a drug procurement plan based on the regional classification attribute combined with the drug preparation recommendation, the drug procurement plan including target drugs and alternative drugs for each genotype.

[0020] According to one aspect of the present application, a device for generating drug preparation suggestions based on classification attributes is proposed, and the device includes: a sample module, used to obtain multiple samples; a detection module, used to determine the detection frequency corresponding to each genotype of a target gene of the multiple samples, and the target gene has a mapping relationship with a target drug; a frequency module, used to determine the reference frequency corresponding to each genotype of the target gene; a strategy module, used to determine whether a preset strategy is met based on the detection frequency and the reference frequency; a classification module, used to classify the multiple samples through a data classification algorithm to generate multiple classification attributes when the preset strategy is met; and a suggestion module, used to generate drug preparation suggestions based on the target gene and the multiple classification attributes, and the drug preparation suggestions include the proportion information of the target drug.

[0021] According to one aspect of the present application, an electronic device is proposed, which includes: one or more processors; a storage device for storing one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors implement the method as described above.

[0022] According to one aspect of the present application, a computer-readable medium is provided, on which a computer program is stored. When the program is executed by a processor, the method described above is implemented.

[0023] According to the method and device for generating drug preparation suggestions based on classification attributes of the present application, multiple samples are obtained; the detection frequency corresponding to each genotype of the target gene of the multiple samples is determined, and the target gene has a mapping relationship with the target drug; the reference frequency corresponding to each genotype of the target gene is determined; based on the detection frequency and the reference frequency, it is judged whether a preset strategy is met; when the preset strategy is met, the multiple samples are classified by a data classification algorithm to generate multiple classification attributes; based on the target gene and the multiple classification attributes, a drug preparation suggestion is generated, and the drug preparation suggestion includes the proportion information of the target drug, which can effectively improve the safety of drug use and the rationality of drug supply, and provide strong technical support for personalized medicine.

[0024] It should be understood that the foregoing general description and the following detailed description are exemplary only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The above and other objects, features and advantages of the present application will become more apparent by describing in detail the exemplary embodiments thereof with reference to the accompanying drawings. The accompanying drawings described below are only some embodiments of the present application, and it is clear to a person skilled in the art that other accompanying drawings can be obtained from these accompanying drawings without creative effort.

[0026] Figure 1 The present invention is a flowchart of a method for generating a drug preparation suggestion based on classification attributes according to an exemplary embodiment.

[0027] Figure 2 yes Figure 1 Flow chart of the method for step S108 in FIG.

[0028] Figure 3 yes Figure 1 Flow chart of the method for step S112 in FIG.

[0029] Figure 4 It is a flow chart of “determining alternative drugs for the target drug and the proportion of the alternative drugs according to the target regional attributes”.

[0030] Figure 5 The present invention is a block diagram of a device for generating medicine preparation suggestions based on classification attributes according to an exemplary embodiment.

[0031] Figure 6 It is a block diagram of an electronic device according to an exemplary embodiment.

[0032] Figure 7 It is a block diagram of a computer-readable medium according to an exemplary embodiment. DETAILED DESCRIPTION

[0033] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this application will be comprehensive and complete and will fully convey the concepts of the example embodiments to those skilled in the art. The same reference numerals in the figures represent the same or similar parts, and thus their repeated description will be omitted.

[0034] In addition, described feature, structure or characteristic can be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present application. However, those skilled in the art will appreciate that the technical scheme of the present application can be put into practice without one or more of the specific details, or other methods, components, devices, steps, etc. can be adopted. In other cases, known methods, devices, realizations or operations are not shown or described in detail to avoid blurring the various aspects of the application.

[0035] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0036] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.

[0037] It should be understood that although the terms first, second, third, etc. may be used herein to describe various components, these components should not be limited by these terms. These terms are used to distinguish one component from another component. Therefore, the first component discussed below can be referred to as the second component without departing from the teachings of the concepts of the present application. As used herein, the term "and / or" includes any one of the associated listed items and all combinations of one or more.

[0038] Those skilled in the art will appreciate that the drawings are merely schematic diagrams of example embodiments, and the modules or processes in the drawings are not necessarily necessary for implementing the present application, and therefore cannot be used to limit the scope of protection of the present application.

[0039] Figure 1 The flowchart of a method for generating a medicine preparation suggestion based on a classification attribute according to an exemplary embodiment is shown. The method for generating a medicine preparation suggestion based on a classification attribute at least includes steps S102 to S112.

[0040] like Figure 1 As shown, in S102, multiple samples are obtained. Multiple test sample information can be collected from different testing institutions. In this application, samples refer to biological samples, including personal DNA information, and drug reactions or disease risks can be analyzed by detecting genotypes in these samples. Testing institutions can perform genetic testing on the collected samples.

[0041] In S104, the detection frequencies corresponding to the respective genotypes of the target genes of the plurality of samples are determined, and the target genes have a mapping relationship with the target drugs.

[0042] More specifically, specific genes associated with the target drug (called target genes) can be found and the frequency of occurrence of genotypes in multiple samples can be calculated.

[0043] In a specific embodiment, the target gene may be a specific gene associated with drug response, which affects the metabolism and response of an individual to a specific drug.

[0044] Taking amoxicillin as an example, amoxicillin is mainly related to the gene HLA-DQB1. When the target drug is amoxicillin, the target gene is the gene HLA-DQB1.

[0045] Abacavir is mainly associated with gene HLA-B. When the target drug is abacavir, the target gene is gene HLA-B.

[0046] Acetaminophen is mainly related to the genes G6PD and UGT1A1. When the target drug is acetaminophen, the target gene is the gene G6PD and UGT1A1.

[0047] Genotype is a specific variation of a gene, reflecting the genetic differences of an individual. For example, the genotype of the RS9274407 site of the HLA-DQB1 gene is AA, AT, TT. The genotype of the HLA-B*57:01:01 site of the HLA-B gene is negative or positive.

[0048] In this application, detection frequency may refer to the ratio of the number of occurrences of a particular genotype in a sample to the total number of samples, which is used to measure the prevalence of a genotype in a particular sample.

[0049] In S106, the reference frequencies corresponding to the various genotypes of the target gene are determined. Existing gene databases or literature may be used to obtain the standard occurrence frequencies of different genotypes of the target gene in the general population for comparison with the detection frequencies.

[0050] In the present application, the reference frequency refers to the average frequency of occurrence of each genotype in the general population determined based on existing medical literature or gene databases, which is used as a benchmark.

[0051] In S108, it is determined whether the preset strategy is met based on the detection frequency and the reference frequency. For example, the detection frequency can be compared with the reference frequency. If the error between the detection frequency and the reference frequency is within a preset range, it is considered to meet the preset strategy. At this point, subsequent analysis can be continued.

[0052] For example, a standard or threshold value can be set according to medical needs or statistical models to assess whether the genotype frequency is within a reasonable range. By calculating the difference between the test frequency and the reference frequency, it is determined whether the sample data meets the specific conditions.

[0053] In S110, when the preset strategy is met, the multiple samples are classified by a data classification algorithm to generate multiple classification attributes. For example, the sample information can be classified by a classification model to generate multiple classification attributes; or for example, the sample information can be classified by a clustering algorithm to generate multiple classification attributes.

[0054] In one embodiment, a data classification algorithm may be used to group samples according to different attributes (such as region, occupation, age, etc.) to generate multiple classification attributes. After classification, the samples can be analyzed and studied more effectively.

[0055] More specifically, the data classification algorithm in the present application may include: clustering algorithm, decision tree, support vector machine, etc., the purpose of which is to classify samples into different categories according to features.

[0056] Classification attributes may include: for example, regional attributes, ethnic attributes, occupational attributes, etc., to help in-depth analysis of the sample group.

[0057] In S112, a drug preparation suggestion is generated based on the target gene and the multiple classification attributes, and the drug preparation suggestion includes the proportion information of the target drug. For example, the target classification attribute can be determined; samples belonging to the target classification attribute are extracted from the multiple samples; the detection frequency corresponding to each genotype of the target gene of the sample of the target classification attribute is determined and the attribute drug preparation suggestion for the target classification attribute is generated based on the detection frequency; the drug preparation suggestion is generated through the attribute drug preparation suggestion corresponding to the multiple classification attributes. Target samples are extracted from multiple classification attributes, the genotype detection frequency of these samples is calculated, and corresponding drug preparation suggestions are generated for these specific classification attributes. Finally, a global drug preparation plan is generated based on the drug preparation suggestions of all classification attributes. According to actual needs, specific group characteristics (such as a certain region or a specific occupation) are selected from multiple attributes. A specific drug preparation plan generated based on a certain classification attribute (such as region or occupation) is used to guide the drug allocation of the group.

[0058] In this application, medication recommendations refer to medication recommendations generated based on genotype and sample classification, which aim to provide the best drug selection and ratio. Drug ratio information refers to quantitative recommendations in drug allocation, which refers to the proportion or allocation of drugs.

[0059] As an optional implementation method, drug preparation recommendations are targeted at specific regions. Since genetic characteristics are often related to regions, the genetic characteristics of people in different regions often vary greatly. For example, the genetic characteristics of northerners and southerners are often quite different. And specific genes affect an individual's metabolism and response to specific drugs. This means that the probability of drug-induced injury / drug hypersensitivity syndrome in different regions for the same drug is different, and even quite different. Among them, the manifestations of drug-induced injury are diverse, such as liver failure, hematopoietic system and liver and kidney damage, etc. For example, when drug A is used in the population of Province A, the probability of liver failure is 3%, while when drug A is used in the population of Province B, the probability of liver failure is 22%. For another example, when drug B is used in the population of Province A, the probability of liver failure is 19%, while when drug B is used in the population of Province B, the probability of liver failure is 6%. Among the generated drug preparation recommendations, the drug preparation recommendations for medical institutions in Province A may include drug A accounting for a higher proportion and drug B accounting for a lower proportion; the drug preparation recommendations for medical institutions in Province B may include drug A accounting for a lower proportion and drug B accounting for a higher proportion.

[0060] The benefits of generating drug stocking recommendations for specific regions are: scientifically and rationally purchasing and preparing drug combinations suitable for the population in the region, avoiding waste caused by excessive purchases of certain drugs, and avoiding shortages caused by insufficient purchases of certain drugs. Currently, medical institutions do not have scientific guidance for drug stocking, and generally rely on experience, but this method is often too rough, and it is often possible that too much of a drug that is prone to drug-induced damage to the population is purchased / prepared, thereby wasting these drugs and causing economic losses.

[0061] The method for generating drug preparation suggestions based on classification attributes of the present application is not for specific individuals. For individuals, it is still necessary to detect the genotype and match the appropriate drug according to the genotype. The method for generating drug preparation suggestions based on classification attributes of the present application is for different classification attributes, especially the commonly used classification attribute is region. The division of regions can be based on provinces, terrain features, etc.

[0062] In this application, the generated drug preparation recommendations include more than two drug combinations. It should be noted that even if the probability of drug-induced injury of a certain drug in a specific region is small, only this drug should not be prepared. When drug A is used in the population of Province A, the probability of liver failure is 3%, which means that 97% of the population in the area can use drug A, but 3% of the population should not use drug A. Then for this 3% of the population, other drugs need to be used.

[0063] The method of the present application can generate drug stocking suggestions for a specific area, and can also generate drug stocking suggestions for a specific medical institution in a specific area. That is, the drug stocking suggestions can include not only the proportion of each type of drug, but also the specific amount of each type of drug. The method of the present application also includes generating drug stocking suggestions for a specific medical institution in a specific area based on a time period and / or a budget of a specific medical institution.

[0064] The time period can be one year, one quarter, or one month, etc., depending on the needs of the medical institution.

[0065] The above-mentioned “budget” refers to the budget of a specific medical institution for a specific disease, not the total budget.

[0066] For example, the drug preparation recommendations for medical institutions in Province A may include 75% of Drug A and 25% of Drug B. A hospital is a medical institution in Province A. Based on the average annual number of consultations in the hospital, the incidence of diseases associated with Drug A in Province A, and the fact that Drug A accounts for 75% and Drug B accounts for 25%, the number of Drug A and Drug B that the hospital needs to prepare in one year (in this embodiment, the time period is one year) can be calculated. Since the shelf life of drugs is generally long, which can reach a shelf life of several years, when preparing drugs in units of years, as long as the purchased drugs have a short ex-factory date, there is no need to worry about the shelf life. Because the drug preparation method of the present application can greatly reduce the waste of drugs and consume these drugs within one year.

[0067] For another example, the drug stocking recommendations for medical institutions in Province A may include 75% of Drug A and 25% of Drug B. A hospital is a medical institution in Province A, and the hospital has a specific budget for diseases associated with Drug A. The budget is allocated based on the drug ratio. The advantage of purchasing drugs based on budget is that even if the hospital is short of funds, it can purchase drugs in a scientific ratio without wasting drugs.

[0068] Patient A suffers from symptoms such as cough, sputum, and runny nose. Patient A comes to the hospital. After the doctor examines Patient A, he finds that Patient A suffers from an upper respiratory tract infection. The commonly used medicine for this disease is amoxicillin. Amoxicillin is mainly related to the gene HLA-DQB1, corresponding to the site rs9274407. There are three genotypes for the corresponding site rs9274407 of the gene HLA-DQB1: TT, AT, and AA. For people whose genotype is TT at the RS9274407 site of the HLA-DQB1 gene, the risk of drug-induced liver injury when using amoxicillin or drugs containing amoxicillin is relatively low. For people whose genotype is AT at the RS9274407 site of the HLA-DQB1 gene, the risk of drug-induced liver injury when using amoxicillin or drugs containing amoxicillin may increase, but it can still be used routinely. For people whose genotype at the RS9274407 locus of the HLA-DQB1 gene is AA, the risk of drug-induced liver injury may increase when using amoxicillin or drugs containing amoxicillin, and they should try to avoid using it. Perform genetic testing on patient A (if patient A has undergone genetic testing before, the results of the previous genetic testing can be used directly). If the genotype of patient A at the RS9274407 locus of the HLA-DQB1 gene is TT or AT, use amoxicillin or drugs containing amoxicillin to treat patient A. If the genotype of patient A at the RS9274407 locus of the HLA-DQB1 gene is AA, do not use amoxicillin or drugs containing amoxicillin, but use alternative drugs to amoxicillin to treat patient A.

[0069] Patient B felt unwell and came to the hospital. After the doctor examined Patient B, he found that Patient B was infected with HIV. The commonly used drug for this disease is abacavir. Abacavir is mainly related to the gene HLA-B, corresponding to the site *57:01:01. There are two genotypes at the HLA-B gene *57:01:01 site: negative and positive. For people whose genotype is negative at the HLA-B gene *57:01:01 site, the probability of drug hypersensitivity syndrome (such as severe drug-induced liver injury, etc.) when using drugs containing abacavir is consistent with that of most people, and the probability of occurrence is relatively low compared with high-risk groups. For people whose genotype is positive at the HLA-B gene *57:01:01 site, the risk of drug hypersensitivity syndrome (such as severe drug-induced liver injury, etc.) when using abacavir is higher than that of most people, and it is recommended that clinical use of drugs containing abacavir should be avoided. Patient B is subjected to genetic testing (if patient B has undergone genetic testing before, the results of the previous genetic testing can be used directly). If the genotype of patient B at the HLA-B gene*57:01:01 site is negative, abacavir or a drug containing abacavir is used to treat patient B. If the genotype of patient B at the HLA-B gene*57:01:01 site is positive, abacavir or a drug containing abacavir is not used, but an alternative drug to abacavir is used to treat patient B.

[0070] Patient C was feeling unwell and came to the hospital. After the doctor examined Patient C, he found that Patient C had symptoms of fever and pain. This symptom can be relieved by acetaminophen. Acetaminophen is mainly related to the genes G6PD and UGT1A1. Among them, the locus corresponding to UGT1A1 is rs1042640. People whose genotype of the gene G6PD is normal enzyme activity and whose genotype of the UGT1A1 gene locus rs1042640 is GG, have normal G6PD enzyme activity and a low risk of hemolytic anemia when using the drug, which is consistent with most people. The UGT1A1 gene rs1042640 locus GG type, the risk of liver failure caused by excessive use of acetaminophen is relatively low, but drugs containing acetaminophen can still be used routinely. Long-term use of acetaminophen can lead to damage to the hematopoietic system and liver and kidney function, and the drug must be used strictly according to the drug instructions. People whose G6PD genotype is uncertain enzyme activity and whose UGT1A1 gene locus rs1042640 genotype is GG are G6PD heterozygotes and need further laboratory tests to determine the activity of the G6PD enzyme. If the further test results show that the enzyme activity is normal, it can be used routinely; if the enzyme activity is absent, medication should be used under the guidance of a doctor. For people with the UGT1A1 gene rs1042640 locus GG type, the risk of liver failure caused by excessive use of acetaminophen may be low, but drugs containing acetaminophen can still be used routinely. Long-term use of acetaminophen can cause damage to the hematopoietic system and liver and kidney functions, and medications must be used strictly according to the drug instructions. People whose G6PD genotype is severely deficient in enzyme activity and whose UGT1A1 gene rs1042640 locus genotype is GG have severe deficiency in G6PD enzyme activity and are at a higher risk of developing hemolytic anemia when using drugs containing acetaminophen than most people. For patients with GG type at rs1042640 of the UGT1A1 gene, the risk of liver failure caused by excessive use of acetaminophen may be low, but drugs containing acetaminophen can still be used routinely. Long-term use of acetaminophen can lead to damage to the hematopoietic system, liver and kidney function, and the drug must be used strictly according to the instructions. Genetic testing is performed on patient C (if patient C has undergone genetic testing before, the results of the previous genetic testing can be used directly). If the genotype of patient C's G6PD gene is normal enzyme activity and the genotype of the UGT1A1 gene rs1042640 site is GG, or if the genotype of the G6PD gene is uncertain enzyme activity and the genotype of the UGT1A1 gene rs1042640 site is GG, acetaminophen can be used for patient C. If the genotype of patient C's G6PD gene is severe enzyme deficiency and the genotype of the UGT1A1 gene rs1042640 site is GG, acetaminophen should not be used for patient C, and an alternative drug to acetaminophen is needed.

[0071] As an optional implementation, the drug preparation suggestion is for a specific occupation, and the classification attribute is occupation. For a military hospital located in a military unit, all of its patients are soldiers. The method for generating drug preparation suggestions in this application generates drug ratio suggestions for diseases that soldiers are prone to, and provides them to military hospitals.

[0072] In this application, specific drug ratio recommendations are generated based on the target genes and classification attributes of the samples to ensure that drugs can be reasonably allocated to different groups. Drug ratio information includes drug usage and alternatives.

[0073] According to the method for generating drug preparation suggestions based on classification attributes of the present application, multiple samples are obtained; the detection frequencies corresponding to each genotype of the target gene of the multiple samples are determined, and the target gene has a mapping relationship with the target drug; the reference frequencies corresponding to each genotype of the target gene are determined; based on the detection frequencies and the reference frequencies, it is judged whether a preset strategy is met; when the preset strategy is met, the multiple samples are classified by a data classification algorithm to generate multiple classification attributes; drug preparation suggestions are generated based on the target gene and the multiple classification attributes, and the drug preparation suggestions include the proportion information of the target drug, which can effectively improve the safety of drug use and the rationality of drug supply, and provide strong technical support for personalized medicine.

[0074] The method for generating drug preparation suggestions based on classification attributes provided in this application can be expressed using the following mathematical formula.

[0075] 1. Calculate the detection frequency

[0076] have N For a specific genotype G i , the genotype in n i Genotype G i The detection frequency can be expressed as: f i = n i / N .

[0077] For example, a gene has three genotypes, namely genotype G 1. Genotype G 2. Genotype G 3. N Among the samples, genotype G 1 in n Occurred in 1 sample; genotype G 2 in n Occurred in 2 samples; genotypeG 3 in n Occurred in 3 samples. Genotype G The detection frequency of 1 is f 1 , f 1 = n 1 / N .genotype G The detection frequency of 2 is f 2 , f 2 = n 2 / N .genotype G The detection frequency of 3 is f 3 , f 3 = n 3 / N .in, f 1 + f 2 + f 3=1, n 1 + n 2 + n 3= N .

[0078] For example, a gene has four genotypes, namely genotype G 1. Genotype G 2. Genotype G 3. Genotype G 4. N Among the samples, genotype G 1 in n Occurred in 1 sample; genotype G 2 in n Occurred in 2 samples; genotype G 3 in n Appeared in 3 samples, genotype G 4 in n Occurred in 4 samples. Genotype G The detection frequency of 1 is f 1 , f 1 = n 1 / N .genotype G The detection frequency of 2 is f 2 , f 2 = n 2 / N .genotype G The detection frequency of 3 is f 3 , f 3 = n 3 / N .genotype G The detection frequency of 4 is f 4 , f 4 = n 4 / N .in, f 1 + f 2 + f 3+ f 4=1, n 1 + n 2 + n 3+ n 4= N .

[0079] 2. Comparison of detection frequency and reference frequency

[0080] Assumptions f i (Standard) is the reference frequency obtained from the gene database or literature. G i , the difference between the detection frequency and the reference frequency can be expressed as:

[0081] d i = f i - f i (reference frequency)

[0082] For multiple groups of genotypes, the error function can be defined E , which is used to measure the overall error:

[0083] E = d 1 2 + d 2 2 +…+ d i 2 +…+ d m 2

[0084] =[ f1- f 1(reference frequency)] 2 + [ f 2 - f 2 (reference frequency)] 2 +…+ [ f i - f i (reference frequency)] 2 +…+ [ f m - f m (reference frequency)] 2

[0085] in, m is the number of different genotypes, m ≥2.

[0086] For example:

[0087] A gene has two genotypes ( m = 2), E = d 1 2 + d 2 2 =[ f 1- f 1(reference frequency)] 2 + [ f 2 - f 2 (reference frequency)] 2

[0088] A gene has three genotypes ( m = 3), E = d 1 2 + d 2 2 + d 3 2 =[ f 1- f 1(reference frequency)] 2 +[ f 2 - f 2 (reference frequency)] 2 + [ f 3 - f3 (reference frequency)] 2

[0089] A gene has four genotypes ( m = 4), E = d 1 2 + d 2 2 + d 3 2 + d 4 2 =[ f 1- f 1(reference frequency)] 2 + [ f 2 - f 2 (reference frequency)] 2 + [ f 3 - f 3 (reference frequency)] 2 + [ f 4 - f 4 (reference frequency)] 2

[0090] A gene has five genotypes ( m = 5), E = d 1 2 + d 2 2 + d 3 2 + d 4 2 + d 5 2 = [ f 1- f 1(reference frequency)] 2 + [ f 2 - f 2 (reference frequency)] 2 + [ f 3 - f 3 (reference frequency)]2 + [ f 4 - f 4 (reference frequency)] 2 +[ f 5 - f 5 (reference frequency)] 2

[0091] 3. Set thresholds and judgment conditions

[0092] Setting Thresholds z , and determine whether the error is within an acceptable range:

[0093] E ≤ z

[0094] If this condition is met, the data is considered to comply with the preset policy.

[0095] 4. Data classification and cluster analysis

[0096] Classify multiple samples through classification algorithms and divide the samples into k Classification (k is the total number of classification attributes). The basis of classification can be region, etc. m j Yes Category C j The mean of x are sample data points. t j Yes Category C j The number of samples. j =1,2,……, k .

[0097] Clustering algorithms usually use an objective function to measure the similarity within a category, for example:

[0098] Similarity within a category J express, J = J 1+ J 2+……+ J k

[0099] J The value of 1 is all the C Samples and categories of 1 C Mean of 1 m The sum of the squares of the differences of 1.

[0100] J The value of 2 is all the C2 samples and categories C Mean of 2 m 2The sum of the squares of the differences.

[0101] …

[0102] J k The value of is all the C k Samples and categories C k The mean m k The sum of the squares of the differences.

[0103] For example, dividing the sample into k Class, respectively, category C 1. Category C 2. Category C k .

[0104] Belongs to category C The sample of 1 has t 1, respectively x 1，1 , x 2，1 , x 3，1 ,……, x t1，1 , belongs to the category C 1 of t The mean of a sample is m 1.

[0105] J 1=( x 1，1 - m 1) 2 +( x 2，1 - m 1) 2 +( x 3，1 - m 1) 2 +……+( x t1，1 - m 1) 2

[0106] Belongs to category C The sample of 2 has t 2, respectively x 1，2 , x 2，2 , x 3，2,……, x t2，2 , belongs to the category C The mean of the t2 samples of 2 is m 2.

[0107] J 2=( x 1，2 - m 2) 2 +( x 2，2 - m 2) 2 +( x 3，2 - m 2) 2 +……+( x t2，2 - m 2) 2

[0108] Belongs to category C j The samples are t j , respectively x 1，j , x 2，j , x 3，j ,……, x tj，j , belongs to the category C j of t j The mean of the samples is m j .

[0109] J j =( x 1，j - m j ) 2 +( x 2，j - m j ) 2 +( x 3，j - m j ) 2 + … + ( x tj，j - m j ) 2

[0110] Belongs to categoryC k The samples are t k , respectively x 1，k , x 2，k , x 3，k ,……, x tk，k , belongs to the category C k of t k The mean of the samples is m k .

[0111] but J k =( x 1，k - m k ) 2 +( x 2，k - m k ) 2 +( x 3，k - m k ) 2 +……+( x tk，k - m k ) 2

[0112] J = J 1+ J 2+……+ J j +……+ J k

[0113] =[( x 1，1 - m 1) 2 +( x 2，1 - m 1) 2 +( x 3，1 - m 1) 2 +……+( x t1，1 - m 1) 2 ]+[ (x 1，2 - m 2) 2 +( x 2，2 - m 2) 2 +( x 3，2 - m 2) 2 +……+( x t2，2 - m 2) 2 ]+……+[( x 1，j - m j ) 2 +( x 2，j - m j ) 2 +( x 3，j - m j ) 2 + …+( x tj，j - m j ) 2 ]+……+[( x 1，k - m k ) 2 +( x 2，k - m k ) 2 +( x 3，k - m k ) 2 +……+( x tk，k - m k ) 2 ]

[0114] For example, the samples are divided into three categories (k=3), namely C 1. Category C 2. Category C 3.

[0115] Belongs to category C There are 1000 samples of 1 ( t 1 =1000), respectively x1，1 , x 2，1 , x 3，1 ,……, x 1000，1 , belongs to the category C The mean of 1000 samples of 1 is m 1.

[0116] Then J1=( x 1，1 - m 1) 2 +( x 2，1 - m 1) 2 +( x 3，1 - m 1) 2 +……+( x 1000，1 - m 1) 2

[0117] Belongs to category C There are 3000 samples of 2 ( t 2 =3000), respectively x 1，2 , x 2，2 , x 3，2 ,……, x 3000，2 , belongs to the category C The mean of the 3000 samples of 2 is m 2.

[0118] Then J2=( x 1，2 - m 2) 2 +( x 2，2 - m 2) 2 +( x 3，2 - m 2) 2 +……+( x 3000，2 - m 2) 2

[0119] Belongs to category C There are 2000 samples of 3 ( t 3 =2000), respectivelyx 1，3 , x 2，3 , x 3，3 ,……, x 2000，3 , belongs to the category C The mean of the 2000 samples of 3 is m 3.

[0120] Then J3=( x 1，3 - m 3) 2 +( x 2，3 - m 3) 2 +( x 3，3 - m 3) 2 +……+( x 2000，3 - m 3) 2

[0121] J = J 1+ J 2+ J 3

[0122] =[( x 1，1 - m 1) 2 +( x 2，1 - m 1) 2 +( x 3，1 - m 1) 2 +……+( x 1000，1 - m 1) 2 ]+[( x 1，2 - m 2) 2 +( x 2，2 - m 2) 2 +( x 3，2 - m 2) 2 +……+( x 3000，2 - m 2) 2 ]+[(x 1，3 - m 3) 2 +( x 2，3 - m 3) 2 +( x 3，3 - m 3) 2 +……+( x 2000，3 - m 3) 2 ]

[0123] 5. Generate medication recommendations

[0124] For each categorical attribute (e.g. region, occupation, etc.), calculate the detection frequency of genotypes within a specific category. C j Medium genotype G i The detection frequency is f i,j , drug proportion information of drug preparation recommendations w j It can be represented by a function based on the detection frequency, for example:

[0125] w j = g ( f i,j )

[0126] in g is a function that maps the detection frequency to drug proportion information. The final drug preparation plan is a comprehensive recommendation based on all classification attributes, which can be generated by weighted average or other methods. W , can be expressed using the following weighted summation:

[0127] W = w 1× p 1+ w 2× p 2+……+ w j × p j +……+ w k × p k

[0128] in p j Is a categorical attribute C jThe weight of .

[0129] Assuming that the demand for drugs in different regions is related to their detection frequency, a function can be introduced to represent the distribution of drug demand in different regions. The following are the specific steps and corresponding formulas:

[0130] 1. Regional drug demand distribution function

[0131] Let the region set be R ={ R 1, R 2, ..., R m},in m Represents the number of regions. Use regional drug demand function D j ( r ) indicates that in the area R j Drugs in the r The demand distribution can be the detection frequency f i A function of the demand for medicines in the region, such as:

[0132] D j ( r ) = h j ( f i,j , P j )

[0133] in:

[0134] f i,j It is a region R j Medium genotype G i The detection frequency;

[0135] P j It is a region R j Population or other demand-related indicators;

[0136] h j is a function that maps testing frequency and population size to a drug demand distribution.

[0137] 2. Drug Demand and Integration of Distribution Functions

[0138] Assuming that drug demand is continuously distributed in the region, the total drug demand can be calculated by integration.R j , assuming that the drug r The demand function is D j ( r , x ),in x is a spatial variable in the region, then the total drug demand in the region can be expressed as:

[0139] ∫ Rj D j ( r , x ) dx

[0140] 3. General medication recommendations

[0141] The drug stocking recommendation should consider the total drug demand of all regions. Assume that the drug demand of all regions is represented by different regional weights. We can express the drug stocking recommendation as follows W ( r ), which is the weighted sum of the drug demand distribution in different regions:

[0142] W ( r ) = w 1 ∫ R1 D 1 ( r , x ) dx +w 2 ∫ R2 D 2 ( r , x ) d x+w j ∫ Rj D j ( r , x ) dx +……+w m ∫ Rm D m ( r , x ) dx

[0143] in w j It is a region R jThe weight represents the relative priority of the region for drug allocation. W ( r ) represents the drug distribution scheme for all regions.

[0144] 4. ANOVA of regional drug distribution differences

[0145] In order to measure the differences in drug demand distribution in different regions, the variance of demand distribution can be calculated. Assume that the weighted average of drug demand in each region is D 平均 ( r ), the variance of drug demand in each region can be calculated using the following formula: 2 :

[0146] σ 2 ={[ ∫ R1 D 1 ( r , x ) dx - D 平均 ( r ) ] 2 +[ ∫ R2 D 2 ( r , x ) dx - D 平均 ( r )] 2 +……+ [ ∫ Rj D j ( r , x ) dx - D 平均 ( r ) ] 2 +……+ [ ∫ Rm D m ( r , x ) dx - D 平均 ( r ) ] 2} / m

[0147] in,

[0148] D 平均 ( r ) = [ ∫R1 D 1 ( r , x ) dx + ∫ R2 D 2 ( r , x ) dx +……+ ∫ Rj D j ( r , x ) dx + …+ ∫ Rm D m ( r , x ) dx ] / m

[0149] This variance measures the differences in the distribution of drug demand between different regions and helps determine whether the drug allocation ratio needs to be adjusted in certain areas.

[0150] Next, let's look at functions f i,j specific form of expression.

[0151] Assumptions f i,j It is a region R j Medium genotype G i The detection frequency of P j It is a region R j We hope h j These two factors can be considered together to express the distribution of regional demand for drugs. Here are several possible expressions:

[0152] 1. Linear Mapping

[0153] The simplest way is to use a linear mapping. We can define the demand as the product of the detection frequency and the population:

[0154] h j ( f i,j ，P j )= αfi,j · P j

[0155] in:

[0156] α is a scaling factor used to adjust the size of the demand.

[0157] Example: If the detection frequency in a certain area f i,j =0.2 (i.e. 20%), population P j =100,000, and α= 1.5, then the drug demand in this area is:

[0158] h j (0.2, 100000) = 1.5 × 0.2 × 100000 = 30000

[0159] Therefore, the drug requirement for this area is 30,000 units.

[0160] In this case, the detection frequency f i,j and population P j If the frequency of testing is high and the population is large, the demand will increase accordingly.

[0161] 2. Index Mapping

[0162] If we assume that the demand for drugs grows exponentially with the frequency of testing and the population, we can use an exponential mapping:

[0163] h j ( f i,j ，P j )= βe γfi,j·Pj

[0164] in:

[0165] β and c is a coefficient used to adjust the growth rate of demand.

[0166] Example: Assuming detection frequency f i,j =0.3, population P j =50,000, and β= 2 and c=0.00001, then the drug demand is:

[0167] h j (0.3, 50000) = 2× e 0.00001×0.3×50000 =2× e 0.15 ≈2×1.1618≈2.3236

[0168] Under this mapping, the drug demand in this area is about 2.32 units.

[0169] This function can describe the more sensitive effects of testing frequency and population size on drug demand.

[0170] 3. Logarithmic Mapping

[0171] If you want the growth of drug demand to be moderate, you can use a logarithmic mapping:

[0172] h j ( f i,j ，P j )= δln (1+ f i,j · P j )

[0173] in:

[0174] d is a coefficient used to adjust the amplification ratio of demand.

[0175] Example: Assuming detection frequency f i,j =0.1, population P j =200,000, and d =500, then the drug demand is:

[0176] h j (0.1, 200000)

[0177] =500× ln (1+ 0.1×200000)

[0178] =500× ln (1+20000)

[0179] =500× ln (20001)

[0180] ≈500×9.9035

[0181] =4951.75

[0182] Under this mapping, the regional drug demand is approximately 4,951.75 units.

[0183] This expression is more suitable for situations where the demand for drugs may increase disproportionately due to high testing frequency and high population size, and is particularly suitable for scenarios where demand shows a decreasing growth trend as the testing frequency increases.

[0184] 4. Comprehensive Linear Combination Mapping

[0185] In order to capture the multiple effects of different factors on demand distribution, an expression for comprehensive linear combination can be used:

[0186] h j ( f i,j ，P j ) = i 1 f i,j + i 2 P j + i 3 f i,j · P j

[0187] in:

[0188] i 1. i 2. i 3 is the weight coefficient, which determines the contribution of each factor based on actual conditions.

[0189] This formula allows for adjustment of testing frequency, population size, and the combined impact of the two separately, thus flexibly reflecting the complexity of regional drug demand.

[0190] Summarize

[0191] These mapping formulas provide different demand distribution models, and the parameters can be adjusted according to actual conditions. a,b,c, d,i 1 ,the 2 ,the 3) to most accurately reflect the impact of testing frequency and population size on drug demand. The specific choice depends on the sensitivity requirements for changes in drug demand in actual applications.

[0192] Through different mapping methods, we can flexibly select or adjust the mapping method according to the actual growth trend of drug demand, so as to effectively reflect the impact of testing frequency and population on the distribution of drug demand. Which formula to use depends on the characteristics of the drug demand being studied and the growth pattern of the demand in different regions.

[0193] The following embodiments do not consider the population size and the total amount of drugs required, but only consider the proportion of drugs in the drug preparation suggestions.

[0194] 1. Simple linear mapping

[0195] If we assume that the drug share is proportional to the frequency of genetic testing, a simple linear mapping can be used. Assume that the testing frequency is f i , then the proportion of drugs g ( f i ) can be expressed as:

[0196] g ( f i )= αf i

[0197] in:

[0198] α It is a proportional coefficient used to adjust the scale of the proportion.

[0199] Example: If f i =0.2 (i.e. the detection frequency is 20%), and α= 0.5, the proportion of drugs is g ( f i )=0.5×0.2=0.1, which means the proportion is 10%.

[0200] 2. Index Mapping

[0201] If the drug share increases at a faster rate with testing frequency, an exponential mapping can be used:

[0202] g ( f i ) = βe γfi

[0203] in:

[0204] β and c is the coefficient, controlling the growth rate of the proportion.

[0205] Example: Assumef i =0.2, β= 0.3, c =2, then the drug ratio is:

[0206] g ( f i ) = 0.3 e 2×0.2 ≈ 0.3 e 0.4 ≈0.3×1.49 ≈0.447

[0207] At this time, drugs accounted for 44.7%.

[0208] 3. Logarithmic Mapping

[0209] Logarithmic mapping can make the drug share slow down as the detection frequency increases:

[0210] g ( f i ) =δln (1+ f i )

[0211] in:

[0212] d is a coefficient used to adjust the output scale.

[0213] Example: Assume f i = 0.5 (i.e. 50%), and δ= 0.7, then the proportion of drugs is:

[0214] g ( f i ) = 0.7 ln (1+0.5) = 0.7 ln (1.5)≈0.7×0.405=0.2835

[0215] At this time, drugs accounted for 28.35%.

[0216] 4. Parabola Mapping

[0217] In order to reflect the situation that the proportion decreases as the detection frequency increases to a certain threshold, a parabola can be used:

[0218] g ( f i ) =-afi 2 + bf i + c

[0219] in:

[0220] a , b , c is a coefficient used to control the opening direction and position of the parabola.

[0221] Example: Assume f i = 0.4, and a = 1, b = 0.5, c = 0.1, then the drug proportion is:

[0222] g ( f i ) =- 1×0.4 2 +0.5×0.4+ 0.1 = 0.14

[0223] Drugs accounted for 14%.

[0224] Summarize

[0225] The mapping method you choose depends on the growth pattern, threshold, growth rate, etc. of drug demand. The above methods provide different assumptions about the relationship between demand and share, and the specific choice can be adjusted according to the actual scenario and data characteristics.

[0226] As an optional implementation, g ( f i ) = i 1×( αf i ) + i 2×( βe γfi )+ i 3×[ δln (1+ f i )]+ i 4×( -af i 2 + bf i + c )

[0227] i 1. i 2. i 3. i 4 is the weight coefficient, which determines the contribution of each factor based on actual conditions.

[0228] These mapping formulas provide different demand distribution models, and the parameters can be adjusted according to actual conditions. α , β , c , d , a , b , c , i 1, i 2, i 3. i 4) to most accurately reflect the impact of testing frequency and population size on drug demand. The specific choice depends on the sensitivity requirements for changes in drug demand in actual applications.

[0229] Through different mapping methods, we can flexibly select or adjust the mapping method according to the actual growth trend of drug demand, so as to effectively reflect the impact of testing frequency and population on the distribution of drug demand. Which formula to use depends on the characteristics of the drug demand being studied and the growth pattern of the demand in different regions.

[0230] In this application, the reference frequency is also called the reporting frequency, which is the frequency corresponding to each genotype of a gene reported by an authoritative organization.

[0231] The reference frequency is the statistical result within a larger sample range and has statistical significance.

[0232] Not all genes have reference frequencies, and some genes still do not have reference frequencies (reference frequencies are unknown).

[0233] The method of the present application is not only applicable to genes that already have reference frequencies, but also to genes that have not yet had reference frequencies.

[0234] The drugs and genes referred to in this application have a mapping relationship, which means that the genes are specific genes related to drug response, and the specific genes affect the metabolism and response of individuals to specific drugs. For example, amoxicillin is mainly related to the gene HLA-DQB1, that is, the drug amoxicillin has a mapping relationship with the gene HLA-DQB1. For another example, abacavir is mainly related to the gene HLA-B, that is, the drug abacavir has a mapping relationship with the gene HLA-B.

[0235] As an optional implementation, multiple samples are obtained; the detection frequencies corresponding to each genotype of the target gene G1 and the target gene G2 of the multiple samples are determined, and the target gene G1 has a mapping relationship with the target drug M1; the target gene G2 has a mapping relationship with the target drug M2; the reference frequencies corresponding to each genotype of the target gene G1 are determined; based on the detection frequency and reference frequency of each genotype of the target gene G1, it is judged whether a preset strategy is met; when the preset strategy is met, the multiple samples are classified by a data classification algorithm to generate multiple classification attributes; based on the target gene G2 and the multiple classification attributes, a drug preparation suggestion is generated, and the drug preparation suggestion includes the proportion information of the target drug M2.

[0236] In this embodiment, the reference frequency of gene G1 is known, and the reference frequency of gene G2 is unknown. If the detection frequency and reference frequency of multiple samples based on gene G1 meet the preset strategy, then these samples are considered to be statistically significant, not only for gene G1, but also for gene G2. Therefore, even if the reference frequency of gene G2 is unknown, these samples can still be used to perform the next step (classification, generating drug preparation suggestions), etc., to generate drug preparation suggestions for drug M2. This means that as long as the reference frequency of a gene is known, it is possible to determine whether the preset strategy is met based on the detection frequency and reference frequency of the gene, that is, to determine whether these samples are statistically significant. As long as these samples are judged to be statistically significant, these samples can be used for the proportioning of drugs related to genes with unknown reference frequencies.

[0237] Figure 2 yes Figure 1 Flow chart of the method for step S108 in FIG. Figure 2 The process shown is Figure 1 A detailed description of S108 "determining whether a preset strategy is satisfied based on the detection frequency and the reference frequency" in the process shown.

[0238] like Figure 2 As shown, in S202, the batch detection frequency corresponding to each genotype of the target gene of each batch of samples in the multiple batches of samples is determined, and the multiple samples include multiple batches of samples. The detection frequency of the target genotype in each batch of samples is calculated. The multiple batches of samples may, for example, come from different testing agencies or test results given by the same inspection agency at different time periods, and the genotype frequency needs to be calculated for each batch separately.

[0239] In a specific embodiment, the target gene is HLA-DQB1, which is related to amoxicillin, and the specific detection site may be, for example, rs9274407.

[0240] Batch 1: 100 samples, of which 30 samples have genotype AA, 50 samples have genotype AT, and 20 samples have genotype TT.

[0241] Batch 2: 200 samples, of which 40 samples had genotype AA, 120 samples had genotype AT, and 40 samples had genotype TT.

[0242] Batch 3: 150 samples, of which 45 samples had genotype AA, 75 samples had genotype AT, and 30 samples had genotype TT.

[0243] The detection frequencies of the three genotypes (AA, AT, TT) in each batch can be calculated.

[0244] For example, in batch 1, the frequency of genotype AA is 30 / 100 = 0.30, AT is 50 / 100 = 0.50, and TT is 20 / 100 = 0.20.

[0245] In S204, it is determined whether the preset strategy is met based on the batch detection frequency and the reference frequency. The genotype detection frequency of each batch is compared with the preset reference frequency. If the difference between the detection frequency and the reference frequency is within the allowed range, it is considered to meet the preset strategy, otherwise further processing is required.

[0246] For example: Assume the reference frequency is as follows:

[0247] The reference frequency of genotype AA is 0.25,

[0248] The reference frequency of genotype AT is 0.55,

[0249] The reference frequency of genotype TT was 0.20.

[0250] Compare the genotype frequencies of batch 1 to the reference frequencies:

[0251] AA: The detection frequency is 0.30, the reference frequency is 0.25, and the difference is 0.05, which meets the preset strategy.

[0252] AT: The detection frequency is 0.50, the reference frequency is 0.55, and the difference is 0.05, which is in line with the preset strategy.

[0253] TT: The detection frequency is 0.20, the reference frequency is 0.20, and the difference is 0, which meets the preset strategy.

[0254] In S206, if the batch detection frequency and the reference frequency of a batch of samples do not meet the preset strategy, the batch of samples is retested. If the difference between the genotype detection frequency and the reference frequency of a batch exceeds the allowable range, the batch of samples needs to be retested to ensure the accuracy of the results.

[0255] Assume that in batch 2, the detection frequency of genotype AT is 0.60, which exceeds the reference frequency of 0.55, and the difference is 0.05, which exceeds the error range allowed by the preset strategy. Therefore, the samples of batch 2 can be retested to retest the distribution of genotypes to ensure the accuracy of the data.

[0256] In S208, if the preset strategy is met, the subsequent steps are processed.

[0257] This application combines genetic testing with drug management through data collection, frequency analysis, classification algorithms, and drug recommendation generation to provide personalized drug use recommendations and improve the rationality and safety of drug use. This application is suitable for large-scale genetic testing projects and helps ensure the reliability of each batch of data, especially when the results will be used for personalized drug recommendations.

[0258] Figure 3 yes Figure 1 Flow chart of the method for step S112 in FIG. Figure 3 The process shown is Figure 1 A detailed description of S112 “generating drug preparation suggestions based on the target gene and the multiple classification attributes” in the process shown.

[0259] like Figure 3 As shown, in S302, the target classification attribute is determined. The specific attribute used to divide the samples is determined, and the attribute may be region, gender, age, occupation, etc. By selecting the target classification attribute, the samples can be grouped more specifically, so as to generate personalized drug preparation recommendations for different groups.

[0260] The following technical description is made by taking the regional attribute as an example. For example, the target classification attribute is the region. For example, it is divided into the northern region and the southern region, or divided into different provincial regions according to the province.

[0261] In S304, samples belonging to the target classification attribute are extracted from the multiple samples. Samples matching the target classification attribute are extracted from the multiple sample data sets.

[0262] Suppose there is nationwide genetic testing sample data, in which the samples record the genetic information, residential area and other demographic characteristics of individuals. If the target classification attribute is the southern region, then we will extract samples from the southern region from all samples, assuming that there are 200 samples from the south.

[0263] In S306, the detection frequency corresponding to each genotype of the target gene of the sample of the target classification attribute is determined, and attribute drug preparation suggestions for the target classification attribute are generated based on the detection frequency.

[0264] For the extracted samples, the genotype detection frequency of the target gene (gene related to the drug) is calculated. Based on the frequency, a drug preparation recommendation for the target classification attribute is generated. This drug preparation recommendation may include the usage ratio of a certain drug, a warning for cautious use, or a recommendation for an alternative drug.

[0265] Assume that the frequency of genotype HLA-B*57:01 is 20% in samples from the southern region, which means that 20% of the population in the southern region may have adverse reactions to abacavir. Based on this test result, a drug reserve recommendation can be generated to prompt hospitals in the southern region to reduce the purchase of abacavir or provide alternative drugs.

[0266] In S308, the drug preparation suggestion is generated through the attribute drug preparation suggestions corresponding to the multiple classification attributes. More specifically, the drug preparation suggestion includes a drug ratio, the classification attribute includes a regional attribute, and the drug preparation suggestion is generated through the attribute drug preparation suggestions corresponding to the multiple classification attributes, including: generating a drug ratio for a target regional attribute, wherein the drug ratio includes a proportion of the target drug.

[0267] Integrate the medication recommendations of multiple classification attributes to generate the final global medication recommendations. Medication recommendations may involve the usage ratio and alternatives of drugs to ensure that individuals with different attributes can obtain reasonable medication allocation.

[0268] For example, if the target classification attribute is not only region but also gender, drug usage recommendations can be generated for southern men and southern women respectively. For example, for men, the frequency of HLA-B*57:01 is 15%, while the frequency for women is 25%. Therefore, the generated drug preparation recommendations need to comprehensively consider these different classification attributes, and finally combine the overall proportion of women in the southern region to provide detailed drug ratios and usage recommendations for hospitals in the southern region.

[0269] In one embodiment, generating a drug ratio for the target regional attribute includes: determining a substitute drug for the target drug for the target regional attribute and a proportion of the substitute drug.

[0270] Figure 4 It is a flow chart of “determining alternative drugs for the target drug and the proportion of the alternative drugs according to the target regional attributes”.

[0271] like Figure 4As shown, in S402, multiple alternative drugs are determined, and the alternative drugs are other drugs under the same drug classification as the target drug. Multiple alternative drugs with similar mechanisms of action as the target drug are selected from the drug classification. The target drug may not be suitable for some patients due to the influence of certain genotypes, so it is necessary to determine alternative drugs with similar functions but higher safety.

[0272] In S402, candidate genes are determined according to the candidate drugs, and the candidate genes have a mapping relationship with the candidate drugs. It can be assumed that the target drug is abacavir (used for HIV treatment). Due to the presence of HLA-B*57:01 genotype in some people, they will have a severe allergic reaction to abacavir. Therefore, possible alternative drugs such as lamivudine, tenofovir, etc. can be selected from the same type of antiretroviral drugs.

[0273] In S406, the detection frequency corresponding to each genotype of the candidate gene of the samples in the target region is determined. The samples in the target region are analyzed to calculate the occurrence frequency of each candidate genotype in the population.

[0274] In S408, alternative drugs for the target drug of the target regional attribute are screened from the multiple alternative drugs according to the detection frequencies corresponding to the genotypes of the candidate genes of the samples of the target regional attribute. Based on the detection frequencies in the previous step, the alternative drugs that are most suitable for the population in a specific region are selected. These alternative drugs should be safer and more effective for patients whose genotypes are common in the region.

[0275] For example, in the southern region, due to the high frequency of HLA-B*57:01 genotype, abacavir is not suitable for patients in this region. By analyzing the gene mapping relationship of other alternative drugs, it was found that tenofovir is more suitable for patients in the southern region because the drug has a low correlation with common gene mutations and a stable metabolic pathway, so it was selected as the main alternative drug.

[0276] In a specific application scenario, it also includes: obtaining the classification attribute of the user; obtaining the target genotype of the user; and preparing drugs for the user according to the classification attribute and the target genotype combined with the drug preparation suggestion. For example, assuming a patient XX in the southern region, the doctor can provide the patient with target drugs and alternative drugs combined with the drug preparation suggestion to avoid adverse reactions.

[0277] In a specific application scenario, it also includes: determining a regional classification attribute; determining a drug procurement plan based on the regional classification attribute combined with the drug reserve recommendation, the drug procurement plan includes target drugs and alternative drugs for each genotype. For example, in hospitals in the southern region, analysis found that the HLA-B*57:01 genotype is more common in the region, so when formulating a procurement plan, the purchase amount of abacavir can be reduced, and the purchase proportion of tenofovir and lamivudine can be increased. For example, the procurement plan may recommend reducing the inventory proportion of abacavir to 10% and increasing the proportion of tenofovir to 30%.

[0278] This application obtains the patient's classification attributes and genotype information, and formulates a drug allocation plan or drug procurement plan based on regional attributes to achieve precision medicine and supply chain optimization. It helps to reduce adverse reactions, optimize drug use strategies, and ensure the rationality and effectiveness of the drug supply chain.

[0279] Those skilled in the art will appreciate that all or part of the steps for implementing the above embodiments are implemented as a computer program executed by a CPU. When the computer program is executed by the CPU, the above functions defined by the above method provided in the present application are performed. The program may be stored in a computer-readable storage medium, which may be a read-only memory, a disk, or an optical disk, etc.

[0280] In addition, it should be noted that the above figures are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present application, and are not intended to be limiting. It is easy to understand that the processes shown in the above figures do not indicate or limit the time sequence of these processes. In addition, it is also easy to understand that these processes can be performed synchronously or asynchronously, for example, in multiple modules.

[0281] The following is an embodiment of the device of the present application, which can be used to execute the embodiment of the method of the present application. For details not disclosed in the embodiment of the device of the present application, please refer to the embodiment of the method of the present application.

[0282] Figure 5 FIG. 1 is a block diagram of a device for generating a drug preparation suggestion based on a classification attribute according to an exemplary embodiment. Figure 5 As shown, the device 50 for generating drug preparation suggestions based on classification attributes includes: a sample module 502 , a detection module 504 , a frequency module 506 , a strategy module 508 , a classification module 510 , and a suggestion module 512 .

[0283] The sample module 502 is used to obtain multiple samples;

[0284] The detection module 504 is used to determine the detection frequency corresponding to each genotype of the target gene of the multiple samples, and the target gene has a mapping relationship with the target drug;

[0285] The frequency module 506 is used to determine the reference frequency corresponding to each genotype of the target gene;

[0286] The strategy module 508 is used to determine whether a preset strategy is satisfied based on the detection frequency and the reference frequency;

[0287] The classification module 510 is used to classify the multiple samples by a data classification algorithm to generate multiple classification attributes when the preset strategy is met;

[0288] The suggestion module 512 is used to generate a drug preparation suggestion based on the target gene and the multiple classification attributes, and the drug preparation suggestion includes the proportion information of the target drug.

[0289] According to the device for generating drug preparation suggestions based on classification attributes of the present application, multiple samples are obtained; the detection frequencies corresponding to each genotype of the target gene of the multiple samples are determined, and the target gene has a mapping relationship with the target drug; the reference frequencies corresponding to each genotype of the target gene are determined; based on the detection frequencies and the reference frequencies, it is judged whether a preset strategy is met; when the preset strategy is met, the multiple samples are classified by a data classification algorithm to generate multiple classification attributes; drug preparation suggestions are generated based on the target gene and the multiple classification attributes, and the drug preparation suggestions include the proportion information of the target drug, which can effectively improve the safety of drug use and the rationality of drug supply, and provide strong technical support for personalized medicine.

[0290] Figure 6 It is a block diagram of an electronic device according to an exemplary embodiment.

[0291] Refer to the following Figure 6 The electronic device 600 according to this embodiment of the present application is described. Figure 6 The electronic device 600 shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0292] like Figure 6 As shown, the electronic device 600 is in the form of a general computing device. The components of the electronic device 600 may include but are not limited to: at least one processing unit 610, at least one storage unit 620, a bus 630 connecting different system components (including the storage unit 620 and the processing unit 610), a display unit 640, etc.

[0293] The storage unit stores a program code, which can be executed by the processing unit 610, so that the processing unit 610 performs the steps described in this specification according to various exemplary embodiments of the present application. For example, the processing unit 610 can perform the following steps: Figure 1 to Figure 4 Follow the steps shown in .

[0294] The storage unit 620 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 6201 and / or a cache memory unit 6202 , and may further include a read-only memory unit (ROM) 6203 .

[0295] The storage unit 620 may also include a program / utility 6204 having a set (at least one) of program modules 6205, such program modules 6205 including but not limited to: an operating system, one or more application programs, other program modules and program data, each of which or some combination may include the implementation of a network environment.

[0296] Bus 630 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0297] The electronic device 600 may also communicate with one or more external devices 600' (e.g., keyboards, pointing devices, Bluetooth devices, etc.) so that a user can communicate with the device that interacts with the electronic device 600, and / or the electronic device 600 can communicate with any device that communicates with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed through an input / output (I / O) interface 650. In addition, the electronic device 600 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) through a network adapter 660. The network adapter 660 may communicate with other modules of the electronic device 600 through a bus 630. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0298] Through the above description of the implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by combining software with necessary hardware. Figure 7 As shown, the technical solution according to the implementation mode of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes a number of instructions to enable a computing device (which can be a personal computer, a server, or a network device, etc.) to execute the above method according to the implementation mode of the present application.

[0299] The software product may be any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, a system, device or component of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination thereof. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0300] The computer readable storage medium may include a data signal propagated in a baseband or as part of a carrier wave, wherein a readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The readable storage medium may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by an instruction execution system, an apparatus, or a device or used in combination with it. The program code contained on the readable storage medium may be transmitted with any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the above.

[0301] Program code for performing the operations of the present application may be written in any combination of one or more programming languages, including object-oriented programming languages, such as Java, C++, etc., and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., using an Internet service provider to connect through the Internet).

[0302] The above-mentioned computer-readable medium carries one or more programs. When the above-mentioned one or more programs are executed by a device, the computer-readable medium realizes the following functions: obtaining multiple samples; determining the detection frequency corresponding to each genotype of the target gene of the multiple samples, and the target gene has a mapping relationship with the target drug; determining the reference frequency corresponding to each genotype of the target gene; judging whether a preset strategy is met based on the detection frequency and the reference frequency; when the preset strategy is met, classifying the multiple samples through a data classification algorithm to generate multiple classification attributes; generating a drug preparation suggestion based on the target gene and the multiple classification attributes, and the drug preparation suggestion includes the proportion information of the target drug.

[0303] Those skilled in the art will appreciate that the above modules can be distributed in the device according to the description of the embodiment, or can be changed accordingly and only used in one or more devices different from the embodiment. The modules of the above embodiments can be combined into one module, or further divided into multiple sub-modules.

[0304] Through the description of the above embodiments, it is easy for those skilled in the art to understand that the example embodiments described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the embodiment of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a mobile terminal, or a network device, etc.) to execute the method according to the embodiment of the present application.

[0305] The exemplary embodiments of the present application are specifically shown and described above. It should be understood that the present application is not limited to the detailed structures, configurations or implementations described herein; on the contrary, the present application is intended to cover various modifications and equivalent configurations included in the spirit and scope of the appended claims.

Claims

1. A method for generating drug preparation suggestions based on classification attributes, characterized in that: include: Obtain multiple samples, where the sample refers to a biological sample, including DNA information of an individual; Determine the detection frequency corresponding to each genotype of the target gene of the multiple samples, the target gene and the target drug have a mapping relationship, wherein the detection frequency refers to the ratio of the number of occurrences of a specific genotype in the sample to the total number of samples; Determining the reference frequency corresponding to each genotype of the target gene; Determining whether a preset strategy is satisfied based on the detection frequency and the reference frequency; When the preset strategy is met, the multiple samples are classified by a data classification algorithm to generate multiple classification attributes, wherein the classification attributes include regional attributes; A drug preparation suggestion is generated based on the target gene and the multiple classification attributes, and the drug preparation suggestion includes proportion information of the target drug.

2. The method according to claim 1, characterized in that The plurality of samples include a plurality of batches of samples, Judging whether a preset strategy is satisfied based on the detection frequency and the reference frequency includes: Determining the batch detection frequency corresponding to each genotype of the target gene of each batch of samples in the multiple batches of samples; Whether the preset strategy is satisfied is determined based on the batch detection frequency and the reference frequency.

3. The method according to claim 2, characterized in that Also includes: If the batch detection frequency based on a batch of samples and the reference frequency do not satisfy the preset strategy, retest the batch of samples; Based on the batch detection frequency of the re-inspection and the reference frequency, it is determined whether the preset strategy is met.

4. The method according to claim 1, characterized in that Generating a drug preparation suggestion based on the target gene and the multiple classification attributes includes: Determine the target classification attributes; Extracting samples belonging to the target classification attribute from the multiple samples; Determine the detection frequency corresponding to each genotype of the target gene of the sample of the target classification attribute and generate attribute preparation suggestions for the target classification attribute based on the detection frequency; The medicine preparation suggestion is generated by using the attribute medicine preparation suggestions corresponding to the multiple classification attributes.

5. The method according to claim 4, characterized in that The drug preparation suggestion includes drug ratio, Generating the drug preparation suggestion by using the attribute drug preparation suggestions corresponding to the multiple classification attributes includes: Generate a drug ratio for the target regional attributes, wherein the drug ratio includes a proportion of the target drug.

6. The method according to claim 5, characterized in that Generate a drug ratio tailored to the target geographical attributes, including: Determine alternative drugs for the target drug and the proportion of the alternative drugs according to the target geographical attributes.

7. The method according to claim 6, characterized in that Determine alternative drugs for the target drug that are specific to the target geographic attributes, including: Determine a plurality of candidate drugs, wherein the candidate drugs are other drugs in the same drug classification as the target drug; Determine an alternative gene according to the alternative drug, wherein the alternative gene has a mapping relationship with the alternative drug; Determining the detection frequency corresponding to each genotype of the candidate gene of the samples in the target region; A substitute drug for the target drug with the target regional attribute is screened out from the multiple candidate drugs according to the detection frequency corresponding to each genotype of the candidate gene of the sample with the target regional attribute.

8. The method according to claim 1, characterized in that The target gene includes a first target gene and a second target gene, wherein the reference frequency of the first target gene is known, and the reference frequency of the second target gene is unknown, Determining the detection frequency corresponding to each genotype of the target gene of the multiple samples, wherein the target gene has a mapping relationship with the target drug, includes: Determine the detection frequencies corresponding to the respective genotypes of the first target gene and the second target gene of the multiple samples, wherein the first target gene has a mapping relationship with the first target drug, and the second target gene has a mapping relationship with the second target drug, Determining the reference frequency corresponding to each genotype of the target gene comprises: Determining reference frequencies corresponding to each genotype of the first target gene; Judging whether a preset strategy is satisfied based on the detection frequency and the reference frequency includes: Based on the detection frequency of the first target gene and the reference frequency of the first target gene, it is determined whether a preset strategy is met, When the preset strategy is met, the multiple samples are classified by a data classification algorithm to generate multiple classification attributes, including: When the detection frequency of the first target gene and the reference frequency of the first target gene meet a preset strategy, the multiple samples are classified by a data classification algorithm to generate multiple classification attributes, Generate a drug preparation suggestion based on the target gene and the multiple classification attributes, wherein the drug preparation suggestion contains information about the proportion of the target drug, including: A drug preparation suggestion is generated based on the second target gene and the multiple classification attributes, and the drug preparation suggestion includes proportion information of the second target drug.

9. The method according to claim 1, characterized in that It also includes calculating the drug proportion according to the following formula: g ( f i ) = θ 1×( αf i )+ θ 2×( βe γfi )+ θ 3×[ δln (1+ f i )]+ θ 4×( -af i 2 + bf i + c ) in, g ( f i ) is the proportion of drugs, f i is the detection frequency, θ 1. θ 2. θ 3. θ 4 is the weight coefficient, α , β , γ , δ , a , b , c as a parameter.

10. A device for generating drug preparation suggestions based on classification attributes, characterized in that: include: A sample module, used to obtain multiple samples, wherein the sample refers to a biological sample, including DNA information of an individual; A detection module, used to determine the detection frequency corresponding to each genotype of the target gene of the multiple samples, wherein the target gene has a mapping relationship with the target drug, wherein the detection frequency refers to the ratio of the number of occurrences of a specific genotype in the sample to the total number of samples; A frequency module, used to determine the reference frequency corresponding to each genotype of the target gene; A strategy module, used to determine whether a preset strategy is satisfied based on the detection frequency and the reference frequency; A classification module, configured to classify the plurality of samples by a data classification algorithm to generate a plurality of classification attributes when the preset strategy is met, wherein the classification attributes include a regional attribute; A suggestion module is used to generate a drug preparation suggestion based on the target gene and the multiple classification attributes, wherein the drug preparation suggestion includes the proportion information of the target drug.

11. An electronic device, characterized in that: include: one or more processors; A storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 9.

12. A computer readable medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 9 is implemented.

Images