Surrogate approach for structure-based virtual screening

against unknown target proteins

Jiwon Choi¹,  Young Do Yoo ², Sukjoon Yoon¹

¹ Department of Biological Sciences, Research Center of Women's Diseases,

Sookmyung women's university, Seoul, Republic of Korea;  ² College of medicine, Korea University,

Seoul, Republic of Korea

Many peptides or small protein modulators play an important role in disease-related biological pathways, thus providing templates in designing drug-like molecules. For this purpose, information about the potential interacting proteins of the modulators will accelerate the discovery of lead compounds.

Unfortunately, binding counterparts (target proteins) of modulators are, in many cases, unknown. Here we present a data mining strategy to retrieve target-relevant 3D structural features from public databases for structure-based virtual screening [1].

Short motif search on non-homologous proteins enabled us to identify relevant structural templates for designing mimetics of given small protein modulators[2-3].The application of this method led to the identification of novel lead compounds in our antioxidants and anticancer drug discovery programs.

[1]   Surrogate Docking: Structure-based virtual screening at high throughput speed, Sukjoon Yoon, Andrew Smellie, Dave Hartsough and Anton Filikov, Journal of Computer-aided Molecular Design, 19(7): 483-497 (2005)

[2]   Rapid assessment of contact-dependent secondary structure propensity: relevance to amyloidogenic sequences, Sukjoon Yoon and William J. Welsh, PROTEINS: Structure, Function and Bioinformatics, 60(1), 110-117 (2005)

[3]   A hidden Markov model for predicting transmembrane helices in protein sequences,In J. Glasgow et al., eds., Proc. Sixth Int. Conf. on Intelligent Systems for Molecular Biology, 175-182. AAAI Press, 1998. 1