D. A. Shulga

Perspectives of Halogen Bonding Description in Scoring Functions and QSAR/QSPR: Substituent Effects in Aromatic Core

Halogen bonding (XB) is a new promising interaction pattern in medicinal chemistry. It has predominantly electrostatic nature – high electrostatic potential anisotropy. However to fully unleash the potential of XB in rational drug design fast and robust empirical methods of XB description should be developed. Current approaches rely heavily on ab initio calculation for each molecule studied. Thus fast prediction of electrostatic parameters for description of XB for arbitrary organic molecules is of paramount importance to promptly establish QSAR/QSPR, virtual screening and molecular docking pipelines suitable for today′s agile development requirements. The two most promising approaches to describe anisotropic electrostatic models – the extra point (EP) charge model and the multipole expansion (ME) model – were studied on their ability (1) to describe ab initio molecular electrostatic potential (MEP) and (2) to produce parameters that can be predicted for each molecule empirically rather than estimated via ab initio calculations. The reference ab initio MEP was calculated for a set of 730 substituted halobenzenes. Parameters for anisotropic electrostatics of both empirical models (EP and ME) studied were extracted from ab initio MEP. The FreeWilson and Hansch type QSPR models relating XB parameters with aromatic substituents were built and analyzed, providing the guidelines for further development.

General Purpose Electronegativity Relaxation Charge Models Applied to CoMFA and CoMSIA Study of GSK-3 Inhibitors

Two fast empirical charge models, Kirchhoff Charge Model (KCM) and Dynamic Electronegativity Relaxation (DENR), had been developed in our laboratory previously for widespread use in drug design research. Both models are based on the electronegativity relaxation principle (Adv. Quantum Chem. 2006, 51, 139–156) and parameterized against ab initio dipole/quadrupole moments and molecular electrostatic potentials, respectively. As 3D QSAR studies comprise one of the most important fields of applied molecular modeling, they naturally have become the first topic to test our charges and thus, indirectly, the assumptions laid down to the charge model theories in a case study. Here these charge models are used in CoMFA and CoMSIA methods and tested on five glycogen synthase kinase 3 (GSK‐3) inhibitor datasets, relevant to our current studies, and one steroid dataset. For comparison, eight other different charge models, ab initio through semiempirical and empirical, were tested on the same datasets. The complex analysis including correlation and cross‐validation, charges robustness and predictability, as well as visual interpretability of 3D contour maps generated was carried out. As a result, our new electronegativity relaxation‐based models both have shown stable results, which in conjunction with other benefits discussed render them suitable for building reliable 3D QSAR models.

The choice of atomic charges calculation scheme in 3D-QSAR modelling of GSK-3β inhibition by paullones

descriptors in the nodes of the threedimensional lat� tice surrounding these molecules. The energy of van der Waals and electrostatic interactions of a probe atom ( with the charge +1) with molecules of the training set (CoMFA) or the electrostatic, van der Waals, hydrophobic, and donor/acceptor similarity indices (CoMSIA) are used as descriptors. The equa� tion for activity prediction is derived using the partial least squares (PLS) method. The ability of graphic representation of PLS model coefficients is the advan� tage of the methods and allows the user to suggest sub� stitutions affecting activity and/or selectivity of the molecules. The electrostatic field rules to the intermolecular interactions and plays major role in CoMFA and CoMSIA modeling. The contribution of the electro� static field is estimated with the help of partial atomic charges of molecules under consideration. The suc� cess of a modeling study strongly depends on proper selection of an adequate method for calculation of atomic charges. Various parameters should be taken into account, such as the calculation time, geometri� cal parameters of molecules, the accuracy and signifi� cance of modeling results, and stability upon small changes of the geometrical parameters of the mole� cule. In this study different charge calculation schemes based on different combinations of the men� tioned parameters were used to build predictive mod� els with the aim to identify one or more schemes pos� sessing an optimal set of parameters for 3DQSAR modeling. The training set comprised of the compounds pre� viously used for CoMSIA modeling (9) along with small series of epoxidecontaining analogues (10). The total number of compounds in the training set was 77. Twentyeight compounds whose inhibitory activity was determined later (11) were used as an external test set. The inhibitory activity of all compounds was determined under identical conditions and expressed as

Description of halogen bonding on the basis of multicenter multipole expansion

139 Halogen bonding (XB) is a type of intermolecular interaction implying that a halogen atom in a mole cule forms a noncovalent bond with electron density donors, thus acting as a Lewis acid. Interest in XB has increased in recent years because of the prospect of its use in design of drugs and functional materials [1, 2]. Halogen bonding has unique properties [2] and results in formation of directional bonds in hydrophobic environment. The energy of these bonds is comparable with the hydrogen bond energy. However, the applied use of XB is prevented by the absence of a reliable, fast, and theoretically correct method for describing its potential [3]. Thus, construction of qualitatively and quantitatively correct empirical potentials of XB for force fields and scoring functions is an important task in both fundamental and applied aspects.

Multipole models of sulphur for accurate anisotropic electrostatic interactions within force fields

Abstract Nowadays, as computing has become much more available, a fresh momentum has been observed in the field of re-visioning and re-parameterizing the usual tools, as well as estimating for the incorporation of new qualitative capabilities, aimed at making more accurate and reliable predictions in drug discovery processes. Inspired by the success of modelling the electrostatic part of the halogen bonding (XB) by means of the distributed multipole expansion, a study is presented which attempts to extend this approach to a tougher case of σ-hole interaction: sulphur-based chalcogen bonding. To that end, 11 anisotropic models have been derived and tested for their performance in the reproduction of reference ab initio molecular electrostatic potential. A careful examination resulted in three models which have been selected for further examination as a part of the molecular mechanics force field (GAFF). The combined force field was used to estimate inter- and intra-molecular interactions for the molecular systems, capable of differentiating the binding from the σ-hole and other directions. The anisotropic models proposed were generally able to correct the wrong predictions of the sulphur models based only on isotropic charges and, thus, are a promising direction for further development of the refined electrostatics force fields.

Quadrupole correction for halogen bonding description in virtual screening and molecular docking

Halogen bonding is electrostatic attraction between halogen atoms in an organic molecule and Lewis bases. It is important to consider halogen bonding during molecular docking and virtual screening, in particular, at early stages of drug development. A new scoring function AutoDock-XB, which takes into account halogen bonding by means of the quadrupole correction, has been constructed. The function has been tested for a series of phosphodiesterase-5 inhibitors.