Helping to Drive Discovery in 23andMe’s Therapeutics Division

Xin Fang’s interest in science started at the dinner table, listening to her parents talking about what they did each day. 

Both her mom and dad are plant scientists. Her mom focuses on genetics, and her father on the environment. They worked long hours, but they would come home enthused. And, sitting with her parents at the table for the family meal, Xin listened as her mom and dad engaged in conversations about the latest discoveries made or problems solved. It fascinated her.

The Wonders of Human Biology

Now a computational biologist in 23andMe’s Therapeutics division, Xin said her parents’ passion for science fueled her own ambition. By high school she knew she wanted to be a scientist, she just wasn’t sure about her path. Moving from China to the United States to attend Johns Hopkins University, Xin studied chemical and biomolecular engineering. It was there that her interest in bioinformatics took root. She was fascinated by the “mystery of the human body and how it works,” she said. Xin was sucked into the fact that data and bioinformatics offered a door to understanding parts of the mystery of human biology.

“I love the fact that we can develop new techniques to understand the mechanisms behind human biology,” Xin said.

Now a little over a year into her work at 23andMe, Xin is deep into collaborating with fellow computational biologists and bench scientists to help find potential drug targets to pursue. Computational biology, data science, and machine learning play a critical role in 23andMe’s drug discovery process. Our scientists have the great fortune of gleaning insights from what is one of the largest genetic and phenotypic datasets in the world.

For Xin, the attraction of coming to work at 23andMe wasn’t just that she’s interested in genetics or that she sees the promise of 23andMe’s data helping to accelerate the development of new treatments, it’s also because drug discovery at 23andMe is driven by data, which allows computational biologists like her to propose new research directions and make major contributions. The size and scale of our dataset enables rapid discoveries.

As part of an occasional series about scientists working at 23andMe, we sat down with Xin to talk about computational biology, 23andMe, and what it’s like being an early career scientist here.

Did you always want to be a scientist?

Pretty early on I did. By high school, I knew I wanted to do something in biochemistry but I wasn’t sure what.

Was your interest in science inspired by a person or an experience?

Definitely it came from my parents. They were both scientists and very passionate about it, and it had a great impact on me. They worked a lot but they were happy about what they were doing. We spent a lot of hours talking about their work over dinner.

You studied chemical and biomolecular engineering at Johns Hopkins, what prompted you to do your doctoral work in bioengineering?

I took a gap year working at Genentech after my undergraduate studies. That and my experience during undergrad helped me realize that I was more interested in bioinformatics and computational modeling.

Consider the human body. It’s this perfect mystery that it works in a way that it does, and we don’t totally understand it. But when you bring in high-throughput data, especially from large genomics datasets or next-generation sequencing, you start to see more clearly different biological mechanisms at play. That’s the most fascinating part. And I love that we can use these new techniques and assays to try and understand it. It’s just awesome.

Why did you focus on studying the gut microbiomes and IBD during your doctoral work at UCSD?

The gut microbiome is a significant part of our body but it hasn’t been well understood before. It interacts with the rest of the human body in a lot of different ways including things like exchanging metabolites and breaking down medicines. And in recent years it has been discovered that the gut microbiome is associated with several diseases and inflammatory bowel disease (IBD) is one of them.  Basically, what you observe is an imbalance in the microbiome where there is a reduction in the beneficial microbes but an increase in pathogens. My thesis looked at characterizing if there are any specific pathogens associated with developing inflammatory bowel disease. The second part of what I did was to look at how the microbiome could affect the treatment for IBD.

What attracted you to working for 23andMe’s Therapeutics team?

I was a 23andMe customer even when I was a broke Ph.D. student. And I’m in bioinformatics and a huge fan of genetics. I’ve always been a strong believer that personal genetics is going to make its way into modern medicine, and become a part of standard practice.

So, when I was looking for a job, 23andMe was a company I was really interested in. But to be honest I didn’t know much back then about 23andMe’s Therapeutics program. I didn’t know about 23andMe’s unique approach to target discovery or the drug pipeline. It wasn’t really until the interview and until I had the opportunity to talk to the Computational Biology team that I fully understood what the Therapeutics program was about and it was really exciting.

A major advantage we have is the size of our database. And this is extremely important in terms of discovery because we have better power to detect genetic signals useful for identifying drug targets that other databases can’t. And of course, our data comes from customers who consented to participate in research and who we can contact in the future if we need to. So, that’s another advantage. 

Describe a little bit about some of what you do in your day-to-day work.

I can divide my current role into two different aspects that support our therapeutics discovery and development process. The first is to help with data analysis in support of identifying biomarkers that could be useful for clinical trials. I am also working on developing automatic reports for later stage programs. 

And the second part of my job is more upstream, on integrating different datasets to improve the variant (GWAS hit) to target gene mapping process. We want to identify genes that are actually affecting the disease and understand the mechanism of action.

What’s different about working here?

For me, I guess what influenced my decision to come here is that 23andMe is doing genetics-based, data-driven discovery. So as part of the Computational Biology team, you can make major contributions to drug discovery whereas in a more traditional biotech/pharma setting Computational Biology would play much more of a supporting role.

As a scientist, I am very interested in doing research and exploratory projects that are intellectually challenging and at 23andMe we have the opportunity to do these things.

How have you managed your work-life balance over the last year during the pandemic?

I’m working primarily from home but I go to the office two or three times a week. At first, it was hard because I started at 23andMe during the shelter in place while we were all working from home. But the team has been really good about meeting each week and making sure questions are answered so I don’t feel like I’m by myself all the time. And there is a good work-life balance.

What interests you outside of work?

I used to play a lot of badminton. I actually trained for it as a kid, but stopped playing in grad school and shifted over to table tennis. In the office we have a table tennis tournament going on after the work hours, so that’s fun. And during the pandemic, I got a puppy so that’s occupying my time right now too.