Pierre Boyeau - postdoctoral fellow at the Broad

I am a postdoctoral fellow at the Eric and Wendy Schmidt Center at the Broad institute. I recently completed my PhD in EECS at UC Berkeley, co-advised by Profs. Nir Yosef and Michael Jordan.

research interests: single-cell omics, statistical machine learning, interpretable deep learning, principled model evaluation.

contact: [first name][last name][at]berkeley.edu

My research focuses on making interpretable and statistically rigorous inferences from single-cell data. Deep generative models are powerful tools for analyzing these complex datasets, but using them to formally test biological hypotheses (for instance, to identify gene programs that drive differences between cell states in development or disease) is not straightforward and requires dedicated statistical methodology.

I develop statistical frameworks that enable rigorous inference within these deep learning models. These methods allow researchers to perform tasks like differential expression analysis or association testing, while providing clear, interpretable results and controlling for statistical error. I have made these methods available as open-source software for analyzing single-cell transcriptomic and multi-omic data, with applications to uncovering sample-level heterogeneity and identifying candidate molecular functional relationships.

I graduated in 2019 with a double degree from Ecole des Ponts ParisTech and Ecole Normale Supérieure de Paris-Saclay. During my PhD, I was a visiting student in Peter Kharchenko's lab at Altos Labs. My prior experiences include internships in finance and retail, and a research visit at the Riken Institute in Japan.