Genome AI Assistant

Inspiration

The are two main problems that we want to fix with our Genome AI Assistant.

More users and developers need to be aware of the opportunity to use their genome information for good. Many people do not know what a genome profile is and what it can tell them.
The genome interpretation model needs feedback data from users so that its accuracy can be improved even further

What it does

Here is how our Genome AI Assistant addresses the two problems described above.

The Genome AI Assistant uses conversation and voice to introduce users to their genome profile in a natural way. The assistant is exploring the genome profile of the user step by step in a fun and interactive game. The assistant is using jokes, asking provocative questions, give points to the user for each correct answer. Users learn a lot in the game and find out how accurately they know themselves.
During the game with the user, we are creating a feedback loop. We ask the user what he thinks about each trait and aptitude before giving them the answer. In this way we can compare what the user thinks with the prediction based on the genome profile. This feedback from the user can be a valuable source of information to further improve the interpretation of the genome data.

How we built it

We have build it using a set of technologies. The front-end is build with Javascript, HTML and CSS. The back-end is developed using Node.js and MondoDB. We used AWS for the voice generation.

Challenges we ran into

The main challenge was to create a good conversational flow and the gamification of the AI assistant. We really tried to create a fun experience for the user so that they can easily explorer their genome profile.

Accomplishments that we're proud of

We are proud of learning how to integrate with the API of Genome Link. Before the hackathon, we were not aware of the awesome opportunities to use genome data in our app. We are also proud of the feedback that we collect which can be used to improve genome interpretations.

What's next for Genome AI Assistant

We want to further develop the product to cover more traits and aptitudes. We would like to also establish a partnership with Genome Link in order to help them collect valuable feedback from users and in this way further improve the accuracy of the interpretation of the genome data.