Abstract
RNA N6-methyladenosine (m6A) methylation is a major epigenetic modification that plays a critical role in regulating gene expression in tumors. Although the regulation of individual genes by m6A methylation has been extensively studied, a systematic quantification of transcriptome-wide associations between RNA methylation and gene expression remains limited. In this study, we analyzed publicly available MeRIP-seq and RNA-seq datasets of paired colorectal cancer (CRC) and adjacent normal tissues from four patients, proposing a statistical model to quantify the cis-regulation between m6A methylation and gene expression in CRC. The results indicated that (1) A total of 46,500 and 31,715 unique m6A peaks were identified in CRC and normal control (NC) tissues, respectively. Compared with NC tissues, 538 genes were upregulated and 3,944 were downregulated in CRC tissues (padj <0.05 and |log(2)FC| > 1). (2) Approximately 66.01% of m6A peaks in CRC are located within genes and 28.78% in promoters, compared to 65.00% and 28.38%, respectively, in NC tissues. CRC tissues exhibited higher methylation levels in exons and 3'UTRs, while NC tissues showed increased methylation in introns. (3) 451 genes exhibited significant cis-regulation between RNA methylation and gene expression. Among these, 371 genes were positively correlated, indicating a promotive effect on gene expression, while 80 genes showed negative correlation. Moreover, 34 genes showing strong correlations (r (2) ≥ 0.9) were identified, including 16 genes previously reported to be associated with CRC. This study provides a transcriptome-wide strategy for quantifying the association between RNA methylation and gene expression in CRC, offering new insights into the potential regulatory roles of RNA methylation in tumor biology.