MLapSVM-LBS: Predicting DNA-binding proteins via a multiple Laplacian regularized support vector machine with local behavior similarity

DNA-binding proteins (DBPs) are of great significance in many basic cellular processes. Experiment-based methods for identifying DBPs costly and time-consuming. To deal with large-scale DBP identification tasks, a variety computation-based have been developed. Inspired by previous work, we propose multiple Laplacian regularized support vector machine local behavior similarity (MLapSVM-LBS) to predict DBP. We serially combine three features that extracted from protein sequences (including PsePSSM, GE, NMBAC) feed them into MLapSVM-LBS. Based on human learning theory, MLapSVM-LBS can better represent the relationship between samples through similarity. introduce new edge weight calculation method takes label information consideration. In addition, distribution parameter reflecting underlying probability sample’s neighborhood is also employed. further improve robustness model, utilize regularization build multigraph model which five graphs constructed changing size. appraise performance our trained tested PDB186, PDB1075, PDB2272 PDB14189 datasets. On two independent testing sets (PDB186 PDB2272), reaches accuracies 0.887 0.712, respectively. The good results both datasets demonstrate reliable model.