Chemical Research in Chinese Universities ›› 2019, Vol. 35 ›› Issue (6): 1111-1118.doi: 10.1007/s40242-019-9183-5

• Articles • Previous Articles     Next Articles

A Strategy to Find Novel Candidate DKAs Inhibitors Using Modified QSAR Model with Favorable Druggability Properties

ZHANG Xiaoyi1, NIU Wenling1, TANG Tang1, HOU Chengfei1, GUO Yajie1, KONG Ren2   

  1. 1. College of Life Science and Bioengineering, Beijing University of Technology, Beijing 100124, P. R. China;
    2. Institute of Bioinformatics and Medical Engineering, School of Electrical and Information Engineering, Jiangsu University of Technology, Changzhou 213016, P. R. China
  • Received:2019-07-03 Revised:2019-08-04 Online:2019-12-01 Published:2019-11-29
  • Contact: ZHANG Xiaoyi E-mail:zhangxiaoyi@bjut.edu.cn
  • Supported by:
    Supported by the Project of the Beijing Municipal Commission of Education, China(No.KM201410005030), the Importation and Development of High-caliber Talents Project of Beijing Municipal Institutions, China and the National Natural Science Foundation of China(No.31100523).

Abstract: The study dealed with quantitative structure-activity relationship(QSAR) to explore the important features of diketo acid(DKA) derivatives for exerting potent HIV-1 integrase inhibitors activity. A three-step screening method was proposed to choose descriptors. Then, additional descriptors were used in the CoMFA and CoMSIA. Lastly, a modified CoMSIA m7 model, constructed by adding Csp2_03_F descriptor, showed better predictive ability. Validation parameters(Q2 and R2) for the models were 0.722 and 0.925, respectively. In addition, external validation for the models using a test group revealed Rpred2=0.892. Contour maps analysis defined favored and disfavored regions of the compounds, and two new compounds with the descriptor structure were designed with better activities than Ralte-gravir(RAL), well drug-likeness and low toxicity. The research provides a base for further DKA development.

Key words: Diketo acid (DKA), Modified quantitative structure-activity relationship (QSAR), Autodock, Drug design, Descriptor screening