Partial atomic charges for proteins

SCHINDLER, Ondřej, Tomáš RAČEK, Aleksandra MARŠAVELSKI, Radka SVOBODOVÁ and Jaroslav KOČA. Partial atomic charges for proteins. In 1st Student Conference in Structural Biology, 2021. 2021.

Basic information

• Original name: Partial atomic charges for proteins
• Name (in English): Partial atomic charges for proteins
• Authors: SCHINDLER, Ondřej, Tomáš RAČEK, Aleksandra MARŠAVELSKI, Radka SVOBODOVÁ and Jaroslav KOČA.
• Edition: 1st Student Conference in Structural Biology, 2021. 2021.

Other information

• Type of outcome: Presentations at conferences
• Confidentiality degree: is not subject to a state or trade secret
• WWW: URL

Abstract

The distribution of the electron density determines many characteristics of the molecule. Electron density is difficult to represent; however, it can be approximated by the partial atomic charges, which are real numbers describing what portion of electron density belongs to each atom. These charges are often used in computational chemistry, chemo- and bioinformatics. Unfortunately, finding the electron density for large proteins is not possible. For this reason, faster but still reasonably accurate empirical methods must be used. These methods try to replicate the results of quantum mechanical (QM) calculations. Nevertheless, the calculation of partial atomic charges for proteins is not yet fully solved. Recently, we have developed the empirical method Split Charges Equilibration with Initial Parameterized Charges (SQE+qp) to tackle this issue. Additionally, we have suggested a new definition of the atomic types (i.e., grouping the atoms into the classes). In particular, each atom is assigned the type based on the elements of the bonded atoms. This fine-grained division ensures the precise assignment of the initial charges. However, such a model has many more parameters than the previous ones, so it was also necessary to design a faster parameterization technique [1]. To demonstrate our results, Figure 1 shows the strong correlation between the reference QM charges and charges of the parameterized SQE+qp model on the dataset of 60 proteins and 60 peptides. We currently work to extend this methodology to support the calculation of partial atomic charges for metalloproteins which we believe is the crucial step to calculate charges for the majority of proteins in the PDB database. The implementation of SQE+qp is included in Atomic Charge Calculator II [2], a web application available freely at https://acc2.ncbr.muni.cz.