m6A-seq Data Analysis Protocol

Data analysis was performed as described (Liu et al., 2020 (link)). Briefly, raw reads were trimmed with Trimmomatic-0.39 (Bolger et al., 2014 (link)) and then aligned to the mouse genome (mm10) and transcriptome (GENCODE, version M19) together with spike-in genomes including unmodified control RNA (Cypridina Luciferase) and m⁶A methylated control RNA (Gaussia Luciferase) (New England Biolabs) using HISAT (version 2.1.0) (Kim et al., 2015 (link)) with “-k 5 --rna-strandness RF” parameters. Mapped reads were separated by strands using samtools (version 1.9) (Li et al., 2009 (link)) and m⁶A peaks were called using MACS2 (version 2) (Zhang et al., 2008 (link)) with parameter “-g 1.3e8 --tsize 150 --extsize 150 --nomodel --keep-dup 5” for each strand separately. Significant peaks with q < 0.01 identified by MACS2 were considered. Peaks identified in at least two biological replicates were intersected using bedtools (v.2.26.0) (Quinlan and Hall, 2010 (link)) and were used in the following analysis. The number of reads mapped to mouse genome divided by number of reads mapped to m⁶A modified spike-in represented whole m⁶A level.