With the ever-increasing potential of new technology and the exponential growth of the life sciences field, researchers are always running into new problems to solve. In this interview series, we get scientists’ opinions on the ‘Big Challenge’ in their field and the steps being taken to address it. From new and unique hurdles to fresh takes on common problems, we dive into the complexities of the research landscape.
In this interview, we chat to Dean Lee (Senior Expert I, Data Science, Novartis) about the big challenges in the life sciences field, particularly those impacting computational biologists and the pharmaceutical industry.
Want to hear more from Dean Lee? Come and hear her speak at The Festival of Genomics and Biodata!
Could you introduce yourself and tell us about your background, your role and what you work on?
I am a Data Scientist at Novartis, Biomedical Research, Oncology. I use large-scale omics data to determine how the tumor microenvironment (TME) contributes to mechanisms of resistance to multiple drug modalities across multiple tumor indications. In my free time I think and write about the complexities in a computational biologist’s work to realize the promise of AI/ML to improve patient outcomes.
What would you say is the big challenge in your field?
There are three big challenges today:
Challenge 1: Cuts to Innovation and Talent Pipeline
The recent cuts to life science research funding (NIH, FDA, NSF, HHS, etc.) will manifest as cuts to our innovation and talent pipeline in a few years’ time. If basic science research and training decreases by, say, 50%, then that translates to a corresponding decrease in academic labs vetting risky ideas to spin off into companies, and in the number of people choosing to study biology. This amounts to losing a generation of innovative companies and scientific talent that this industry has taken for granted for decades. How will we cope with fewer players at the earliest, riskiest stages of drug development and with fewer highly trained scientists?
Challenge 2: China Rising
In stark contrast to the decrease in federal support for life science innovation in the US is the maturation of the Chinese biopharma ecosystem. In 2024, one-third of pharma licensing deals were with Chinese biotechs. They offer cheaper, faster, and equal-quality science. What do we do with this reality? Which parts of our business make sense to outsource to China, and which parts make sense to keep in-house? This poses a strategic question to large swathes of US biotech/pharma.
Challenge 3: AI Strategy
Another challenge is coming up with an AI strategy that meets biotech/pharma’s needs. After the initial rush for every company to signal that they are doing something flashy with AI, I hope the discussion becomes much more concrete and granular. Boring, even. The discussion needs to be broken down into several distinct sub-discussions about: data infrastructure; data governance; AI as point solutions versus AI as whole-process transformations; AI for business and operations problems versus AI to reveal novel biological insights, etc. Actually, to speak of a single AI strategy is a sign that we have not thought deeply enough about what it entails. Every company needs a suite of AI strategies for a suite of problems.
What needs to be done to help address those challenges?
Addressing Challenge 1 and 2 requires a forum in which biotech/pharma leaders can speak openly about the issues they are facing and coordinate action. To the best of my knowledge, that forum does not yet exist. Creating that space would be the first step.
Addressing Challenge 3 requires biotech/pharma leaders, especially those not from a computational background, to spend enough time listening to their data science team. Just commissioning them to come up with an “AI strategy” is not enough. Even better, you might want to place some data scientists at the highest levels of the company.
What advice do you have for somebody trying to break into this field?
To break into computational biology work, focus more on demonstrating skills with projects rather than on degrees or certificates. Pick a language (Python or R), find a GitHub repo associated with a recent publication, and try to implement the authors’ analyses. As you reproduce others’ observations, challenge yourself to tweak their analyses to give a more nuanced interpretation of the data. In the end, you should have a slide deck that walks the audience through your biological question, the data and method you used to answer that question, your computational analysis results, and why your findings matter to the business.
What are you looking forward to most about the Festival of Genomics and Biodata, and why would you encourage people to come to the event?
I love that there is a good balance between academic and industry talks. It is rare to find a forum that has ample representation from both groups. Come for the professional development workshops and the networking. Come to get a snapshot of the mood of the industry in real-time.
Want to hear more from Dean Lee? Come and hear her speak at The Festival of Genomics and Biodata!
