Description
Quantitative structure-activity relationship (QSAR) analysis is a methodology that works under the assumption that certain activities of a chemical compound can be related to its structure through a computational model. These models provide a cheaper and quicker means to evaluating drug compounds, when compared to current in vivo evaluation methods, and can help expedite screening of large chemical libraries. Unfortunately, these computational models can sometimes be very time consuming to develop due to the complex nature of the problem and lack of simple analytical solutions. In this thesis, we have significantly enhanced a novel QSAR toolkit developed by Ko et al. called evoQSAR, by adding two additional components, parallel computation and a graphic user interface (GUI). Parallelization helps to decrease computation time by increasing efficiency through the division of independent tasks, normally run sequentially, in parallel. This is achieved by leveraging Python’s built-in Multiprocessing module to harness the power of all the processors on one’s local machine. Due to architectural constraints, the implementation used in this thesis is limited to Linux and OS X operating systems (OS). A user-friendly GUI was implemented, in lieu of its default python scripting interface, to open the functionalities of evoQSAR to a greater audience, including chemists, other pharmaceutical companies, and fellow researchers. The GUI is implemented through the PyQt framework which leverages off of the popular Qt framework. The addition of parallelization and a GUI in the evoQSAR toolkit will enable chemists, pharmaceutical companies, academia, and others who could not previously utilize evoQSAR, either due to the time or learning curve associated with using it, the ability to use it for identifying novel drug compounds and expedite screening of large chemical libraries.
