Fig. 1From: A deep learning method to more accurately recall known lysine acetylation sitesThe computational framework of the predictor. Step 1, a peptide of the length of 31 with a center lysine (K) was used to extract sequences from the acetylated proteins. Step 2, six different encoding schemes that are described in Section 2.2 were utilized to encode fragments. Step 3, these six groups of encoded features were used to the train model in two ways. Step 4, the predicted results of the samplesBack to article page