Basic algebrra seems to lead to this function:
rarederatk -
function (x, k)
{
x - x[x0]
J - sum(x)
d - digamma(J-k+1) -digamma(J-x-k+1)
g - lgamma(J-x+1) + lgamma(J-k+1) - lgamma(J-x-k+1)-lgamma(J+1)
d - d*exp(g)
sum(d[is.finite(d)])
}
Here 'x' must be a vector of species
Hi all,
I've made a network of OTUs using phyloseq. Now, I'd like to correlate
environmental variables to the clusters I observed. Anybody here who has
an idea on how to do this exactly? Perhaps there are multiple ways, or
if it is possible using another package (network, sna, ...), please
Apologies for cross-posting
We would like to announce the following statistics course in Palm Cove,
Australia.
Course: Introduction to zero inflated models with R
Location: Palm Cove, Australia
Date: 17-21 August 2015
Price: 550 GBP
Course website: http://www.highstat.com/statscourse.htm
Hi Bjorn,
You could check out Local Similarity Analysis (http://meta.usc.edu/softs/lsa/)
or alternatively, network assortativity e.g. in {igraph}
http://www.inside-r.org/packages/cran/igraph/docs/assortativity
Hope it helps,
Cheers,
Johannes
On May 11, 2015, at 4:40 PM, Bjorn Tytgat wrote: