Binding free energy calculation of peptides to carbon nanotubes using molecular dynamics with a linear interaction energy approach

Peptides interaction with carbon nanotubes is of potential importance in a wide range of biological and chemical applications. Coating carbon nanotubes (CNTs) with peptides solubilizes nanotubes in water. To consider the affinities of CNTs for different peptides, binding free energies of peptides to single-walled carbon nanotubes (SWCNTs) were calculated using linear interaction energy (LIE) model in conjugation with molecular dynamics simulation. This simple model represents the linear relationship between van der Waals energy as the dominant interaction in binding process and the energy of binding. The model yields an average error of 0.83 kcal/mol for the binding free energy prediction of five peptides considered as training set. The predictive power of the model was revealed by high value of Q2 (0.97) and low value of sPRESS (1.07) from the leave-one-out cross validation. The calculated binding free energies of peptides in the training set were in good agreement with the experimental data and the ranking of binding affinities of peptides in both training and test sets was well reproduced. The findings of our study can be utilized in the estimation of ranking the binding affinities of new peptides to CNTs to open the door in exploring the novel ways for increasing the solubility of CNTs in water.