RU2790285C1 - Method and device for determination of gene association degree - Google Patents

Abstract

FIELD: biotechnology.

SUBSTANCE: method for determination of a degree of association between a disease description record and a gene is described. By means of a unit for determination of a disease description record, the disease description record in a case description text is determined. For each of a set of given association databases: by means of a unit for determination of an association indicator, record data for a target association record(s) is determined in the given association database in accordance with the disease description record and gene identifiers, each of which corresponds to one of a set of genes, wherein each of given association databases stores records of association of the disease description record and gene identifiers corresponding to at least one of the set of genes. By means of the unit for determination of an association indicator, record data is entered into a given record matrix of gene association to determine an association indicator of the disease description record with the corresponding each of the set of genes from the given association database. By means of a unit for determination of a degree of association, degrees of association between the disease description record and each of the set of genes are determined in accordance with association indicators of the disease description record with gene identifiers corresponding to the set of genes from the set of given association databases. A corresponding device for determination of a degree of association between a disease description record and a gene, containing the above-mentioned units, is also described. In addition, a storage medium for determination of a degree of association between a disease description record and a gene is described. According to the invention, a program is stored on the storage medium, and, when executed by a processor, it implements a method for determination of a degree of gene association. A processor is disclosed for determination of a degree of association between a disease description record and a gene. According to the invention, the processor is used to execute a program, wherein the program, when executed, implements a method for determination of a degree of gene association. An electronic device for determination of a degree of association between a disease description record and a gene is presented, containing at least one processor, at least one processor-associated memory, and a bus. In this case, the processor communicates with memory via the bus, and the processor is used to call program commands from memory to implement a method for determination of a degree of gene association.

EFFECT: solution allows for an increase in a speed of determination of a degree of association between a disease description record and each of a set of genes.

15 cl, 6 dwg

Description

The present application claims priority over Chinese Invention Application No. 202011535972.2 filed with the Patent Office of the People's Republic of China on December 23, 2020 under the title "METHOD AND DEVICE FOR DETERMINING A DEGREE OF GENE ASSOCIATION", the entire content of which is hereby incorporated by reference.

The field of technology to which the invention belongs

The application relates to the field of gene association technology and, in particular, to a method and device for determining the degree of gene association.

State of the art

At present, due to the deepening of medical research, the identification of pathogenic genes associated with hereditary diseases will play an important supporting role for the relevant personnel in the investigation and subsequent treatment of these hereditary diseases.

Hereditary disease is usually associated with numerous pathogenic genes. A lot of clinical information about the phenotype described in case reports may exist in the text of case reports for patients with hereditary diseases. Various clinical information about the phenotype may be associated with one or more pathogenic genes. Pathogenic genes associated with different clinical phenotypic information may be the same or different. Currently, 5181 types of hereditary diseases and 15428 types of genes have been identified. How to quickly determine the association between disease description records and genes in the text of a patient's case report has become an urgent technical problem for the relevant personnel.

The essence of the invention

In terms of the problems described above, the present application provides a method and apparatus for determining the degree of gene association in order to overcome the above problem or at least partially solve the above problem. The corresponding technical solution is presented below.

A method for determining the degree of gene association, comprising the steps of:

determining a disease description entry in the case description text;

for a set of given association databases:

determining record data for the target association record(s) in the given association database according to the disease description record and gene identifiers, each of which corresponds to one of the plurality of genes, where each of the given association databases stores association records of the disease description record and gene identifiers, corresponding to at least one of the plurality of genes;

entering the entry data into a predetermined gene association entry matrix to determine an association score of the disease description entry with each of the plurality of genes in the predetermined association database; And

determining the degree of association between the disease description record and each of the plurality of genes according to the association scores of the disease description record with the gene identifiers corresponding to the plurality of genes from the plurality of predetermined association databases.

Alternatively, determining entry data for the target association entry(s) in a given association database according to a disease description entry and gene identifiers, each corresponding to one of a plurality of genes, comprises the steps of:

for any gene of the plurality of genes, record data is determined for the target association record(s) of the disease description record and the gene identifier corresponding to the gene from the given association database, respectively, where the record data for the target association record(s) contains: the first number of records associations containing both the gene identifier of the gene and the disease description record, a second number of association records containing the disease description record, a third number of association records containing the gene gene identifier, and the total number of association records in a given association database.

Alternatively, a given gene association matrix entry has the form:

,

,

where M is the indicator of the association of a disease description record with a gene identifier corresponding to a gene in a given information gene database; T ₁ - the first quantity number, T ₂ - the second quantity, T ₃ - the third quantity; T _S - total.

Alternatively, determining the degree of association between the disease description record and each of the plurality of genes according to the association score of the disease record with gene identifiers corresponding to the plurality of genes from the plurality of given association databases, for any gene, comprises the steps of:

weighting association scores of the disease description record with a gene identifier corresponding to a gene from a plurality of predetermined gene association information databases to obtain an association value between the disease description record and the gene; And

determining the degree of association between the disease description record and the gene according to the association value.

Alternatively, determining the degree of association between a disease description record and a gene according to the association value comprises:

, где L - степень ассоциации между записью описания болезни и геном, e – основание натурального логарифма и z - значение ассоциации.determining the degree of association between a disease description record and a genome according to

, where L is the degree of association between the disease description record and the genome, e is the base of the natural logarithm, and z is the value of the association.

Alternatively, the learning process for a given record-gene association matrix includes the steps of:

input as training information the record data for target association records containing a plurality of disease description records and gene identifiers of a plurality of genes from a plurality of given gene association information databases into the current gene association training record matrix and determine the predicted association indicators corresponding to the training information and the set genes in the given gene association information databases, respectively;

determining a predictive degree of association between the plurality of disease description records and the plurality of genes, respectively, according to certain predictive association scores;

determining errors between predicted association rates and actual association rates for the plurality of disease description records and the plurality of genes;

determining if the errors are below a predetermined error threshold;

if so, determining that the current training gene association matrix record is a gene association matrix record; And

otherwise, adjust the matrix parameters in the current training record-gene association matrix by means of a descending gradient algorithm, using the adjusted training record-gene association matrix as the current training record-gene association matrix, and returning to the input step as training information, writing data of the target association record(s) containing a plurality of disease description records and gene identifiers of a plurality of genes from a plurality of predetermined gene association databases into the current gene association training record matrix, and determining association predictive scores corresponding to the training information and the plurality of genes in the predetermined gene association information databases, respectively.

A gene association degree determination device, comprising: a disease description recording unit, an association indicator determination unit, and an association degree determination unit;

wherein the disease description entry definition block is used to determine the disease description entry in the case description text;

the association indicator determining unit is used for the plurality of predetermined association databases to determine entry data for the association target record(s) in the predetermined association database in accordance with the disease description record and gene identifiers, each of which corresponds to one of the plurality of genes, where each from the given association databases, stores association records of the disease description record and gene identifiers corresponding to at least one of the plurality of genes; and inputting the entry data into a predetermined gene association entry matrix to determine an association score of the disease description entry with a corresponding each of the plurality of genes from the predetermined association database; And

the association degree determining unit is used to determine the degree of association between the disease description entry and each of the plurality of genes according to the association score of the disease description record with the gene identifiers corresponding to the plurality of genes from the plurality of predetermined association databases.

Alternatively, the association indicator determination unit is specifically used for any gene of the plurality of genes, respectively determining the entry data for the target association entry(s) containing the disease description entry and the gene identifier corresponding to the gene from the predetermined association database, the entry data for the target association the association record(s) contain a first number of association records containing both the gene gene identifier and a disease description record, a second number of association records containing a disease description record, a third number of association records containing the gene gene identifier, and the total number of association records in given association database.

Alternatively, a given gene association matrix entry has the form:

,

,

where M is the indicator of the association of a disease description record with a gene identifier corresponding to a gene from a given information gene database; T ₁ - the first number, T ₂ - the second number, T ₃ - the third number; T _S - total.

Alternatively, the association degree determination block comprises an association value obtaining sub-block and an association degree determination sub-block,

wherein the association value obtaining sub-unit is used for any gene, weighting association scores of the disease description entry with a gene identifier corresponding to a gene from a plurality of predetermined gene association information databases to obtain an association value between the disease description entry and the gene;

the association degree determination sub-unit is used to determine the degree of association between the disease description record and the gene according to the association value.

, где L - степень ассоциации между записью описания болезни и геном, e – основание натурального логарифма и z - значение ассоциации.Alternatively, the degree of association determination subunit is used to determine the degree of association between a disease description record and a gene according to

, where L is the degree of association between the disease description record and the genome, e is the base of the natural logarithm, and z is the value of the association.

Alternatively, the device further comprises: a predictive association index determination unit, a predictive degree of association determination unit, an error determination unit, an error comparison unit, a gene association matrix entry determination unit, and a matrix parameter correction unit,

wherein the association predictor determination unit is used to input, as training information, record data for target association records containing a plurality of disease description records and gene identifiers of a plurality of genes from a plurality of predetermined gene association information databases into the current training gene association matrix record, and to determine predictive association scores corresponding to the training information and the plurality of genes in the given gene association information databases, respectively;

a predictive degree of association determining unit is used to determine a predicted degree of association between the plurality of disease description records and the plurality of genes, respectively, in accordance with certain association predictive scores;

an error determination unit is used to determine an error between the predicted degrees of relationship and the actual degrees of relationship of the plurality of disease description records and the plurality of genes;

an error comparer is used to determine if the error is below a predetermined error threshold, and if so, run a gene association matrix entry determiner, or otherwise run a matrix parameter adjuster;

a gene association matrix entry determining unit is used to determine the current training gene association matrix entry as the gene association matrix entry;

the matrix parameter adjustment block is used to adjust the matrix parameters of the current gene association matrix training record by a descending gradient algorithm, using the adjusted gene association training record as the current gene association training record matrix, and to start the predicted association indicator determination block.

A storage medium that stores a computer program that, when executed by a processor, performs a method for determining the degree of gene association according to any of the methods described above.

A processor for executing a program, wherein the program, when executed, performs a method for determining the degree of gene association in accordance with any of the methods described above.

An electronic device comprising at least one processor, at least one memory associated with the processor, and a bus; in which the processor communicates with the memory through the bus; the processor is used to recall program instructions from memory to perform the method for determining the degree of gene association in accordance with any of the methods described above.

A computer program product that, when executed on an electronic device, is configured to execute a program to initialize the steps of a method for determining the degree of gene association in accordance with any of the methods described above.

Using the technical scheme described above, the method and device for determining the degree of gene association presented in this application is able to determine the record of the description of the disease in the text of the case description. For a plurality of given association databases, record data for the target association record(s) in a given association database is determined according to a disease description record and gene identifiers, each of which corresponds to one of the plurality of genes, where each of the given association databases stores records association of the disease description record and gene identifiers corresponding to at least one of the plurality of genes; the entry data is entered into a predetermined gene association entry matrix to determine an association score of the disease description entry with a corresponding each of the plurality of genes in the predetermined association database; and the degree of association between the disease description entry and each of the plurality of genes is determined according to the association scores of the disease description entry with gene identifiers corresponding to the plurality of genes in the plurality of predetermined association databases. In this application, by determining the record data for the target association record(s) of the disease description record and a plurality of genes from each given association database and entering the record data into a given gene association record matrix, the association score of the disease record with the corresponding each of the set is determined. genes in a given association database, and the degree of association between a disease description record and a plurality of genes can be quickly obtained.

The above is only a brief overview of the technical solution in this application. In order to understand the technical means of the present application more clearly, it can be implemented in accordance with the record of this description, and in order to make the above and other objects, features and advantages of the present application more clear, specific embodiments of the present application will be presented below.

Brief description of the drawings

In order to more clearly explain the embodiments of the present application or the technical solutions of the prior art, a brief introduction to the drawings will be given below, necessary for use in describing the embodiments or the prior art. Obviously, the drawings shown below are just some of the embodiments of the present application. A person skilled in the art can also obtain other drawings corresponding to such drawings, provided that no inventive efforts are applied.

Fig. 1 is a flowchart of the method for determining the degree of gene association provided in some embodiments of the present application;

fig. 2 is a flowchart of a method for determining the degree of gene association provided in some embodiments of the present application;

fig. 3 is a diagram of a learning process for a given gene association matrix record as provided in some embodiments of the present application;

fig. 4 is a flowchart of yet another method for determining the degree of gene association provided in some embodiments of the present application;

fig. 5 is a block diagram of a device for determining the degree of gene association provided in some embodiments of the implementation of the present application;

fig. 6 is a block diagram of another device for determining the degree of gene association provided in some embodiments of the implementation of the present application.

Detailed Description of the Preferred Embodiments

Exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. Although the drawings show exemplary embodiments of the present disclosure, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments presented here. Rather, these embodiments are set forth to allow a more complete understanding of the present disclosure and to fully convey to those skilled in the art the scope of the present disclosure.

As shown in FIG. 1, the method for determining the degree of gene association provided in some embodiments of the present application may comprise the following steps, where:

S100: define the record of the description of the disease in the text of the description of the case.

The text of the case description may be a record of medical activities such as examination by medical staff and diagnosis of patients' illnesses. Alternatively, the case description text may be sentences or multi-word paragraphs. The disease description record may be a record describing clinical information about a patient's disease phenotype.

Alternatively, the disease description record may be an HPO manifestation or a hereditary disease manifestation. HPO (human phenotype ontology, ontology of the human phenotype) is a set of standard vocabulary describing pathological phenotypes caused by human diseases. The standard vocabulary set contains numerous manifestations of HPO. It is understood that manifestations of HPO may be expressed in Chinese words or words in other languages such as English. A hereditary disease manifestation may be a trade name, and/or an abbreviation and/or one of the other user-defined disease names where genes are the main cause of the diseases. It should be noted that user-defined industry names may be colloquial terms for diseases. For example, the professional name for the disease "bronchial asthma" can be shortened to "asthma" and is often expressed as a colloquial expression for "wheezing". It should be understood that due to the continuous deepening of research into hereditary diseases in the field of research and the different actual needs in different fields of application, industry personnel can create their representations of hereditary diseases in accordance with the needs of the study or the actual needs of the application.

Alternatively, in some embodiments of the present application, the text of the case description can be segmented and structured representations of the description of the disease can be extracted from the unstructured text of the case description through the extraction of named records (named entity recognition, NER) in natural language processing (NLP) . Embodiments of the present application can accurately extract a disease description entry from a case description text using said named entry extraction technology.

It should be understood that embodiments of the present application may also extract the disease description entry from the case description text via predetermined disease description keywords.

It should be understood that the case description text may contain one or more manifestations of the disease description entry, and embodiments of the present application may perform the method for determining the degree of gene association provided by the embodiments of the present application on any disease description record determined from the case description text.

S200: for a plurality of given association databases, determining record data for one or more target association records from a given association database, corresponding disease description records and gene identifiers, each corresponding to one of the plurality of genes, each of the given association databases storing records association of the disease description record and gene identifiers corresponding to at least one of the plurality of genes; and entering the entry data into a predetermined gene association entry matrix to determine an association score of the disease description entry with a corresponding each of the plurality of genes in the predetermined association database.

The predetermined association database may be a database derived from an industry-wide disease database to associate displayable records with gene identifiers. Alternatively, industry-wide disease databases may contain: OMIM (disease-related gene) database, ClinVar (genetic variation) database, HGMD (human gene mutation) database. genes), the HPO (standard terms of human phenotype) database, and the Orphanet (rare disease-related) database. Embodiments of the present application may accordingly create association databases for various industry-wide disease databases. Alternatively, a given association database may store association records that represent an association relationship between disease description records and gene identifiers.

Alternatively, embodiments of the present application may determine entry data for the target association entry(s) in a given association database in accordance with the disease description entry and gene identifiers corresponding to the plurality of genes. Alternatively, the record data for the target association record(s) may be the number of association records containing the disease description record and any of the plurality of genes from the given association database, or the number of association records containing the gene identifier of any of the plurality of genes in the given database. associations.

Alternatively, step S200 may comprise, for any gene of the plurality of genes, determining entry data for the target association entry(s) of the disease description entry and the gene identifier corresponding to the gene in the predetermined association database, respectively. The record data for the target association record(s) comprises a first number of association records containing the gene gene identifier and a disease description record, a second number of association records containing a disease description record, a third number of association records containing the gene gene identifier, and the total number of association records. in the given association database.

It should be understood that the target association record(s) may be an association record(s) in a given association database, and may be an association record(s) containing a gene identifier of a gene and a disease description record in a given association database, record( associations) containing a disease description record in a given association database, or an association record(s) containing the gene identifier of a gene.

Alternatively, a given gene association matrix record may look like:

,

,

where M is the indicator of the association of the disease description record with the gene identifier corresponding to the gene in the given gene information database; T ₁ is the first quantity, T ₂ is the second quantity, T ₃ is the third quantity, and T _S is the total quantity.

In embodiments of the present application, by inputting entry data for a specific target association entry(s) into a given gene association entry matrix, association scores derived from the given gene association entry matrix can be obtained by calculation. It should be noted that the number of association scores of a particular disease description record with respect to any one of the plurality of genes is the same as the number derived from the plurality of given association databases. For example, assuming that there are five given association databases for a disease description record and any gene, in some embodiments of the present application, association scores of a disease description record with a gene can be determined based on the five given association databases using a given matrix record. gene association, that is, five indicators of the association of a disease description record with a gene can be determined. It should be understood that the five association measures may be the same or different.

S300: determining the degree of association between the disease description entry and each of the plurality of genes according to the association scores of the disease description entry with the gene identifiers corresponding to the plurality of genes from the plurality of predetermined association databases.

Alternatively, in embodiments of the present application, for any gene, association scores corresponding to a disease description record and a gene identifier corresponding to a gene from a plurality of predetermined gene association information databases are summed to obtain an association value of the disease description record and the gene.

Alternatively, based on the method shown in FIG. 1, another method for determining the degree of gene association provided in some embodiments of the present application is shown in FIG. 2. Step S300 may comprise:

S310: For any gene, association scores of the disease description record with a gene identifier corresponding to a gene from a plurality of predetermined gene association information databases are weighted to obtain an association value between the disease description record and the gene.

In some embodiments of the present application, a weight may be set for each given gene association information database in advance so that a more reliable association value between the disease description record and the gene can be obtained after the weighting procedure. For ease of understanding, this is illustrated here with an example. The weight of the given infobase A of the gene association data is set to w1, the weight of the given infobase B of the gene association data is set to w2, the weight of the given infobase C of the gene association data is set to w3, the weight of the given infobase D of the gene association data is set to w4, and the weight of the given infobase C of the gene association data is set to w4, and the weight of the given the information base E of the gene association data is set to w5 in advance; the association score of the disease description record with the gene identifier of the gene corresponding to the given association data gene information base A is x1, the association score is x2, the association score corresponding to the given gene base information B in the association data is x2, the association score corresponding to the given gene information base database C is x3, the association score corresponding to the predetermined gene association information database D is x4, and the association score corresponding to the predetermined gene association information database E is x5. Through a weighting process, the association value between the disease description record and the gene is obtained as follows:

S320: determining the degree of association between the disease description record and the gene according to the association value.

Alternatively, step S320 may comprise:

, где L - степень ассоциации между записью описания болезни и геном, e – основание натурального логарифма и z - значение ассоциации.determining the degree of association between a disease description record and a genome according to

, where L is the degree of association between the disease description record and the genome, e is the base of the natural logarithm, and z is the value of the association.

The present application provides a method for determining the degree of gene association, capable of determining the record of the description of the disease in the text of the case description. For a plurality of predetermined association databases, record data for an association record(s) in a predetermined association database is determined according to a disease description record and gene identifiers, each corresponding to one of the plurality of genes, each of the predetermined association databases storing association records disease description records and gene identifiers corresponding to at least one of the plurality of genes; the entry data is entered into a predetermined gene association entry matrix to determine an association score of the disease description entry with a corresponding each of the plurality of genes in the predetermined association database; and the degree of association between the disease description entry and each of the plurality of genes is determined according to the association scores of the disease description entry with the gene identifiers corresponding to the plurality of genes from the plurality of predetermined association databases. In the present application, as a result, record data is determined for the target association records of the disease description record and a plurality of genes in each given association database, and the record data is entered into a given record-matrix of the gene association, the indicator of the association of the disease description record with each of the plurality of genes from the given database is determined. association data and the degree of association between a disease description record and a plurality of genes can be quickly obtained.

Alternatively, as shown in FIG. 3, the learning process for a given gene association matrix record, provided in some embodiments of the present application, may include the steps of:

S10: input as training information record data for target association records containing a plurality of disease description records and gene identifiers of a plurality of genes from a plurality of predetermined gene association information databases into the current gene association training record matrix, and determining association predictive indicators corresponding to the training information and a plurality of genes in the given gene association information databases, respectively.

Embodiments of the present application may initialize the gene association matrix training record at the start of training. Step S10 is similar to step S200. Embodiments of the present application may sequentially input record data for association target records of a plurality of disease description records and gene identifiers corresponding to a plurality of genes from a plurality of predetermined gene association information databases into the current training record gene association matrix to obtain a predicted association score result from the current training record-matrix of gene association. Likewise, a certain number of predictable indicators of association of a disease description record with any one of a plurality of genes is the same as the number of a plurality of given association databases.

S20: Predictable degrees of association between the plurality of disease description records and the plurality of genes are determined according to certain predictive rates of association.

Alternatively, in embodiments of the present application, for any gene, association scores corresponding to a disease description record and a gene identifier corresponding to a gene from a plurality of predetermined gene association information databases are summed to obtain an association value of the disease description record and the gene.

Alternatively, in embodiments of the present application, for any gene, association scores corresponding to a disease description record and a gene identifier corresponding to a gene from a plurality of predetermined gene association information databases are weighted to obtain an association value between the disease description record and the gene.

In embodiments of the present application, the degree of association between a disease description record and a gene may be determined according to an association value.

S30: determine the errors between the predicted degrees of association and the actual degrees of association of the plurality of disease description records and the plurality of genes.

The actual degrees of association may be degrees of association between a plurality of disease description records and a plurality of genes as determined by those skilled in the art. In embodiments of the present application, whether the training of the current gene association matrix training record has been completed can be determined according to the errors between the actual association rates and the predicted association rates.

S40: determining if the errors are below a predetermined error threshold; if yes, then step S50 is performed, or otherwise, step S60 is performed.

The given error threshold can be determined in the embodiments of the present application in accordance with actual needs, which is not an additional limitation here.

S50: It is determined that the current training gene association matrix record is a gene association matrix record.

S60: Adjusting the matrix parameters in the current training record gene association matrix by a descending gradient algorithm using the adjusted training record gene association matrix as the current training record gene association matrix, and returning to step S10.

Embodiments of the present application adjust the matrix parameters through a descending gradient algorithm and can check and regenerate the training record gene association matrix after adjusting the matrix parameters so that using the final training record gene association matrix, more accurate association scores can be obtained, further reducing errors between actual association rates and predicted association rates.

It should be understood that embodiments of the invention may also adjust the weight of each given gene association information database by a downward gradient method when it is determined that the errors are not below a given error threshold. By adjusting the weight of each given gene association information database, errors between actual association rates and predicted association rates can be reduced.

Alternatively, based on the method shown in FIG. 1 in FIG. 4 shows another method for determining the degree of gene association provided in some embodiments of the present application. After step S300, the method may further comprise the steps of:

S400: The plurality of genes are sorted in descending order of the degree of association between the disease description record and the plurality of genes.

It should be understood that the higher the degree of association between the gene and the disease description record, the closer the clinical phenotype of the disease described by the disease description record to the phenotype of the gene. Sorting the set of genes after determining the degrees of association between the disease description record and the set of genes is useful for the appropriate technicians to identify and screen out pathogenic disease genes corresponding to the disease description record.

In accordance with the above-mentioned embodiments of the method, in some embodiments of the implementation of the present application, a device for determining the degree of gene association is additionally provided. The structure of the device is shown in Fig. 5 and may include: a disease description record determination unit 100, an association index determination unit 200, and an association degree determination unit 300.

The disease description entry determination unit 100 is used to determine the disease description entry in the case description text.

The text of the case description may be a record of medical activities such as examination by the medical staff and diagnosis of the patient's illnesses. Alternatively, the case description text may be sentences or multi-word paragraphs. The disease description record may be a record describing clinical information about a patient's disease phenotype.

Alternatively, the disease description record may be an HPO record or a hereditary disease record. HPO (human phenotype ontology, ontology of the human phenotype) is a set of standard vocabulary describing pathological phenotypes caused by human diseases. The standard dictionary set contains numerous HPO entries. It is understood that the HPO entries may be Chinese words or words in other languages such as English. An inherited disease entry can be a trade name and/or an abbreviation and/or one of the other user-defined disease names that have genes as the main cause of diseases. It should be noted that user-defined names may be industry colloquial expressions for diseases.

For example, the professional name for the disease "bronchial asthma" can be shortened to "asthma" and is often expressed as a colloquial expression for "wheezing". It should be understood that due to the continuous deepening of research on hereditary diseases in the field of research and the different actual needs in different fields of application, industry personnel can create their records of hereditary diseases in accordance with the needs of the study or the actual needs of the application.

Alternatively, the disease description determination unit 100 may segment the case description text and then extract structured disease description records from the unstructured case description text via natural language processing (NLP) named entity recognition (NER) extraction. Embodiments of the present application can accurately extract a disease description entry from a case description text using said named entry extraction technology.

It should be understood that the disease description record determining unit 100 may also extract the disease description record from the case description text by setting disease description keywords in advance.

The association indicator determining unit 200 is used for a plurality of predetermined association databases: determining the record data for the target association record(s) in the predetermined association database between the disease description record and the gene identifiers of each of the corresponding one of the plurality of genes, each of the predetermined databases the association stores association records of the disease description record and gene identifiers corresponding to at least one of the plurality of genes; and entering the entry data into a predetermined gene association entry matrix to determine an association score of the disease description entry with a corresponding each of the plurality of genes from the predetermined association database.

The predetermined association database may be a database created based on an industry-wide public disease database to associate mapped records with gene identifiers. Alternatively, industry-wide disease databases may contain: OMIM (disease-related gene) database, ClinVar (genetic variation) database, HGMD (human gene mutation) database. genes), the HPO (standard terms of human phenotype) database, and the Orphanet (rare disease-related) database. Alternatively, a given association database may store association records that represent an association relationship between disease description records and gene identifiers.

Alternatively, the association indicator determining unit 200 may determine the entry data for the target association entry(s) in the predetermined association database according to the disease description entry and the gene identifiers corresponding to the plurality of genes. Alternatively, the record data for the target association record(s) may be the number of association records containing the disease description record and any of a plurality of genes from a given association database, or the number of association records including the gene identifier of any of a plurality of genes from a given database. associations.

Alternatively, the association indicator determination unit 200 is specifically used for any gene of the plurality of genes, determines the entry data of the target association entry(s) of the disease description entry and the gene identifier corresponding to the gene in the predetermined association database, respectively. The record data for the target association record(s) comprises a first number of association records containing the gene gene identifier and a disease description record, a second number of association records containing a disease description record, a third number of association records containing the gene gene identifier, and the total number of association records. in the given association database.

It should be understood that the target association record(s) may be an association record(s) in a given association database, and may be an association record(s) containing a gene identifier of a gene and a disease description record in a given association database, wherein the record The association(s) contains a disease description record in a given association database, or an association record(s) containing the gene identifier of the gene.

Alternatively, a given gene association matrix record may look like this:

,

,

where M is the indicator of the association of the disease description record with the gene identifier corresponding to the gene from the given gene information database; T ₁ is the first number, T ₂ is the second number, T ₃ is the third number, and T _S is the total number.

The association score determination unit 200 is used to, by inputting the entry data for the determined target association entry(s) into the predetermined gene association entry matrix, output the predetermined gene association matrix entry by calculation. It should be noted that the number of indicators of association of a particular disease description record with any of the plurality of genes is the same as the number of given association databases.

The association degree determining unit 300 is used to determine the degree of association between the disease description entry and each of the plurality of genes corresponding to the association scores of the disease description record with gene identifiers corresponding to the plurality of genes from the plurality of predetermined association databases.

Alternatively, the association degree determination unit 300 may be used for any gene: the association scores corresponding to the description entry and the gene identifier corresponding to the gene from a plurality of predetermined gene association information databases are summed to obtain the association value of the disease description entry and the gene.

Alternatively, the association degree determination block 300 comprises an association value obtaining sub-block and an association degree determination sub-block.

The association value derivation sub-unit is used for any gene, weighting the association scores of a disease description record with a gene identifier corresponding to a gene from a plurality of predetermined gene association information databases to obtain an association value between the disease description record and the gene.

In embodiments of the present application, a weight may be set for each given gene association information database in advance so that after weighting processing, a more reliable association value between the disease description record and the gene can be obtained.

The association degree determination sub-unit is used to determine the degree of association between the disease description entry and the gene corresponding to the association value.

, где L - степень ассоциации между записью описания болезни и геном, e – основание натурального логарифма и z - значение ассоциации.Alternatively, the association degree determination subunit is specifically used to determine the degree of association between a disease description record and a gene according to

, where L is the degree of association between the disease description record and the genome, e is the base of the natural logarithm, and z is the value of the association.

In the present application, a device for determining the degree of gene association is provided, which is configured to determine the record of the description of the disease in the text of the case description. For a plurality of given association databases, record data for the target association record(s) in a given association database is determined according to a disease description record and gene identifiers, each corresponding to one of the plurality of genes, each of the given association databases storing records association of the disease description record and gene identifiers corresponding to at least one of the plurality of genes; the entry data is entered into a predetermined gene association entry matrix to determine an association score of the disease description entry with each of the plurality of genes in the predetermined association database; and determining the degree of association between the disease description entry and each of the plurality of genes according to the association scores of the disease description entry with the gene identifiers corresponding to the plurality of genes from the plurality of predetermined association databases. In the present application, by determining the record data for the target association record(s) of the disease description record and a plurality of genes from each given association database and entering the record data into a given gene association record matrix, the association score of the disease description record with each of the plurality of genes is determined. in a given association database, and the degree of association between a disease description record and a plurality of genes can be quickly obtained.

Alternatively, the gene association degree determination device may further comprise: a predictive association indicator determination unit, a predicted association degree determination unit, an error determination unit, an error comparison unit, a gene association matrix record determination unit, and a matrix parameter adjustment unit.

The association predictive score determining unit is used to input, as training information, record data for target association records containing a plurality of disease description records and gene identifiers of a plurality of genes from a plurality of predetermined gene association information databases into the current gene association training record matrix, and determine predictable indicators. associations corresponding to the training information and the set of genes in the given gene association information databases, respectively.

The predictive degree of association determination unit is used to determine the predicted degree of association between the plurality of disease description records and the plurality of genes in accordance with the determined association scores.

Alternatively, the predicted degree of association determination unit can be used for any gene: the association scores corresponding to the disease description record and the gene identifier corresponding to the gene from a plurality of predetermined gene association information databases are summed to obtain the association value of the disease description record and the gene.

Alternatively, the predicted degree of association determination unit can be used for any gene: the association scores corresponding to the disease description record and the gene identifier corresponding to the gene from a plurality of predetermined gene association information databases are weighted to obtain the association value of the disease description record and the gene.

The predicted association degree determining unit may be used to determine the degree of association between the disease description record and the gene according to the association value.

An error determination block is used to determine an error between the predicted degrees of association and the actual degrees of association of the plurality of disease description records and the plurality of genes.

The actual degrees of association may be the degrees of association between a plurality of disease description records and a plurality of genes as determined by those skilled in the art.

The error compare block is used to determine if the error is below a predetermined error threshold, if so, the gene association matrix record determination block is called, or otherwise, the matrix parameter adjustment block is called.

The gene association matrix record determination unit is used to determine the current training gene association matrix record as a gene association matrix record.

The matrix parameter adjuster is used to adjust the matrix parameters in the current training record gene association matrix by a downward gradient algorithm, using the adjusted training record gene association matrix as the current training record gene association matrix and calling the predicted association score determination block 200 .

Alternatively, based on the device shown in FIG. 5 in FIG. 6 shows another device for determining the degree of gene association provided in some embodiments of the present application. The device may further comprise a gene sorting unit 400 .

The gene sorter 400 may be used to sort the plurality of genes in descending order of the degree of association between the disease description record and the plurality of genes, after the degree of association determiner 300 determines the degree of association between the disease description record and the plurality of genes, respectively.

The device for determining the degree of gene association contains a processor and a memory. The disease description record determination unit 100, the association index determination unit 200, and the association degree determination unit 300 are all stored in memory as program blocks. The above program blocks are executed by the processor to perform the respective functions.

The processor contains a core that calls the appropriate program blocks from memory. One or more cores may be provided that determine record data for target association records containing a disease description record and a plurality of genes in each given association database, enter the record data into a given gene association record matrix, and determine an association score of the disease description record with the corresponding each of the plurality of genes from a given association database, and by adjusting the basic parameters, the degree of association between the disease description record and the plurality of genes can be quickly obtained.

Some embodiments of the present application provide a storage medium that stores a program that, when executed by a processor, performs a method for determining the degree of gene association.

Some embodiments of the present application provide a processor for executing a program that, when executed, implements a method for determining the degree of gene association.

Some embodiments of the present application represent an electronic device containing at least one processor, at least one memory associated with the processor, and a bus; wherein the processor communicates with the memory via a bus; the processor is used to call program instructions from the memory to carry out the above method for determining the degree of gene association. The electronic device here may be a server, a personal computer, a personal computer assistant or a mobile phone, etc.

The present application further provides a computer program product that, when executed on an electronic device, is capable of executing a program that initializes the method steps for determining the degree of gene association.

The present application has been described with reference to flowcharts and/or flowcharts of methods, devices, electronic devices (systems), and computer program products corresponding to embodiments of the present application. It should be understood that each process and/or block in the flowcharts and/or flowcharts and in combinations of processes and/or blocks in the flowcharts and/or flowcharts may be executed by computer instructions. programs. Computer program instructions may be provided to the processor of a general purpose computer, a special purpose computer, an embedded processor, or other programmable device forming such a machine, such that instructions executed by the processor of the computer or other programmable data processing device produce a means that performs the functions specified in one or more threads. flowcharts and/or in one or more blocks of flowcharts.

In a typical configuration, an electronic device contains one or more processors (CPUs), memory, and a bus. The electronic device may further comprise an input/output interface, a network interface, or the like.

The memory may include non-persistent memory on a computer-readable medium, random access memory (RAM) and/or non-volatile memory such as read only memory (ROM) or flash RAM. The memory contains at least one memory chip. Memory is an example of a computer-readable medium.

Computer-readable media includes persistent and non-permanent media, removable and non-removable media, and information storage may be implemented in any method or technology. The information may be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), others. types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD -ROM, digital versatile disc (DVD) or other optical storage device, magnetic cassettes, magnetic tape, magnetic disk storage device or other magnetic storage devices or any other durable storage media that can be used to store information, which can be accessed by computer devices oistva. As defined herein, computer-readable media does not include transmittable media such as modulated data signals and carrier waves.

Additionally, the terms "comprises", "comprises", or any other variations thereof are intended to cover a non-exclusive combination such that a process, method, product, or apparatus containing a number of factors may contain not only those factors, but also other factors not expressly listed. , or factors intrinsic to that process, method, product, or device. Without limitation, a factor defined by the words "comprises..." does not preclude the existence of other such factors in a process, method, product, or device containing such a factor.

Those skilled in the art will understand that, embodiments of the present application may be presented as a method, system, or computer program product. Therefore, embodiments of the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment containing both hardware and software elements. In addition, the present application may take the form of one or more computer program products containing computer executable codes that may be implemented on a computer executable storage medium (including, but not limited to, discs, CD-ROMs, optical discs, etc. ).

The above are only examples of the present application and are not used to limit the present application. For those skilled in the art, the present application may have various modifications and changes. Any modifications, equivalent substitutions or improvements made within the scope of the notation and principle of this application shall be included within the protection scope of the claims of this application.

Claims (46)

1. A method for determining the degree of association between a disease description record and a genome, characterized in that it contains the steps at which: determining, by means of a disease description record determination unit, a disease description record in the case description text; for each of the set of given association databases: determining, by means of an association indicator determining unit, record data for the target association record(s) in the given association database in accordance with the disease description record and gene identifiers, each of which corresponds to one of the plurality of genes, each of the given association databases storing association records of the disease description record and gene identifiers corresponding to at least one of the plurality of genes; inputting, by means of an association score determining unit, the record data into a predetermined gene association record matrix to determine the association score of the disease description record with a corresponding each of the plurality of genes from the predetermined association database; And determining, by the association degree determining unit, the degree of association between the disease description record and each of the plurality of genes according to the association scores of the disease description record with the gene identifiers corresponding to the plurality of genes from the plurality of predetermined association databases. 2. The method according to claim 1, characterized in that determining the entry data for the target association entry(s) in a given association database, the corresponding disease description entry and gene identifiers, each of which corresponds to one of the plurality of genes, comprises the steps of which: for any gene from a plurality of genes, record data is determined for the target association record(s) of the disease description record and the gene identifier corresponding to the gene from the given association database, respectively, and the record data for the target association record(s) contains the first number of association records containing both the gene identifier of the gene and the disease description record, a second number of association records containing the disease description record, a third number of association records containing the gene gene identifier, and the total number of association records in the given association database. 3. The method according to claim 2, characterized in that the given record-matrix of the gene association has the form:

, where M is the association score of the disease description record with the gene identifier corresponding to the gene in the given association database; T ₁ - the first number, T ₂ - the second number, T ₃ - the third number; T _S - total. 4. The method according to any one of paragraphs. 1-3, characterized in that the degree of determination of the association between the disease description record and each of the plurality of genes corresponding to the indicators of the association of the disease description record with gene identifiers corresponding to the plurality of genes from the plurality of given association databases contains, for any gene from the plurality of genes, steps, where: weighting association scores of the disease description record with a gene identifier corresponding to a gene from a plurality of predetermined association databases to obtain an association value between the disease description record and the gene; And determining the degree of association between the disease description record and the gene according to the association value. 5. The method according to claim 4, characterized in that determining the degree of association between the disease description record and the gene in accordance with the association value comprises the steps of: determine the degree of association between a disease description record and a gene according to

, where L is the degree of association between the disease description record and the genome, e is the base of the natural logarithm, and z is the value of the association. 6. The method according to any one of paragraphs. 1-5, characterized in that the method further comprises the steps of: entering as training information the record data for the target association record(s) containing a plurality of disease description records and gene identifiers of a plurality of genes from a plurality of given association databases into the current gene association training record matrix, and determining association predictive indicators corresponding to the training information and a plurality of genes from predetermined gene association information databases, respectively; determining a predictable degree of association between the plurality of disease description records and the plurality of genes in accordance with certain predictive association scores; determining errors between the predicted degree of association and the actual degree of association of the plurality of disease description records and the plurality of genes; determining if the errors are below a predetermined error threshold; if so, deciding that the current training gene association matrix record is a gene association matrix record; And otherwise, adjust the matrix parameters in the current training record-gene association matrix by a descending gradient algorithm, using the adjusted training record-gene association matrix as the current training record-gene association matrix, and return to the input step as training information, write the data of the target association records containing a plurality of disease description records and gene identifiers of a plurality of genes from a plurality of predetermined association databases into the current gene association training record matrix and determine predictive association scores corresponding to the training information and the plurality of genes from the predetermined association databases, respectively. 7. A device for determining the degree of association between a disease description record and a gene, characterized by the content of a disease description record determination unit, an association indicator determination unit, and an association degree determination unit; wherein the disease description entry definition block is used to determine the disease description entry in the case description text; the association indicator determining unit is used for each of the plurality of predetermined association databases to determine the entry data for the target association entry(s) in the predetermined association database in accordance with the disease description entry and gene identifiers, each corresponding to one of the plurality of genes, wherein each of the given association databases stores association records of the disease description record and gene identifiers corresponding to at least one of the plurality of genes; and entering the entry data into a predetermined gene association entry matrix to determine an association score of the disease description entry with a corresponding each of the plurality of genes from the predetermined association database; And the association degree determining unit is used to determine the degree of association between the disease description entry and each of the plurality of genes according to the association scores of the disease description record and the gene identifiers corresponding to the plurality of genes from the plurality of predetermined association databases. 8. The device according to claim 7, characterized in that the association indicator determination unit is specifically used for any gene of the plurality of genes, respectively determines the entry data for the target association entry(s) containing the disease description entry and the gene identifier corresponding to the gene from the given association databases, where the record data for the target association record(s) comprises: a first number of association records containing both the gene identifier of the gene and a disease description record, a second number of association records containing a disease description record, a third number of association records containing the gene ID of the gene, and the total number of association records in the given association database. 9. The device according to claim 8, characterized in that the given record-matrix of the gene association has the form:

, where M is the association score of a disease description record with a gene identifier corresponding to a gene from a given association database; T ₁ is the first number, T ₂ is the second number, T ₃ is the third number, and T _S is the total number. 10. The device according to any one of paragraphs. 7-9, characterized in that the association degree determination block comprises an association value obtaining sub-block and an association degree determination sub-block, wherein the association value obtaining sub-unit is used for any gene of the plurality of genes to weight association scores of the disease description record with the gene identifier corresponding to the gene from the plurality of predetermined association databases to obtain an association value between the disease description record and the gene; the association degree determination sub-unit is used to determine the degree of association between the disease description record and the gene according to the association value. 11. The device according to claim 10, characterized in that the degree of association determination sub-unit is used to determine the degree of association between the disease description record and the gene in accordance with

, where L is the degree of association between the disease description record and the genome, e is the base of the natural logarithm, and z is the value of the association. 12. The device according to any one of paragraphs. 7-11, characterized in that it additionally contains: a block for determining the predicted indicator of association, a block for determining the predicted degree of association, an error determination block, an error comparison block, a gene association matrix record determination block, and a matrix parameter correction block, wherein the association predictor determination unit is used to input, as training information, the record data for the target association records containing the plurality of disease description records and the gene identifiers of the plurality of genes from the plurality of predetermined association databases into the current training gene association matrix record, and determine the predicted association scores corresponding to the training information and a plurality of genes from the predetermined association databases, respectively; a predictive degree of association determination unit is used to determine a predicted degree of association between the plurality of disease description records and the plurality of genes in accordance with certain predictive association rates; an error determination unit is used to determine an error between a predicted degree of association and an actual degree of association of the plurality of disease description records and the plurality of genes; an error compare block is used to determine if the errors are below a predetermined error threshold, if so, call a gene association matrix record determination block, or otherwise call a matrix parameter adjustment block; a gene association matrix entry determining unit is used to determine the current training gene association matrix entry as the gene association matrix entry; the matrix parameter adjustment block is used to adjust the matrix parameters in the current gene association training record matrix by a descending gradient algorithm using the adjusted gene association training record matrix as the current gene association training record matrix, and to call the predicted association indicator determination block. 13. A storage medium for determining the degree of association between a disease description record and a gene, characterized in that the program is stored on the storage medium and, when executed by the processor, performs a method for determining the degree of gene association in accordance with any one of paragraphs. 1-6. 14. A processor for determining the degree of association between a disease description record and a gene, characterized in that the processor is used to execute a program, wherein the program, when executed, performs the method for determining the degree of gene association in accordance with any one of paragraphs. 1-6. 15. An electronic device for determining the degree of association between a disease description record and a gene, characterized in that it contains at least one processor, at least one memory associated with the processor, and a bus; wherein the processor communicates with the memory via a bus, and the processor is used to call program instructions from the memory to perform the gene association degree determination method according to any one of claims. 1-6.

Images