Skip to main content
. Author manuscript; available in PMC: 2022 Nov 1.
Published in final edited form as: Genomics. 2021 Sep 23;113(6):3864–3871. doi: 10.1016/j.ygeno.2021.09.016

Figure 1. EditPredict model schema. The two left panels with light grey shading correspond to the training (top) and application (bottom) processes, respectively.

Figure 1.

A. Approximately, flanking sequences of ~4.6 million known RNA editing sites were extracted as the positive set, and matched negative sites were constructed. One-hot encoding for a segment of 11 nucleotides was illustrated. B. The CNN architecture handles bi-directional long flanking sequences of potential A-to-I editing sites. The number of layers and the number of neurons per layer are depicted. The convolution process has been omitted for clarity. C. An example of a personalized genome approach. Multiple flanking sequences for a real subject may be generated based on mutation and INDELs.