GenAMap: Genetic Analytics Software

By providing an easy to explore interface and data visualizations, GenAMap has the potential to be used by doctors for personalized medicine.

Overview

GenAMap began as a project by a Carnegie Mellon graduate student to create a genetic analytics software that uses machine learning algorithms to help with the exploration of genome and phenome-wide association mappings. By providing users with an interface that uses multiple coordinate views (overview of analysis first, then zoom and filter, and details on demand), this software can be used by a larger audience than just a technical audience to study diseases. My role for this project is to understand exactly who will be using this software (i.e. doctors, bioinformaticians, computational biologists, or some other group), how this group will use the software, and develop GenAMap into a system that can be accessible for this group.

Interviews

Two types of interview guides were developed to learn from experts in bioinformatics and computational biology, who focus on SNP analysis and RNA sequencing. Semi-structured interviews allowed for the team to gather details about the work the experts are currently conducting, their perspective of the field, and the specific processes they take when conducting their research. We conducted four Think Aloud user studies to provide information about how these experts interact with and learn to use GenAMap.

Findings

One of the most interesting findings was that the workflow taken by bioinformaticians and computational biologists can be highly complex, requiring a constant shift between different softwares, creating an inconvenience when iterating through an analysis. Opportunities to reduce workflow for users became very clear through these interviews.

Additionally, we uncovered many usability issues with the software, and developed methods to allow researchers to better focus on data exploration, compare results with exisiting literature and algorithms, and save specific areas of interest for later analysis.

Proposed diagram of the workflow of bioinformaticians who use GenAMap along with additional software used at each stage.

Future Work

Working with another student to help make informed design decisions, we began by developing prototypes through an iterative design process, where at each stage we conducted heuristic evaluations of the prototype. The software will be updated accordingly, and future user research is recommended.

Screenshots of an interactive prototype created from the user research.


designed and developed by me!