This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Widening the distances between PCA vectors in PCA a biplot

Hello.

I'm running a PCA analysis on R with 62 variables and when creating the biplot some vectors are getting on top of others and i cannot read their labels (see pic). Is there any way to make them more readable.

enter image description here

I used the factoextra package to plot it by running

fviz_pca_biplot(res.pca)

Thank you.

piblot bioplot pca r vector
• 3,944 views • link
updated by Alex Reynolds Alex Reynolds • written by arronar arronar

you could remove the black dot labels with fviz_pca_biplot(res.pca, label ="var") or you can simply get rid of black dots with fviz_pca_biplot(res.pca, invisible ="ind"). once you remove black dots, if you use xlim and ylim that only encircles your blue arrows. say your arrows are within the x axis range -3,2 while y axis range is 1,5 to -2 then do fviz_pca_biplot(res.pca, invisible ="ind") + xlim(-3,2) + ylim(-2,1.5) . I think this might widen the spaces.

• 0 views • link
written by firatuyulur firatuyulur

Thank you very much. Seems that it works.

• 0 views • link
written by arronar arronar

1 answer

In addition to using a smaller type size, you can render the plot to PDF, open with Adobe Illustrator, and add a stroke color of white and a thickness of 0.25pt. Adding a thin outline around text helps distinguish one label from another by keeping labels from blending into one another.

• 0 views • link
written by Alex Reynolds Alex Reynolds

Log in to answer this question.

More posts like this