Beck, DavidGuo, Jiayuan2018-07-312018-07-312018-07-312018Guo_washington_0250O_18998.pdfhttp://hdl.handle.net/1773/42224Thesis (Master's)--University of Washington, 2018Machine learning is a powerful technique to analyze massive *omics data. Combined with visualization approaches, such as gene expression profile graphs, machine learning algorithms have found great use in exploring the hidden mechanism in *omics field. This work presents a user-friendly web application, called DashOmics to efficiently compute unsupervised machine learning algorithms on *omics data and interactively visualize machine learning results and gene expression profiles to provide insight into underlying gene expression patterns. The functionality of DashOmics includes K-Means clustering algorithms, Elbow Method and Silhouette Analysis as model evaluation methods to explore clustering analysis on *omics data, and also Principal Component Analysis (PCA) to reduce dimensionality and visualize data in an intuitive way. It is open-source and freely available on GitHub at https://github.com/BeckResearchLab/DashOmicsapplication/pdfen-USnoneclusteringgene expression profilemachine learningomics dataChemical engineeringBiochemistryChemical engineeringThesis