Viruses use very complex interaction networks to interface and hijack host machinery in order to do their bidding. Mapping protein interaction topologies represents a fundamental step towards understanding these biological processes. Over the last 3 years, I have developed quantitative high-throughput microfluidic tools and applied them to study protein interaction networks. These tools allow us to shed light on difficult biological questions with clinical implications. In many aspects, these questions could not be readily addressed with conventional methods (membrane protein expression, etc.). Viruses modulate host networks on the genome, transcriptome and proteome levels. I would like to discover such viral-host interfaces by mapping viral protein interaction networks with the host genome, transcriptome and proteome. My research is guided by several questions: (1) What are the common "tools" viruses use to hijack host network? (2) How do these "tools" evolve? (3) How can we modulate or "turn off" these interfaces? (4) Can we mimic these "tools" in order to interface with cellular networks?  

A proof of concept of this methodology can be found in Einav and Gerber et al., 2008. In this paper, we used the same microfluidic platform to characterize a new function for a membrane protein from Hepatitis C virus, as well as find inhibitors to this new function. In less than 2 years, we now already have a compound that successfully passed preliminary clinical trials.