Good Morning,
first let me thank you very much for answering my first two questions on
this list.
Currently, i do vegan's EnvFit to simple PCA ordinations. When drawing
the biplot, one can set a cutoff to just fit the parameters with
significant p-values (via p.max=0.05 in the plot command).
There is already sufficient coverage on the net for biplotting this kind
of data with ggplot2 (with the problem being the arrow length).
http://stackoverflow.com/questions/14711470/plotting-envfit-vectors-vegan-package-in-ggplot2
However, what the solution is not covering is the exclusion from
insignificant environmental parameters, as the score extraction process
as described in the link only works with 'display="vectors"':
|Envfit_scores<- as.data.frame(scores(list_from_envfit, display=
"vectors"))
||||
|
envfit creates lists like this:
PC1 PC2 r2 Pr(>r)
param1 -0.70882 0.70539 0.0994 0.000999 ***
param2 -0.60122 0.79908 0.0593 0.000999 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05
'.' 0.1 ' ' 1
P values based on 999 permutations.
The list contains a vector called $pval containing the pvalues.
So, i need to reduce the list created by envfit to rows meeting a
criterion in $pval (via "unlist" and "which", i suppose).
However, i
have difficulties to work out the correct code.
Any help is much appreciated!
--
Tim Richter-Heitmann (M.Sc.)
PhD Candidate
International Max-Planck Research School for Marine Microbiology
University of Bremen
Microbial Ecophysiology Group (AG Friedrich)
FB02 - Biologie/Chemie
Leobener Straße (NW2 A2130)
D-28359 Bremen
Tel.: 0049(0)421 218-63062
Fax: 0049(0)421 218-63069
[[alternative HTML version deleted]]