Sponsored Insight
Over the past decade, an explosion of innovative therapeutic approaches has significantly improved outcomes for people with previously intractable illnesses. From mRNA vaccines, to immunotherapies, to cell and gene therapies, scientists have advanced medical breakthroughs at a speed and scale unimaginable twenty years ago. However, the process of discovering and developing new medicines remains incredibly challenging, and the questions we need to address are more complex than ever. How does the biology of disease differ from person to person? How can we dissect the molecular circuitry of individual cells and apply this information to better understand and treat patients? And how can we make the process of drug discovery and development more predictable so we can get a “multiplicative” benefit for patients rather than simply an “additive” one?
One of the most promising opportunities to address these questions lies in computational approaches such as machine learning. Alongside therapeutic advances, the past decade has also brought significant advances in computation and data analysis. Most notably, machine learning has gotten better at the “trifecta” of making more accurate predictions, generating new entities, and being interpretable, meaning you can understand why a certain prediction is made or why a certain thing, like a particular antibody sequence, is generated.
Beyond having the right computational approaches and algorithms, we are also now at a unique point in time when we have the right volume and type of biological and medical data and the right computational power to harness the true potential of machine learning for R&D. For the first time, machine learning is poised to enhance every aspect of drug discovery and development — from identifying new biological targets to building new molecules to designing faster, more robust, and more equitable clinical studies.
At Genentech and our parent company Roche, we are using machine learning and other computational methods to accelerate drug discovery and development and build better models that reflect the diversity of all patients. In addition to building in-house capabilities and infrastructure, bringing on diverse talent, and exploring new technologies and platforms, we are forging partnerships and building interdisciplinary networks across academia and industry to drive a convergence of biology and computation that we believe will transform scientific discovery. Realizing this vision will be a marathon, not a sprint — but in the end, we believe this massive effort will pay off for patients, now and for decades to come.
View open positions at Genentech here and additional opportunities at Roche here.