Goknur will be hosting WEHI Bioinformatics Seminar at WEHI on May 26th on behalf of WEHI Bioinformatics Division
Time: 10:45am Tuesdays.
Venue: Level 7 Seminar Room 2, WEHI1
Speaker: Jerry Gao – WEHI Molecular Medicine
Visualising high-dimensional data using tSNE: Not just a new kid on the block, but a good kid on the block
Recent technological advances have enabled detailed interrogation of biological systems by simultaneously measuring a large number of parameters at single-cell resolution. However, this explosion of data generation presents new challenges in visualization and interpretation of the data in a meaningful manner. In this talk I introduce the t-Distributed Stochastic Neighbor Embedding (tSNE) algorithm, which is a recently developed dimensionality reduction technique for visualizing and exploring high-dimensional data sets. I show examples of tSNE having superior performance over other dimensionality reduction techniques such as PCA and also present my application of this algorithm to study aspects of haematopoiesis. This includes using tSNE to unmask complex traits and heterogeneity when assessing cell fate in cellular barcoding data as well as to link cellular phenotype with cell fate at the single-cell level.