We invite you to read a new scientific publication "Assessing how residual errors of scoring functions correlate with ligand structural features" co-authored by our statistician Nikita Ivanov.
The paper was published in the International Journal of Molecular Sciences as part of the special issue of QSAR and Chemoinformatics in Molecular Modeling and Drug Design 4.0.
The work is devoted to estimation functions that are used in drug design in the computational modelling phase. They are used to quickly estimate and predict binding energy of a ligand (future drug, active ingredient) to a target (enzymes, receptors, etc.). They are often used in virtual screening, where a library of hundreds of thousands or millions of structures are evaluated for affinity to a particular target. The rapid estimation of binding energy affects the accuracy of the prediction. In our work, we investigate the relationship between the presence of certain ligand functional groups and the reproduction error of binding energy estimation functions. In the future this will help to account for these errors and increase the accuracy of the estimated functions, which will have a very positive impact on the speed and quality of new drug development.
Nikita Ivanov is a scientist who is a member of a research group that works as part of the Laboratory of Medicinal Chemistry at the Chemistry Department of Moscow State University.
The group has several areas: creating new approaches in drug development (drug design) and developing methods for evaluating charges. The group works with analytical centre Exacte Labs on drug design. Read the article here.