Computational tools for spatially resolved transcriptomic data analysis (Brendan Miller)

About this Class

Dr. Brendan Miller is a post-doctoral research fellow at Johns Hopkins University in the Department of Biomedical Engineering. On Friday Feb 17, 1:00-2:00 PM, he will be discussing some tools he has recently helped develop for the analysis of spatially resolved transcriptome data. Delineating the spatial organization of transcriptionally distinct cell types within tissues is critical for understanding the cellular basis of tissue function. Recent technological advancements have enabled spatially resolved transcriptomic profiling, but new computational approaches are needed to take advantage of this new spatial information. In this talk, Dr. Miller will highlight two recently published computational tools for uncovering spatially resolved gene expression and cell type spatial organizational patterns in tissues. The first is MERINGUE, an approach to characterize significant spatial gene expression heterogeneity in spatially resolved molecular resolution data that is also robust to cellular density. The second is STdeconvolve, an approach to deconvolve cell types and their transcriptional profiles in spatially resolved multi-cellular pixel resolution data without a reference. Taken together, these tools can enable identification of cell type organizational patterns and distinct transcriptional states within poorly characterized tissues, such as tumors or other perturbations where the cell type composition and spatial organization remains unclear. MERINGUE and STdeconvolve are both available as open-source R software packages with code and tutorials available at https://jef.works/MERINGUE/ and https://jef.works/STdeconvolve/. Meeting ID: 160 400 8994 Passcode: 082861