Importance of interface and surface areas in protein-protein binding affinity prediction: A machine learning analysis based on linear regression and artificial neural network

Yang, Yong; Wang, Pan; Zhu, Bao Ting

doi:10.1016/j.bpc.2022.106762

Cited by 8 publications

(30 citation statements)

References 87 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Two subsets containing 52 or 24 protein–protein complexes with reliable experimental binding affinity values were used for testing. As shown in Figure A, the best Pearson’s correlation coefficient ( R ) between the predicted and experimental binding affinities for these 60 models is 0.87 for the 52 complexes and 0.92 in another subset containing 24 oligomers with reliable experimental binding affinity values . The performances of the best models are superior to those of PRODIGY , and LISA (Figure A).…”

Section: Resultsmentioning

confidence: 98%

“…The primary components of area-based descriptors are calculated using Q contact (area of interface residue pair) and dr_sasa (solvent accessible surface area of atom) in our recent studies . According to the physicochemical properties, the 20 amino acid types are categorized into four groups: basic AA s (basic amino acids: HIS, ARG, and LYS), nonpolar AA s (nonpolar, hydrophobic amino acids: ILE, PHE, LEU, TRP, ALA, MET, PRO, and VAL), polar AA s (polar but uncharged amino acids: CYS, ASN, GLY, SER, GLN, TYR, and THR), and acidic AA s (acidic amino acids: ASP and GLU).…”

Section: Methodsmentioning

confidence: 99%

“…The interface area is divided into 10 kinds: basic AA s ∼ basic AA s ( A 9 ), nonpolar AA s ∼ nonpolar AA s ( A 10 ), polar AA s ∼ polar AA s ( A 11 ), acidic AA s ∼ acidic AA s ( A 12 ), basic AA s ∼ nonpolar AA s ( A 13 ), basic AA s ∼ polar AA s ( A 14 ), basic AA s ∼ acidic AA s ( A 15 ), nonpolar AA s ∼ polar AA s ( A 16 ), nonpolar AA s ∼ acidic AA s ( A 18 ). Additionally, the total RSA ( A 19 ), total LSA ( A 20 ), and total interface area ( A 21 ) are also employed to construct the nonlinear models with explicit formations …”

Section: Methodsmentioning

confidence: 99%

“…Three types of models, i.e. , linear, nonlinear, and mixed models, have been trained, generated, or constructed by employing area-based descriptors . Linear models are generated using linear regression, nonlinear models are constructed using trial-and-error methods or trained using neural network , or random forest, , and mixed models are generated using linear regression based on nonlinear models.…”

Section: Methodsmentioning

confidence: 99%

“…, PRODIGY (PROtein binDIng enerGY) . In our recent work, the area-based protein–protein binding affinity prediction methods were developed using different interface and surface areas present in the structure of a protein–protein complex. , The performances of the area-based methods were found to be superior or at least comparable to those of PRODIGY and LISA (Local Interaction Signal Analysis), which are considered the best existing methods based on linear and nonlinear models, respectively, for prediction of protein–protein binding affinity.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

AREA-AFFINITY: A Web Server for Machine Learning-Based Prediction of Protein–Protein and Antibody–Protein Antigen Binding Affinities

Yang

Huang

Wang

et al. 2023

J. Chem. Inf. Model.

Self Cite

View full text Add to dashboard Cite

Protein–Protein binding affinity reflects the binding strength between the binding partners. The prediction of protein–protein binding affinity is important for elucidating protein functions and also for designing protein-based therapeutics. The geometric characteristics such as area (both interface and surface areas) in the structure of a protein–protein complex play an important role in determining protein–protein interactions and their binding affinity. Here, we present a free web server for academic use, AREA-AFFINITY, for prediction of protein–protein or antibody–protein antigen binding affinity based on interface and surface areas in the structure of a protein–protein complex. AREA-AFFINITY implements 60 effective area-based protein–protein affinity predictive models and 37 effective area-based models specific for antibody–protein antigen binding affinity prediction developed in our recent studies. These models take into consideration the roles of interface and surface areas in binding affinity by using areas classified according to different amino acid types with different biophysical nature. The models with the best performances integrate machine learning methods such as neural network or random forest. These newly developed models have superior or comparable performance compared to the commonly used existing methods. AREA-AFFINITY is available for free at: .

show abstract