StackPDB: Predicting DNA-binding proteins based on XGB-RFE feature optimization and stacked ensemble classifier

DNA-binding proteins (DBPs) not only play an important role in all aspects of genetic activities such as DNA replication, recombination, repair, and modification but also are used key components antibiotics, steroids, anticancer drugs the field drug discovery. Identifying DBPs becomes one most challenging problems domain proteomics research. Considering high-priced inefficient experimental method, constructing a detailed prediction model urgent problem for researchers. In this paper, we propose stacked ensemble classifier based method predicting called StackPDB. Firstly, pseudo amino acid composition (PseAAC), pseudo-position-specific scoring matrix (PsePSSM), position-specific matrix-transition probability (PSSM-TPC), evolutionary distance transformation (EDT), residue probing (RPT) applied to extract protein sequence features. Secondly, extreme gradient boosting-recursive feature elimination (XGB-RFE) is employed gain excellent subset. Finally, best features composed XGBoost, LightGBM, SVM construct After applying leave-one-out cross-validation (LOOCV), StackPDB obtains high ACC MCC on PDB1075, 93.44% 0.8687, respectively. Besides, independent test datasets PDB186 PDB180 84.41% 90.00%, The 0.6882 0.7997, results training dataset show that has great predictive ability predict DBPs.