This article is the fourth in a series of bi-weekly columns focusing on the area of "high-throughput biology," or new techniques for high-volume cell based screening and imaging that biopharma is using to validate targets generated through genomics or proteomics, screen for toxicity, and replace other traditional assays. Links to previous columns in this series can be found below.
NEW YORK, Jan. 26 (GenomeWeb News) - Microplates are just about the least intelligent devices in the drug discovery armamentarium. Little more than miniaturized muffin pans with a coating, they're designed for high-volume screening rather than high-value data.
But Wafergen wants to change all that. The startup, launched earlier this month in Fremont, Calif. by a small group of telecom entrepreneurs and a bioinformatics veteran, is designing an "intelligent" silicon chip platform with wells on top of the chip.
This system, the K2 platform, consists of a chip the size of a pentium, which sits on a board "just like you put in the back of your computer, and it's connected to a PC," said Frank Eeckman, the company's CEO The chip itself has wells on it made of SU8 polymer, with photodiodes at the bottom of the wells. These photodiodes have "filters, lenses, or any other sensor you like to put there," as well as heating and cooling electrodes, Eekman said. "So you pipette from your microtiter plate into those wells," he explained. "As soon as you are done with one well, you can locally control it. You heat it up. You run your reaction. You collect the data using the photodiodes. The data go onto your computer disk. When you are done, you take the chip and throw it away."
Along with these well-chips, the company has designed a pipetter and environmental control chamber for delivery of reagents into the wells, but Eeckman said the chips are compatible with other fluid delivery systems.
The company's founders initially sought to develop an ultra-high throughput microplate, but discovered after speaking to scientists that they were more interested in precision than throughput.
"What we had been hearing from the scientists, "is 'we don't have individual well control so I cannot tell what the reaction is,'" said Alnoor Shivji, one of the founders. "So we thought [that] the natural thing to do is to make sure that we can come up with individual well control where they can actually determine how much reagent to put in, what the compound would be, control the temperature, the pH, as well as the reaction. And get readings--many readings at once--and using different means."
They also have label-free detection capabilities, but have made their platform compatible with current off-the-shelf fluorescent assays in order to fit better the current needs of the screening sector.
While Wafergen has not even finished its website, it has already made a deal with a big pharma company to be a beta-tester for the K2 platform, and is planning to ship its 32-well prototype to that company in February. The next step is to scale up to 96-, 384- and even larger versions of these chips, according to Eeckman. But the emphasis at every step, he said, will be on making sure the reactions are well-controlled. "We heard that lesson loud and clear that people are no longer interested in ultra-high throughput," he said.
This is the third startup for Eeckman, an MD-PhD who headed up Lawrence Berkeley National Lab's bioinformatics team on the Human Genome Project before leaving to cofound Gene/Networks and GeneTrace. Eeckman's bioinformatics background has helped him to market the K2 platform in a data-driven discovery environment. With the ability to do multiple measurements using the silicon chips, "What we can offer you is those results in the same well, obtained at the same time," he said. This combination "makes the whole data integration thing so much easier. Because now you're recording two different things at the same time. Your statistics are the same because they happen in the same well."
The other members of Wafergen's startup team include Shivji, a successful telecom entrepreneur-turned venture capitalist; Victor Joseph, a mechanical engineer who had worked at Hewlett Packard in designing inkjet printing technology, and then Cisco; and Amjad Huda, who holds an MBA and worked at Nortel Networks.
The Wafergen concept began with a series of patent applications for the underlying IP that Joseph wrote between his jobs at Hewlett-Packard and Cisco, and which are still pending, said Shivji. A couple of years later, Joseph and Huda approached Shivji, a general at Global Asset Capital in Menlo Park, and founder of three telecom firms, the last of which he sold for $2.6 billion. While this is Shivji's first effort in the life sciences, he said there are parallels between some of the challenges he faced in the telecom industry and the ones in drug discovery " At the telecoms, he said, the challenge was to transition from "very ancient technology providers" and old systems to new equipment, without disrupting the customers. Similarly, "there is some very ancient technology which is being used in this whole process of drug discovery, and what we see [at Wafergen] is an opportunity to really help in improving that whole process with high tech. And, I have been pretty good at building teams that have been able to put systems together."
Shivji has helped the team raised an initial round of about a "half million to a million" dollars in seed financing, and is planning to help it seek additional funding--either a small tranche of $2-$3 million or a $7 to $10 million round, he said.
While Wafergen's underlying IP covers other areas of the life sciences, the company chose to to initially target the screening market because it is a relatively cohesive group of people who are familiar with automation, know what they want and are "always looking for new technology," said Eeckman.
Wafergen's goal is to be "clearly the market leader in this space," said Shivji. "We not only replace many of the mictotiter plates and the glass slides, but also many of the plate readers. Because you don't need the plate readers any more. All the intelligence is built into the well."
Other High-Throughput Biology Columns: