QSPR modeling for enthalpies of formation of organometallic compounds by means of SMILES-based optimal descriptors

A quantitative structure–property relationship (QSPR) model for a predicting gas-phase enthalpy of formation have been developed, using as chemical information descriptors based on the simplified molecular input line entry system (SMILES). The model is one-variable equation. The SMILES-based descriptors calculated with correlation weights of SMILES attributes which are obtained by the Monte Carlo method. The model addressed organometallic compounds. Statistical characteristics of the model are the following: n = 104, R2 = 0.9943, Q2 = 0.9940, s = 19.9 (kJ/mol), F = 17701 (training set); n = 28, R2 = 0.9908, Q2 = 0.9892, s = 29.4 (kJ/mol), F = 2788 (test set).