Causality
Causality in machine learning is essential for understanding the true drivers of outcomes, making reliable predictions, and informing decision-making processes in various biological applications.
Generative Modeling
We strive to improve generative model theory and work on applications like predicting protein binding to design antibodies.
Optimal Transport
Optimal transport (OT) provides a flexible and powerful framework for comparing and transforming probability distributions. Our lab uses OT with generative models, such as autoencoders, to generate pseudo-lineages of cells and study regulatory drivers of cell state transitions.
Disease Diagnostics
We innovate multimodal representation learning techniques tailored for clinical datasets, aiming to enhance disease classification and propel the frontier of precision medicine.
Drug Discovery
The lab advocates for the integration of single cell genomic, proteomic, and imaging techniques to obtain nuclear morphometric biomarkers for early cancer diagnostics.
Reprogramming
We are developing robust strategies for iteratively planning, performing, and learning from interventions for example to reprogram cells or direct the differentiation of pluripotent cells towards a specific cell type.
