Presented by:
Wen-Chieh Hsieh and McKeith Pearson

Abstract:
The initial step for creating effective therapeutic approaches is tounderstand the disease-causing mechanism in detail, which in many casesinvolves acquiring and analyzing massive amounts of biological data. However,conventional, human-involved analysis methods present serious issues whenprocessing high-information content image data, such as user-introducederror/biases and low throughput. For instance, examining protein subcellularlocalization and cell morphology typically requires quantifying hundreds of cellsper experiment through visual inspection and manually outlining areas withinmicroscopy images collected. The procedure is not only laborious but alsoimprecise restricting the efficiency of the data analysis process, and moreimportantly, affecting the accuracy of the interpretation after the data analysis.

To address such limitations of current methods, we are developingalgorithms to extract and quantitatively analyze biological information frommicroscopy images. Specifically, the algorithms were designed to quantitativelyanalyze the distribution of "green fluorescent protein" (GFP) tagged membraneproteins or morphological differences of cells. Importantly, we envision thesealgorithms will contribute to the development of therapeutic approaches againstdiseases.