Fig. 2

AStruct is valuable in assisting with the identification of functional variants. a, b The cumulative distribution curve of the phastCon100 score for different groups. a for AStrcut, b for RNAsnp. c The bubble plot shows the comparison of the number of ASRS SNPs annotated as allele-specific RBP binding events between different groups. d A G-to-A SNP (rs74488735) in the ‘High’ group shows allele-specific pattern of a PPIG binding site in PXD1 of HepG2 cell line. It is illustrated by the different proportion of reads count covering each allele in the eCLIP-Seq data and the control RNA-Seq data (left), and the significant structure difference with each allele (right). e The bubble plot shows the comparison of the number of ASRS SNPs annotated as m6A related variants between different groups. f A A-to-G SNP (rs3820677) in the ‘High’ group shows allele-specific pattern of a m6A modification site in KIRREL1 of HeLa cell line. It is illustrated by the different proportion of reads count covering each allele in the MeRIP-Seq IP data and the control Input data (left), and the significant structure difference with each allele (right). g, h The cumulative distribution curve of the FATHMM-XF score for different groups. g For AStrcut, h for RNAsnp. i The bubble plot shows the comparison of the number of disease-related ASRS SNPs annotated by ClinVar between different groups. The P value of cumulative distribution curve was calculated using Kolmogorov–Smirnov (KS) test. The P value of bubble plot was calculated using Fisher’s exact test