<p>As one of the most common and abundant post-transcriptional modifications, N⁶-methyladenosine (m⁶A) has been extensively studied for its essential regulatory role in gene expression and cell functions. The location of m⁶A RNA modification sites, however, remains a challenging problem, because of the inability to characterize m⁶A modified sites at a multi-scale level in their native RNA context. Here, we introduce an interpretability-guided invertible neural network (m⁶A-IIN), a deep learning model to accurately identify m⁶A RNA modification sites by integrating both primary and secondary structure information under an invertible coupling framework. Compared to existing methods, m⁶A-IIN achieves state-of-the-art performance in the prediction of m⁶A RNA modification sites across 11 benchmark datasets collected from different species and tissues. Furthermore, we find evidence indicating high consistency in methylation-related regions between primary and secondary structure of RNA, providing novel insights into m⁶A biology from the phylogenetic perspective. By analyzing conserved methylation-related regions identified by m⁶A-IIN across tissues, m⁶A-IIN facilitates the identification of novel pan-cancer genes, providing valuable contributions to cancer biology. Our results underscore the interpretability and predictive accuracy of m⁶A-IIN, opening an avenue towards the understanding of m⁶A RNA modification mechanisms.</p>