Attribute combination discovery for predisposition determination of health conditions
A method, software, database, and system in which a query attribute is used as the basis for accessing stored attribute combinations and their frequencies of occurrence for individuals; and tabulating, based on frequencies of occurrence, those attribute combinations that are most likely to co-occur with the query attribute.
1. A computer-implemented method comprising:
receiving, by a server device and from a client device, a query regarding body mass index (BMI) with respect to an individual, wherein genomic data of the individual is stored in one or more databases;
based on a set of genomic markers that are associated with BMI in the genomic data of the individual, determining, by the server device, a predicted BMI of the individual;
comparing, by the server device, the predicted BMI to BMI of a population of individuals, wherein genomic data and behavioral data of the population of individuals are stored in the one or more databases;
based on comparing the predicted BMI to BMI of the population of individuals, determining, by the server device, a BMI predisposition for the individual; and
providing, by the server device and to the client device, the BMI predisposition for the individual in response to the query.
2. The computer-implemented method of claim 1 , further comprising:
obtaining a set of behavioral attributes having strengths of association above a threshold value with BMI, wherein the set of the behavioral attributes are stored in the one or more databases; and
providing, by the server device and to the client device, the set of the behavioral attributes.
3. The computer-implemented method of claim 1 , wherein
providing the set of the behavioral attributes comprises arranging the set of the behavioral attributes in a rank-ordered listing based on their strengths of association with BMI.
4. The computer-implemented method of claim 1 , wherein determining the predicted BMI of the individual is based on a trained linear regression model of the genomic markers.
5. The computer-implemented method of claim 1 , wherein age, sex, or ethnicity of the individual are also stored in the one or more databases, and wherein the predicted BMI of the individual is also based on the age, sex, or ethnicity of the individual.
6. The computer-implemented method of claim 5 , wherein the BMI predisposition for the individual is also based on the age, sex, or ethnicity of the individual.
7. The computer-implemented method of claim 1 , wherein comparing the predicted BMI to BMI of the population of individuals comprises determining a degree of genomic overlap between the genomic markers of the individual and those of the population of individuals, and wherein the BMI predisposition for the individual is based on the degree of genomic overlap.
8. The computer-implemented method of claim 1 , wherein the BMI predisposition for the individual corresponds to a genetic predisposition to exhibit certain BMI values.
9. The computer-implemented method of claim 1 , further comprising:
providing, by the server device and to the client device, the predicted BMI of the individual also in response to the query.
10. The computer-implemented method of claim 1 , wherein the genomic markers are single nucleotide polymorphisms (SNPs).
11. The computer-implemented method of claim 1 , wherein the genomic markers are selected from an entire genome.
12. The computer-implemented method of claim 1 , wherein the genomic markers are single nucleotide polymorphisms (SNPs) and are selected from an entire genome.
13. A computer-implemented method comprising:
based on a set of genomic markers that are associated with body mass index (BMI) in genomic data of a population of individuals, training a model to predict BMI values from the genomic markers;
based on behavioral attributes of the population of individuals, determining a set of the behavioral attributes having strengths of association above a threshold value with BMI;
receiving, from a client device, a query regarding BMI with respect to an individual;
based on a set of genomic markers in genomic data of the individual, determining a predicted BMI of the individual;
comparing the predicted BMI to BMI of the population of individuals;
based on comparing the predicted BMI to BMI of the population of individuals, determining a BMI predisposition for the individual; and
providing, to the client device, the BMI predisposition for the individual and the set of the behavioral attributes in response to the query.
14. The computer-implemented method of claim 13 , wherein providing the set of the behavioral attributes comprises arranging the set of the behavioral attributes in a rank-ordered listing based on their strengths of association with BMI.
15. The computer-implemented method of claim 13 , wherein the model is a linear regression model of the genomic markers.
16. The computer-implemented method of claim 13 , wherein the predicted BMI of the individual is also based on age, sex, or ethnicity of the individual.
17. The computer-implemented method of claim 16 , wherein the BMI predisposition for the individual is also based on the age, sex, or ethnicity of the individual.
18. The computer-implemented method of claim 13 , wherein comparing the predicted BMI to BMI of the population of individuals comprises determining a degree of genomic overlap between the genomic markers of the individual and those of the population of individuals, and wherein the BMI predisposition for the individual is based on the degree of genomic overlap.
19. The computer-implemented method of claim 13 , wherein the BMI predisposition for the individual corresponds to a genetic predisposition to exhibit certain BMI values.
20. A non-transitory computer-readable medium having stored thereon program instructions that, upon execution by a computing device, cause the computing device to perform operations comprising:
receiving, from a client device, a query regarding body mass index (BMI) with respect to an individual, wherein genomic data of the individual is stored in one or more databases;
based on a set of genomic markers that are associated with BMI in the genomic data of the individual, determining a predicted BMI of the individual;
comparing the predicted BMI to BMI of a population of individuals, wherein genomic data and behavioral data of the population of individuals are stored in the one or more databases;
based on comparing the predicted BMI to BMI of the population of individuals, determining a BMI predisposition for the individual; and
providing, to the client device, the BMI predisposition for the individual in response to the query.