Hi Floris,

If you used the multiple threshold version of the method the different colours represent different thresholds (timing of shift between diversification and coalescent processes). (At least an earlier version plotted output like this, whereas single threshold plotted all clusters in red.) If this is what you have used, make sure you have the latest version of the code as it has been revised since the original implementation. See Fujisawa & Barraclough 2013 for details and for how to determine whether the multiple threshold is a statistical improvement on the single threshold version: http://sysbio.oxfordjournals.org/content/early/2013/05/16/sysbio.syt033

//Aelys

Dear James Pettengill

Thank you for your answer. I have checked the figure in illustrator and the colours are identical. R clearly reused the colours. I was surprised it would do some directly after it was used and not after it went through the entire cycle of colours. If you think it is usefull I can upload the pdf as well and you can more clearly view the different clusters.

Sincerely

Floris

Hi Floris,

Are you using package "splits" and running the species delimitation method described by Pons et al. 2006? If so, the best way of checking which groups are considered different species by the analysis is using the "spec.list" command. So, just to make it clear, your whole analysis would look something like:

require(splits)  #load the package

tree

Dear Fabricius.

Thank you very much for this detailed despription

I have done what you wrote before. I have indeed used the splits package and the tree I have uploaded is the result from this. So the cluster colours are randomly assigned then. I will continue with that information.

Sincerely

Floris

Hi Floris,

If you used the multiple threshold version of the method the different colours represent different thresholds (timing of shift between diversification and coalescent processes). (At least an earlier version plotted output like this, whereas single threshold plotted all clusters in red.) If this is what you have used, make sure you have the latest version of the code as it has been revised since the original implementation. See Fujisawa & Barraclough 2013 for details and for how to determine whether the multiple threshold is a statistical improvement on the single threshold version: http://sysbio.oxfordjournals.org/content/early/2013/05/16/sysbio.syt033

//Aelys

Dear Aelus Humphreys

Thanks for you contribution. Yes I have used multpile thresholds. The single treshold approach was lacking (multpile thresholds performed much better) .I have done these analysis rather recently and have used the most recent codes as far as I know. 

Sincerely

Floirs

Hi Floris,

for your goal you can ignore the various colours. The most important fact for: monophyletic groups of of a single color represent a single species. Various monophyletic groups may share identical colors, but they are nevertheless different species.

Martin

Thank you everybody for your contributions. I think I can go on with my work. You even clarified a few things I did not think about before.

Sincerely

Floris

Hello everyone

I have a question... I use the confset

Hi Camila,

confset.gmyc() will give you a list of models retained in your 95% confidence set of models.

If you want support values for each node, based on the confidence set of models, use gmyc.support()

You can then use those values as nodelabels() to add them to your plot of the tree obtained using plot(tr) .

Aelys