Skip to main content

Conference Program

Subpage Hero

Loading

Validation and Implementation of Emedgene Variant Interpretation and Reporting Software for Clinical Genomic Testing 

Laboratory Genetics and Genomics
  • Primary Categories:
    • Laboratory Genetics
  • Secondary Categories:
    • Laboratory Genetics
Introduction:
Variant interpretation and reporting are critical components of the clinical genetic testing workflow. This novel validation study at the Stanford Medicine Clinical Genomics Laboratory focused on transitioning from the Agilent Alissa Interpret and Sunquest Mitogen LAB tools to Illumina Emedgene software, integrating artificial intelligence (AI) with filter-based analyses for germline variant prioritization and reporting. This validation was essential to ensure the analytical and reporting performance of Emedgene to support clinical genome-based exome and panel analysis workflows.

 

Methods:
The validation plan assessed reportable range, clinical performance, and reproducibility across 37 sample datasets, including 7 commercially available reference samples from the Genome in a Bottle (GIAB) consortium and 30 clinical control cases with 88 previously reported germline variants. Emedgene was evaluated across multiple workflow points, including data transfer from the sequencing pipeline, automated and manual case initiation, variant curation, and report generation. Annotation accuracy, filtration performance, customized preset filter performance, and reporting integrity were validated through a combination of automated comparisons and manual reviews, with additional precision testing by transferring data from staging to the production environment.

 

Results:
Annotation and filtration accuracy were confirmed by concordant results with truth sets across 13 annotation components and 12 filtration criteria. Preset filters demonstrated consistent accuracy, capturing concordant variants across all selected samples. Report generation was verified for all custom templates and fields. Repeatability and reproducibility were validated using GIAB samples, with all workflows yielding robust results. Clinical performance, assessed by detecting clinically significant variants among control samples, demonstrated concordance with previously reported results and appropriate classification updates.

 

Conclusion:
Clinical variant interpretation and reporting using the Emedgene platform demonstrated robust analytical and reporting capabilities, meeting all validation benchmarks. The AI-assisted variant prioritization and curation functionalities integrate effectively into the Clinical Genomics Laboratory’s workflow for both genome-based exome and panel testing, supporting comprehensive genomic analysis and report generation for clinical diagnostics. Taken together, this comprehensive validation supports the implementation of an AI- and filtering-based tool to establish a reliable, efficient, and accurate interpretation system for clinical genomic testing.

Agenda

Sponsors