[R-sig-eco] Use of PCoa in varipart

[Fernando Rodrigues da Silva]

Hi Torsten,

Thank you for your answer.

I am aware that varpart function (vegan package) performs dbRDA when I use
data from PCoA, but I am really interested in the msr function that adjusts
for spurious contributions due to spatial autocorrelation. In the example
of Clappe et al. (2008 - Ecology 99, 1737-1747), the results of varipart
function (ade4 package) are used directly in the msr function. However, I
am unable to use the varpart results in the msr function.

I tried to contact Stephane Dray but his e-mail seems to block my messages.

All the best,


*Prof. Dr. Fernando Rodrigues da Silva *
*===========================================================**===*
Laboratório de Ecologia Teórica: Integrando Tempo, Biologia e Espaço (
LET.IT.BE),
Departamento de Ciências Ambientais (DCA),
Universidade Federal de São Carlos (UFSCar), campus Sorocaba

Personal Page: *http://fernandoecologia.wix.com/fernandorodrigues
<http://fernandoecologia.wix.com/fernandorodrigues>*

*       https://orcid.org/0000-0002-0983-3207
<https://orcid.org/0000-0002-0983-3207>    *


*===============================================================*


On Mon, Jun 8, 2020 at 10:49 AM Torsten Hauffe <torsten.hauffe using gmail.com>
wrote:

> Hi Fernando,
>
> using a PCoA for variation partitioning should be possible. This should
> equal a distance-based redundancy analysis (plus some magic to make the
> PCoA euclidean).
> I don't know ade4 very well but alternatively you could use the vegan
> function varpart. You can subject your Sörensen distance matrix directly to
> varpart.
>
> Cheers!
>
> On Sun, Jun 7, 2020 at 11:32 PM Fernando Rodrigues da Silva <
> fernando.ecologia using gmail.com> wrote:
>
>> Dear R users,
>>
>> I would like to know if there is a possibility to perform the same
>> analysis
>> of Clappe et al. 2018 (Ecology, 1737-1747), but instead of using a
>> Principal Component Analysis (PCA) in the function *varipart*, I use a
>> Principal Coordinates Analysis (PCoA) based on the Simpson index of
>> dissimilarity.
>>
>> I tried running the example of the article, but I get an error saying it
>> is
>> not yet available.
>>
>>
>> *comp_dist <- beta.pair (mafragh$flo, index.family =
>> "sorensen")pcoa.species <- dudi.pco (comp_dist$beta.sim, scannf = F)vprda
>> <- varipart (pcoa.species, environment, eigenfw, type = "parametric")*
>> # Error in dudi.type(Y$call) : Not yet available
>> *vprdaMSR <- msr (vprda, mafragh.lw, nrepet = 999)*
>>
>> Does anyone know if there is any script available for this type of
>> approach?
>>
>> All the best,
>>
>> *Prof. Dr. Fernando Rodrigues da Silva *
>> *===========================================================**===*
>> Laboratório de Ecologia Teórica: Integrando Tempo, Biologia e Espaço (
>> LET.IT.BE),
>> Departamento de Ciências Ambientais (DCA),
>> Universidade Federal de São Carlos (UFSCar), campus Sorocaba
>>
>> Personal Page: *http://fernandoecologia.wix.com/fernandorodrigues
>> <http://fernandoecologia.wix.com/fernandorodrigues>*
>>
>> *       https://orcid.org/0000-0002-0983-3207
>> <https://orcid.org/0000-0002-0983-3207>    *
>>
>>
>> *===============================================================*
>>
>>         [[alternative HTML version deleted]]
>>
>> _______________________________________________
>> R-sig-ecology mailing list
>> R-sig-ecology using r-project.org
>> https://stat.ethz.ch/mailman/listinfo/r-sig-ecology
>>
>

	[[alternative HTML version deleted]]