Dry cell biology
The increasing complexity and volume of microscopy data presents an opportunity for deciphering fundamental biological processes by analyzing image data. Discovery of complex dynamic patterns in such data require deep knowledge of the biological process, the experimental possibilities, the type of information that can be extracted and the computational tools to extract it. The field is desperately in need of computational (or ‘dry’) cell biologists, interdisciplinary scientists who can drive biological inquiry by bridging cell biology and computer science.
We work at the boundaries of quantitative disciplines and cell biology, identify scientific problems, steer (but usually do not perform) experiments, import, adapt and apply quantitative tools, and interpret the data to conceive testable numerical predictions. Read our "dry cell biologist" essay (2016) and our perspective on data science in cell imaging (2021).
Visual interpretability of deep learning models
The success of deep learning in biomedical imaging comes at the expense of lacking biologically meaningful interpretations. We are developing explainable artificial intelligence (XAI) methods for bioimaging data (Viewpoint 2022, Comment 2024) and demonstrate applications in predicting melanoma metastatic aggressiveness (paper 2021), in vitro fertilization (paper 2024), high-content phenotypic screening (paper 2025), and in silico organelle localization (paper 2025, stay tuned for more!).
Bottom-up modular characterization of tissue state
How multicellular patterns emerge from the heterogeneous behavior and interactions of individual cells? We are developing computational methodologies for deciphering the information propagation between single cells in multicellular systems, driving an emergent collective cell intelligence, and applying it to diverse biological systems such as long-range cell-cell mechanical communication (paper 2023), collective cell death (paper 2020), multicellular calcium synchronization in vitro (paper 2022) and ex vivo (paper 2024), and intracellular adhesion-mediated communication (paper 2024). Spatial biology technologies (e.g., spatial proteomics) enable us to study aspects of cell-cell interactions (stay tuned!), higher-order interactions (bioRxiv 2025) and tissue organization (bioRxiv 2025b) in the context of human disease.
In silico organelle localization
Computationally translating label-free microscopy images into virtual fluorescent images could enable minimally invasive long-term live-cell imaging (Review 2024). However, this technology is limited by structural inaccuracies, limited generalization, and a lack of interpretability. We are developing "in silico labeling" models that achieve highly accurate and uncertainty-aware predictions (stay tuned!), context-dependent models that generalize to out-of-distribution data (paper 2025), and post-processing methods to enhance interpretability and trustworthiness toward practical applications (stay tuned!).
Measuring continuous cellular state transitions across scales
Obtaining a holistic mechanistic understanding of biological processes relies on the ability to continuously measure the physiological states of cells and molecular machines through time. We are leveraging live cell imaging and computation to systematically measure cellular state transitions across scales: multicellular signaling synchronization (paper 2022), single cell differentiation (paper 2024) and death (stay tuned!), and biomolecular complexes structural characterization (stay tuned!).
Multimodal bioimaging integration
Each bioimaging modality provides only a partial view of a biological sample's complex architecture. Multimodal bioimaging provides a more complete view by acquiring complementary imaging modalities from the same tissue specimens (Perspective 2024). We execute this vision by building computational methods that align and fuse multimodal bioimaging data, moving beyond single-modality boundaries to capture complex cross-modality relationships and enhance predictive performance (stay tuned!).
