RNA modifications such as m6A methylation form an additional layer of complexity in the transcriptome. Nanopore direct RNA sequencing can capture this information in the raw current signal for each RNA molecule, enabling the detection of RNA modifications using supervised machine learning. In this presentation I will introduce m6Anet, a neural-network-based method that leverages the multiple instance learning framework to obtain read-level and site-level m6A modification probabilities. The m6Anet method outperforms existing computational methods, shows similar accuracy as experimental approaches, captures the underlying read-level stoichiometry, and generalizes with high accuracy to different cell lines and species. Finally, I will provide an update on m6A identification using RNA004 with m6Anet.
10 окт 2023
