A penalized quantitative structure–property relationship study on melting point of energetic carbocyclic nitroaromatic compounds using adaptive bridge penalty

A penalized quantitative structure–property relationship (QSPR) model with adaptive bridge penalty for predicting the melting points of 92 energetic carbocyclic nitroaromatic compounds is proposed. To ensure the consistency of the descriptor selection of the proposed penalized adaptive bridge (PBridge), we proposed a ridge estimator (β^Ridge) as an initial weight in the adaptive bridge penalty. The Bayesian information criterion was applied to ensure the accurate selection of the tuning parameter (λ). The PBridge based model was internally and externally validated based on Q² _int, Q² _LGO, Q² _Boot, ^CCC _train, ^MAE _train, ^MSE _train, the Y-randomization test, Q² _ext, ^CCC _train, ^MAE _train, (^MSE _train and the applicability domain. The validation results indicate that the model is robust and not due to chance correlation. The descriptor selection and prediction performance of PBridge for the training dataset outperforms the other methods used. PBridge shows the highest Q² _int of 0.959, Q² _LGO of 0.953, Q² _Boot of 0.949 and ^CCC _train of 0.959, and the lowest ^MAE _train and ^MSE _train. For the test dataset, PBridge shows a higher Q² _ext of 0.945 and ^CCC _test of 0.948, and a lower ^MAE _test and ^MSE _test, indicating its better prediction performance. The results clearly reveal that the proposed PBridge is useful for constructing reliable and robust QSPRs for predicting melting points prior to synthesizing new organic compounds.