How do we use AI to make biological discoveries? While AI is a powerful tool for scientific discovery, its successes are where conditions align with where AI already excels: ample data, abundant prior knowledge to turn into labels, and where problems are already defined to be computable. I focus on using AI to empower the long tail of biology that remains exploratory, uncharacterized, and descriptive.

I focus on several themes:

• I precisely characterize limitations of current AI for biology methods. My work has exposed that biological foundation models may not function well in zero-shot exploratory applications, learn universally relevant signal, or outperform bioinformatics baselines.
• I develop biologically grounded methods that unlock new capabilities. I’ve shown that self-supervised methods incorporating biological priors for genomes, proteins, and microscopy can unearth new insights. I also develop generative models. Within genomics, these models can design enhancers without prior knowledge of function and in microscopy, overcome experimental scaling limitations.
• I translate insights to widespread impact in science. The challenges I study are not exclusive to biology. I collaborate with domain experts to translate the core insights I develop within biology to other domains, including linguistic study of low resource sign languages, AI models for pediatric healthcare, and probabilistic weather forecasting.