Integrating mass spectrometry with Nanopore direct RNA sequencing for de novo modification profiling of bacteriophage MS2

RNA modifications contribute to the regulatory and structural complexity of RNA molecules, influencing key biological processes. Nanopore direct RNA sequencing (DRS) enables the detection of these modifications at single-molecule resolution without chemical conversion. Oxford Nanopore Technologies recently introduced RNA004 sequencing chemistry and the Dorado basecaller, which improves accuracy and enables the identification of eight RNA modifications. However, the reliability of these predictions requires careful validation using orthogonal approaches. Here, we profiled RNA modifications in the MS2 bacteriophage genome using both LC-MS/MS and Nanopore DRS. LC-MS/MS analysis revealed that Ψ is the predominant modification, present at approximately one site per RNA molecule, while all other modifications were found at low levels (< 0.04 modifications per RNA). In contrast, Dorado-based modification calling not only confirmed Ψ but also predicted multiple m ⁶ A, m ⁵ C, 2’OMe, and inosine sites not supported by LC-MS/MS data. To refine modification calling, we used in vitro transcribed (IVT) RNA as a negative control and subtracted IVT-derived false positive rates from native RNA predictions. This adjustment reduced overcalling and improved confidence in site-specific predictions. These data demonstrate that while Dorado can detect RNA modifications de novo , its predictions require careful filtering and validation. Our studies of provide a rare benchmark dataset for assessing the ability of transcriptome-wide methods to accurately identify RNA modifications and estimate modification stoichiometry. These findings support the use of orthogonal approaches, including LC-MS/MS and IVT controls, alongside Nanopore sequencing to provide a more reliable and interpretable strategy for studying RNA modification patterns.