Making a better-looking barplot with the CSVs the program gives you.

Hi forum,

I've got raw CSV outputs from the barplot program and need to redo these barplots in something else very quickly because my committee didn't think the QIIME2view SVG output was up to snuff. What are your favorite ways to start with CSVs of feature counts that come out of the barplot program and turn them into relative abundance plots? Beyond turning those values into percentages I'm not actually sure what I'm doing since I haven't used R in ages, and from what I've seen this is likely to be quicker in R. (also, if you've got code you're willing to post I'd be extra thankful.)

I like making plots in R using the Tidyverse

Phyloseq is also popular, see plot_bar(), and you may find qiime2R helpful for importing data.

1 Like

If you need quick and dirty, excel can be good. Is the issue the number of colors? If so, you can filter to discard anything less than either 1% on average or just retain the top 7-11 items. Leave the remainder as "other" and let that be your last color. (A human brain can see distinguish, on average, about 8-12 distinct colors).

You may need to invest some extra time making the plots look like they didnt come from excel; I'd suggest turning off shading, hiding lines, and making sure that you use a non-default color map. (Colorbrewer2 is a great resource; you can set a hex color in powerpoint or just use the dropper).


1 Like