Decoding the epitranscriptional landscape from native RNA sequences

Abstract

Traditional epitranscriptomics relies on capturing a single RNA modification by antibody or chemical treatment, combined with short-read sequencing to identify its transcriptomic location. This approach is labor-intensive and may introduce experimental artifacts. Direct sequencing of native RNA using Oxford Nanopore Technologies (ONT) can allow for directly detecting the RNA base modifications, although these modifications might appear as sequencing errors. The percent Error of Specific Bases (%ESB) was higher for native RNA than unmodified RNA, which enabled the detection of ribonucleotide modification sites. Based on the %ESB differences, we developed a bioinformatic tool, epitranscriptional landscape inferring from glitches of ONT signals (ELIGOS), that is based on various types of synthetic modified RNA and applied to rRNA and mRNA. ELIGOS is able to accurately predict known classes of RNA methylation sites (AUC > 0.93) in rRNAs from Escherichiacoli, yeast, and human cells, using either unmodified in vitro transcription RNA or a background error model, which mimics the systematic error of direct RNA sequencing as the reference. The well-known DRACH/RRACH motif was localized and identified, consistent with previous studies, using differential analysis of ELIGOS to study the impact of RNA m6A methyltransferase by comparing wild type and knockouts in yeast and mouse cells. Lastly, the DRACH motif could also be identified in the mRNA of three human cell lines. The mRNA modification identified by ELIGOS is at the level of individual base resolution. In summary, we have developed a bioinformatic software package to uncover native RNA modifications.

RNA sequencing: new technologies and applications in cancer research

“…This shortcoming can be overcome by directly sequencing native RNA molecules using methods such as nanopore sequencing. Transcript modifications could be inferred from the current signal as the modified RNA molecules passing nanopore cause a characteristic temporary current blockade, which enables the detection of diverse modifications such as m6A or 5-methylcytosine (m5C) [196][197][198].…”

Analysis of RNA Modifications by Second- and Third-Generation Deep Sequencing: 2020 Update

“…The bioinformatic tool, called Epitranscriptional Landscape Inferring from Glitches of ONT signals (ELIGOS), was trained on various types of synthetic modified RNA and applied to rRNA and mRNA sequencing. ELIGOS is able to accurately predict known classes of RNA methylation sites (AUC > 0.93) in rRNAs from E. coli, yeast, and human cells [158]. Model-based base calling from ionic current signal levels is certainly required for reliable analysis [159].…”

ModPhred: an integrative toolkit for the analysis and storage of nanopore sequencing DNA and RNA modification data

“…While many tools have been developed in the recent years to detect DNA and RNA modifications from nanopore sequencing datasets (Yuen et al , 2020; Stoiber et al , 2017; Ni et al , 2019; Leger et al , 2019; H. Liu et al , 2019; Jenjaroenpun et al , 2021; Q. Liu, Georgieva, et al , 2019; Pratanwanich et al , 2020; Begik et al , 2021), there are limited tools allowing retrieval, storage, manipulation and visualisation of modification information (De Coster et al , 2020; Leger, 2020).…”