Natural products have been successfully used to improve human health as is evident by the number of products available on the market. However, existing methods used to identify and understand the bioactivity mechanisms of these compounds are extremely time-consuming and this limits the development of new natural products. To develop new drug molecules, we plan to develop Quantitative Structure Activity Relationship (QSAR) models for the prioritized bioactivity assays (e.g. those relevant to anxiolytic mechanisms in our recent project). This will be done in order to construct the original mechanism profiles of the target compounds of interest. Then, we will use these resulting models to virtually screen the natural product compounds without experimental data. The models may also generate the mechanism profiles needed to evaluate the MOA of target natural product compounds. We will select those compounds with the desired bioactivity and new chemical scaffolds that are different from the existing well-known pharmacophores. The identified natural product compounds will be experimentally tested using animal models.