![]() The biochemical algorithms library (BALL) software from Dehof et al. developed a heuristic method to determine bond orders based on arbitrary penalty scores . The methods of Froeyen and Herdewijn and Labute could theoretically be used on structures with only atom type and connectivity information, but they were developed primarily for use when 3D coordinate information, albeit for only the heavy atoms, is provided. All of these methods, however, require accurate 3D coordinate information. assign bond orders based on characteristic bond lengths, bond angles and torsion angles, as do Hendlich et al., who also include small functional group identification to help avoid incorrect assignments due to erroneous or ambiguous geometrical data . Baber and Hodgkin follow a similar scheme, but can also assign bond orders . IDATM from Meng and Lewis can be used to determine the connectivity and hybridisation state of atoms based on input three-dimensional (3D) coordinates. The COBRA program of Leach et al. uses a backtracking search algorithm to automatically assign bond orders. ![]() As such, over the last few decades, a number of such schemes have been developed. However, this approach can be computationally expensive, and with the advent of large databases of organic molecules, such as the Protein Data Bank (PDB) and Cambridge Structural Database (CSD), the need for fast automated schemes became pertinent. The best way to determine the Lewis structure of a molecule is to calculate the actual electronic density distribution and then use the Natural Bond Orbital method to obtain bond orders and formal charges. Both properties can be easily deduced from the Lewis structure of a molecule, which shows how the valence electrons are arranged amongst the atoms and bonds of the molecule. The formal charge is the charge assigned to an atom in a molecule assuming that electrons in all chemical bonds are shared equally between atoms, and the bond order of a bond is the number of chemical bonds between a pair of atoms. The ability to assign chemical characteristics such as bond orders and formal charges is crucial to many higher-order algorithms in computational chemistry and cheminformatics. We show that our method can assign bond orders and formal charges at a high degree of accuracy across a wide range of molecules from two different databases, and that the FPT algorithm provides the best combination of speed and accuracy. We tested three different optimisation methods-local optimisation, an A* pathfinding method, and an FPT optimisation method utilising tree decompositions-for finding the best electron position assignment, from which the bond orders and formal charges are extracted. Each electron position assignment is scored according to lookup tables of atomic and bond dissociation energies derived from quantum chemical calculations. Our method utilises a graph theoretical description of electron positions. Here we describe a method for determining the bond order and formal charge assignments from only the atom types and connectivity. In cheminformatic applications it is necessary to be able to assign these properties to a given molecular structure automatically, given minimal input information. Bond orders and formal charges are fundamental chemical descriptors.
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |