Accelerating rare disease diagnostics by linking DNA and RNA through an explainable and interactive RNA-guided...

Area of Science:

• Genomics
• Bioinformatics
• Medical Genetics

Background:

• RNA sequencing (RNA-seq) faces challenges in genome diagnostics due to biological and technical variation.
• Interpreting RNA-seq data from diverse sources over time is complex, hindering clinical application.
• Existing machine learning methods offer partial correction but don't fully address interpretation complexities.

Purpose of the Study:

• To develop a comprehensive RNA-guided workflow to address variation in RNA-seq data.
• To enable accurate gene-disease association analysis in rare disease patients.
• To streamline variant interpretation for clinical decision-making.

Main Methods:

• Developed a novel RNA-guided workflow integrating OUTRIDER, FRASER, Borzoi, and MOLGENIS VIP.
• Implemented a streamlined process for handling biological and technical variation in RNA-seq data.
• Utilized genomic, phenotypic, and segregation analysis for rare disease cohorts.

Main Results:

• The workflow successfully identifies gene-disease associations by managing data variation.
• Interactive reports visualize outlier genes and prioritize patient-level variants for clinical interpretation.
• Analysis of 144 cases demonstrated enhanced variant interpretation and aided clinical decision-making.

Conclusions:

• The RNA-guided workflow effectively handles variation, facilitating gene-disease association discovery.
• It accelerates the prioritization and reclassification of genetic variants, including variants of unknown significance.
• This approach supports clinical interpretation and mainstream adoption of RNA-seq in diagnostics.