Ordering comp_barplot() by genus of interest

[sbachas]

Hi,

Is there an easy way to order samples within comp_barplot() for a specific genus of interest (i.e., if you're interested in a Prevotella, to have samples ordered in the plot by largest to smallest % Prevotella)? Is that an option within sample_order?

Thanks!

[david-barnett]

Hi @sbachas !

This is possible, by first aggregating and transforming the taxa to compositions, then arranging the samples with ps_arrange, then plotting comp_barplot with "default" sample_order, which keeps the order as already set.

Here's an example, but beware that this might not look so good with a less abundant taxon. Hope this helps!

data("dietswap", package = "microbiome")
dietswap %>% 
   tax_agg("Phylum") %>% 
   tax_transform("compositional") %>% 
   ps_arrange(Bacteroidetes, .target = "otu_table") %>% 
   comp_barplot(tax_level = "Phylum", sample_order = "default")

[sbachas]

Thank you, this is great! One more question: when plotting the heat map using cor_heatmap(), the prevalence isn’t plotting (prev box has no data points), though abundance does plot. Do you know what could be going on? Thanks again! On May 6, 2021, at 2:42 PM, David Barnett ***@***.******@***.***>> wrote: Hi @sbachas<https://github.com/sbachas> ! This is possible, by first aggregating and transforming the taxa to compositions, then arranging the samples with ps_arrange, then plotting comp_barplot with "default" sample_order, which keeps the order as already set. Here's an example, but beware that this might not look so good with a less abundant taxon. Hope this helps! data("dietswap", package = "microbiome") dietswap %>% tax_agg("Phylum") %>% tax_transform("compositional") %>% ps_arrange(Bacteroidetes, .target = "otu_table") %>% comp_barplot(tax_level = "Phylum", sample_order = "default") [11d6f3ff-1264-4c6e-948b-483f2a90431d]<https://user-images.githubusercontent.com/33695702/117368314-a5e47280-aec3-11eb-9640-9b58157e3480.png> — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub<#24 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AT7ILSAKJD3OISDZUB5WN6DTMMEMNANCNFSM44H5S77Q>.

[david-barnett]

no problem 🙂

I'm not sure about your problem with cor_heatmap, the prevalence plot should show a barchart.
Can you post the code?