Research
Cellular Responses to Pathogen Attack and Environmental Stimuli
Our lab applies systems biology approaches to investigate spatial and temporal features of molecular and cellular processes against external perturbations. We are interested in integrating high-dimensional data sets to explore alternative hypotheses, to explain observed patterns, and ultimately to form testable hypotheses that can be validated experimentally. The goal is to obtain a quantitative understanding of complex biological systems.
DNA damage response network
Cells experience a high frequency of environmental perturbations that can give rise to genome instability. Most human cancers are associated with genetic instability, ranging from elevated mutation rates to alteration of chromosomes. Little is known about mechanisms and pathways that constitute genetic robustness against instability in normal cells, and whether defects in this robustness underlie genome instability seen in many cancers. We are developing computational algorithms to discover functional modules in the DNA damage response network. Based on topological models and prior biological knowledge, we can further develop dynamic models to include feedback loops for investigating the mechanisms of DNA damage response pathways.
Innate immune signaling pathways
The innate immune response is the first line of defense against invading pathogens. Despite complexity, it is relatively conserved from invertebrates to vertebrates and across different pathogen infection scenarios. The innate immune signaling starts from a set of pattern-recognition receptors, which recognize pathogen-associated molecular patterns. We are developing a Bayesian approach to learn the topological structure of signaling pathways with causal links. Insights from this study will improve our understanding of common initial responses by the hosts to various infectious agents.
Microbial gene function prediction
Over 500 microbial genomes are currently available in public domains. However, many genes have uncharacterized functions, which is a major bottleneck for microbial research. Other genome-wide measurements have emerged in the past years, including gene expression profiles, gene regulations, protein characterizations, and protein-protein interactions. We are developing computational algorithms to infer functional modules by merging multiple datasets. This will help predict genes of unknown functions, and identify genes essential for microbial growth and virulence.