t-distributed Stochastic Neighbor Embedding (t-SNE)

How to run t-SNE analysis on Pluto!

What is t-SNE?

t-distributed Stochastic Neighbor Embedding (t-SNE) is another common method for dimensionality reduction for visualizing data with multiple dimensions. The primary goal is to maintain pairwise similarities between data points so close neighbors remain nearby and distant points remain distant when reducing the data set down to two or three dimensions.

t-SNE visualization will produce distinct clusters and is particularly useful when you want to get an idea of how the data points are arranged when you have data points that relate to each other in a non-linear manner.

Set Up Your Analysis

  1. Navigate to your experiment and click on the + Analysis button on the right of the screen
  2. Scroll down on the sidebar and select t-SNE under "Dimensionality Reduction"
  3. Hit the Next button to generate analysis parameters. Here, you can select which samples you'd like to include or exclude in your analysis if applicable.
  4. Then hit "Run Analysis"

 

t-SNE_SelectAnalysisParameters



Customize Your Plot

Edit t-SNE plot

  1. Use the plot tab to customize your plots by clicking the three dots on the top right of your newly generated plot and selecting "Edit"
  2. Here, you can select what projections you want on your axes, the variable to group points by and your color scheme.
  3. Select "View Changes" to apply modifications to your plot.