Skip to main content
Premium Trial:

Request an Annual Quote

Frameshift Genomics Wins NHGRI Grant to Commercialize U of Utah's Iobio Platform


UPDATE: This article has been updated to link to a paper with some benchmarks for the Iobio platform.

NEW YORK (GenomeWeb) – Bioinformatics company Frameshift Genomics has been awarded a Small Business Technology Transfer grant of an undisclosed amount from the National Institutes of Health's National Human Genome Research Institute to commercialize Iobio, a computational platform developed by researchers at the University of Utah that offers tools for visualizing and analyzing genomic data.

According to two Frameshift co-founders, the company's apps will make genomic analysis more accessible to clinicians who want to use bioinformatics tools but lack the expertise to do so. "Our main goal is to make exploratory analysis of big data a reality," Frameshift co-founder Alistair Ward told GenomeWeb in an email. "We are aiming to bring experts in disease genetics, working directly with patients, into the analysis process ... [and] get the best eyes looking at DNA data, hopefully improving the diagnostic rate achieved from DNA sequencing studies." Importantly, the platform marries data visualization with real-time analysis to help researchers "ask more questions, and quickly follow up on their hypotheses," Chase Miller, also a company co-founder, said in a statement. 

Specifically, the NHGRI funds will support the development of a commercial app for data quality control and quality assurance, in particular for cleaning up sequence alignment data, Miller told GenomeWeb this week. The planned app, called Multibam.iobio, will help users quickly analyze Bam files and will provide several metrics that highlight particular data quality issues and how these can be corrected. It helps researchers identify areas of data insufficiency, for example, portions of the genome that are missing, which could result in missed mutations.

A portion of the funds will support the development of the core Iobio platform so that it can support multiple users running multiple analyses in parallel. "The way we typically run an Iobio app right now is [by] sampling data from a file so ... we can jump in and pull out small chunks of data and look at a region in particular, like a gene region," Ward explained. But in the commercial application, "what we actually want to do is read through the entire file and do a full analysis of the files. So there's a secondary goal of building the backend infrastructure that allows us to do the full analysis of files and the storing and managing of the data that we generate. It allows us to really build up the kind of service behind the scenes ... to support the commercial user."

Meanwhile, the company is developing a premium version of an existing app on the open-source Iobio platform, namely the Gene.iobio application, which helps users investigate likely causative variants in rare and Mendelian disorders. The tool lets researchers search for variants in genes known to be associated with diseases of interest as well as mutations in genes that are suspected to have some connection to the disorder in question. 

More specifically, the premium version of Gene.iobio "will have some power-hungry features that, while very useful, are not feasible to be offered for free," Miller told GenomeWeb. "It allows us to provide, for example, simultaneous analysis of a user's entire gene list, for those that are willing to pay for the extra compute." Premium customers will also have the benefit of installation, maintenance, and support contracts with Frameshift, he added. It is not clear exactly when Frameshift will launch its apps or how much it will charge for access to them.

The target market for these apps will be research hospitals and diagnostic labs. The company is currently working with some unnamed collaborators in these contexts to ensure that they are comfortable with Iobio's results. In many cases, the algorithmic tools that underlie the Iobio apps are standard ones used by the community, which makes users more confident of the results they generate. According to Miller, internal tests using clinical data show that the Iobio apps generate similar results to existing clinical pipelines — details about these benchmarks are provided here.

The Iobio platform was developed a few years ago in the laboratory of Gabor Marth, a professor of human genetics at the University of Utah. It offers access to apps such as Bam.iobio, which was described in an article in Nature Methods, as well as to the open-source version of the soon- to-be-commercialized Gene.iobio. Also available is Taxonomer.iobio, which offers metagenomics classification and analysis tools, and Vcf.iobio a tool for sampling VCF files and generating metrics about them, such as variant density, base changes, and allele frequencies. 

The decision to commercialize parts of the platform was based in part on cost, Miller said. Iobio works by sampling sections of sequence rather that uploading entire files to the system. "So we only analyze the exact data that we need to answer the specific question," he explained. However, some users want to explore that selected subsection of data in the context of the larger genome, but the costs of the compute resources that would be needed to support such analysis would be prohibitive for an academic institution. By making some apps commercial, the company will be able to expand the range of features and support that it can offer to clients, he said.

Another reason for commercializing was to cater to clients that prefer to run private iterations of the Iobio platform on their own infrastructure but would rather pay someone to manage, maintain, and update the platform as needed, Ward noted. That's a responsibility that Frameshift would take on, he said.

The company is choosing to maintain the app model used by the open-source Iobio platform because it believes that by focusing on specific questions, "we can create simpler, more intuitive user interfaces and craft custom visualizations that better suit not only the type of data generated but also the specific answers the user may be after," Miller told GenomeWeb. But it does have plans in place to integrate apps in some fashion to provide a more seamless analysis experience. So, for example, a clinical researcher could use the Multibam.iobio web application to obtain project- or study-level quality assurance or control metrics for their sequence data and then link to Bam.iobio to get sample-level quality assurance metrics, Miller explained. Furthermore, all the apps will eventually "make use of a single login and data management infrastructure that will provide authentication [or] authorization and data access across all our web apps," he said.

Frameshift sees opportunities to collaborate with other players in the market, such as Illumina's BaseSpace. "We believe the Iobio tools are complementary to those available on BaseSpace, providing functionality that is not otherwise available," Miller said. In fact, the company is currently working with BaseSpace developers to add Bam.iobio, Vcf.iobio, and Gene.iobio to the Illumina app store. The move would allow existing BaseSpace customers who already have their data stored on the cloud to analyze the data with the Iobio tools. Also, "all the data access and authentication steps are handled and we can concentrate on providing effective quality control and analysis tools," Miller said.

Frameshift currently has three employees but there are several additional people working on the open-source Iobio platform, which is helping to push commercial development efforts forward. The company also plans to continue offering free versions of Iobio apps. In fact, it might not create commercial versions of some apps. Miller told GenomeWeb that Bam.iobio and Vcf.iobio will likely always be available for free. Meanwhile, Frameshift is currently seeking additional grant funding for Gene.iobio and other applications that it will develop later on. Future apps could focus on analyzing somatic mutations in tumors, Ward said. "One of our biggest problems is not what things we should be tackling next [but] how much time we have," he said.