Trait prediction coordination for genomic application environment
Systems and methods are provided for trait prediction coordination. One embodiment is a method that includes launching a genomics application at a user device, receiving a command to present a partner application within the genomics application, selecting characteristics to predict for an individual, based on an identifier that distinguishes the partner application from other partner applications, operating polygenic models that generate predictions for the characteristics based on genetic records for the individual, acquiring media for the partner application, based on the predictions, and operating the partner application to update a display at the user device with the media, thereby providing the predictions in a format specific to the partner application.
1 . A system comprising:
a user device and an application coordination server communicatively coupled via a network; and
a genomics server configured to operate trained machine-learning polygenic models to make predictions about individuals based on one or more genetic records that report genetic variants for the individuals;
the user device comprising:
a memory configured to store a genomics application comprising an executable file, and partner applications;
a controller configured to launch the genomics application by executing the executable file, receive a selection by a user to present a partner application within the genomics application on a display of the user device, and generate a prediction request from the partner application that includes a target identifier for an individual and an application identifier that distinguishes the partner application from other partner applications; and
an interface configured to transmit the prediction request toward the application coordination server via the network; and
the application coordination server comprising:
an interface configured to receive the prediction request via the network; and
a controller configured to select characteristics to predict for the individual based on the application identifier, direct the genomics server to operate at least one of the trained machine-learning polygenic models to generate the predictions for the characteristics based on the genetic records stored at the genomics server for the individual, receive the predictions from the genomics server without receiving the genetic records, and generate a response to the prediction request that includes the predictions provided to the user device through the interface of the application coordination server;
wherein the controller of the user device is additionally configured to acquire media for presentation within the partner application based on the predictions, and operate the partner application to update the display at the user device with the media, thereby providing the predictions in a format specific to the partner application.
2 . The system of claim 1 wherein:
the genomics server comprises:
an interface configured to receive the target identifier from the application coordination server; and
a controller configured to retrieve the genetic records stored for the individual based on the target identifier, and input the genetic records to the at least one of the trained machine-learning polygenic models to generate the predictions for the characteristics;
wherein the interface of the genomics server is configured to provide the predictions to the application coordination server.
3 . The system of claim 1 wherein:
the media varies by partner application.
4 . The system of claim 1 wherein:
the controller of the user device is further configured to acquire layout information indicating an arrangement of the media at the display.
5 . The system of claim 1 wherein:
the trained machine-learning polygenic models use different combinations of predetermined genetic loci as input.
6 . The system of claim 1 wherein:
the controller of the user device is further configured to acquire the media by:
reporting the application identifier and the predictions to a content management server; and
receiving the media, from the content management server, that is associated with the partner application and the predictions.
7 . The system of claim 6 wherein:
the memory of the user device is further configured to store a Uniform Resource Identifier (URI) for each of the partner applications; and
the controller of the user device is further configured to use the URI corresponding with the partner application to acquire the media from the content management server.
8 . The system of claim 1 wherein:
the controller of the user device is further configured to receive multiple commands to present a partner application within the genomics application, each command referring to a different partner application.
9 . The system of claim 1 wherein:
the memory of the user device is further configured to store icons for the partner applications; and
the controller of the user device is further configured to present the icons after the genomics application is launched, and determine the selection of the partner application by the user based on a selection of an icon corresponding with the partner application.
10 . A method comprising:
storing, at a user device communicatively coupled to an application coordination server via a network, a genomics application comprising an executable file, and partner applications;
launching, at the user device, the genomics application by executing the executable file;
receiving a selection by a user at the user device to present a partner application within the genomics application on a display of the user device;
transmitting, from the user device to the application coordination server via the network, a prediction request from the partner application that includes a target identifier for an individual and an application identifier that distinguishes the partner application from other partner applications;
selecting, at the application coordination server, characteristics to predict for the individual based on the application identifier;
operating, at a genomics server in response to direction from the application coordination server, one or more trained machine-learning polygenic models to generate predictions for the characteristics based on genetic records that report genetic variants for individuals and that are stored at the genomics server for the individual;
providing, via the genomics server, the predictions to the application coordination server without providing the genetic records to the application coordination server;
providing the predictions from the application coordination server to the user device;
acquiring, at the user device, media for presentation within the partner application based on the predictions; and
operating the partner application to update the display at the user device with the media, thereby providing the predictions in a format specific to the partner application.
11 . The method of claim 10 wherein the operating at the genomics server comprises:
receiving the target identifier from the application coordination server;
retrieving the genetic records stored for the individual based on the target identifier; and
inputting the genetic records to the one or more trained machine-learning polygenic models to generate the predictions for the characteristics.
12 . The method of claim 10 wherein:
the media varies by partner application.
13 . The method of claim 10 further comprising:
acquiring layout information indicating an arrangement of the media at the display.
14 . The method of claim 10 wherein:
the trained machine-learning polygenic models use different combinations of predetermined genetic loci as input.
15 . The method of claim 10 wherein:
acquiring the media comprises:
reporting the application identifier and the predictions to a content management server; and
receiving the media, from the content management server, that is associated with the partner application and the predictions.
16 . The method of claim 15 wherein:
the storing further comprises storing a Uniform Resource Identifier (URI) for each of the partner applications; and
the acquiring the media comprises using the URI corresponding with the partner application to acquire the media from the content management server.
17 . The method of claim 10 wherein:
selecting the characteristics to predict is performed by the application coordination server that is located remotely from the user device.
18 . The method of claim 10 wherein:
the storing comprises storing icons for the partner applications; and
the method further comprises:
presenting the icons after the genomics application is launched; and
determining the selection of the partner application by the user based on a selection of an icon corresponding with the partner application.
19 . A non-transitory computer readable medium embodying programmed instructions which, when executed by a processor, are operable for performing a method comprising:
storing, at a user device communicatively coupled to an application coordination server via a network, a genomics application comprising an executable file, and partner applications;
launching, at the user device, the genomics application by executing the executable file;
receiving a selection by a user at the user device to present a partner application within the genomics application on a display of the user device;
transmitting, from the user device to the application coordination server via the network, a prediction request from the partner application that includes a target identifier for an individual and an application identifier that distinguishes the partner application from other partner applications;
selecting, at the application coordination server, characteristics to predict for the individual based on the application identifier;
operating, at a genomics server in response to direction from the application coordination server, one or more trained machine-learning polygenic models to generate predictions for the characteristics based on genetic records that report genetic variants for individuals and that are stored at the genomics server for the individual;
providing, via the genomics server, the predictions to the application coordination server without providing the genetic records to the application coordination server;
providing the predictions from the application coordination server to the user device;
acquiring, at the user device, media for presentation within the partner application based on the predictions; and
operating the partner application to update the display at the user device with the media, thereby providing the predictions in a format specific to the partner application.
20 . The medium of claim 19 wherein the operating at the genomics server comprises:
receiving the target identifier from the application coordination server;
retrieving the genetic records stored for the individual based on the target identifier; and
inputting the genetic records to the one or more trained machine-learning polygenic models to generate the predictions for the characteristics.
21 . The medium of claim 19 wherein:
the media varies by partner application.
22 . The medium of claim 19 wherein the method further comprises:
acquiring layout information indicating an arrangement of the media at the display.
23 . The medium of claim 19 wherein:
the trained machine-learning polygenic models use different combinations of predetermined genetic loci as input.
24 . The medium of claim 19 wherein:
acquiring the media comprises:
reporting the application identifier and the predictions to a content management server; and
receiving the media, from the content management server, that is associated with the partner application and the predictions.
25 . The medium of claim 24 wherein:
the storing further comprises storing a Uniform Resource Identifier (URI) for each of the partner applications; and
the acquiring the media comprises using the URI corresponding with the partner application to acquire the media from the content management server.
26 . The medium of claim 19 wherein:
the storing comprises storing icons for the partner applications; and
the method further comprises:
presenting the icons after the genomics application is launched; and
determining the selection of the partner application by the user based on a selection of an icon corresponding with the partner application.