TADeus2

TADeus2 evaluates the pathogenicity of structural variants (SVs) that disrupt three-dimensional chromatin architecture to support interpretation of SVs in clinical and research genomics.

Key Features:

• SV pathogenicity assessment: Quantifies and ranks the pathogenic potential of structural variants affecting chromatin organization.
• 3D chromatin and TAD analysis: Assesses impacts of SVs on topologically associating domains (TADs) and chromatin conformation.
• Hi-C matrix analysis: Utilizes Hi-C matrices to evaluate chromatin contact changes associated with SVs.
• Flanking region evaluation: Evaluates effects of SVs on flanking coding and non-coding genomic regions.
• Breakpoint and genome-context perspectives: Analyzes SVs from a continuous genome perspective and via rearrangement breakpoints.
• Integration with external classifiers: Integrates outputs from TADA (Topologically Associating Domain Analysis) and ClassifyCNV for variant interpretation.
• Sampling-based P-value estimation: Applies a sampling-based P-value approach to support statistical assessment of pathogenicity.

Scientific Applications:

• Clinical pathogenicity interpretation: Support assessment of clinical significance for structural variants in diagnostic genomics.
• Non-coding variant interpretation: Interpret SVs in non-coding regions by linking genomic rearrangements to 3D chromatin alterations.
• Prioritization of SVs: Rank structural variants for follow-up based on combined chromatin-impact and classifier-derived scores.
• TAD disruption studies: Investigate how SVs alter topological domain boundaries and chromatin folding.
• Integration of multi-source evidence: Combine Hi-C-derived chromatin changes with TADA and ClassifyCNV outputs for comprehensive evaluation.

Methodology:

Analyzes Hi-C matrices to assess chromatin conformation and TAD disruption, evaluates effects on flanking coding and non-coding regions from continuous genome and rearrangement-breakpoint perspectives, and quantifies and ranks pathogenicity by integrating TADA and ClassifyCNV outputs with a sampling-based P-value approach.