DNAstack Launched for Google Cloud Genomics
October 18, 2016 | This week DNAstack, a Toronto-based genomic software company, launched its Google Cloud platform to accelerate genetic disease research and precision medicine.
The company is headed by Marc Fiume, who chairs the Beacon Project for GA4GH. Last summer, Fiume spoke with Bio-IT World about the Beacon Project, and hinted then at DNAstack’s aspirations to use the GA4GH APIs to manage and store data. Now, he says, DNAstack offers “a secure and cost-effective solution with global standards in genomic analysis and sharing.”
“We developed this open standard saying here’s how you share data. With our launch we’re actually providing the tools so that people can actually participate in the network without needing a PhD or a computer science degree. We’re actually providing software that’s drag and drop,” Fiume said. “It’s lowering the barriers significantly in terms of the complexity of the solution.”
DNAstack is hosted in Google’s Cloud, and is the first commercial platform for genomics built on Google Genomics. (It’s one of the first built with GA4GH APIs.) Researchers can, Fiume explained, simply drag and drop FASTQ files from their sequencers to DNAStack.com and complete their analysis using best practices or custom bioinformatics workflows.
DNAstack is working with GA4GH to help define a language for specifying bioinformatics workflows, Fiume said. The language lets users accurately specify algorithms to be run as part of the sequence analysis.
“We’re enabling a bioinformatics workflow execution engine that can perform bioinformatics execution on any workflow that’s defined in that common workflow language,” he said.
But for users who just want the DNAstack “best practices” pipeline, a few clicks gets the process started.
“The beauty of the system is you’re not required to have training in genomic best practices in order to access the APIs,” Fiume said. This gives all users the opportunity to take advantage of best practices and open standards without even knowing it, he said.
The results of the analysis can be shared on the Google Cloud. Users can invite collaborators to work on datasets and/or share results more broadly through GA4GH API.
DNAstack is available now under a “freemium” model, and that’s a distinguishing feature of DNAstack’s version of cloud genomic analysis.
“We’re trying to provide the simplest and most cost-effective solution to adopt global standards,” Fiume said. “I think our focus on developing networks of shared genomic datasets is also an innovation. Promoting a culture of data sharing is core to the vision of DNAstack.”
The basic DNAstack platform is free, providing at-cost access to bioinformatics. The company says the platform can process genomes at less than $10 per genome. Users simply pay for their storage and compute on the Google Cloud. DNAstack does handle billing for its users, white-labeling the whole process so that users don’t have to interact with the Google Cloud separately.
Fiume outlined several features he expects to come soon. In the next evolution of the software, users will be able to create and share within private networks—separate organizations that allow the same Beacon protocol to work between them. “That can really accelerate genotype-phenotype association, for example, or trial recruitment,” Fiume said. He mentioned in particular proprietary technologies that may help users share and match patients.
Although Fiume stressed that users don’t need bioinformatics training to use the platform, it might be hard to use the results without some expertise. Fiume foresees developing downstream applications to serve the needs of those customers.