Anvita Gupta has believed since childhood in a destiny to help human health. Today she applies her skills in computing, technology and artificial intelligence to health research to fulfill this personal and professional mission. Anvita is equally passionate about ending the gender divide in tech through her non-profit LITAS for Girls, which seeks to inspire girls to pursue STEM careers by teaching them to code with a purpose. SEEMA asked Anvita to tell us more…
Your work straddles two very different fields—computer science and biology. Which one piqued your interest first?
I was not initially interested in computer science as a career, but I grew up around computers. My parents are both chemical engineers by training, and they were setting up their IT business when I was growing up. What I actually wanted to be was a pediatrician. This was inspired by my own pediatrician, who I really looked up to as a child. I really wanted to make a difference in children’s lives and health.
When did these two diverse interests come together for you?
Around 10th grade, I wanted to do something in biology for my school science fair. But the local labs wouldn’t really take a 15-year-old, and my parents couldn’t drive me to more distant labs everyday. So I started thinking about the computer I’d grown up with and how I can use the tools that were available to me to make a difference in medicine rather than take the traditional route.
What was the first successful idea that propelled you further down this career path?
Going back to this science fair, I started looking at pancreatic cancer. I started delving into the existing standard of care for treating cancer and how scientists find the important targets to go after. What I saw at the time was that scientists see that certain proteins are over-expressed in pancreatic cancer and decide to knock them out. Coming from a very analytical background, I wondered why they would pick these proteins–maybe they’re just signals that show that something has gone wrong in the cell, that the cell has become cancerous. My idea, which I ended up doing for this project and got me interested in bioinformatics, was to look at applying graph theory to this problem. I used it to look at which proteins have the most interactions with other proteins in these cancer cells instead of looking at what’s over-expressed in the cell.
The idea worked—we found some of the Holy-Grail gene targets for cancer therapy, like KRAS and TP53 ,with our approach. That was my first introduction to the world of cancer. It taught me to take a problem that people have been approaching with brute force and look at it more elegantly to find a solution much more quickly. And that’s what I see as the real contribution of computer science to biology.
Can you expand a little bit on that? What does the computer science approach do differently to effect change in medicine or health care?
There are two different ways in which computers are used to change biology. The first type of computer models work with immense amounts of data coming from experiments in genomics or proteomics. Feeding data into a computer helps us to draw correlations to key endpoints like disease outcomes or progression. The other way computers are impacting medicine right now is particularly exciting. We ask how to take data and use it for synthetic biology. How do we create molecules that have properties that we want? Rather just looking through nature for a needle in a haystack, how do you create your own needle? That’s something that I’m particularly excited about.
Can you give an example of what can be created?
We’re having this huge problem with plastics. Can we create an enzyme that can break down plastic waste to degradable materials? Instead of me combing through every bacteria on the planet with that enzyme, is it possible for me to look at all the proteins that already exist and use a computer to combine their different parts to come up with an enzyme that will break down plastic?
In our most recent work, we have just shown how we can edit genes to have specific functions that they didn’t have before. It’s early days, but there’s a lot of progress being made.
Apart from your research work, you have also developed a program called LITAS for Girls. Can you tell us what it does and how it came about?
In my first computer science class, the class started out 50-50 in gender, and by the time the end of the year came around, three fourths of the girls had dropped out of the class. What was really astonishing to me was that the girls who dropped out or were calling themselves non-technical had higher grades than the boys.
I was drawn into computer science by the idea that I could use a computer to make a difference in the world. I started LITAS in 2014, to teach girls computer science using a project-based approach. We started at a middle school to teach the basics of computers. They took to the approach so well that today we are an international non-profit.