Entering edit mode
3.0 years ago
Claire
•
0
How can i replace the p.adj.threshold by the normal pvalue and to plot it in the heatmap ? What i want it is to plot the p value and not the p.adj
# Define data sets to cross-correlate
y <- log10(abund_table)
x <- as.matrix (meta_table)
# Cross correlate data sets
correlation.table <- associate(x, y, method = "pearson", mode = "table", p.adj.threshold = 0.05, n.signif = 1)
## map figure
p <- heat(correlation.table, "x", "y", fill = "Correlation", star = "p.adj", p.adj.threshold = 0.05)
print(p)
What packages are you using? Always show your
sessionInfo()unless you're being explicit about the package you have an issue with.i used microbiome package.
You should have mentioned that in your post and added it as a tag. Their manual does not seem to give explicit examples on how to skip p-val adjustment. Try not giving the
p.adj.threshold = 0.05argument - it is defaulted toInfand that might work in your favor. From a cursory glance at the source code, raw pvalue is only returned when there is no significant adjusted p-value.You've also copy-pasted lines from the manual here but omitted a few lines.
correlationsis not the same object/format ascorrelation.table. Look at these objects and see if either of them have a raw pvalue field.Given that this is a bioConductor package, you may have better luck if you post on bioC Support. If you create a post there, edit your post here and add a link that points to your new post.
p.adj.threshold = 0.05 is an argument by defaulted. I corrected the script.
What do you mean by this? See the function definition (by running
?associate) - the default value for this argument isInf.i mean that this argument it could not be changed in this function.
Why can it not be changed?