<img height="1" width="1" style="display:none" src="https://www.facebook.com/tr?id=1052243361488264&amp;ev=PageView&amp;noscript=1">

Data Visualization – t-SNE Plots Explained

Posted on: May 30, 2019

Scientist in lab reading data

Flow cytometry has moved into the era of “big data,” and data plots have also shifted from scatter plots and histograms toward visualization methods that can handle this complex data. This era of computational cytometry has ushered in the use of T-distributed Stochastic Neighbor Embedding (t-SNE) data analysis, a machine learning algorithm applied to complex flow cytometry data. Check out these three reasons why t-SNE data analysis is a valuable data visualization tool for flow cytometry.   

  1. Managing Multiple Parameters: t-SNE data analysis has been widely used for flow cytometry analysis of multiple parameters. Flow cytometry staining panels go way beyond four colors currently, and some panels may stain for 20 or more markers. This means flow cytometry data analysis will need to generate plots for multiple markers on several different cell types. Manual analysis is not appropriate in this setting, but t-SNE data analysis is a type of dimensionality reduction method that can make a lower-dimensional plot, like a single bivariate plot, while preserving the structure of the high dimensional data. This results in a plot for a cell subset, such as CD4+ T cells, clustered into groups based on the intensity of staining for different markers.

  2. Data Reveals the Biology: t-SNE data analysis does not rely on users having preconceived parameters that define cell subsets. As such, new subsets of cells may be defined that were not previously revealed by less sophisticated analysis. This means t-SNE data analysis can reveal new biological insights about how cell types develop and differentiate.

  3. Untangling Terabytes of Data: The highly complex staining panels used currently for flow cytometry data analysis can generate very large datasets. Conventional analysis techniques cannot handle data at this scale, but t-SNE data analysis can use downsampling, or analysis of a representative subset of data, to analyze data within a reasonable time frame.

The application of t-SNE data analysis to flow cytometry data should involve an expert in computational cytometry or data visualization. Consider working with an experienced analyst as you design these complex experiments in order to assure that you are measuring appropriate parameters for t-SNE data analysis.

New call-to-action

Posted in: , , , , ,

Computational Cytometry | Flow Cytometry Data Analysis in the Era of Quantitative Data Science
 

Recent Blog Posts

Receive Updates By Email

Download Accelerated CAR T Cell Therapy with Flow Cytometry
Flow Cytometry & Immunology Research
Flow Cytometry for Clinical Trials