Fig. 6: m6A identification in 7SK RNA.
From: RNA modifications detection by comparative Nanopore direct RNA sequencing
A On the left, the secondary structure of 7SK showing positions of known protein binding sites and structural conservation. On the right, the secondary structure of 7SK with the Nanocompore p-value (METTL3-KD vs WT, GMM-logit test) overlaid as a colour scale. For each nucleotide the colour indicates the lowest p-value among those of the 5 kmers that overlap it. Only p-values<0.01 are shown in colour. B m6A profile of 7SK, showing the Nanocompore GMM-logit p-value (y axis, -log10) across the transcript length. C Scatter plot showing the scaled median intensity vs the scaled log10 dwell time for each read covering kmer 41 of 7SK. Each point shows data for a distinct read colour coded according to the sample. The contour lines show the kernel density estimates for the two samples. For visualisation purposes the x- and y- axis are truncated at -4 and +3 respectively. D Violin plots showing the distributions of median intensity (top) and scaled log10 dwell time (bottom) for the Hexim1 binding sites and neighbouring kmers. All coordinates refer to the first nucleotide of each kmer relative to ENST00000636484. The cross mark indicates the intensity and dwell time value of the kmer according to the unmodified model. E m6A RIP-qPCR results in three non-overlapping regions of 7SK in WT and METTL3 KD MOLM13 cells. Bars show the mean of 6 independent experiments. Vertical bars show the standard error of the mean. The p-values were calculated using a one-sided Welch’s t-test. Full uncropped scans of Western Blots confirming METTL3 KD are shown in Figure S14.
