Skip to main content
Premium Trial:

Request an Annual Quote

Sophia Genetics Adds Cystic Fibrosis Analysis to Clinical Bioinformatics Portfolio


NEW YORK (GenomeWeb) – Sophia Genetics, a clinical genomic data analysis services provider, this week launched a new CE-IVD-marked software pipeline that lets clinical researchers identify and characterize genetic variants associated with cystic fibrosis using data from a single next-generation sequencing experiment.

According to the company, it's the first solution of this nature on the market and enables clinical labs to run fewer experiments than the status quo while still obtaining accurate results, Jurgi Camblong, co-founder and CEO of Sophia Genetics, told BioInform.

Current testing for CF mutations requires labs to sequence samples as well as run additional reflex and multiplex ligation-dependent probe amplification experiments in order to accurately capture and characterize all of the associated variants, he explained. With Sophia's solution, sold under a software-as-a-service model, labs only need to have data from a single NGS experiment, cutting down on time and conserving resources that would otherwise have been expended on additional separate experiments.

Specifically, this particular solution uses an internally developed statistical technique to detect CF-relevant copy number variants in NGS data. It also has a unique realignment algorithm that accurately identifies variants in the CF transmembrane conductance regulator (CTFR) gene, and distinguishes between various combinations of repeat lengths. Furthermore, the new clinical pipeline provides a regulation-compliant alternative to the research-centric algorithms that some labs use to try to identify these variants, Camblong noted.

The CF-specific solution runs on Sophia DDM, the underlying informatics infrastructure on which all of Sophia's analysis pipelines are based. In addition to its CF pipeline, Sophia has validated pipelines for genetic testing of familial Mediterranean fever, and hypercholesterolemia, among others. The company uses essentially the same pattern-recognition and machine-learning algorithms in each of its pipelines but tailors them for specific gene panels as needed, Camblong explained.

The company has been developing its algorithms, he said, since it launched in 2011 and worked with about 30 institutions to hone and refine them. From the beginning, the founders' goal was to "bring NGS to routing labs in a sustainable way," he said, and it chose to do so by building a secure system whose algorithms would pool and use NGS data from multiple sources to identify the roots of genetic disease, then incorporate and learn from new information as it becomes available in order to more accurately call variants in future tests.

But beyond that, from the get-go Sophia focused on servicing the clinical lab market rather than beginning with research-centric solutions and transitioning to clinics later on, Camblong said. Part of that process meant learning to navigate the details of the regulations that govern wet lab work and infrastructure use in clinical settings and figuring out an efficient and cost-effective path to validation. The company believes it has worked out that path and is offering its expertise to clinical labs trying to file for ISO 15189 accreditation and CE-IVD marking for their NGS-based genetic testing methodology and pipelines — currently it takes roughly a month to gain the required approval using Sophia's approach.

Sophia has both ISO 13485 accreditation and CE-IVD marking for Sophia DDM and intends to seek CE-IVD marking for each genetic test-specific bioinformatics pipeline it includes in its portfolio — it has already done so for its CF, familial Mediterranean fever, hypercholesterolemia, and breast cancer pipelines. Moreover, the company is willing to work with customers to apply its pipeline and constituent algorithms to their specific genes of interest, taking into account factors such as sample type, gene panel, NGS platform, and amplification or capture kit, Camblong said. It takes about a week to get a customer's pipeline ready for use, and then depending on the number of samples, it can take a few minutes to several hours to complete the analysis and generate a report, he told BioInform.

Sophia charges on a per-sample basis but it does not disclose the exact dollar amount publicly. It believes that this approach sets it apart from other firms in the space that have chosen to sell licenses to their solutions. According to Camblong, licensing-based business models tend to discourage the sort of data sharing and information exchange needed for the algorithms underlying Sophia and other similar systems to perform well and improve over time. Some other companies in the space focus on providing fast analysis algorithms but these solutions aren't quite right for the clinical testing space, which emphasizes accuracy over speed.

Sophia is headquartered in Lausanne, Switzerland, with branches in the UK and France. This past July, the company raised $13.75 million in a Series B financing round led by Invoke Capital. The investment, Sophia said at the time of the announcement, gave it access to Invoke's portfolio company Genalys, which also focuses on applying mathematical approaches to genomic information. The combination means that clinicians and hospitals across Europe gain access to very accurate clinical analytics service for early diagnosis and optimized treatment of cancer and genetic diseases, Sophia said.

Sophia's platform is used at multiple European institutions including hospitals in Italy, Switzerland, and France, including St. Anthony Hospital in Paris.