SXGBsite: Prediction of Protein–Ligand Binding Sites Using Sequence Information and Extreme Gradient Boosting

Abstract

The prediction of protein–ligand binding sites is important in drug discovery and drug design. Protein–ligand binding site prediction computational methods are inexpensive and fast compared with experimental methods. This paper proposes a new computational method, SXGBsite, which includes the synthetic minority over-sampling technique (SMOTE) and the Extreme Gradient Boosting (XGBoost). SXGBsite uses the position-specific scoring matrix discrete cosine transform (PSSM-DCT) and predicted solvent accessibility (PSA) to extract features containing sequence information. A new balanced dataset was generated by SMOTE to improve classifier performance, and a prediction model was constructed using XGBoost. The parallel computing and regularization techniques enabled high-quality and fast predictions and mitigated overfitting caused by SMOTE. An evaluation using 12 different types of ligand binding site independent test sets showed that SXGBsite performs similarly to the existing methods on eight of the independent test sets with a faster computation time. SXGBsite may be applied as a complement to biological experiments.

A Comprehensive Review of Computation-Based Metal-Binding Prediction Approaches at the Residue Level

“…Figure 5 illustrates the predictive performance on two benchmark test datasets. Details are provided in Table S1 and Table S2 in Supplementary Materials, respectively; the corresponding results are sourced from [ 56 , 57 , 81 , 92 ]. We notice that the predictors show relatively big differences in recognizing various types of metal-binding residues.…”

“…Besides that, Figure 5(a) indicates that all five methods show a decent performance on recognizing Fe 3+ -binding residues (MCC values close or higher than 0.4), compared with MCC close or less than 0.2 on Ca 2+ binding residues. Figure 5(b) draws the bars of AUC values for SXGBsite [ 92 ], EC-RUS [ 95 ], and TargetS [ 101 ], respectively. These three predictors all achieve high AUC scores (close or higher than 0.9) on Zn 2+ - and Fe 3+ -binding residues.…”

“…Another available docking-based method is mFASD, which distinguishes different types of metal-binding sites according to the interaction distances. MPLs-Pred [ 91 ], SXGBsite [ 92 ], MIonSite [ 90 ], OSML [ 100 ], EC-RUS [ 95 ], and TargetS [ 101 ] are all sequence-based predictive tools, which accepts FASTA-formatted input and produced the results of putative metal-binding residues. ZinCaps [ 94 ], SSWPNN [ 93 ], and ZincBinder [ 103 ] are specially designed for the identification of zinc-binding residues.…”

DTLM-DBP: Deep Transfer Learning Models for DNA Binding Proteins Identification

“…α i α j y i y j x i x j (6) The mapping of training vectors xi into the higher dimensional space uses a function called kernel function K(x i , x j ) ≡ (x i ) (x j ). There are several SVMs kernel functions, such as: Linear kernel:…”

“…Several experimental technical methods can be used for identifying DNABPs, but they are time-consuming and expensive [6]. Therefore, there is a significant need to find a suitable and efficient computational method for replacing these experimental methods.…”