NEW YORK (GenomeWeb) – Numerate said this week that it has completed a round of Series A financing, during which it raised $8 million that it will use to push forward internal and collaborative drug discovery projects in metabolic, cardiovascular, and neurodegenerative diseases.
Guido Lanza, Numerate's president and CEO, said in a statement that the funds would support two internal early stage drug discovery projects. The first is a metabolic disease program where "we are targeting the free fatty acid receptors to address diabetes, obesity, and related co-morbidities," and the second focuses on cardiovascular disease, in which the firm is targeting the Ryanodine receptor 2 to discover improved drugs for heart failure and arrhythmia. The funds will also support Numerate's ongoing participation in a neurodegenerative disease project being conducted in collaboration with the Gladstone Institutes with funding support from the Wellcome Trust, he told BioInform.
"The goal is to take these programs forward and demonstrate the value that [Numerate's] platform can add" and to go "as fast as we can to get one of these two to a development candidate" that can be licensed to interested pharmaceutical companies, he said. The company has a number of potential partners in mind, according to Lanza, but it is not disclosing who they are at present.
Part of the funds will also go towards building out the privately held San Bruno, Calif.-based company's computational platform, Lanza said. This is in addition to a grant from the US Department of Defense's Defense Threat Reduction Agency that it has been using for algorithm development. Efforts on this front include creating algorithms for absorption, distribution, metabolism, and excretion (ADME) and drug toxicity studies, according to Lanza. These additions to the system will enable the company to pick candidate compounds based not just on "their activity and their functional effects but also to prioritize scaffolds based on their predicted ADME and potential off-target effects," Lanza said.
Numerate's system — dubbed Numatix — is made up of proprietary machine-learning algorithms that are used to generate in silico small molecule drug leads for specific disease targets. These algorithms, which run on either Amazon's Elastic Cloud Compute infrastructure or the Google cloud, work by using large collections of compounds from public and private sources to create virtual libraries of billions of compounds. It then creates "virtual assays" for specific drug design properties, activity and toxicity, for example, that are applied to the libraries to identify the compounds most likely to satisfy the specific design criteria.
Initially, Numerate's business model was based on forming risk-sharing partnerships with pharmaceutical and biotechnology companies and academic groups. Under the terms of these partnerships, it would only charge partners when it successfully created and delivered compounds that worked. It established one such partnership with Boehringer Ingelheim in 2011, partnering with the pharma company to generate drug targets for an unnamed infectious disease.
Numerate has since modified its approach to the market, adopting a more traditional biotech model, where it internally generates and cultivates leads and then offers them for licensing, alongside its existing partnership model. It chose to do so because, under the previous approach, it proved far more difficult to "control the value that [we were] generating long term," Lanza told BioInform.
By that, "I mean if you are working on somebody else's [drug discovery] program … you may have some downstream payments, but the odds that you reach those payments for something that you contributed in somebody else's pipeline are fairly small," he explained. "The way to really capture a lot more in terms of milestones, royalties, [and so on] is the more standard biotech model where you are starting programs and licensing them later."
The 12-person company has also set aside development on BitMill data analytics, a new next-generation sequence data analysis product that Numerate introduced and opened up for beta-testing last September with a planned launch for sometime this year. That product, which would have run on the Numatix infrastructure, was intended to provide researchers with access to commonly used open-source tools for NGS analysis such as TopHat, SAMTools, and Cufflinks, as well as some of Numerate's own internally developed capabilities.
Early reactions to Bitmill were "positive," and "some very interesting collaborations came out of [the beta]," Lanza said. However, "once the opportunity [came] to work with Atlas and Lilly … it was clear that we needed to put every ounce of energy into these [firms]," Lanza said. That's not to say that Numerate might not continue developing BitMill in the future, he said, but for now, its focus is on cultivating projects that will generate the most value long term.
The current financing round was co-led by Atlas Venture and Lilly Ventures with participation from unnamed existing Numerate investors. As part of their respective investments in the company, Bruce Booth, a partner at Atlas Venture, and Steve Hall, a venture partner from Lilly Ventures, have both joined the Numerate's board of directors.