A Long-Context RNA Foundation Model for Predicting Transcriptome Architecture

Explore a groundbreaking talk on LoRNASH, an innovative RNA foundation model that predicts transcriptome architecture using large-scale single-molecule transcriptome sequencing and advanced machine learning techniques. Discover how this model tackles the challenge of linking DNA sequences to genomic function by learning how pre-mRNA nucleotide sequences determine the relative abundances and molecular structures of mRNA isoforms. Learn about the StripedHyena architecture that enables LoRNASH to process extremely long sequence inputs of up to 65 kilobase pairs, allowing for quantitative, zero-shot predictions of various aspects of transcriptome architecture. Gain insights into the model's potential applications in RNA biotechnology and its use as a foundation for fine-tuning transcriptome-related downstream prediction tasks, including cell-type specific gene expression, splicing, and general RNA processing. Access the related research paper and connect with the speaker and community through the provided links to further explore this cutting-edge development in genetics and genomics.

A long-context RNA foundation model for predicting transcriptome architecture | Ali Saberi