Date: Friday Feb. 3rd at 3:00-4:00 pm EST
Place: Zoom https://gwu-edu.zoom.us/j/95224177633
Speaker:Yutong Sha (UC Irvine)
Title: Reconstruct Growth and Dynamic Trajectories from Single-cell Transcriptomics Data
Abstract: Time series single cell RNA-sequencing (scRNA-seq) datasets provide unprecedented opportunities to learn dynamic processes of cellular systems. Due to the destructive nature of sequencing, how to link the scRNA-seq collected at different time points remains a major challenge. Here we present a deep learning based dynamic unbalanced optimal transport to reconstruct the growth and dynamic trajectory simultaneously as well as the underlying gene regulatory network (GRN). We validate the framework on simulated data generated by GRN, one lineage tracing and two single-cell datasets, showing its accuracy of predicting cell fate, growth and underlying regulatory mechanisms. In general, it can be extended to high dimensional unpaired time series snapshots.