Comparison of tools for identification of rRNA Modifications in Escherichia coli using Oxford Nanopore direct RNA Sequencing

RNA modifications are known to play roles in gene expression, RNA life cycle. However, detection of these modifications is difficult. In most cases, RNA modifications are identified using mass spectrometry, enzymatic pulldown, or chemical conversion, followed by short-read sequencing. On a transcriptome-wide basis, identifying RNA modifications with these methods would be incredibly laborious and difficult. However, with the advent of long-read sequencing technologies, especially Oxford Nanopore Technologies' (ONT) direct RNA sequencing, there is now potential to detect modifications in RNA without any preprocessing. Several tools have been developed to detect RNA modifications from direct RNA sequencing data. Although there have been few systematic comparisons of their performance, Here we test the sensitivity and specificity of four of these tools to identify modifications in the ribosomal RNA (rRNA) of Escherichia coli. We find that for nanoDoc and tombo, the accuracy/areas under ROC/AUROC increases with higher coverage while for other tools, increase in coverage did not increase AUROC.