New Advances in QSPR/QSAR Analysis of Nitrocompounds: Solubility, Lipophilicity, and Toxicity

Abstract

This chapter discusses QSAR/QSPR applications of the simplex representation of molecular structure (SiRMS) methodology. It has been determined that SiRMS proves to be quite an efficient tool for analyzing nitroaromatic aqueous solubility, lipophilicity, and toxicity. Using multiple linear regression (MLR) and random forest (RF) statistical methods at the 2D level of representation of molecular structure, models possessing high statistical characteristics (MLR: R 2=0.85, Q 2=0.83; RF: R 2=0.99, \( R_{\text{OOB}}^2 = 0.{88} \)) were obtained for aqueous solubility of more than 2,800 organic compounds. The external validation set of 301 compounds (including 47 nitro-, nitroso-, and nitrogen-rich military compounds) was used for evaluation of the models’ predictive ability.