Revealing the epitranscriptomic landscape of m6A using deep neural networks
- LecturerMr. Bo-Yi Yu (The University of Tokyo, Department of Advanced Interdisciplinary Studies)
Host: Chung-Yen Lin - Time2023-01-06 (Fri.) 10:30 ~ 12:00
- LocationAuditorium 106 at IIS new Building
Abstract
N6-methyladenosine (m6A) has been one of the most abundant and well-known modifications in mRNA since its discovery in 1970s. Recent studies have demonstrated that m6A gets involved in various biological processes such as alternative splicing and RNA degradation, playing an important role in all kinds of diseases. To have a better understanding of the role of m6A, transcriptome-wide m6A profiling data is indispensable. In these years, the third-generation Oxford Nanopore Technology (ONT) direct RNA sequencing (dRNA-seq) platform has shown promise in RNA modification detection based on current disruptions measured in transcripts. However, decoding the current intensity data to m6A modification profiles remains a challenge.
Here, we developed a novel deep-learning based approach for m6A detection at single-nucleotide resolution by using dRNA-seq. We demonstrated that our model outperformed other Nanopore-based m6A detection tools by using different validation datasets. In addition, we showed that our tool can be applied to transcriptome-wide m6A detection in several human cell lines. The long-term goal is to depict the m6A profiles of patients and pave the way for precision medicine in epitranscriptomics.
Here, we developed a novel deep-learning based approach for m6A detection at single-nucleotide resolution by using dRNA-seq. We demonstrated that our model outperformed other Nanopore-based m6A detection tools by using different validation datasets. In addition, we showed that our tool can be applied to transcriptome-wide m6A detection in several human cell lines. The long-term goal is to depict the m6A profiles of patients and pave the way for precision medicine in epitranscriptomics.