PIVOTAL

PIVOTAL prioritizes variants of uncertain significance (VUS) using spatial genomic patterns in three-dimensional protein structures to improve pathogenicity prediction and candidate disease association at proteome scale.

Key Features:

• Spatial neighborhood-based analysis: Applies a spatial neighborhood-based method on 3D protein structures to assess the structural context of variants.
• Integration of known disease mutations: Utilizes the spatial distribution of known disease mutations on three-dimensional protein models.
• Evolutionary conservation: Incorporates evolutionary conservation data to inform variant assessment.
• Co-evolution data: Integrates co-evolution data alongside conservation and mutation distribution.
• Proteome-wide prioritization: Operates at a proteome-wide scale to prioritize variants across many proteins.
• Large-scale pathogenicity predictions: Has been applied to generate pathogenicity predictions for over 140,000 VUS.
• Candidate disease etiology inference: Aids identification of potential disease etiologies associated with specific candidate variants.

Scientific Applications:

• VUS prioritization: Prioritizes variants of uncertain significance based on three-dimensional structural context.
• Pathogenicity prediction: Provides pathogenicity assessments across proteome-scale variant datasets.
• Candidate disease association inference: Infers potential disease etiologies for candidate variants, including in genes linked to multiple distinct diseases.
• Large-scale genomic studies: Supports large-scale genomic analyses for variant interpretation and discovery.

Methodology:

PIVOTAL uses a spatial neighborhood-based method on three-dimensional protein models that integrates the spatial distribution of known disease mutations with evolutionary conservation and co-evolution data and has been applied proteome-wide to predict pathogenicity for over 140,000 VUS.