The landscape of A-to-I RNA editome in mouse embryonic development

A-to-I editing, as a post-transcriptional modification process mediated by ADAR, plays a crucial role in many biological process in metazoans.The study of RNA editing in mouse early embryonic development can understand the dynamic changes of RNA editing during mouse early embryonic development, and further reveal the role of RNA editing in mouse early embryo development.We found that as embryonic developing, RNA editome in mice is getting more and more. Besides, RNA editome in mouse embryonic development is stage specified. We collected 29 mouse embryonic samples RNA-seq from GSE71434 and GSE66582. Then we identified RNA editome by three identification method. We integrated these RNA editome to eight stages and record their position, reference, alter, strand and RNA editing level. mRNA-seq reads were aligned to the mm10 genome assembly using STAR(version 2.5.2b) with default parameter. Duplicated reads that mapped to the same location were removed using Picard(version 1.141)(https://broadinstitute.github.io/picard/) with default parameter. Reads with a mapping quality <20 were removed using samtools(version 1.3.1) with parameter '-q 20'. Variants were called using the GATK(version 3.5.0) HaplotypeCaller tool with options stand_call_conf of 20 and stand_emit_conf of 0. Candidate RNA editing site were called using the following filtering steps: (1) all variants present in dbSNP (except SNPs of molecular type “cDNA”, database version 137, http://www.ncbi.nlm.nih.gov/SNP/); (2) variants located in the first 6 bp from either end of a sequence read were discarded; (3) in non-Alu regions, we removed sites in simple repeats according to RepeatMasker annotation, discarded intronic candidates if they were located within 4 bp of all known splicing junctions according to Ensembl gene annotations, and removed sites in homopolymer runs of ≥5 bp; (4) variant sites in regions of high similarity to other parts of the genome were discarded using the BLAST-like alignment tool (BLAT); (5) the editing type of each site was inferred based on the strand of overlapping annotated genes and only A-to-I base changing editing sites were retainned. Tab-delimited text files include basic information of identified A-to-I editing sites for each sample; basic information includes chromosome, position, strand, reference base, altered base, edting level of each A-to-I editing site.