Skip to main content
Fig. 4 | Genome Biology

Fig. 4

From: DENA: training an authentic neural network model using Nanopore sequencing data of Arabidopsis transcripts for detection and quantification of N6-methyladenosine on RNA

Fig. 4

The m6A identification for A. thaliana and human using DENA. a Venn diagram shows the common m6A sites identified by DENA among Col-0, VIRc and vir-1 samples. b Jointplot shows the correlation of m6A rates at 28299 overlapped sites between Col-0 and vir-1. c Jointplot shows the correlation of m6A modification rate at 35,048 overlapped sites between Col-0 and VIRc. d Venn diagram shows the common m6A sites identified by DENA from wild-type (WT) and METTL3-knockout (M3KO) human cells. e The m6A distribution on transcripts in WT and M3KO, respectively. f Jointplot shows the correlation of m6A rates at 7036 overlapped sites between WT and M3KO. g Pie charts display the performance of DENA on 4201 miCLIP-identified m6A sites that are supported at least 50 reads, and 3768 sites out of them are regarded as m6A modification sites by DENA. h Flowcharts illustrating the identification of m6A sites using Bst 2.0 DNA polymerase Splint ligase with qPCR. Dp and Up are down probe and up probe, respectively. i The real-time fluorescence curves produced at Site_7340 and Site_7207 on the mRNA of PRP8A, respectively. A is Site_7207 and m6A is Site_7340 on PRP8A. j Bar plot demonstrates consistency of m6A rate at the four known modified sites between DENA detection and SCARLET identification, and the significant overestimation in the prediction from Nanom6A

Back to article page