Skip to main content
Fig. 3 | Genome Biology

Fig. 3

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. 3

Extracted native training data and trained a neural network model. a Flowcharts illustrating the classification of modified and unmodified native reads and the procedures of training DENA. b The ROC curves of DENA in the 12 consensus sequences from “RRACH” motif. c Bar plots demonstrate the reduction of m6A rates identified by DENA at the ACTB-1217 site from direct RNA-Seq data of wild-type and METTL3-knockout human cells, respectively. d Venn diagram of the number of m6A sites identified by DENA from these sites that are supported at least 50 reads in all replicates of wild-type A. thaliana, respectively. e The cross-correlation coefficient of m6A rates identified by DENA among four replicates of wild-type A. thaliana

Back to article page

Genome Biology

ISSN: 1474-760X

Contact us