Almost 10 years ago, Ben Liu, an associate professor of forestry biotechnology at NC State, realized that there was a serious problem with the quality of gene and protein expression data. While researching a book on statistical genetics, Liu found that many of the papers in the literature contained “too low quality data, at less than rigorous treatment,” he says. So he resolved to do something about it.
In his spare time away from his research on statistical genetics, Liu investigated how he might develop a better biochip for genomics applications, and came up with two inventions related to biochip design and fabrication. The patents, filed initially in 2000, describe a “devilishly simple” concept, he says. By expanding the geometry of a biochip into three dimensions, Liu could stack a series of chips one on top of the other, drastically increasing the amount of information he could extract from one assay. Stacking several hundred chips into a cube, Liu says, allows researchers to both coat the chips and assay samples in parallel.
As an example of how this works, Liu says individual chips can be assigned to a particular gene probe, so that every one of the 384 wells on the chip is filled with the same gene probe. By stacking 200 of the chips together into a cube, a researcher can construct an experiment investigating 200 different genes. And because the design allows samples to pass through them without contaminating adjacent wells, a researcher can inject 384 different samples through the cube, in the process determining whether the samples contain any of the 200 genes of interest. “You stack them on top of each other, and the fluid can flow through the whole stack of the chip,” Liu says. “That’s how you can do it in parallel.”
Liu started a company, Cary, NC-based Plexigen, to commercialize the technology, and for the past three years Liu and his colleagues have kept a low profile while they optimized their approach. Now, he says, the company is searching for core facilities to test the platform, called the GeneCube, as well as partners to help develop novel applications while the company focuses on providing research services.
In fact, Liu says the GeneCube platform may find early success with proteins, given the demand for parallelizable ELISA-type assays. Together with collaborators at the Duke University School of Medicine, Liu and his group at Plexigen are planning to publish their first paper describing how to apply the platform to protein research in the Journal of Immunology. “We got very comprehensive data on antibody-antigen detection,” he says. “Even though we got started on proteins late [compared to gene expression], we think this could be a very large application, because of the lack of this kind of system on the market.”
— John S. MacNeil