Skip to main content
Premium Trial:

Request an Annual Quote

Seeking an Edge in Clinical Market, German Startup Preps Community-Based Variant Annotation Platform


GeneTalk, an informatics startup based in Berlin, is hoping to make its bread and butter by providing customers with algorithms to prioritize disease variants and a forum through which they can find additional information about these variants from others in the community.

GeneTalk, which will officially open its doors in a few weeks, will offer access to its web-based platform of the same name for free to academic and commercial users for one year, after which it will begin charging an annual license fee to cover the cost of additional services and support, Peter Krawitz, one of GeneTalk’s founders, told BioInform.

The company will offer two kinds of licenses — one for academics and another for commercial companies — and each will be priced differently, though Krawitz declined to provide pricing details.

"We feel that genetic information is … something society pays so much for and so we want to give back that analysis tool for as low cost as possible," he said.

Krawitz, who is a physician at the Charité University of Medicine in Berlin, said that the company will support itself during the first year with research funding that it has already received.

Krawitz told BioInform that he and co-founder Tom Kamphans launched the company to commercialize algorithms that they had developed at the university. Last month, the company's co-founders published a paper in Bioinformatics that describes applications of the GeneTalk platform.

At present, GeneTalk only handles exome sequence data analysis but the company plans to extend the platform to include capabilities for whole-genome analysis at a later date, Krawitz said.

For now, GeneTalk expects its platform to appeal to customers in both clinical and research settings where exome sequencing is used to detect mutations responsible for rare genetic diseases, among other applications, he said.

A Crowded Field

When the company brings its product to market in the next few weeks, it will face competition from a number of informatics vendors developing genome interpretation software for the clinical market.

For example, NextBio is preparing to launch a version of its informatics software for interpreting genomic and clinical data in clinical cases. The tool, dubbed NextBioClinical, is currently being used in a series of private projects including one with the Cancer Care Institute (BI 6/29/2012).

Another startup, GenoSpace, offers a cloud-based platform for storing and linking genomic and clinical data and recently joined forces with Thomson Reuters to develop an annotated gene variant database for use in clinical and research environments (see related story this issue).

Other potential competitors for GeneTalk include GenomeQuest, Knome, Omicia, Cypher Genomics, and Personalis, all of whom offer tools for detecting variants in clinical settings (BI 12/22/2011).

Meanwhile, ongoing efforts in academia to develop infrastructure specifically for clinical use of genomic information could also pose a threat to GeneTalk.

These include the Medical College of Wisconsin's bioinformatics team, which has built a software system called CarpeNovo that is used to identify and annotate disease-causing variants in whole-genome sequence data. Separately, a team at the University of Toronto has developed a similar system, dubbed MedSavant, that is currently used for clinical research into genetic factors associated with autism (BI 11/18/2011).

Furthermore, researchers at Harvard Medical School's Partners HealthCare Center for Personalized Genetic Medicine have developed GeneInsight, which provides tools to help genetic testing laboratories store and manage genetic variant information and create interpretative reports (BI 8/26/2011).

There are also collaborative efforts such as the Human Variome Project (BI 12/16/2011), ClinVar, and MutaDataBase (BI 8/5/2011) that aim to develop comprehensive databases of human variation to improve sequence interpretation in clinics.

However, Krawitz believes that a key selling point for GeneTalk's platform is its communication mechanism, which is "mutation specific" and unique to the company.

"You can start a [conversation] specifically for the mutation [in which] you are interested … that has no competitor," he said.

In order to advertise their expertise on particular genes and variants, users create profiles on the GeneTalk site that include this information.

Krawitz said that GeneTalk's ease of use should also set it apart in the market. In building the platform, the developers paid close attention to usability —particularly for inexpert users. Krawitz noted that this can be an issue with academic software where the focus is primarily on addressing research questions rather than implementation.

When users upload exome files into GeneTalk, the system runs an algorithm that checks the quality of the data prior to analysis. Users can then prioritize variants that are relevant to their disease using a series of filtering tools.

These include a functional filter that looks for synonymous and nonsynonymous variants; a frequency filter that selects rare variants; an inheritance filter for identifying homozygous and heterozygous variants; and more. The platform also includes an annotation database that contains more than 32,000 entries from the National Center for Biotechnology Information's database of single nucleotide polymorphisms.

Once the user has pared the variants down to a few dozen relevant candidate mutations, the next step is to "find someone in the scientific community who knows something about the gene[s] or … mutation[s]," Krawitz said.

As an example of a potential use case, he explained that a clinician could request exome sequencing for a particular patient that has an intellectual disability stemming from a rare genetic disorder. Using GeneTalk, the clinician could select candidate variants as well as request additional annotation and interpretation information from participating experts.

So far, the company has used GeneTalk in research studies at the University Hospital, Berlin, where it was developed and at other centers in Europe, Krawitz said. It has also picked up customers in North America, he said without providing specific names.

The company hopes to collaborate with other bioinformatics groups and build links to their tools as well as build interfaces for research-focused groups to add their bioinformatics tools to GeneTalk’s portfolio, he said.

When it launches, GeneTalk will have five employees working out of its office in Berlin.