Dihydrouridine (D) is a modified pyrimidine nucleotide universally found in viral, prokaryotic and eukaryotic species. The modification can serve as a metabolic modulator for various pathological conditions, and its elevated levels in tumors are associated with a series of cancers. Precise identification of D sites on RNA is vital for understanding its biological function. Previous studies proposed computational methods to predict D-sites in tRNA. Here we present DPred, the first computational method for predicting D modification in S.cerevisiae mRNA. The sequence data was encoded by the combination of nucleotide chemical property and nucleotide density, which outperformed three other encoding schemes. The model framework was mainly built upon the additive local self-attention and convolutional neural network architecture.
