Pangolin

Pangolin predicts splice site strength from DNA sequences using deep learning to assess effects of genetic variants on RNA splicing across multiple tissues.

Key Features:

• Deep Learning Model: Employs a deep learning architecture trained to predict splice site strength from DNA sequences across multiple tissues.
• Variant Impact Prediction: Predicts the impact of genetic variants, including common, rare, and lineage-specific variants, on RNA splicing.
• Loss-of-Function Mutation Identification: Identifies loss-of-function mutations with high accuracy and recall, including splice-disrupting variants not classified as missense or nonsense.

Scientific Applications:

• Variant interpretation in genomics and personalized medicine: Quantifies splice-altering effects of variants to inform molecular mechanisms and clinical variant interpretation.
• Identification of pathogenic loss-of-function variants: Supports discovery of novel pathogenic variants by detecting splice-disrupting loss-of-function mutations.
• Study of tissue-specific splicing: Enables analysis of splice site strength changes across multiple tissues to investigate tissue-specific splicing regulation.

Methodology:

Processes VCF or CSV files of variants, predicts changes in splice site strength for each variant, outputs results in the same file format, and applies models to custom DNA sequences.