A deep learning approach to automate whole‐genome prediction of diverse epigenomic modifications in plants

Epigenetic modifications function in gene transcription, RNA metabolism, and other biological processes. However, multiple factors currently limit the scientific utility of epigenomic datasets generated for plants. Here, using deep-learning approaches, we developed a Smart Model Epigenetics Plants (SMEP) to predict six types modifications: DNA 5-methylcytosine (5mC) N6-methyladenosine (6mA) methylation, (m6A) three histone modification. Using from japonica rice Nipponbare, SMEP achieved 95% prediction accuracy 6mA, also around 80% 5mC, m6A, modification based on 10-fold cross-validation. Additionally, > 6mA peaks detected after heat-shock treatment were predicted. We successfully applied examining indica 93-11 even B73 maize line. Taken together, show that deep-learning-enabled can reliably mine diverse plants yield actionable insights about sites. Thus, our work opens new avenues application predictive tools facilitate functional research, will almost certainly increase efficiency genome engineering efforts.