Computer implemented predisposition prediction in a genetics platform
A method, software, database and system for attribute partner identification and social network based attribute analysis are presented in which attribute profiles associated with individuals can be compared and potential partners identified. Connections can be formed within social networks based on analysis of genetic and non-genetic data. Degrees of attribute separation (genetic and non-genetic) can be utilized to analyze relationships and to identify individuals who might benefit from being connected.
1. A computer-implemented method comprising:
obtaining, by a genetics platform, self-reported user data for a particular user, wherein the genetics platform includes software executing on one or more computing devices;
applying, by the genetics platform, heuristic rules to the self-reported user data to predict that the particular user exhibits a particular phenotype;
based on genetic data and phenotypic data of a plurality of individuals, determining, by the genetics platform and for the particular phenotype, an attribute-positive population of the individuals and an attribute-negative population of the individuals;
based on the genetic data and the phenotypic data of the attribute-positive population and the attribute-negative population, determining, by the genetics platform, a set of genetic attributes that are statistically associated with the particular phenotype, wherein determining the set of genetic attributes that are statistically associated with the particular phenotype comprises training a linear-regression-based or logistic-regression-based model on the genetic data and the phenotypic data to predict predispositions for the particular phenotype;
determining, by the genetics platform, a predisposition for the particular user to exhibit the particular phenotype, wherein determining the predisposition for the particular user to exhibit the particular phenotype comprises applying the linear-regression-based or logistic-regression-based model to particular genetic data or particular phenotypic data of the particular user; and
providing, by the genetics platform and to a client device, an indication of the predisposition.
2. The computer-implemented method of claim 1 , wherein the genetic attributes are non-monogenetic.
3. The computer-implemented method of claim 1 , wherein the genetic attributes are single nucleotide polymorphisms.
4. The computer-implemented method of claim 1 , wherein the genetic attributes are selected from an entire genome.
5. The computer-implemented method of claim 1 , wherein individuals in the attribute-positive population exhibit the particular phenotype, and wherein individuals in the attribute-negative population do not exhibit the particular phenotype.
6. The computer-implemented method of claim 1 , wherein determining the set of genetic attributes that are statistically associated with the particular phenotype comprises:
determining statistical associations between combinations of the genetic attributes and the particular phenotype; and
selecting the combinations of the genetic attributes having at least a minimum statistical association with the particular phenotype.
7. The computer-implemented method of claim 1 , wherein determining the set of genetic attributes that are statistically associated with the particular phenotype comprises:
determining statistical associations between combinations of the genetic attributes and the particular phenotype; and
filtering, from the set of genetic attributes, the combinations of the genetic attributes having less than a minimum statistical association with the particular phenotype.
8. The computer-implemented method of claim 1 , wherein determining the predisposition for the particular user to exhibit the particular phenotype comprises determining that particular genetic data of the particular user exhibit the set of genetic attributes that are statistically associated with the particular phenotype.
9. The computer-implemented method of claim 1 , wherein the genetics platform also determines sets of phenotypic attributes, physical attributes, behavior attributes, or environmental attributes that are statistically associated with the particular phenotype, and wherein determining the predisposition for the particular user to exhibit the particular phenotype comprises determining that data relating to the particular user exhibits at least some of the sets of phenotypic attributes, physical attributes, behavior attributes, or environmental attributes that are statistically associated with the particular phenotype.
10. The computer-implemented method of claim 1 , wherein the genetics platform also determines ethnicity data or sex-type data that are statistically associated with the particular phenotype, and wherein determining the predisposition for the particular user to exhibit the particular phenotype comprises determining that data relating to the particular user exhibits at least some of the ethnicity data or sex-type data that are statistically associated with the particular phenotype.
11. The computer-implemented method of claim 1 , wherein the genetics platform also determines weight data that are statistically associated with the particular phenotype, and wherein determining the predisposition for the particular user to exhibit the particular phenotype comprises determining that data relating to the particular user exhibits at least some of the weight data that are statistically associated with the particular phenotype.
12. The computer-implemented method of claim 1 , wherein determining the predisposition for the particular user to exhibit the particular phenotype is also based on timestamped historical genetic attributes, timestamped historical physical attributes, or timestamped historical behavioral attributes of the particular user.
13. The computer-implemented method of claim 1 , wherein attributes of the self-reported user data are timestamped, and wherein determining the predisposition for the particular user to exhibit the particular phenotype comprises predicting onset or progression of disease based on the attributes as timestamped.
14. The computer-implemented method of claim 1 , wherein determining the predisposition for the particular user to exhibit the particular phenotype comprises calculating a risk associated with the particular user exhibiting the particular phenotype.
15. The computer-implemented method of claim 1 , wherein the genetic data and the phenotypic data of the plurality of individuals are anonymously linked to identities of these individuals using non-descriptive alphanumeric identifiers.
16. The computer-implemented method of claim 1 , wherein reception of the indication of the predisposition causes the client device to display, in a graphical representation, changes to the predisposition over time.
17. The computer-implemented method of claim 16 , further comprising:
receiving, from the client device, attribute modifications relating to the particular user;
based on the attribute modifications, determining an updated predisposition for the particular user to exhibit the particular phenotype; and
providing, to the client device, an indication of the updated predisposition.
18. A non-transitory computer-readable medium storing program instructions that, when executed by a processor, cause a computing system to perform operations comprising:
obtaining self-reported user data for a particular user;
applying heuristic rules to the self-reported user data to predict that the particular user exhibits a particular phenotype;
based on genetic data and phenotypic data of a plurality of individuals, determining, for the particular phenotype, an attribute-positive population of the individuals and an attribute-negative population of the individuals;
based on the genetic data and the phenotypic data of the attribute-positive population and the attribute-negative population, determining a set of genetic attributes that are statistically associated with the particular phenotype, wherein determining the set of genetic attributes that are statistically associated with the particular phenotype comprises training a linear-regression-based or logistic-regression-based model on the genetic data and the phenotypic data to predict predispositions for the particular phenotype;
determining a predisposition for the particular user to exhibit the particular phenotype, wherein determining the predisposition for the particular user to exhibit the particular phenotype comprises applying the linear-regression-based or logistic-regression-based model to particular genetic data or particular phenotypic data of the particular user; and
providing, to a client device, an indication of the predisposition.
19. The non-transitory computer-readable medium of claim 18 , wherein determining the set of genetic attributes that are statistically associated with the particular phenotype comprises:
determining statistical associations between combinations of the genetic attributes and the particular phenotype; and
selecting the combinations of the genetic attributes having at least a minimum statistical association with the particular phenotype.
20. A computing system comprising:
one or more processors;
memory; and
program instructions stored in the memory that, when executed by the one or more processors, cause the computing system to perform operations comprising:
obtaining self-reported user data for a particular user;
applying heuristic rules to the self-reported user data to predict that the particular user exhibits a particular phenotype;
based on genetic data and phenotypic data of a plurality of individuals, determining, for the particular phenotype, an attribute-positive population of the individuals and an attribute-negative population of the individuals;
based on the genetic data and the phenotypic data of the attribute-positive population and the attribute-negative population, determining a set of genetic attributes that are statistically associated with the particular phenotype, wherein determining the set of genetic attributes that are statistically associated with the particular phenotype comprises training a linear-regression-based or logistic-regression-based model on the genetic data and the phenotypic data to predict predispositions for the particular phenotype;
determining a predisposition for the particular user to exhibit the particular phenotype, wherein determining the predisposition for the particular user to exhibit the particular phenotype comprises applying the linear-regression-based or logistic-regression-based model to particular genetic data or particular phenotypic data of the particular user; and
providing, to a client device, an indication of the predisposition.