Molecular properties modelling
Tools for modeling conformational and electrostatic properties
The energy and properties of a chemical system depend on the exact three-dimensional molecular structure. Subtle variations in functional groups can result in dramatic differences in behavior. Force field methods that represent the potential energy of a molecule as simple functions of atomic distances and bond angles have proven to be an efficient approach in obtaining accurate relative energies for chemical systems. The efficiency of force field-based calculations allows the exploration of large portions of the conformational space, revealing the detailed relationship between structure and energy.
Force field-based molecular modeling is routinely applied to examine molecular conformations, molecular motion, and intermolecular interactions, such as those in a ligand-receptor complex. Reproducing bioactive ligand geometries in minimally sized conformer sets, accurate results from high-performance conformation generations save time and effort in downstream applications.
Conformer generation is useful in many aspects of both molecular modeling in general and drug discovery in particular. The relative energies of small molecule conformations play a crucial role in determining shape, function, and activity. Moreover, the ability to generate a bioactive conformer is a vital pre-requisite to any successful computer-aided drug design project.
While it is impossible for a conformer search algorithm to determine a flexible ligand’s bioactive conformer with absolute confidence, carefully considered search criteria do allow an algorithm to reject conformers likely to be of high energy or inactive. Beyond merely expediting the conformer search process, this approach creates efficiently sized conformer sets that nevertheless contain a reasonable approximation of the bioactive geometry. Efficient conformer sets have wide-ranging ramifications in downstream applications. For example, with fewer irrelevant conformations to process, virtual database screens and shape-based similarity searches run to completion in a fraction of the time without sacrificing accuracy.
The calculation generates the stable 3D structures, i.e. conformers of a given molecule. The stability of the conformers is estimated through comparing their energy content, calculated in the framework of a molecular mechanics force-field. The 3D structure (conformation) strongly affects the properties and the reactions of molecules.
Besides generating conformer ensembles of a compound, the Conformation plugin can be used to create a lowest energy conformer only. For this purpose different levels of accuracy (i.e. convergenc criteria and the involvement of post-optimization steps) can be used. The lowest energy conformers can be generated using the Dreiding (ref) or the Merck Molecular force-fields (MMFF94) (ref).
Flexible and rigid 3D Alignment
The 3D Alignment plugin supports the calculation of the maximal overlap of two molecules by two different approaches. The first, rigid approach restores the input conformation, whereas the second, flexible approach maximizes the overlap of atoms by an on-the-fly adjustment of the 3D coordinates.
In both cases atoms of the same type are mapped on each other to add chemical relevance to the alignment process. The mapping can be based on different approaches, these include extended atom types, pharmacophore points, common scaffold patterns or user defined atoms.
Types can differentiate atomic number, hybridization state and aromaticity, eg. ethene and benzene cannot be aligned. The input compounds (two or more) can be defined in 2D or 3D; in the former case 3D coordinates will be generated automatically.
Molecular dynamics simulations aim to describe the molecular structure not as rigid snapshot in time, but rather as a dynamically changing entity. The structure of flexible molecules can be described much more accurately via trajectories from molecular dynamics simulations than via single conformers. The Molecular Dynamics Plugin propagates the positions of the atomic nuclei on the potential energy surface defined by a molecular mechanics force-field, the Dreiding force-field. The resulting trajectory can be reviewed as a molecule array or animation.
The variable charge distribution of polarizable systems plays a key role in the physicochemical and binding properties of these compounds. ChemAxon’s Charge bundle contains a collection of calculations to estimate partial point charges and polarizability of molecules.
Partial charge calculations
Partial charge distribution determines many physico-chemical properties of a molecule, such as ionization constants, reactivity and pharmacophore patterns. The Charge plugin computes the partial charges on each atom based on a modified method of Gasteiger (J.Gasteiger and M.Marsili: Tetrahedron Vol. 36. , pp. 3219-3288 (1980), M.Marsili and J.Gasteiger: International Symposium on Aromaticity, Dubrovnik, Yugoslavia , Sept (1979), Croat.Chim.Acta. (1979) ). Total charge is calculated from sigma and pi charge components, any of these three charge values can be displayed on a molecule surface in MarvinSpace.
The electric field generated by partial charges of a molecule spread through intermolecular cavities and the solvent that the molecule is solved within. The induced partial charge (induced dipole) has a tendency to diminish the external electric field which effect is termed polarizability. Our calculation takes into account the effect of partial charges upon atomic polarizability as well as 2D and 3D geometries. The more stable each ionized site is, the more polarizable is its vicinity. This is why atomic polarizability is an important factor in the determination of pKa and why it is also considered in our pKa calculation plugin.
Atomic polarizability is altered by partial charges of atoms. Our calculation takes into account the effect of partial charge upon atomic polarizability as well as the 2D and 3D geometries.
Orbital electronegativity is the measure of the ability of atoms to attract electrons in the context of chemical bonds. As the partial charge distribution of a molecule is governed by the orbital electronegativity of its atoms, this measure can be predicted together with the partial point charges of the atoms in an iterative approach, where the atoms’ ionization potential and electron affinity are taken into consideration.