Speaker: Shuxiong Wang, UC Irvine
Sequencing the transcriptomes of single cells has greatly advanced our understanding of the cellular composition of complex tissues. In many of these systems, the role of heterogeneity has risen to prominence as a determinant of cell type composition and lineage transitions. While much effort has gone into developing appropriate tools for the analysis and comprehension of single cell sequencing data, further advances are required. Here we present SoptSC: an optimization-based algorithm for the identification of subpopulation structure, transition paths, marker genes, and pseudotemporal ordering within a cell population. Based on a measure of similarity between cells, SoptSC identifies cell populations via non-negative matrix factorization and infers pseudotime by computing shortest path of the cell-cell interaction graph based on binary adjacent matrix. We compare our methods with current state of the art on clustering and pseudotime inference on several published single cell datasets. It follows that SoptSC can identify cell populations and recapitulates known trajectories with high accuracy.
Wednesday, March 7 at 1:10pm to 2:00pm
You may login with your UCR NetID
In order to submit events, users must have a UCR email address (@ucr.edu). All events submitted by non-UCR users WILL be deleted.