Title: "Analysis of single-cell transcriptome measurements of complex tissues"
Speaker: Peter V. Kharchenko; Center for Biomedical Informatics, Harvard Medical School, Boston, MA

Place: Lilly (LILY) Hall G126
Date: December 2, 2014; Tuesday
Time: 4:30pm


Recent extensions of the RNA-seq methods provide means for examining the overall transcriptional state of many individual cells. A key advantage of such single-cell assays is the ability to identify subpopulations comprising complex tissues and cell mixtures. The analysis of such data, however, is complicated by high levels noise stemming from both technical factors, as well as intrinsic biological variability. I will describe a computational method for combined identification and interpretation of heterogeneity within groups of cells assayed using single-cell RNA-seq. The method will be applied to characterize subtypes of neuronal progenitor cells found at mid-embryonic stage in the mouse brain, identifying multiple groupings of progenitor cells based on the independent aspects of their transcriptional state.

The transcriptional state of a cell reflects a variety of biological processes, including persistent regulatory configuration traditionally associated with a cell type, transient processes such as cell cycle stages, local metabolic demands, or extracellular interactions. Our approach aims to decompose this complex transcriptional signature by identifying known or newly-discovered gene sets that are linked to statistically significant heterogeneity within the measured collection of cells. This provides important clues about likely functional interpretation of the detected heterogeneity, allowing one to focus on relevant cell groups and differential expression signatures, while controlling for unrelated sources of heterogeneity.

Associated reading:
1. Bayesian approach to single-cell differential expression analysis. 2014. Nature Methods.

2. Bayesian approach to single-cell differential expression analysis. Supplement. 2014. Nature Methods."

Click here www.stat.purdue.edu/~doerge/BIOINFORM.D/FALL14/sem.html for a full schedule of BIOINFORMATICS SEMINARS, past and present.