A hybrid deep learning-based method for rice protein lysine acylation (PLA) site prediction using both sequence and structure information
As a main subtype of post-translational modification (PTM), protein lysine acylations (PLAs) play crucial roles in regulating diverse functions of proteins. With recent advancements in proteomics technology, the identification of PTM is becoming a data-rich field. A large amount of experimentally verified data is urgently required to be translated into valuable biological insights.
The developed hybrid deep learning framework integrates both sequence-based and structure-based information for lysine acylation site prediction. First, we incorporated traditional sequence-based feature encoding schemes (i.e., word embedding, one-hot, and BLOSUM62 encoding) to encapsulate evolutionary and contextual relationships, alongside atomic structural descriptors extracted from AlphaFold structural models. The framework consists of three core modules: a multi-track encoder for sequence and structural features, a graph neural network-based structure encoder to capture spatial interactions, and a decoder with attention and MLP for binary classification. RicePLA effectively leverages both evolutionary constraints and 3D structural information to achieve accurate predictions of multitype lysine acylations.
