Alice Zhang, Verge Genomics — Founder Q&A

The most important thing about starting and running a successful company is persevering. There were many times where it would have been easy to give up, but I just needed to keep my head down and get through it. It’s all about grit.

With a lot of hard work — and tutoring from her parents — Alice Zhang flowered intellectually, ultimately graduating with honors from Princeton and then enrolling in the MD/PhD program at UCLA’s School of Medicine.

But to the surprise of her peers and professors, Zhang decided to leave the program at UCLA before graduating and decamped in Silicon Valley where she wanted to pursue a very big idea: to find a way to merge machine learning, neuroscience, and experimental biology in the pursuit of speeding up the process of drug discovery for complex diseases like Parkinson’s and Alzheimer’s.

In 2015, Zhang co-founded Verge Genomics with the idea of shattering the decades-long limitations of the drug discovery system, putting in practice her ideas for an interdisciplinary team that would combine top PhDs in machine learning with veteran drug developers to more rapidly understand the gene interactions behind complex diseases.

Q: Your folks are both immigrants from China, right?
Yes. My dad was a political activist in China and my mom wound up getting her master’s degree in electrical engineering.

Q: Was it hard for them to adjust to a new country and a new life?
It’s hard to even begin to imagine what it was like for my parents. But they came here because they wanted more opportunities for me so that I could have the freedom to do what I wanted and get the education that I needed.

Q: We all inherit a certain amount of baggage from our parents. How do you think your family background impacted the person you are today?
One of the big lessons that you learn as an immigrant is to be very humble about what you know and what you do not know — and when to bring in the right resources to help you. That’s been important for me, particularly as a CEO. A couple of other takeaways I inherited was to never take anything for granted and understand that you can overcome many obstacles with grit and perseverance.

Q: So, what motivated you to pursue a career in medicine?
I first became interested in medicine after I visited China when I was in high school to work at a human rights organization to provide medical care to AIDS villages. I saw first-hand the devastating impact that the lack of healthcare can have on people.

Q: How do you think that affected your subsequent career path?
My life’s mission has always been to impact as many people as possible, with the greatest delta possible per person, by the end of my life. The experience I had in China made me realize the importance of healthcare as a fundamental human right. This eventually led me to medical school and graduate school. I realized that making fundamental scientific discoveries could be one of the biggest ways of influencing healthcare.

Q: What did your family and friends think when you told them that you were dropping out of the MD/PhD program to move to Silicon Valley?
I’d say there was a lot of surprise.

Q: Did you always have the proverbial entrepreneurial itch?
No. In fact, I never thought that I would become an entrepreneur. I’m not one of those people who wanted to start a company from an early age. Until recently, I had always thought I’d end up as a physician-scientist.

Q: Did your professors try to talk you out of dropping out?
I wouldn’t say that they tried to talk me out of it, but most were probably taken aback. It was a pretty uncommon path at the time. But by then, I had gotten to know the faculty pretty well and I think they probably trusted my decisions. They knew that I was a thoughtful person and saw my passion for commercializing this work.

Q: So, you land in Silicon Valley. Now what? Were there moments when you ever second-guessed yourself? Or thought, `Hey, I should have stuck with being a doctor’?
That’s an interesting question. I think that as a person, I rarely have had regrets about the things that I’ve done. More often, I have had regrets about things that I haven’t done.

Q: OK, but did the process of starting up a company ever seem daunting?
I took it in stages. First, we went to Y Combinator to focus on the idea for a few months and see if it could get traction. It did, and then we got funding! We learned that there was a real need in the field. After that, we were able to build a fantastic team. So, there wasn’t necessarily a specific moment in time where I thought we’d need to give up everything to do this one thing that’s unproven. It was more of, ‘OK, let’s build this step by step and get evidence to see whether this is a worthwhile investment of time.’

Q: So, no self-doubts?
Every founder has self-doubts. What’s been really important for me to get through the lowest of the lows has been to have a really strong sense of mission and a really strong drive to get drugs out to patients. It makes it much easier — not just for me but also for the team — to be able to push through the low points because you have the patient in mind.

Q: How was this different for you? Did it demand different skills compared with what you knew from being a grad student or in academia?
The number one lesson that I immediately had to learn was how to communicate effectively with the audience. I had become used to writing scientific jargon. Now I needed to think about how I’d present pretty complex ideas to someone like my Mom — and that means understanding the subject so well that you can peel back different layers and get to the essence. The second lesson I learned was that you absolutely have to be able to think outside of the box. Academia is very much based on step-by-step career ascension. It’s only at the very top of the ladder after you’ve finished everything, that you can start getting independent funding for your lab. With a company, you need the confidence to know that you’re uniquely positioned to spot and solve a problem and think creatively about ways to do it. Lastly, I think the most important thing about starting and running a successful company is persevering. There were many times where it would have been easy to give up, but I just needed to keep my head down and get through it. It’s all about grit.

Q: What’s the biggest obstacle in the way of using artificial intelligence (AI) and human genomics to do more effective drug discovery?
One major hurdle is breaking down the silos that still exist between computer scientists, biologists, drug hunters, and academics. Traditionally, computer scientists and biologists have existed in silos, which has precluded AI from making meaningful traction in the drug discovery space. A lot of AI companies just have a machine learning platform while a lot of drug discovery companies just have biologists. But there are very few that have been successful at combining the two.

Q: How did you get around that?
If we were to truly make this succeed, it was clear to me that we would need to start from day one with an integrated team, where we could put neurobiologists right next to computer scientists and not only build a software platform but also build our own labs and our own drug discovery facilities. Having both the computational platform and drug discovery labs allows us to rapidly test computational predictions in the lab and generate an enormous amount of proprietary, validation data that retrains our models and improves our algorithms. That way, we not only produce predictions in a black box, but we can quickly move those predictions from the computer to the lab where we test to see whether they are working.

Q: When you think about future drug discovery and medical advances, are algorithms going to be smart enough to take a long list of compounds and then make the right correlations with genomics data to identify the winners?
We think a major barrier is actually in the data. Applications of AI in biology are different from tech, because there’s a lot of missing data. This occurs because we don’t fundamentally understand most of human biology yet. It’s irrelevant how sophisticated your machine learning algorithms are if they don’t have the right type or amount of data from which they can train and learn. And that’s related to my earlier point about integrating the two because without a lab, you can’t generate data. That closes the loop to feed the algorithms and actually retrain them. What’s so exciting about the field right now is that the major change happening has not been the sophistication of the algorithms, it’s been in the availability of patient data and genomic data, and that has been the consequence of major breakthroughs in the hardware needed to generate that data.

Q: Does success boil down to data handling — the idea that all we need to do is pull together enough data and correlate it correctly? Or is it a function of coming up with the understanding and insights that will unlock the puzzle?
I think both are necessary. Obviously, you need to have the right algorithms in place. But the major thing that will lead to a breakthrough is having the right type of data. And that’s harder to find. And once you get the data, the question is what kind of insights can you produce and also, importantly, what do you do with those insights.

Q: Are we still far removed from the point where researchers are going to be able to deploy machine learning to identify the next generation of treatments for diseases like Parkinson’s or Alzheimer’s?
I don’t think so. The last few years have seen an exciting convergence of new machine learning technologies, an explosion of genomic data, and a deeper understanding of the brain. We are in the midst of an unprecedented time where cures for complex diseases once thought to be incurable are now within reach. New neuroscience tools have enabled a better understanding of the brain and increased quantity, diversity, and higher resolution datasets not possible before. And it comes at the right time. Neurodegeneration is now the world’s greatest unmet medical need. Alzheimer’s disease is the only disease with growing death rates over the last decade among the most prevalent diseases. But, I have hope because we finally have the right tools to start tackling the complexity of these diseases.

Q: I imagine you put in a lot of hours every week. Outside of work, what do you like to do? How do you blow off steam?
I like traveling, but I don’t do it enough. I try to exercise every day and I do a lot of yoga. It’s also important for me to have one day each week where I disconnect completely from email and all work. I like to read and I also go to the farmer’s market on Sunday. I love cooking.

Q: So, what do you hope you’ll have accomplished by the time you reach 40?
I would love to be the company that puts the first drug from a computational platform into a human and shows that it actually slows or reverses disease. To develop a transformative treatment for patients suffering from Parkinson’s, Alzheimer’s, or ALS. The ultimate dream would be to help usher in a new age for taking a completely different systems-wide, data-driven approach to drug discovery across the industry.

Originally published November 16, 2018.