Small molecules such as metabolites, signaling molecules, and disease-treating drugs are essential for life; they have even been called the "missing link in the central dogma" of biology. There has been a recent explosion of data measuring whole-genome responses to small molecule perturbations in many organisms, but there has been a lack of research combining bioinformatics and chemical informatics to elucidate gene-drug relationships from these data. This thesis describes computational methods for finding and characterizing both functional (indirect) and physical (binding) interactions between small molecules and genes or proteins. In one project, we analyzed ∼6,000 single-gene deletion strains in yeast, grown in the presence of several hundred small molecule treatments. Interestingly, nearly all gene deletion strains revealed a defective growth phenotype in some condition, suggesting that nearly all genes are required for growth, and genetic redundancy is limited. We also identified a large set of multi-drug resistance genes, which were surprisingly involved primarily in membrane trafficking. In a second project, we refined predictions of functional interactions in this dataset into predictions of physical interactions, utilizing the chemical structures of the compounds and features of the genes in the phenotypic assay. We found that incorporating knowledge of functional interactions improved the predictions of physical interactions. We predicted novel binding interactions and found external evidence for predictions that certain FDA-approved psychoactive compounds may have a secondary target, Cox17. Finally, using publicly-available small molecule screening data from humans and other organisms, we learned small molecule binding sites on proteins, using only 1-dimensional protein sequence and small molecule structure. This allows inclusion of any sequenced protein, the vast majority of which do not have solved 3-dimensional structures, and it identifies the actual sites of interaction. In all three projects, the learned interactions between genes and small molecules reveal gene functions and small molecule targets, and they should improve understanding of both basic biology and drug discovery.
|Advisor:||Davis, Ronald W.|
|School Location:||United States -- California|
|Source:||DAI-B 70/01, Dissertation Abstracts International|
|Keywords:||Drug discovery, Protein-ligand interactions|
Copyright in each Dissertation and Thesis is retained by the author. All Rights Reserved
The supplemental file or files you are about to download were provided to ProQuest by the author as part of a
dissertation or thesis. The supplemental files are provided "AS IS" without warranty. ProQuest is not responsible for the
content, format or impact on the supplemental file(s) on our system. in some cases, the file type may be unknown or
may be a .exe file. We recommend caution as you open such files.
Copyright of the original materials contained in the supplemental file is retained by the author and your access to the
supplemental files is subject to the ProQuest Terms and Conditions of use.
Depending on the size of the file(s) you are downloading, the system may take some time to download them. Please be