SpliceGPT is a transformer-based deep learning model designed to analyze and predict alternative exon splicing events. By leveraging the power of transformers, SpliceGPT aims to capture the complex regulatory interactions that dictate exon inclusion or exclusion, providing a comprehensive understanding of splicing mechanisms and their implications for disease and therapeutic development.
Transformer-Based Architecture: Uses self-attention mechanisms to process and understand sequential RNA data.
Predictive Modeling: Identifies alternative splicing events across various tissues and conditions.
Alternative exon splicing is a post-transcriptional process that enables a single gene to produce multiple mRNA transcripts by selectively including or excluding certain exons. This increases proteomic diversity and plays a crucial role in development, cellular differentiation, and adaptation to environmental changes. Misregulation of splicing is associated with diseases such as cancer, neurodegenerative disorders, and genetic syndromes.
Winter 2025 Drexel AI project