SVInterpreter

SVInterpreter interprets structural variants (SVs) to predict their potential clinical outcomes by leveraging topologically associated domains (TADs) and integrated gene-centric annotations for balanced and unbalanced SVs.

Key Features:

• TAD-based analysis: Uses topologically associated domains (TADs) as analysis units to contextualize SVs within 3D genome architecture.
• Phenotype Similarity Scoring: Computes phenotype similarity scores to link SV-impacted genes with clinical phenotypes.
• CNV Scoring Metrics: Applies copy number variant (CNV) scoring metrics to evaluate effects of deletions and duplications on gene dosage.
• Position Effect Prediction: Predicts position effect events where genomic rearrangements may alter gene regulation due to relocation within TADs.
• Integration of Functional Annotations and Dosage Sensitivity: Integrates gene functional annotations and dosage sensitivity information to prioritize candidate disease genes.
• Support for SV Types: Analyzes balanced (translocations, inversions) and unbalanced (deletions, duplications, insertions) structural variants.
• Candidate Gene Identification: Identifies candidate genes potentially implicated in disease based on combined scoring and annotations.

Scientific Applications:

• Clinical interpretation of SVs: Interpreting clinical significance of structural variants identified by genomic sequencing.
• Candidate gene prioritization: Prioritizing candidate genes affected by SVs using dosage sensitivity and functional annotations.
• Assessment of regulatory disruption: Assessing potential position effect events and disruption of regulatory architecture due to SVs.
• Application to balanced and unbalanced datasets: Applied retrospectively to 97 balanced (translocations and inversions) and 125 unbalanced (deletions, duplications, insertions) SVs from published studies and to 145 SVs from 20 clinical samples to evaluate predictions.
• VUS reduction and validation: Demonstrated a 40% reduction in variants of uncertain significance (VUS) and confirmation of over half of original study predictions in retrospective analyses.

Methodology:

Performs TAD-based analysis; integrates gene functional annotations and dosage sensitivity; computes phenotype similarity scores and CNV scoring metrics; and identifies potential position effect events.