This is the dissertation I submitted in partial fulfillment of the requirements for the degree of Bachelor of Arts in Computer Science. In it, I use a customized version of the Correlated Topic Model (a Bayesian network usually used for natural language processing and text analysis) to predict drug-pathway interactions (essentially trying to make a judgment on the effect of a drug based on its gene expression profile).
The PDF version of the dissertation is available at https://www.dropbox.com/s/fgqkmz3hzudj5b4/dissertation.pdf?dl=1.