QSAR Studies on the IC50 of a Class of Thiazolidinone/Thiazolide Based Hybrids As Antitrypanosomal Agents

Bo Yang,Hongzong Si,Honglin Zhai
DOI: https://doi.org/10.2174/1570180817999201102200015
2020-01-01
Letters in Drug Design & Discovery
Abstract:Background:: Trypanosomiasis is a widespread zoonotic disease and the existing drugs are not enough to prevent and treat it. Objective:: This study aimed to build a quantitative structure-activity relationship model by the chemical structures of a class of thiazolidone/thiazolidamide based hybrids. The model was used to screen new antitrypanosomal agents and predict the properties of composite molecules. Methods:: All compounds were randomly divided into a training set and a test set. A large number of descriptors were calculated by the software, then some of the best descriptors were selected to build the models. The linear model was built by the heuristic method and the nonlinear model was built by gene expression programming method. Results:: In the heuristic method, the correlation coefficients ,R2, R2cv, F and S2 were 0.581, 0.457, 14.053 and 15.311, respectively. In gene expression programming, the R2 and S2 were 0.715, 10.997 in the training set and 0.617, 22.778 in the test set. The results showed that the relative number of S atoms and the minimum bond order of an H atom had a significant positive contribution to IC50. Meanwhile, the relative number of double bonds and the count of hydrogen-bonding acceptor sites had a great negative impact on IC50. Conclusion:: Both the heuristic method and gene expression programming had a good predictive performance. By contrast, the gene expression programming method fitted well with the experimental values and it was expected to be beneficial in the synthesis of new antitrypanosomal drugs.
What problem does this paper attempt to address?