CAPICE

CAPICE predicts and prioritizes pathogenic single nucleotide variants (SNVs) and short insertions or deletions (InDels) in clinical exome sequencing data using a consequence-agnostic pathogenicity scoring approach to identify Mendelian disease-causing variants.

Key Features:

• Consequence-agnostic scoring: Generates pathogenicity scores independent of variant consequence type to enable cross-consequence prioritization.
• Machine-learning-based classification: Employs a machine-learning approach to distinguish pathogenic from benign variants in exome data.
• Targets SNVs and short InDels: Specifically addresses single nucleotide variants (SNVs) and short insertions and deletions (InDels) detected in clinical exome sequencing.
• Performance versus established predictors: Demonstrates improved performance compared with general and consequence-type-specific predictors such as CADD, GAVIN, REVEL, and ClinPred.
• Prioritization of rare and ultra-rare variants: Effectively prioritizes rare and ultra-rare genetic variation for downstream interpretation.

Scientific Applications:

• Clinical variant interpretation: Prioritizes candidate pathogenic variants for Mendelian disease diagnosis from clinical exome sequencing data.
• Research on rare disease genetics: Facilitates identification of rare and ultra-rare variants for follow-up studies in genetic research.
• Benchmarking of pathogenicity predictors: Serves as a comparator in performance evaluations against tools like CADD, GAVIN, REVEL, and ClinPred.
• Variant triage for diagnostic pipelines: Enables ranking of exome variants to support downstream expert review and validation workflows.

Methodology:

Uses a machine-learning-based, consequence-agnostic approach to compute pathogenicity scores for SNVs and short InDels in clinical exome sequencing data.