Exhaustive QSPR Studies of a Large Diverse Set of Ionic Liquids:  How Accurately Can We Predict Melting Points?

Abstract

Several popular machine learning methods--Associative Neural Networks (ANN), Support Vector Machines (SVM), k Nearest Neighbors (kNN), modified version of the partial least-squares analysis (PLSM), backpropagation neural network (BPNN), and Multiple Linear Regression Analysis (MLR)--implemented in ISIDA, NASAWIN, and VCCLAB software have been used to perform QSPR modeling of melting point of structurally diverse data set of 717 bromides of nitrogen-containing organic cations (FULL) including 126 pyridinium bromides (PYR), 384 imidazolium and benzoimidazolium bromides (IMZ), and 207 quaternary ammonium bromides (QUAT). Several types of descriptors were tested: E-state indices, counts of atoms determined for E-state atom types, molecular descriptors generated by the DRAGON program, and different types of substructural molecular fragments. Predictive ability of the models was analyzed using a 5-fold external cross-validation procedure in which every compound in the parent set was included in one of five test sets. Among the 16 types of developed structure--melting point models, nonlinear SVM, ASNN, and BPNN techniques demonstrate slightly better performance over other methods. For the full set, the accuracy of predictions does not significantly change as a function of the type of descriptors. For other sets, the performance of descriptors varies as a function of method and data set used. The root-mean squared error (RMSE) of prediction calculated on independent test sets is in the range of 37.5-46.4 degrees C (FULL), 26.2-34.8 degrees C (PYR), 38.8-45.9 degrees C (IMZ), and 34.2-49.3 degrees C (QUAT). The moderate accuracy of predictions can be related to the quality of the experimental data used for obtaining the models as well as to difficulties to take into account the structural features of ionic liquids in the solid state (polymorphic effects, eutectics, glass formation).

New Approach for Accurate QSPR Modeling of Metal Complexation: Application to Stability Constants of Complexes of Lanthanide Ions Ln3+, Ag+, Zn2+, Cd2+ and Hg2+ with Organic Ligands in Water

“…One consensus model combines predictions issued from a multitude of individual models originated from different types of the SMF descriptors and variable selection algorithms. [23,28,35,73,74] Thus for every compound from the test set, the target property is computed as an arithmetic mean of values obtained by individual models excluding those leading to outlying values according to Tompson's rule. [75] If a test compound is identified as being outside an applicability domain (AD) of individual model, the prediction by given model for a given compound is not included in CM.…”

“…[40,74] In this procedure, an entire dataset is divided in 5 non-overlapping pairs of training and test sets. Predictions are prepared for all molecules (n) of the initial dataset, since each of them belongs to one of the test sets.…”

Estimation of thermal conductivity of ionic liquids using quantitative structure–property relationship calculations

“…Several authors use the concepts of QSPR to calculate diverse properties of several substances [11]. But for ILs, these methods have commonly been applied for: density [10], melting point [12][13][14], electrical conductivity [15,16], surface tension [17] and viscosity [15,16]. Recently, Das and Roy present a complete review with several QSPR applications for ILs [18].…”

Physico-Chemical Properties of Task-Specific Ionic Liquids

“…Decision trees and neural networks thus constitute suitable methodologies for modelling purposes. In 2007 Varnek et al [31] published a comprehensive study testing the performance of different linear and non-linear machine learning methods in performing QSPR modelling. A large set of 717 nitrogen-containing cations, all paired with bromide anions, was studied, as well as subsets of it.…”

“…It is necessary to know some experimental values in order to develop QSPR models, but for some ILs several T m values are reported. [31] This can have to do with the occurrence of polymorphs or, sometimes, with poor quality measurements due to, by instance, the presence of impurities.…”