No evidence for ac4C within human mRNA upon data reassessment.

ABSTRACT

Cytidine acetylation (ac4C) of RNA is a post-transcriptional modification catalyzed by Nat10. Recently, an approach termed RedaCT was employed to map ac4C in human mRNA, relying on detection of C>T mutations in WT but not in Nat10-KO cells. RedaCT suggested widespread ac4C presence. Here, we reanalyze RedaCT data. We find that mismatch signatures are not reproducible, as C>T mismatches are nearly exclusively present in only one of two biological replicates. Furthermore, all mismatch types-not only C>T-are highly enriched in WT samples, inconsistent with an acetylation signature. We demonstrate that the originally observed enrichment in mutations in one of the WT samples is due to its low complexity, resulting in the technical amplification of all classes of mismatch counts. Removal of duplicate reads abolishes the skewed mismatch patterns. These analyses account for the irreproducible mismatch patterns across samples while failing to find evidence for acetylation of RedaCT sites.