Any input file containing molecular structures is parsed and interpreted at several levels.
In the ACC environment, a molecule represents the sum of all structural elements in the input file, regardless of other annotations within the file (e.g., ATOM vs HETATM records, chain identifiers marking different components of a complex, etc.). In other words, molecule is used here as an umbrella term for everything from simple compounds to biomacromolecular complexes made up of proteins, nucleic acids, ligands, ions, water, etc. ACC thus assigns a unique identifier (molecule ID) to each molecule according to the input file name.
Each molecule is made up of a set of atoms. In the ACC environment, each atom is uniquely identified in the input file by its chemical element and atom serial number. Additional useful information can be used in the characterization of each atom, if such information is available in the input file (atom name and residue details).
We generally use the term residue to refer to any component of a biomacromolecule. Within the ACC environment, any collection of atoms bound by chemical bonds (covalent, coordinative or ionic) is considered a residue if this fact is appropriately indicated in the input file. Specifically, all the atoms that make up a residue should have the same residue name (3-letter code) and residue serial number. Not all file formats include residue information, but this has no bearing over the charge calculation procedure in ACC.
ACC operates with charges at several levels of resolution.
The smallest resolution visible in ACC is at the atom level. We use the term atomic charge to refer to a single value which represents the algebraic sum between the atom's nuclear charge (given by the number of protons in the nucleus), and the amount of molecular electron density associated to this atom. In space, the position of the atomic charge is given at the coordinates of the atomic nucleus.
The next level of resolution in ACC refers to residues. We use the term residue charge to refer to the sum of atomic charges over all atoms which belong to one residue. The origin of each atom is established according to residue name and serial number, provided this information is present in the input file.
The total molecular charge is a formal value based on the molecular structure and the protonation state of all ionizable groups. The process of atomic charge calculation in ACC consists of distributing the amount of electron density given by the total molecular charge across the molecule, to each atom based on certain criteria and algorithms. Thus, the sum of atomic charges over all atoms in a molecule equals the total molecular charge. Similarly, the sum of residue charges over all residues in a molecule equals the total molecular charge.
The work flow of atomic charge calculation in ACC is based on a few key terms.
Many modeling projects employ the concept of atom type to describe a set of atoms with common properties. Atom types are generally defined based on chemically relevant properties, or characteristics which are believed to correlate strongly with the chemical properties. Examples of atom types could be carbon atoms, non-polar hyrogen atoms, carbonyl oxygen atoms, etc.
Atom type scheme
An atom type scheme attempts to group atoms according to atom type by using some unified set of descriptors. ACC is currently able to distinguish two types of atom type schemes, namely according to chemical elements (C,H,O...), and according to chemical elements and bond order (C1, C2, H1, O1, O2...).
The Electronegativity Equalization Method (EEM) is the procedure by which atomic charges are calculated. EEM is an empirical approach, and employs special parameters for each atom type (see also the theoretical background). EEM parameters are generally developed for certain target molecules (organic molecules, zeolites, etc.) by fitting to reference data. ACC implements an EEM formalism where each atom type is characterized by 2-3 EEM parameters.
EEM parameter sets
A set of EEM parameters, or EEM parameter set, represents a collection of EEM parameters that has been developed for a certain group of target molecules, using a certain kind of reference data and a certain kind of fitting procedure. An EEM parameter set generally covers atom types for H,C,N,O, but also halogens, metals, etc, depending on the molecules targetted during the development of the EEM parameters. Reference data generally comes from high level Quantum Mechanical (QM) calculations. The applicability domain of an EEM parameter set is closely related to the applicability domain of the reference QM data used during the development. The accuracy of an EEM parameter set is thus evaluated as the ability of the resulting EEM charges to reproduce the reference QM data.
The computation method refers to the EEM implementation that will be used during the atomic charge calculation (see also the theoretical background), along with related computation options specific to each method. ACC implements the classical EEM formalism, along with two EEM approximations which increase time and memory efficiency, and allow real time calculations for extremely large biomacromolecular complexes.
An atomic charge calculation job in ACC is defined based on the molecule and its total charge, along with the EEM parameter set and computation method used. A single ACC computation may consist of multiple jobs, and all output files are thus named according to the job ID.
Each job produces a set of charges identified by a unique ID encoding the EEM parameter set, method and total molecular charge. If the original input file contained values of atomic charges, these are also stored in a charge set named according to the input file, and marked by the tag ref. All charge sets are available for analysis in ACC.
Return to the Table of contents.