EXPLORING THE SPACE OF COMPUTER-AIDED DRUG DESIGN - MODULATOR DESIGNS TARGETING PFKFB3 AND COMPUTATIONAL WORKFLOW DEVELOPMENTS

This thesis involves works relating to the application and development steps within the computer-aided drug design (CADD) process. Tackling atherosclerosis via the path of metabolism modulation by targeting PFKFB3 is a novel idea regarding atherosclerosis treatment. The disease has been connected to abnormal up-regulation in the cellular metabolism process, predominantly through an elevation in glycolysis. Part of the work elucidated here applied a multi-strategy approach targeting an important bifunctional enzyme, PFKFB3, which regulates the glycolysis process. With the object of modulating PFKFB3 activity, multiple strategies have been attempted, resulting in several potential hit compounds for further testing and developments. Moreover, a virtual screening workflow was developed and tested. This workflow incorporated an additional rescoring procedure to re-evaluate the screening outcome. A series of test sets were generated and used for benchmark study. The rescoring procedure was further dissected, and statistical analyses were performed. Furthermore, it was suggested that water molecules play important roles in bridging ligand-receptor interactions. A C++ program was developed to select important binding site water positions in both holo and apo protein structures. The program can be easily incorporated into a virtual screening process and initial testing has shown promising performance. Additionally, the applicability of network analysis in biomarker detection was also preliminarily investigated. The early stage of a network analysis workflow development was carried out and was demonstrated with promising results. This thesis can provide further insights into, and also showed bright future prospects of, CADD studies.