Missing `we.eBH` from the aldex2 differential.qza results

I find I cannot identify we.eBH from the aldex2 differential results; to plot a volcano plot from the tutorial Tutorial: Integrating QIIME2 and R for data visualization and analysis using qiime2R

I know when you run Aldex in R you get aldex.ttest wchih get you that output as a column, however, I cannot access it through the qiime output, maybe is on the object but I cannot find it?

results<-read_qza("songbird_test/differentials.qza")$data does not have a column called we.eBH, I have the following columns: Feature.ID, Intercept, and C(group1:group2), C(group2:group3)

results %>%
left_join(taxonomy2) %>%
mutate(Significant=if_else(we.eBH<0.1,TRUE, FALSE)) %>%
mutate(Taxon=as.character(Taxon)) %>%
mutate(TaxonToPrint=if_else(we.eBH<0.1, Taxon, “”))%>% #only provide a label to signifcant results
ggplot(aes(x=diff.btw, y=-log10(we.ep), color=Significant, label=TaxonToPrint)) +
geom_text_repel(size=1, nudge_y=0.05) +
geom_point(alpha=0.6, shape=16) +
theme_q2r() +
xlab(“log2(fold change)”) +
ylab("-log10(P-value)") +
theme(legend.position=“none”) +
scale_color_manual(values=c(“black”,“red”))

Just to check, this differentials artifact was generated using aldex and not songbird correct? Asking due to your directory (songbird_test) and the column titles don’t match up to anything I would expect from aldex (although I haven’t tried to use the glm function via qiime2). Can you export the data from differentials.qza using qiime tools export and see if the table imported to R matches what was inside the artifact?

Hi Jordan, the arctifact was done using

qiime songbird

as in https://github.com/biocore/songbird

I can’t seem to be able to export ?
qiime tools export --input-path differential.qza --output-path test-differentials

.py", line 62, in parse_type
type_expr = eval(string, {‘builtins’: {}}, locals_)
File “”, line 1, in
NameError: name ‘Differential’ is not defined

" may be needed to define it." % name)

qiime2.sdk.util.UnknownTypeError: Name ‘Differential’ is not a defined QIIME type, a plugin may be needed to define it

Oh ok, I could be wrong, but I don’t think the differentials type is strictly defined and as such you probably would not expect the we.eBH column that shows up in aldex output in songbird output. I don’t have first hand experience with songbird, but a quick look at the documentation would suggest that your differentials.qza file contains exactly what it is supposed to. For visualization, see section 8 on the songbird github repo. I’m not sure a volcano plot is the best way to represent the results.

1 Like