Transfer learning enables identification of multiple types of RNA modifications using nanopore direct RNA sequencing

• Published in: Nature communications Vol. 15; no. 1; pp. 4049 - 19
• Main Authors: Wu, You, Shao, Wenna, Yan, Mengxiao, Wang, Yuqin, Xu, Pengfei, Huang, Guoqiang, Li, Xiaofei, Gregory, Brian D., Yang, Jun, Wang, Hongxia, Yu, Xiang
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 14.05.2024; Nature Publishing Group; Nature Portfolio
• Subjects: 49/91; 631/114/1305; 631/114/1314; 631/208/514/1949; 631/337/1645/2570; 631/61/212/2019; Cell lines; Datasets; Deep Learning; Gene sequencing; Humanities and Social Sciences; Humans; Identification methods; Inosine - genetics; Inosine - metabolism; multidisciplinary; N6-methyladenosine; Nanopore Sequencing - methods; Nanopores; Oryza - genetics; RNA - genetics; RNA - metabolism; RNA modification; RNA Processing, Post-Transcriptional; Science; Science (multidisciplinary); Sequence Analysis, RNA - methods; Transcriptome - genetics; Transfer learning
• ISSN: 2041-1723; 2041-1723
• DOI: 10.1038/s41467-024-48437-4

Nanopore direct RNA sequencing (DRS) has emerged as a powerful tool for RNA modification identification. However, concurrently detecting multiple types of modifications in a single DRS sample remains a challenge. Here, we develop TandemMod, a transferable deep learning framework capable of detecting multiple types of RNA modifications in single DRS data. To train high-performance TandemMod models, we generate in vitro epitranscriptome datasets from cDNA libraries, containing thousands of transcripts labeled with various types of RNA modifications. We validate the performance of TandemMod on both in vitro transcripts and in vivo human cell lines, confirming its high accuracy for profiling m 6 A and m 5 C modification sites. Furthermore, we perform transfer learning for identifying other modifications such as m 7 G, Ψ, and inosine, significantly reducing training data size and running time without compromising performance. Finally, we apply TandemMod to identify 3 types of RNA modifications in rice grown in different environments, demonstrating its applicability across species and conditions. In summary, we provide a resource with ground-truth labels that can serve as benchmark datasets for nanopore-based modification identification methods, and TandemMod for identifying diverse RNA modifications using a single DRS sample. Simultaneous profiling of multiple RNA modifications is a promising yet understudied field of research. Here, authors develop a transferable deep learning framework capable of detecting multiple types of RNA modifications in single nanopore sequencing sample.