How does one effectively convey data? How do different visualizations affect understanding? What if the readers are experts? Or alternatively, non-experts? These are the questions the Wellesley PGP team is asking. At the moment, tabular views are the most common method for presenting personal genomic data. But while such a method is familiar and pragmatic, it is far from the best approach when it comes to not only readability, but also the display of relationships and trends. As a result, the Wellesley HCI lab has been exploring alternative visualizations that could potentially increase the accessibility of genomic data. Previous to this summer, we developed a set of new visualizations that were reviewed by individuals who boast a strong background in either the biological sciences and/or health care. From their feedback, the PGP team improved the visualizations, and are in the process of launching a new study to determine the understandability of these visualizations to non-experts rather than experts. In particular, we are interested in discovering how the presentation of data affects the average user’s understanding of their personal genomic data.
To create this study, the PGP team spent much of their time building a new website on which to run their study. In addition, careful thought was given to the drafting of each question, for wording is paramount when it comes to user understanding. After two weeks of hard work, we were ready to begin our pilot. During a joint brainstorm with the BostonU iGEM Team, we asked that our users check for confusing wording and functionality. In addition, throughout the rest of the week, the PGP team tracked down other members of the HCI lab, students at Wellesley College, students at the NEGEM Conference, and family members to test and rework our study.
Finally, on June 24, the PGP team successfully launched our first study of the summer through Amazon Mechanical Turk. As the results pour in, we are excited for the next stage!
Exploring Personal Genomics
The Personal Genome Project
Questions about how people make sense of and engage with their personal genomic information, and how comfortable they feel about sharing it in order to advance scientific and biomedical research, are not only of paramount importance for society and policy makers, but also a pressing issue for HCI researchers. This summer, in collaboration with the Personal Genomics Project out of the Harvard Medical School, the Wellesley Human Computer Interaction Lab is researching how to design effective interaction techniques for non-experts with personal genomic information. The Wellesley PGP team is also exploring whether user interface design interventions impact users’ willingness to share their personal genomic data.
The Wellesley PGP team consists of four students, Laura Ascher, Joanna Bi, Claire Schlenker, and Elizabeth Stowell, under the direction of Dr. Orit Shaer. Wellesley PGP has also been collaborating with Claire Cerda, a rising senior, sociology major, and student researcher for Dr. Darakhshan Mir, to brainstorm about privacy concerns related to the sharing of genomic data. In addition, we have also collaborated with Dr. Oded Nov and his team from the Polytechnic Institute of New York University to brainstorm about the direction of our project and to launch the initial phases of our research. They have been a great resource as we have pushed towards deploying our first Amazon Mechanical Turk study, exploring how different visualizations of PGP data affect user’s understanding of their results. This wide-reaching and interdisciplinary team has really enjoyed collaborating together and coming up with unique routes of exploration for this project. Having a diverse group of students and professors has allowed this team to expand the scope of their study while focusing on one of the most central questions in Human Computer Interactions: how to make information accessible and interactive.
The Human Genome Project
The Personal Genomics Project