RNA editing in the human ENCODE RNA-seq data.

Area of Science:

• Genomics
• Molecular Biology
• Bioinformatics

Background:

• RNA sequencing (RNA-seq) allows identification of sequence differences from reference genomes.
• Distinguishing RNA editing from genomic variations like single nucleotide polymorphisms (SNPs) is crucial.
• Previous studies have explored RNA editing events across various tissues.

Purpose of the Study:

• To analyze ENCODE Project RNA-seq data for candidate RNA editing events across 14 human cell lines.
• To characterize the types, locations, and confidence levels of RNA variants.
• To investigate the biological significance and reproducibility of RNA editing events.

Main Methods:

• Analysis of deep RNA-seq data from 14 human cell lines.
• Identification and filtering of RNA sequencing variants not present in dbSNP.
• Utilized genome resequencing and ChIP-seq data to exclude genomic variations.
• Applied Gene Ontology enrichment analysis.

Main Results:

• Approximately 43% of non-dbSNP RNA variants within gene boundaries are A-to-G(I) RNA editing candidates.
• Most A-to-G(I) edits occur in introns and 3' UTRs; only 123 are in coding sequences.
• After filtering, up to 85% of high-confidence variants are A-to-G(I) editing candidates.
• Genes with A-to-G(I) edits are enriched in terms related to cell division, viral defense, and translation.
• Non-A-to-G variants resemble SNPs, and reproducible nonsynonymous A-to-G(I) edits were not found in lymphoblastoid cell lines.

Conclusions:

• A-to-G(I) RNA editing is prevalent in human cell lines, particularly in non-coding regions.
• Genomic variation filtering is essential for accurate RNA editing identification.
• The study suggests transcript-level editing regulation is more significant than individual site editing, with implications for gene function.