SAI-10k-calc

SAI-10k-calc predicts splicing aberrations and their sequence- and frame-level consequences for genetic variants by extending SpliceAI analysis to a 10 kilobase (kb) window.

Key Features:

• SpliceAI integration: Leverages SpliceAI maximum delta scores to assess variant impact on splicing.
• 10 kilobase analysis window: Evaluates variants within a 10 kilobase (kb) region to detect distal splicing effects.
• Comprehensive aberration prediction: Predicts pseudoexonization, intron retention, partial exon deletion, and multi-exon skipping.
• Sequence and frame impact analysis: Predicts sizes of inserted or deleted sequences, determines reading-frame consequences, and infers resultant amino acid sequence changes.
• Automated amino acid prediction: Generates altered amino acid sequences and flags variants likely to cause nonsense-mediated decay or truncated proteins.
• Performance validation: Reported sensitivity of 95% and specificity of 96% based on validation with 1,212 single-nucleotide variants (SNVs) with curated splicing assay results.
• Customizable thresholds: Allows adjustment of default scoring thresholds to target specific performance trade-offs.

Scientific Applications:

• Variant interpretation in genomic medicine: Inform pathogenicity assessment by predicting functional consequences of variants on mRNA splicing and protein sequence.
• Splicing disorder research: Identify and characterize splicing aberrations such as pseudoexonization and partial intron retention in disease-associated genes.
• Molecular consequence analysis: Evaluate effects on reading frame, amino acid sequence, and likelihood of nonsense-mediated decay or protein truncation for experimental follow-up.

Methodology:

SAI-10k-calc leverages SpliceAI’s maximum delta score approach and expands the analysis window to 10 kilobases (kb) around variants to predict splicing aberrations and downstream sequence/frame consequences.