NEW YORK (GenomeWeb) – Virginia Tech spinout GenoFAB is mulling options for making GenoCAD, a rules-based DNA design web application developed by researchers at Virginia Tech's Virginia Bioinformatics Institute, commercially available. To help with the decision, the company is seeking industry and academic partners to run pilot projects using the software and is engaging with its target community to get a better sense of its needs.
GenoFAB was launched in 2013 to commercialize GenoCAD, software for designing DNA constructs from libraries of genetic parts. Commercializing the software would create revenue streams that would be unavailable to an academic research project and that could be used to maintain and develop new features for the tool.
"GenoCAD is now a mature platform that has outgrown the resources available in an academic environment," Jean Peccoud, professor at VBI, principal investigator of the GenoCAD project, and CEO of GenoFAB, explained in a statement. He further noted that GenoCAD has grown from early days as a research project into a resource that has been supporting other research projects. Its continued use makes it crucial to procure a steady source of income that can ensure its sustainability long term.
Over the next several months, Peccoud told GenomeWeb, he and his colleagues at GenoFAB will work on understanding the needs of current and potential users in both academic and industry settings by providing software demonstrations. They will also develop resources that will help customers use the software effectively in their projects, including training materials, tutorials, and videos. In addition, they will launch pilot projects with partners, for instance in the biologics and biosimilars arena, who may be interested in using GenoCAD's model-driven approach to design constructs that optimize their research and development processes.
Besides demonstrating the value of the software — and also hopefully providing some early revenue for the company through potential industry partnerships — these activities will in part inform GenoFAB's decision about the most appropriate commercialization model for GenoCAD, Peccoud said. The company is currently mulling potential options, including a subscription-based approach or more of a strategic partnership-based approach where the software itself is free and the company charges for its services. In the meantime, GenoFAB is talking to investors to try to raise funds for the company, he said.
For now, the software remains free to both academic and corporate users.
GenoCAD began its days as a VBI research project in 2007. According to its developers, the synthetic biology software offers a rules-based design methodology that breaks down complex DNA sequences into parts, which can then be recombined using specific sets of rules to ensure the resulting DNA constructs function properly in cells. It provides a structured workflow, accessible via a user-friendly point-and-click interface that makes it easy for researchers to design genetic constructs that can be used in efforts to develop new medical treatments and gene therapies, as well as in plant biotechnology and synthetic biology projects.
"GenoCAD helps researchers and organizations develop sets of rules applicable to the design of expression vectors orgenetic constructs" by providing tools that help users organize and annotate the genetic sequences available to them, Peccoud explained. "GenoCAD also provides them with tools to define rules describing how different categories of genetic sequences can be combined."
Sample studies that demonstrate the tool's capabilities include one published by researchers at VT in 2013, which focused on designing constructs for expressing genes in Chlamydomonas reinhardtii, a single-cell green alga. A second study conducted by VT and collaborators at the Massachusetts Institute of Technology devised a language for designing synthetic transcription factors in eukaryotic cells, and another study, performed in collaboration with the National Institute of Biology in Ljubljana, Slovenia, focused on creating a design strategy for multi-genic expression plasmids. Both of these have been implemented in GenoCAD.
In the past, researchers funded GenoCAD's development and maintenance with grants from the National Science Foundation and the National Institutes of Health, the first of which was awarded in 2009. In 2010, developers released its source code under an Apache 2.0 license, a move meant to ensure that the software would continue to remain available in some fashion to the scientific community well beyond the lifetime of its grants. Going forward, new versions of GenoCAD will no longer be released under Apache licensing.
Funding for GenoCAD ran out last year. With no additional grants coming in from the federal government, Peccoud and his GenoFAB team — who are not from VBI — spent the intervening months contemplating ways of staying afloat, including considering a sponsorship model which developers of some other academic resources have tried, before ultimately settling on commercialization. "We had [a] user base [and] we started seeing a number of publications that used GenoCAD to provide some information about DNA sequences and designers in different application domains [so] the choice was either shutting down the server or trying to find a new home," he said.
It is a difficult decision that a number of other resources developed with the help of academic grants have had to come to terms with in recent years amid tight research budgets and increasing competition for funds. In 2011, developers of the Kyoto Encyclopedia of Genes and Genomics began charging users for subscriptions to its FTP site when its funding ran out, though accessing the data via the KEGG website remains free.
In 2010, developers of the Arabidopsis Information Resource (TAIR) tried a corporate sponsorship program to sustain the database after its funds dried up. In 2013, investigators involved with TAIR adopted a subscription model under which users would be charged varying fees depending on data usage. They launched an independent non-profit organization that year called Phoenix Bioinformatics that allowed them to explore subscriptions and other user-based funding mechanisms that could work for TAIR. Interestingly, the J. Craig Venter Institute received an NSF grant, also in 2013, to create a separate resource dubbed Araport that would house TAIR data. In a conversation earlier this year, Eva Huala, TAIR's principal investigator, told GenomeWeb that Phoenix maintains a collaborative relationship with JCVI and its partners and provides Araport with data that has been housed in TAIR for at least a year. At the time, both groups were exploring mechanisms of making more current datasets available through the JCVI portal.
"Transferring [GenoCAD] to a company essentially makes it possible to access sources of funding that we just cannot access very easily in an academic environment," Peccoud said. "The plan is to keep that important research resource available to the community through a more diversified source of funding than just NSF and NIH grants."
As part of their commercialization efforts, Peccoud and his team of four part-time employees have reinstalled the roughly 800,000 lines of code that make up the GenoCAD application on new computing servers outside of VT, he said. Customers can access the software there but in cases where users are hesitant to put private data on third party compute resources, GenoFAB can provide virtual images of GenoCAD that clients can install and run on their local infrastructure, he said.
They have also added a new combinatorial design tool to GenoCAD that improves researchers' ability to design constructs that optimize protein expression. Instead of trying to build DNA sequences one at a time, the new feature lets users create sequence templates and select the genetic parts that fit the template. GenoCAD then automatically creates constructs, each of which contain a part from the defined range and match the rules set out by the template. This way, a user can generate up to 20,000 DNA designs with slight variations that allow for better optimization. These designs can be exported as Fasta, GenBank, or plain text files or saved in GenoCAD accounts.
Earlier this summer, the VT researchers also released GenoLIB, a repository of biological parts derived from roughly 2,000 plasmid parts, which is available to GenoCAD users. In addition to providing building blocks to design new plasmids, GenoLIB also provides reference sequences that can be used to revisit the annotation of existing plasmids available from sources such as iGEM's Registry of Standard Biological Parts.
Peccoud told GenomeWeb that the company has additional features lined up in its development pipeline, including a new sequence editing feature. As GenoFAB goes to market, it will have to compete with offerings from companies like Genome Compiler — which also recently began offering a plasmid library — that have similar products and target similar kinds of clients.
However, Peccoud believes that GenoCAD offers distinct features from its competitors. "I think that GenoCAD is really the only knowledge capture solution [that] allow[s] users to specify and share design strategies applicable to different industries, organizations, or even projects," he told GenomeWeb. Besides saving them time and money, "it helps teams think of all the different ways to achieve a particular objective or phenotype" and "provides a foundation to collect quantitative data that can be used to optimize gene expression."
Furthermore, GenoCAD is not a sequence editing tool — although it will soon include a tool for this purpose — and does not provide capabilities such as restriction site analysis and plasmid mapping that would be available from programs like VectorNTI, a product from ThermoFisher Scientific, he said. "GenoCAD is focused upstream of that type of bioinformatic analysis."