Characterization and comparison of human nuclear and cytosolic editomes.

摘要: We developed a robust computational statistical framework to identify RNA editing events from RNA-Seq data with high specificity. Our approach handles several outstanding challenges of genome-wide analyses, including the effect on read alignment and utilization redundant reads. By applying this framework, we characterized nuclear cytosolic editomes seven human cell lines. found that 93.8–99.2% are A-to-G (or A-to-I). Nuclear transcriptomes contain many more than transcriptomes. Most sites exhibiting nucleus-specific in introns or novel intergenic transcripts preferentially localized nucleus regardless their status, arguing against role retention. In contrast, exhibit cytosol-specific show comparable expression, suggesting differential subcellular compartmentalization edited unedited alleles. is globally associated modification microRNA regulation 3′ untranslated regions, whereas coding regions rare tend be synonymous. Interestingly, at derived alleles lineage tends result reversion back ancestral forms level. This suggests can mediate memory evolutionary time-scales maintain genetic information.