Quantitative Structure-property Relationship Method Research Articles

Neural Network Based on Quantum Chemistry for Predicting Melting Point of Organic Compounds

The melting points of organic compounds were estimated using a combined method that includes a backpropagation neural network and quantitative structure property relationship (QSPR) parameters in quantum chemistry. Eleven descriptors that reflect the intermolecular forces and molecular symmetry were used as input variables. QSPR parameters were calculated using molecular modeling and PM3 semi-empirical molecular orbital theories. A total of 260 compounds were used to train the network, which was developed using MatLab. Then, the melting points of 73 other compounds were predicted and results were compared to experimental data from the literature. The study shows that the chosen artificial neural network and the quantitative structure property relationships method present an excellent alternative for the estimation of the melting point of an organic compound, with average absolute deviation of 5%.

Study on QSPR method for theoretical calculation of heat of formation for some organic compounds

Quantitative Structure ‐ Property Relationship (QSPR) models based on molecular descriptors derived from molecular structures have been used for the prediction of heat of formation using a set of 20 organic compounds. The molecular and quantum chemical descriptors used to represent molecular structure include topological indices and constitutional descriptors. Forward stepwise regression was used to construct the QSPR models. Multiple linear regressions are utilized to construct the linear prediction model. The prediction results are in good agreement with the experimental values of these properties.

Estimation of enthalpies of formation of organometallic compounds from their molecular structures

A quantitative structure–property relationship (QSPR) was developed, aiming to estimate the gas-phase enthalpies of formation (Δ f H 0) of a set of 132 organometallic compounds of general formula MR n X n− m , where M is a metal or a semimetal from groups 12 to 16, R is an alkyl, aryl, alkenyl, or alkynyl group, and X is Cl, Br, I, or H. The proposed model, derived from multilinear regression, contains nine descriptors that can be readily calculated from molecular structures. Correlations with R 2 and RMSE of 0.988 (29.1) and 0.990 (30.2) for the training and prediction sets, respectively, are obtained. The ability of QSPR methods to estimate reliable values of enthalpies of formation has been confirmed by the results obtained with a set of 168 organic compounds, which contain the same type of groups of the organometallic compounds. The nine descriptors-derived model, containing only descriptors of the constitutional, topological, and geometrical types, predicts Δ f H 0 with accuracies comparable to well established additive methods.

A drug release study from hydroxypropylmethylcellulose (HPMC) matrices using QSPR modeling

This investigation is aimed at characterization of the mode of release from two different substitution types of HPMC and the effect of chemical structure of drugs using the QSPR (Quantitative - Structure-Property Relationship) technique. To this end, release profiles of HPMC matrices of several drugs containing the same formulation and compressed at a constant pressure were studied. QSPR method was used to establish statistically significant relationships between release parameters and the structural descriptors. Structural descriptors consisted of molecular mechanical, quantum mechanical and graph-theoretical parameters, as well as the partition coefficient and the aqueous solubility of the drugs. The results showed that the most important factors determining the release profile from both HPMC K4M and HPMC E4M matrices were the aqueous solubility of drugs (which could be substituted efficiently by dipole moment) and the size of the drug molecules. Comparison of drug release from matrices prepared using the two grades of HPMC showed very distinct differences for some drugs, as evaluated by the similarity factor. The results indicated that the source of the difference could be sought in the drug properties (as exemplified by the aqueous solubility and surface area) as well as the rate of erosion (that depends mainly on the polymer type).

Quantitative structure–property relationship prediction of permeability coefficients for some organic compounds through polyethylene membrane

In this work artificial neural network was constructed and trained for the prediction of the permeability coefficients of various organic compounds through low-density polyethylene, based on quantitative structure–property relationship method. The inputs of this network were theoretically derived molecular descriptors which were selected by the stepwise multiple linear regressions technique. These descriptors are; the number of oxygen atoms in a molecule, area-weighted surface charge of hydrogen bonding donor atoms (HA-dependent HDCA-2), molecular transform index lag 11 weighted by atomic van der Waals volume (Morse-11v), molecular transform index lag 10 weighted by atomic polarizability (Morse-10p) and polarity parameter. In order to assess the accuracy and predictability of the proposed model, the cross- validation and Y-scrambling test were employed. The statistical parameters for cross-validation tests are; R 2 = 0.964, PRESS = 0.221. The results obtained showed the ability of developed artificial neural network to prediction of permeability coefficients of various compounds. Also result reveals the superiority of the artificial neural network over the multiple linear regressions model.

Prediction of micelle–water partition coefficient from the theoretical derived molecular descriptors

The micelle–water partition coefficients of 81 organic compounds in SDS solution were predicted by quantitative structure–property relationship method. The multiple linear regression (MLR) and artificial neural network (ANN) techniques were used to build linear and nonlinear model, respectively. In this work the proposed QSPR models, both by MLR and ANN, contain identical descriptors which are zero order of Kier–Hall index, count of Hydrogen donors site [Zefirovs PC], average valency of a C atom, atomic charge weighted by partial positively charged surface area and minimum one electron reaction index for a C atom. The MLR model gave a root mean square (RMS) of 0.166, 0.25, and 0.289 for training, prediction and test sets, respectively, whereas ANN gave an RMS error of 0.06, 0.21, and 0.20 for training, prediction, and test sets, respectively. Comparison the results of these two methods reveals that those obtained by the ANN model are much better.

Prediction of dielectric constants and glass transition temperatures of polymers by quantitative structure property relationships

The use of quantitative structure property relationships (QSPRs) is proposed for the calculation of the dielectric constants ( ε) and glass transition temperatures ( T g) for polymer. The descriptors involved in these models were calculated from the structures of the repeat units by the density functional theory (DFT) method. Subsets of these descriptors are used to build models in an attempt to find the best possible correlation between chemical structural parameter and dielectric constants or glass transition temperatures. Model with three quantum-chemical descriptors was selected for the prediction of ε ( R 2 = 0.9086, s = 0.00103), and model with two quantum-chemical descriptors was selected for T g( R 2 = 0.9212, s = 21.7378). These results encourage the further application of QSPR methods to other classes of polymer.

Physical Properties of Ionic Liquids: Database and Evaluation

A comprehensive database on physical properties of ionic liquids (ILs), which was collected from 109 kinds of literature sources in the period from 1984 through 2004, has been presented. There are 1680 pieces of data on the physical properties for 588 available ILs, from which 276 kinds of cations and 55 kinds of anions were extracted. In terms of the collected database, the structure-property relationship was evaluated. The correlation of melting points of two most common systems, disubstituted imidazolium tetrafluoroborate and disubstituted imidazolium hexafluorophosphate, was carried out using a quantitative structure-property relationship method.

Correlating Polymer-Carbon Composite Sensor Response with Molecular Descriptors

We report a quantitative structure-activity relationships (QSAR) study using genetic function approximations to describe the activities of a polymer-carbon composite chemical vapor sensor using a novel approach to selecting a molecular descriptor set. The measured sensor responses are conductivity changes in polymer-carbon composite films upon exposure to target vapors at parts-per-million concentrations. The descriptor set combines the basic analyte descriptor set commonly used in QSAR studies with descriptors for sensing film-analyte interactions. The basic analyte descriptors are obtained using a combination of empirical and semiempirical quantitative structure-property relationships methods. The descriptors for the sensing film-analyte interactions are calculated using molecular modeling and simulation tools. A statistically validated QSAR model was developed for a training data set consisting of 17 analyte molecules. The applicability of this model was also tested by predicting sensor activities for three test analytes not considered in the training set.

The Fragment Constant Method for Predicting Octanol–Air Partition Coefficients of Persistent Organic Pollutants at Different Temperatures

The octanol–air partition coefficient (KOA) is a key physicochemical parameter for describing the partition of organic pollutants between air and environmental organic phases. Experimental determination of KOA is costly and time consuming, and sometimes restricted by lack of sufficiently pure chemicals. There is a need to develop a simple but accurate method to estimate KOA. In the present study, a fragment constant model based on five fragment constants and one structural correction factor, was developed for predicting logKOA at temperatures ranging from 10 to 40°C. The model was validated as successful by statistical analysis and external experimental logKOA data. Compared to other quantitative structure–property relationship methods, the present model has the advantage that it is much easier to implement. As aromatic compounds that contain C, H, O, Cl, and Br atoms, were included in the training set used to develop the model, the current fragment model applies to a wide range of chlorinated and brominated aromatic pollutants, such as chlorobenzenes, polychlorinated naphthalenes, polychlorinated biphenyls, polychlorinated dibenzo-p-dioxins and dibenzofurans, polycyclic aromatic hydrocarbons, and polybrominated diphenyl ethers, all of which are typical persistent organic pollutants. Further study is necessary to expand the utility of the method to all halogenated aliphatic and aromatic compounds.

Testing the Potential for Computational Chemistry to Quantify Biophysical Properties of the Non-Proteinaceous Amino Acids

Although most proteins of most living organisms are constructed from the same set of 20 amino acids, all indications are that this standard alphabet represents a mere subset of what was available to life during early evolution. However, we currently lack an appropriate quantitative framework with which to test the qualitative hypotheses that have been offered to date as explanations for nature's "choices." Specifically, although many indices have been developed to describe the 20 standard amino acids, few or no comparable data extend to prebiotically plausible alternatives because of the costly and time-consuming bench experiments that would be required. Computational chemistry (specifically quantitative structure property relationship methods) offers a potentially fast, cost-effective remedy for this knowledge gap by predicting such molecular properties in silico. Thus, we investigated the use of various freely accessible programs to predict three key amino acid properties (hydrophobicity, charge, and size). We assessed the accuracy of these predictions by comparisons with experimentally determined counterparts for appropriate test data sets. In light of these results, and factors of software accessibility and transparency, we suggest a method for further computational assessments of prebiotically plausible amino acids. The results serve as a starting point for future quantitative analysis of amino acid alphabet evolution.

Structurally “Targeted” Quantitative Structure−Property Relationship Method for Property Prediction

Prediction of unknown properties for a target compound using quantitative structure−property relationships (QSPRs) is reliable only if the compound is within the model applicability domain. To improve the prediction reliability, a “targeted” QSPR (TQSPR) method is developed in which a training set containing only compounds structurally similar to the target compound is first identified. Similarity is measured by the partial correlation coefficients between the vectors of the molecular descriptors of the target compound and each potential predictive compound. Available property data in the training set are then used in the usual manner to select molecular descriptors for QSPRs, predicting the properties of the target and the rest of the compounds in the set. Preliminary results show that the method proposed yields predictions within the experimental error for compounds well represented in the database and fairly reliable estimates for complex compounds that are sparsely represented. The cutoff value of the partial correlation coefficient provides an indication of the expected prediction error.

Quantitative Structure-Uptake Relationship of Metal-Organic Frameworks as Hydrogen Storage Material

ABSTRACTWe reported the relationship between the structure of metal-organic frameworks (MOFs) and the capability of hydrogen uptake. The QSPR (quantitative structure-property relationship) method was used to find out the factor which affects the adsorption amount of hydrogen molecule on the MOFs. The derivatives which were substituted by functionalized aromatic rings showed the effect of polarization within the identical topology of the frame and similar lattice constants. And the typical series of MOFs with different topology of the frames were investigated to examine the influence of topological change. For the consideration of saturation of hydrogen adsorption amounts, the result of fitting the adsorption curve with Langmuir-Freundlich equation was used to the QSPR approach additionally. We found out that the polar surface area plays a key role on the adsorption amount of hydrogen molecule into the MOFs and the specific value of electrostatic potential surface was calculated to indicate the interaction between hydrogen molecule and MOF.

Comparison of subcooled liquid vapor pressures of polychlorinated dibenzo-p-dioxins and dibenzofurans predicted by QSPR and GC-RI methods

Subcooled liquid vapor pressures (P L) are of great importance for assessing the persistent behavior of organic pollutants. As P L cannot be determined by direct experiments, it is of interest to develop and evaluate various predictive methods. In the current study, gas chromatography retention index (GC-RI) and quantitative structure-property relationship (QSPR) methods were used to develop predictive models for P L of polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs). The model development was based on P L values converted from consistent experimental solid vapor pressures (P S). The P L values predicted by the two methods are highly consistent with each other, and in-between sets of values predicted by others. Since the QSPR method can be regarded as independent of experiments, and can be used to interpret intermolecular interactions that govern the magnitude of P L, it may be superior to the GC-RI method.

Prediction of GC Retention Indexes for Insect-Produced Methyl-Substituted Alkanes Using an Artificial Neural Network and Simple Structural Descriptors

A quantitative structure-property relationship (QSPR) study based on the artificial neural network (ANN) technique was performed for the prediction of gas chromatography retention indexes of methyl-substituted alkanes produced by insects. Simple descriptors such as the total number of carbons in the backbone, the number of the multiple methyl groups attached to the carbon chain, and their relative positions were selected, and an ANN with a 9 : 8 : 1 architecture was generated using the nine descriptors in the input layer. The average relative error was 3.3%. The method was also compared with the QSPR method, which utilizes topological and quantum chemical descriptors.

Partition coefficients in food/packaging systems: a review

Food contamination can result from various interactions between food and packaging materials. Migration of volatiles, additives, monomers and oligomers from packaging materials into food or adsorption of volatile compounds from the food by the polymer are important considerations from safety, hygienic and economic points of view. The term ‘migration’ includes two phenomena (partition and diffusion) that can be important in determining the concentration of contaminants in a food system at any time. An estimation of the partition coefficient, K, in food/packaging systems has been the major objective of numerous different studies. Various parameters can influence K such as temperature, pH, the chemical structure of the migrant, molecular size and structure, fat content, and degrees of crystallinity. Some theoretical approaches such as the quantitative structure–property relationship method could be of interest in the near future.

Calculation of the Enthalpy of Sublimation by the QSPR Method with the Use of a Fragment Approach

The enthalpies of sublimation of organic compounds belonging to various classes were studied for the first time in terms of the fragment approach based on the QSPR (Quantitative Structure-Property Relationships) method. The applicability of this technique to calculation of this parameter was demonstrated and a model that makes it possible to predict the enthalpy of sublimation of compounds on the basis of descriptors taking into account the fragment composition of a molecule was suggested.

Amorphous cell studies of polyglycolic, poly( l-lactic), poly( l, d-lactic) and poly(glycolic/ l-lactic) acids

In this paper static amorphous state properties (solubility parameter, free volume (using the Voorintholt method and the Voronoi tessellations) and pair correlation functions, the last ones also by including water molecules in the cells), which can be related to the probability for water uptake, have been studied for polyglycolic (PGA), poly( l-lactic) (PLLA), poly( l, d-lactic) (PLLA/PDLA) and poly(glycolic/ l-lactic) (PGA/PLLA) acids, known to be biodegradable polymers. The polymer consistent force field, as modified by the authors, has been used in the calculations. The main purpose of this paper is to investigate, which of the amorphous state properties would be relevant for water uptake. We also discuss the validity of th6e methods used for these kinds of studies, and the related reliability of the computed results. Chain flexibilities of the studied polyesters in the amorphous phase have been analyzed, and the intermolecular interactions are found to cause the most significant variations in the distributions of the adjacent chain dihedral angle pairs and in the related populations of the low-energy regions of the comonomers. The solubility parameters, as calculated from the cohesion energy densities of the constructed models, suggest PGA being most compatible with water, in agreement with experiments. On the other hand, the quantitative structure–property relationships method ‘Synthia’ suggests a very similar solubility in water for all particular polyesters. In the PLAs and PGA/PLLA, however, a larger number of hydrogen bonds is formed between the water molecules and the carbonyl oxygen atoms of the chains showing a better possibility of PLLA and its copolymers to break into shorter chains. As an explanation, the hydrophobic methyl groups of the lactide units are suggested to push the water molecules closer to the carbonyl groups than in homo-PGA.

Relativity Study on Topological Index of Methylalkane Structures and Chromatographic Retention Index

Quantitative structure-property relationships (QSPR) have been demonstrated to be a powerful tool in chromatography. QSPR have been used to obtain simple models to explain and predict the chromatographic behavior of various classes of compounds. The study of quantitative structure and retention index relationship (QSRR) is an important subject in chromatographic field. One hundred twenty-seven topological descriptors of 207 methylalkane structures are calculated. GAPLS method, which is a variable selection method combining with genetic algorithms (GA), back stepwise and partial least squares (PLS), is introduced in the variable selection of quantitative structure gas chromatographic (GC) retention index relationship. Seven topological descriptors are selected from 127 topological descriptors by GAPLS method to build QSRR model with high regression quality: squared correlation coefficient (R2) of 0.99998, standard deviation (S) of 2.88. The error of the model is similar to the experimental error. The validation of the model is checked by leave-one-out cross-validation technique. The result of leave-one-out cross-validation indicates that the built model is reliable and stable with high prediction quality, such as squared correlation coefficient of leave-one-out (R2cv) of 0.99997 and standard deviation of leave-one-out predictions (Scv) of 2.95. A successful interpretation of the complex relationship between GC retention indexes of methylalkanes and the chemical structure is achieved using QSPR method. The seven variables in the model are also rationally interpreted, which indicates methylalkane retention index are precisely represented by topological descriptors.