Category: Assay Development and Screening
Understanding biology in situ at the tissue level is important for understanding human disease. High resolution imaging of tissue reveal the relationships and interactions between individual cells, and new methods are probing the transcritome and proteome of these at greater depth and in greater detail. In combination with advances in the automation of microscopy and computing power, biological images have reached unprecedented abundance and detail. The analysis of these images lags behind their creation. Recent efforts have been made by the Broad Imaging Platform to meet these demands. The open source software CellProfiler is developed and maintained by the Broad Imaging Platform, and its functionality has been expanded to quantify tissue image data. An example will be presented showing the analysis of a tissue microarray of human small intestine to quantify epithelial ion transport via DRA expression. The quantification of DRA along the proximal and distal axis of the small intestine reveals a gradient that varies between healthy and inflamed tissue. The method of analysis includes using machine learning to train models that classify crypts, villi, and connective tissue. These identified regions of the small intestine are then further processed to segment individual cells within using CellProfiler. CellProfiller tools also create visualizations of the gradient. A tutorial of this analysis will be made available online.
Kyle Karhohs– Postdoc, Broad Institute, Cambridge, Massachusetts
Kyle obtained a BS in Electrical Engineering from the University of Nevada, Reno and a PhD in Systems Biology from Harvard University. Since 2016 he has been a postdoc at the Broad Institute Imaging Platform performing the role of an assay developer. His work focuses on quantifying bioimages, especially large datasets such as high-content screening and slide scans of tissue sections. He makes contributions to the open source project CellProfiler, a software for analyzing bioimages.