We reconstruct biological histories to enable better understanding of development, aging, and disease. To do this, we develop novel computational and experimental methods operating across spatial and temporal domains at single-cell and single-molecule resolution.
Extracting Latent Information from Data
All available information in experimental data should be fully embedded in the space determined by both the data themselves and their respective latent prior. To maximize the recovery of information, we develop novel computational methods to compress the observation together with the prior, obtaining an approximation of the probability distribution which maximizes the likelihood of such observation. Such tools enable us to recover the latent biological information from the data with improved signal preservation.
Resolving Dynamic Biological Processes
Delineating the history of evolution, development and disease processes requires tracking and locating molecular events over time. In most cases, what we can obtain are end-point samples instead of time-resolved snapshots of dynamically transforming processes. Starting from the first principles, we develop novel experimental and computational methods to track and model molecular events in time and space, particularly those previously untraceable by classical methods.
Dissecting Molecular Mechanism of Epigenome Replication
All cells in an individual multicellular organism share largely identical genomic sequence. Epigenomic modification enables cell fate determination and subsequent stable inheritance of phenotype across cell division. Unfaithfulness in epigenome replication results in aging, oncogenesis, and developmental disorders. We track the replication of the epigenome between parental and daughter copies to determine its key underlying molecular mechanisms, using a combination of experimental and computational approaches, particularly on the single-molecule level.