Would you like to introduce yourself and what you do?
My name is Lia Petronio, and I’m currently a data visualization designer and developer working at a genomics research institute called Broad Institute. I work on a team of data visualization designers and developers, and we specialize in coming up with ways of visualizing really complex genetic data in order to understand the data and find patterns in it. I focus primarily on translating relationships in data into a visual form and understanding the underlying data structures and math that is required in order to calculate these visual attributes, such as position, shape, color, and density. I’m really interested in coming up with new visualizations instead of what is typically used (like scatter plots or heat maps).
Were you always in data visualization?
Before this, I was very much a fine arts type of person, working with physical material. I was really interested in installations that are multimedia with video. The reason I got into data visualization was due to my interest in technology and art.
It’s pretty interesting that you jumped from fine arts to data visualization. Can you tell me more about that journey?
From a young age, I was really interested in art and drawing. I got a lot of encouragement from my parents, especially my dad, who taught me a lot about drawing and designing. I ended up doing a Bachelor in Fine Arts at Mason Gross, and I had a dual concentration in drawing and design. Eventually, I started getting really interested in creating more environments or immersive experiences, so I started moving towards sculpture and performance, focusing on the absurdities of human-technology relationships. I started making handmade kinetic sculptures, which eventually led to building tower installations with metal and electronics. By the time I was graduating, I was talking with one of my professors, and he recommended this visualization master’s program at Northeastern.
What was it like transitioning from fine arts to the program?
It was a natural transition because it was essentially the type of work I was interested in. At the time, I don’t think I fully understood how big data visualization would become or imagined myself designing interactive visualization interfaces. I just thought, “I’m gonna learn how to code, do my homework, and build robots.” Once I got to school, I started learning about all of the applications of data visualization. I saw that it’s more of understanding large and complex systems of information and trying to figure out how to connect the way that we see things.
I learned just how important a skill design thinking is in adjusting and understanding information. You might receive content in a certain way, but once you start thinking about the relationships or hierarchies in that information and formulating a storyline, the design really helps you to go back and edit the information you have.
Any advice you have or things you learned from that program?
Try to use your educational environment to advance projects that you’re already interested in. Have a bigger plan, not just meeting the requirements. Take the assignment and skills you need to learn and apply that to perhaps a project that does not quite fit the assignment that you’re really passionate about. For example, there were cases when I coded part of a project by my own volition, even if it was not necessary for the assignment. I learned a lot of programming, but not necessarily from my classes. I think that that was one of the highlights of my education—just the collaboration with peers and having resources like the professors to guide your creativity.
Take the assignment and skills you need to learn and apply that to perhaps a project that does not quite fit the assignment that you’re really passionate about.
From graduation and getting your first job in data visualization to now, were there any challenges or differences from now at your current company?
The place that I work now is in some ways like a university since what we do is science and research. I think I am lucky to work here because I do like academia and being able to come up with good solutions. I’m definitely in a space that encourages doing things you’re excited about and following your curiosity, even if it’s risky or time-consuming.
But in my previous position, I was working in a university newsroom where the data was less complicated, and the data visualizations I was making were much simpler. The challenge I faced was having no other specialists in visualization. There was not as much support for thinking about visual storytelling or complex, interesting visualizations. There was definitely this mentality of making data look pretty, even if the data itself was not very good. It was hard for me since I care for using visualization in a strong way that makes sense.
What would you say to people interested in data visualization?
I think you should like math because there will be times you may need to use geometry or linear algebra to translate numbers into visuals. Data visualization is like a mathematical visual bridge, you are thinking mathematically about how you can design for a certain set of data. Though, I do know people design visualizations and information-dense interfaces without being super math-heavy. It depends on if you want to do the development—if you do, then you would definitely need to understand the math.
Are there any projects you’re particularly proud of that you would like to talk about?
I worked on a design that was chosen for the cover of Nature, a leading multidisciplinary science journal. We weren’t sure if the article would be considered for the cover since getting chosen is pretty difficult and uncommon. We had a lot of other designs to prioritize, so I was told I could do some designs if I had time. I was really excited about my ideas, so I spent two full weeks focusing almost the whole day on it. They loved my initial pencil sketch so much that they said they were going to choose the design for the cover.
I worked with my colleague Andrew Tang to come up with this idea called the petal plot. The petal plot is a visualization looking at different cancer cell lines. Each cell line comes from a different tissue and targets different organs. For example, the cancer is in my liver, but it will metastasize to go to my brain. “Metastasize” means it basically starts in one place but affects another place. So the flowers are showing where that cancer will go. The bigger the petal, the more likely it will go to that place. The number of petals also is determined based on a categorical attribute of data. I actually developed a rotatable 3D version of this design, so this is an example of doing tons of development work to come up with this image. It was nice to see this flower field on a gray-silver grid, pushing that juxtaposition of the organic and rigid.
Lia Petronio is a data visualization designer at Broad Institute. She has a passion for visual storytelling and using data visualization to convey information in a useful and powerful way.
Visit http://liapetronio.com/ to learn more about Lia Petronio and her work.