# PCA, UMAP and tSNE scatter plots

After performing [exploratory analyses](https://help.connected.illumina.com/partek/partek-flow/user-manual/task-menu/exploratory-analysis) such as PCA, UMAP and t-SNE is is helpful to visualize the results on a scatter plot. This can help visually assess the source of variation affecting the results of an experiment, [classify](https://help.connected.illumina.com/partek/partek-flow/tutorials/analyzing-cite-seq-data/classifying-cells) cells and select samples for downstream analysis. Here we have a PCA scatter plot generated from the analysis of 12 samples from a scRNA sequencing study. The first three most informative PCs are plotted by default and the percentage of variation explained is stated next to each one of them.

![Figure 1. Example of a 3D PCA scatterplot](https://1384254481-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FJVEESmJAPppJ3ijFq5aR%2Fuploads%2Fgit-blob-12f92b021d5e9a14a0a0ecd8333e9e4442846453%2FScreenshot%202022-12-20%20at%2015.38.31.png?alt=media)

The *Configure >* *Style* menu on the left can then be used to color the features in the scatter plot based on an attribute (Figure 2). In this case, Figure 3 shows the cells being colored based on their cell-type.

![Figure 2. Customization menu](https://1384254481-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FJVEESmJAPppJ3ijFq5aR%2Fuploads%2Fgit-blob-740a19be1d5d73661ba16a25e71e153acb123f75%2FScreenshot%202022-12-20%20at%2015.43.59.png?alt=media)

![Figure 3. PCA scatterplot colored by cell-type](https://1384254481-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FJVEESmJAPppJ3ijFq5aR%2Fuploads%2Fgit-blob-9f237936d99a85be9942744c18f70f56e0d05733%2FScreenshot%202022-12-20%20at%2015.42.22.png?alt=media)

Additionally, you can adjust the opacity of the points to better assess the density across the groups (Figure 4). It is also possible to split the plot based on the same attribute in the *Configure > Grouping* menu (Figure 5).

![Figure 4. Adjusted opacity shows point density more accurately](https://1384254481-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FJVEESmJAPppJ3ijFq5aR%2Fuploads%2Fgit-blob-f030aaa3e8b8a857388623cadffd90d613fb2b68%2FScreenshot%202022-12-20%20at%2016.09.05.png?alt=media)

![Figure 5. Splitting by an attribute can help better visualize their effect on the data](https://1384254481-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FJVEESmJAPppJ3ijFq5aR%2Fuploads%2Fgit-blob-8ddf7a8a2a8a5303c57ef43818be9083bb8b33cd%2FScreenshot%202022-12-20%20at%2016.30.14.png?alt=media)

Click the **Save image** button <img src="https://1384254481-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FJVEESmJAPppJ3ijFq5aR%2Fuploads%2Fgit-blob-128a4b43bc88cc56946d07e3051884f8e7328af2%2Fimage2019-8-23_13-16-43.png?alt=media" alt="image2019-8-23_13-16-43" data-size="line"> to save a PNG, SVG, or PDF to your machine.

Click the **Send to notebook** button <img src="https://1384254481-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FJVEESmJAPppJ3ijFq5aR%2Fuploads%2Fgit-blob-bdb91ec221d7b883501afad19f115b6449f8f31b%2Fimage2019-8-23_13-17-9.png?alt=media" alt="image2019-8-23_13-17-9" data-size="line"> to send the image to a page in the Notebook.

These same options apply to [UMAP](https://help.partek.illumina.com/partek-flow/user-manual/task-menu/exploratory-analysis/umap) and [t-SNE](https://help.partek.illumina.com/partek-flow/user-manual/task-menu/exploratory-analysis/t-sne). Open the t-SNE results in the data viewer by double clicking the t-SNE data node. If there are already cell level attributes published to the project the Data viewer will automatically color the plot by the first cell level attribute (Figure 6).

<figure><img src="https://1384254481-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FJVEESmJAPppJ3ijFq5aR%2Fuploads%2Fgit-blob-6c4d4a66a8ef2316e478da5b46c7cc807a80508e%2Ft-sne%203d%20plot.png?alt=media" alt=""><figcaption><p><em>Figure 6. t-SNE results in the Data viewer are automatically plotted in a 3D scatter plot</em></p></figcaption></figure>

The t-SNE can also be plotted in a 2D scatter plot. Use **New plot** to select 2D Scatter plot then drag it to the canvas and select the t-SNE data node from the analyses pipeline (Figure 7).

<figure><img src="https://1384254481-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FJVEESmJAPppJ3ijFq5aR%2Fuploads%2Fgit-blob-bf3a530a59d0da29997a7821c1d5df239bf4a406%2Ft-snde%20add%202d%20scatter%20plot.png?alt=media" alt=""><figcaption><p><em>Figure 7. Choose 2D Scatter plot to add the t-SNE in 2D space</em></p></figcaption></figure>

This will plot the t-SNE in a 2D visualization which can be exported to your machine (Figure 8).

<figure><img src="https://1384254481-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FJVEESmJAPppJ3ijFq5aR%2Fuploads%2Fgit-blob-62da1092044fbf5c61f3e8b2216c11c93eb3fd13%2F2D%20scatter%20plot%20t-sne.png?alt=media" alt=""><figcaption><p><em>Figure 8. The t-SNE and related visualizations can be plotted as 3D or 2D scatter plots</em></p></figcaption></figure>

Use **New plot** or **Get data** to add the UMAP data node to the canvas (Figure 9).

<figure><img src="https://1384254481-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FJVEESmJAPppJ3ijFq5aR%2Fuploads%2Fgit-blob-0ac087c2fc44c6e1703d94f972653b1247685cb8%2Fumap%203d%20plot.png?alt=media" alt=""><figcaption><p><em>Figure 9. Plot as many visualizations in the Data viewer canvas as required</em></p></figcaption></figure>

The entire Data viewer canvas can be exported to your machine as one visualization using the left menu **Export image** <img src="https://1384254481-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FJVEESmJAPppJ3ijFq5aR%2Fuploads%2Fgit-blob-7ad8fccf6a677bf219e51c73a2a66291883f70fc%2Fexport%20image%202.png?alt=media" alt="image2019-8-23_13-16-43" data-size="line"> option.

## Additional Assistance

If you need additional assistance, please visit [our support page](https://www.illumina.com/company/contact-us.html#/other/technical-support) to submit a help ticket or find phone numbers for regional support.