Density Functional Theory Computation and Machine Learning Studies of Interaction between Au₃ Clusters and 20 Natural Amino Acid Molecules

The optimal adsorption sites and the binding energiesof neutralAu(3) clusters with 20 natural amino acids under the gasphase and water solvation were systematically investigated based ondensity functional theory (DFT) calculations. The calculation resultsshowed that in the gas phase Au-3 tends to bind with N atomsof amino groups in amino acids, except methionine, which tends tobind with Au-3 through S atoms. Under water solvation, Au-3 clusters tended to bind to N atoms of amino groups and Natoms of side chain amino groups in amino acids. However, methionineand cysteine bind more strongly to the gold atom through the S atom.Based on the binding energy data of Au-3 clusters and 20natural amino acids under water solvation calculated by DFT, a machinelearning model (gradient boosted decision tree) was proposed to predictthe optimal binding Gibbs free energy (Delta G)of the interaction between Au-3 clusters and amino acids.The main factors affecting the strength of the interaction betweenAu(3) and amino acids were uncovered by the feature importanceanalysis.