Insilicos, a Seattle-based biomedical software company, is partnering with the University of Washington to market a cloud-based version of the university’s Rosetta protein and molecular modeling software.
The partners have launched a new company, called [email protected], that will offer molecular modeling and related services and storage on a pay-per-use basis via Amazon Web Services. Target customers are biotech and pharmaceutical firms interested in using the software for drug discovery.
The cloud-based service combines the Rosetta suite and its associated community of developers with Insilicos expertise in providing software services in the commercial arena to provide a way for customers to access molecular modeling software and hardware they need relatively cheaply, Erik Nilsson, Insilicos’ president, told BioInform. This option might be good for groups, such as small biotechs, that are looking to test-drive such tools on their data.
These companies “are poorly served by the available molecular modeling systems,” which are expensive and geared towards more “mature” firms, Nilsson said. “I’m certainly interested in customers like that but I think that if we are really going to achieve what we want to achieve in biomedicine … we need to put these kinds of tools in the hands of smaller, more nimble kinds of organizations.”
The new offering also addresses a need expressed by “many potential users who would like to avoid the trouble of installing Rosetta,” David Baker, a professor of biochemistry at UW and one of the developers of the software, told BioInform in an e-mail. “[email protected] could make it much easier for them to use Rosetta to help with their research.”
Currently [email protected] is available as a limited beta and includes tools for de novo structure prediction, comparative protein structure modeling, protein-ligand docking, protein-protein docking, and antibody modeling.
He also said that the company will launch a second beta this month that will test additional features, including a set of web interfaces for a broad range of Rosetta capabilities.
Nilsson declined to share details on pricing for the beta version of the service, but said it will be “attractive.”
“There are a lot of people who have experience with the Rosetta system from an academic or commercial setting and … we are interested in those people early in the beta process because … they are going to be more familiar with Rosetta and … they’ll be able to work it harder and give us good feedback,” he said.
The company is also interested in clients “who have no experience with Rosetta” because “we [also] want to know how approachable the service is,” he said.
The partners haven't set a definite timeline for the beta. “We’ll end it when we are satisfied with the results. We are using a stable release of Rosetta so the beta could go very quickly,” Nilsson said.
The company also isn’t disclosing how much the service will cost when it’s fully launched although Nilsson did say that pricing would be different from the beta. Customers will be charged based on the amount of compute time needed and there will be different pricing options for academic and commercial users.
Meanwhile, UW will continue to offer the existing commercial and academic licenses for non cloud-based versions of Rosetta, Nilsson said. The fee for commercial use of Rosetta is $40,000 per site with discounts available for licenses that cover more than two sites; the academic license is free.
With its focus on biotech and pharma, the newly minted company will have to compete with products from more established companies such as Accelrys, Schrödinger, Tripos, and Molegro, which was acquired by CLC Bio this week (see related story this issue).
However, [email protected]’s emphasis on small biotechs and its use of cloud infrastructure are factors highly in its favor, Nilsson said.
“We are pursuing a customer that those companies are less interested in … so we are competing by addressing an underserved part of the market … that doesn’t have a solution available to [it] right now,” he said.
Furthermore, because the cloud-based service eliminates upfront investments in local hardware, “we have a lower cost of business” and, as a result, “we are going to be more efficient” than the competition, he added.
Rosetta at a Glance
UW’s Rosetta is a software suite for predicting and designing protein structures, protein folding mechanisms, and protein-protein interactions. It offers tools for things like de novo protein structure prediction; fragment assembly of RNA; small molecule-protein docking; enzyme design; and membrane protein ab initio modeling, among other functionalities.
Recently, the software was used in a UW project that focused on improving two in silico-designed influenza antibodies so that they could better fight different viral strains (BI 6/15/2012).
Last fall, players of Foldit — an online protein-folding game from UW that is based on the Rosetta algorithm — solved the structure of a protein-cutting enzyme from an AIDS-like virus whose configuration had eluded researchers for more than a decade (BI 9/23/2011).
In 2009, a distributed version of the software package, [email protected], was selected by Intel for its Progress Thru Processors initiative, an application built on Facebook to help recruit participants for distributed computing projects (BI 8/7/2009).
The software was also used in a research effort funded by the Bill & Melinda Gates Foundation that tried to design synthetic molecules that could trigger antibodies against HIV (BI 7/21/2006).