Fig. 4From: DeepM6ASeq: prediction and characterization of m6A-containing sequences using deep learningSignificant learned motifs (E-value < 0.01) in human, mouse and zebrafish. The learned motifs from the first CNN layer of each species model are aligned with known motifs by means of TOMTOM. For each aligned result, the upper panel is the known motif, while the bottom panel is the learned motif. The names of known motifs and the significant scores (E-value) are shown on the sideBack to article page