Coworking: Ron Hause employs AI to treat genetic diseases

Coworking is a weekly segment where we spotlight Tech Brew readers who work with emerging technologies. Click here if you’d like a chance to be featured.

How would you describe your job to someone who doesn’t work in tech?

I use computers and genomic technologies to work toward developing safer and more effective treatments for human disease. We’re at the stage in human health where there’s an awesome convergence of a lot of biological data with artificial intelligence. I’m part of a team trying to use that data to figure out how to not only create new medicines, but also deliver them to where they need to go in the body to be effective and safe, and manufacture them at an efficient scale that makes them materially more affordable to patients. For context, gene therapies can currently cost millions of dollars for a single treatment, so making progress on accessibility is important.

What’s the most compelling tech project you’ve worked on, and why?

I had the privilege of being part of developing the CAR-T cell therapies Abecma and Breyanzi to treat multiple myeloma and lymphoma, respectively. CAR T-cell therapy is a new approach to treat cancer by genetically modifying a patient’s own T-cells to target and kill cancer cells. These therapies are an exciting advancement in personalized cancer treatment, and we and others are using machine learning on clinical trial data to continually make these therapies even more effective.

What technologies are you most optimistic about? Least? And why?

I’m most optimistic about the potential of diffusion algorithms and large language models in biological sequence design. New algorithms in this space are revolutionizing how we design new biological elements, from RNAs to proteins, dramatically accelerating the pace of drug development. This allows scientists to focus more on experimentation and the optimization of promising drug development leads. We’re beginning to learn the “language” of biology akin to how ChatGPT learned the lexicon of human written language, and Shape Therapeutics is using these methods on a wealth of high-throughput genomics screening data to design novel guide RNAs to specifically correct disease mutations and deliver them where they need to go in the body.

One of the biggest challenges (and largest opportunities to innovate) will be how quickly we, as a biotech industry, can continue to scale the experimental and preclinical structure to test, validate, and reinforce these AI models. And ultimately, how we can ensure that generative AI approaches continue to yield safe solutions that are intended for good. Regulation and ethics problems in this space are much more complicated than the AI methods themselves and require more time to flesh out.

What’s the best tech-related media you read/watch/listen to?

I really enjoy Kyle Polich’s Data Skeptic podcast, as it covers a wide range of applications of AI and statistics across different domains, while dipping into some nontechnical elements such as ethics and culture. I also enjoy Lex Fridman’s deep-dive interviews and Two Minute Papers to stay abreast of new ML developments outside of my field.

What’s something about you we can’t guess from your LinkedIn profile?

I’m a first-gen college grad and played competitive Ultimate Frisbee from my college days through my early 30s. As I near 40, I’m beginning to realize that the body holds up less well to laying out for discs, so I’ve transitioned into your stereotypical PNW dad: Hikes and bikes on the weekend, CrossFit and coffee in the mornings, and Boulevardiers in the evenings.

What do you think about when you’re not thinking about tech?

When not at work, I tend to ponder how families with two or more kids manage any semblance of work-life balance. I am constantly trying to figure out the multi-objective optimization problem of raising my 4- and 2-year-olds while also being a supportive partner and managing a career at a biotech startup.