Single-cell RNA sequencing can provide data on a potentially large number of individual cells. The resulting expression data is a matrix with cells and genes, and — because of its sheer size and potential information content — is challenging for data analysis. Data filtering, projection, clustering, cell trajectory discovery, the discovery of marker genes, and cell-type classification are just a few approaches we can apply to scRNA data. This talk will argue that despite this jungle of methods and data, we can train any biologists, within a few hours, to apply the latest algorithms of data science and get insight into their single-cell data. The key to this endeavor is the easy-to-use tool for data mining that features interactive visualizations and intuitive, visual programming approaches for the construction of data analysis workflows. We will demonstrate the use of one such tool, scOrange (https://singlecell.biolab.si
), on several recent single-cell data sets. We will show that single-cell data analysis does not necessarily require in-depth knowledge of computer science, statistics, and machine learning, and that construction of advanced data analysis workflows can be as easy as playing with Lego bricks.
Contact Person: C. Combi