Sophia Genetics, global leader in data-driven medicine, announced at the 50th Anniversary Conference of the European Society of Human Genetics (ESHG) the release of its Whole Exome Solution (WES) and Clinical Exome Solution (CES).

Accessible through Sophia DDM, the company’s analytical platform for clinical diagnostic, these solutions are powered by new knowledge-driven DNA sample preparation kits, which allow clinicians to fully leverage the power of SOPHiA Artificial Intelligence (AI) to get access to broader, deeper and more meaningful clinical insights.

Today, only 25% of rare diseases are accurately diagnosed. With SOPHiA’s applications for exome analysis, this is set to change.

The exome is the protein-coding region of the human genome, which represents just 1% of the genome but contains approximately 85% of known disease-causing genetic variants. SOPHiA takes exome sequencing to new heights, allowing for unmatched analytical performances to detect and annotate disease-related genetic variants over all protein-coding regions of the human genome.

Exome sequencing generates large amounts of sequenced data. The problem is that reporting variants depends on the algorithms used to sift through this data, leading to a substantial discordance in variants description between the different annotation tools and their descriptions in functional and clinical databases.

Consequently, clinicians can confound variant matching, which is a critical step in variant classification. SOPHiA now overcomes variant annotation tools’ limitations by annotating variants in a way that helps clinicians better interpret genomic data and achieve more accurate and precise clinical care.

Based on pattern-recognition technologies, SOPHiA features a database search engine that guarantees the identification and retrieval of the matching variants regardless of their representations, such as Indels (insertion or the deletion of bases in the DNA) aligned differently, or complex variants.

Variants interpretation is one of the most complex challenges in exome sequencing. But this is also an area where AI can deliver superior insights. SOPHiA processes raw genomic data to detect, annotate, and preclassify variants to facilitate and accelerate clinician’s decisions. Through SOPHiA, knowledge silos are broken in a collective effort to democratise diagnosis of rare genetic disorders.

Already being used by more than 830 experts (biologists, pathologists and geneticists) across 285 hospitals, SOPHiA’s unique approach to knowledge sharing makes the experience of an expert in one hospital scalable for the diagnosis of patients in other hospitals.

In fact, the more variant interpretations are performed, the more SOPHiA is trained, and the better genetic variants are preclassified according to the disease types. With the steady increase of both clinically relevant genes and genetic variants to consider in diagnostic, this technology is a now must to help ensure all patients are being diagnosed at the same level, no matter their location.

Dr Reinhard Hiller from the Centre for Proteomic and Genomic Research Artisan Biomed Laboratory based in Cape Town, South Africa, explained how WES helps save precious time and resources: “Sophia Genetics’ Whole Exome Solution (WES) serves as an excellent benchmark for our laboratory as it detects and validates a more comprehensive variant list. The Sophia DDM analytical platform is user-friendly and easy to navigate, making it possible for a user to be hands-on from benchtop to variant calling.”

In less than 3 years, SOPHiA has made genomics for routine clinical diagnostic a reality in more than 285 hospitals from 50 countries across the globe.

Using AI and leveraging knowledge-sharing to create a collective intelligence, Sophia Genetics continues its mission to democratise data-driven medicine.

With faster, more affordable, and more accurate results, the AI-powered exome solutions by Sophia Genetics represent the next frontier in data-driven medicine, and the most efficient diagnostics tool for clinicians faced with unclear phenotypic data.