In the summer of 2000, Caltech PhD student Derek Debe and his advisor — Molecular Simulations and Schrödinger co-founder Bill Goddard — launched the computational structural and comparative proteomics firm Eidogen, but chances are you have never heard of them.
That’s entirely by design, according Debe. “We have been completely under the radar,” he said. For the last three years, “we’ve been a skunkworks.”
Eidogen — whose name was inspired by eidos, Greek for “shape” or “form” — was founded on Debe’s StructFast protein structure prediction algorithm, which he developed at Caltech. However, Debe said it was immediately clear that the company wouldn’t succeed on the strength of a single piece of software. “Our perspective was very different than many companies,” he said. “We felt that the bar was going to be extremely high, so we needed to be a skunkworks and develop each piece of the puzzle to make something that was going to really be useful for a pharmaceutical researcher.”
Three years later, Debe and his colleagues at the Pasadena, Calif.-based firm are bringing the fruits of that development effort to market. The company’s Target Informatics Platform (TIP) offers four different algorithms in addition to StructFast — SeqFast for remote homology detection, MemFast for membrane protein structure determination, SiteSeeker for small-molecule binding site determination, and SiteSorter for binding site comparison. In addition, TIP includes a database of “the universe of all [predicted] protein structures, the universe of [predicted] binding sites, and the universe of similarities between all of those,” and a suite of visualization tools for manipulating protein structures for comparative structural analysis.
Eidogen is marketing the platform as a system “where you can literally analyze binding sites and targets as easily as running Blast or doing comparative gene sequence analysis,” Debe said. ”While there have been many enterprise platforms in the genomics space, there are not for downstream drug discovery applications. So that's where we felt we could really change things,” he said.
Some of the components of the system, such as SeqFast, are available as stand-alone packages. In
addition, the company’s database of predicted structures and binding sites — in the neighborhood of 650,000
and 490,000, respectively — can be divvied up for smaller companies that are only interested in particular target families. However, Eidogen sees large pharmaceutical companies as the primary customer for the “soup to nuts” platform because they will be able to upload their own proprietary structural and sequence data, “and the database will evolve into something new based on that data,” Debe said. The platform also works with any sort of small-molecule data, he added. “Customers can download the structures from the database, dock any small molecule they want with it, and re-upload into the database.”
In addition to increasing the efficiency of lead discovery and optimization by identifying novel binding sites, Debe said that an important application area for the platform is in finding new opportunities for compounds that are already validated. “Our knowledgebase has many examples of important targets where we found an allosteric, or secondary, binding site that should be interrogated,” he said.
Additionally, he said, the platform’s ability to compare the structures of binding sites can also be very valuable for drug discovery. As an example, he noted that the molecular target for the antifungal Diflucan, lanosterol demethylase, “would never be discovered by comparative genomics” because it is found in both humans and fungi. “What you need is to be able to go in and not ask, ‘Is it in fungus but not in human?’ but, ‘Is it in both but are the differences in the binding pockets of these molecules significant enough that you can selectively kill the fungus without having it bind to the human?’ That’s an example of a very important application that’s opened up by our platform,” he said.
Eidogen has raised two rounds of financing led by Tavistock Life Sciences since its launch. The financing supported development of TIP and helped the company assemble a staff of 25. Now, Eidogen is ready to market its platform in the US and Japan. “So far,” Debe said, “all the feedback has been extremely positive.”
Debe did not disclose pricing information for TIP.