bPTP interpretation

[jj.ja...@gmail.com]

I have been using a server based bPTP.

Have question regarding the interpretation of the output

Is there any cut-off which we can use to identify the species?

If there is support of 0.95 vs 0.49? how does one interpret 0.49? does it mean it has multiple species? or not resolved? can we treat support values as posterior probability?

what does zero mean at any node?

Thanks in advance

Jahnavi

[Alexandros Stamatakis]

Dear Jahnavi,

Please have a look at:

http://species.h-its.org/help/

under point 10:

Does that help?

Alexis

--
Alexandros (Alexis) Stamatakis

Research Group Leader, Heidelberg Institute for Theoretical Studies
Full Professor, Dept. of Informatics, Karlsruhe Institute of Technology
Adjunct Professor, Dept. of Ecology and Evolutionary Biology, University
of Arizona at Tucson

www.exelixis-lab.org

[Luis Javier Chueca]

Hi! 

I have the same question and the point 10 didn't resolve my doubts, If the estimated number is species is between 16 to 25, but I don't have 16 species with PP values upper to 0.9, what it means?

thanks!

[Alexandros Stamatakis]

So the PP values mean that, if you see a value of 0.95 on a branch/node,
all descendants of this branch/node are more likely to be from a single
species.

In other words, for the above example, 95% of the delimitations in the
posterior sample of delimitations (after the burnin) identified all
leafs below this node/branch as being a single species.

So if you don't have 16 nodes with PP values > 0.9 this means that the
delimitation of some of these is difficult.

Alexis

[Jahnavi Joshi]

Thanks for the explanation. 

But in many cases where it is 0, we know we are not dealing with species anymore, right? at many higher level nodes it shows support '0'. or branches the once in red should only be considered and the in blue are not part of the species delimitation process? 

If it will help to explain it better, I can send my tree file as example file. 

Thanks again for your advice. 

--
You received this message because you are subscribed to a topic in the Google Groups "PTP species delimitation" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/ptp-species-delimitation/bABK5APN5mo/unsubscribe.
To unsubscribe from this group and all its topics, send an email to ptp-species-delimitation+unsub...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

Jahnavi Joshi

Fellow, 
National Centre for Biological Sciences
Bangalore 560065, India.

[Alexandros Stamatakis]

yeah, please send the tree to me that will be easier,

cheers,

alexis

>> ptp-species-delimi...@googlegroups.com.

>> For more options, visit https://groups.google.com/d/optout.
>>
>
>
>

--

[Alexandros Stamatakis]

I looked at the tree

On 05/07/2014 11:54 AM, Jahnavi Joshi wrote:

> Thanks for the explanation.
>
> But in many cases where it is 0, we know we are not dealing with species
> anymore, right?

Branches with 0 indicate that in none of the tree sampled by the MCMC
process, the taxa below that branch were found to be just one species

> at many higher level nodes it shows support '0'. or
> branches the once in red should only be considered and the in blue are not
> part of the species delimitation process?

No I think you shouldn't worry about the color coding, the important
thing are really the posterior probabilities

alexis

>> ptp-species-delimi...@googlegroups.com.

>> For more options, visit https://groups.google.com/d/optout.
>>
>
>
>

--

[Victoria Johnson]

Hi I'm having a similar problem analysing my results. Just to clarify the 0 at higher nodes - you say that means no branches below that are considered 1 species? which would make sense as they are higher levels of classification right? so one should really just look at the lower nodes for individual species? and the posteriors values show support for these nodes containing one species. 

Okay I think I explained that to myself. Please correct me if I'm misunderstanding.

Thank you

[Alexandros Stamatakis]

Dear Victoria,

> Hi I'm having a similar problem analysing my results. Just to clarify
> the 0 at higher nodes - you say that means no branches below that are
> considered 1 species?

This means that out of the delimitations that were sampled via MCMC,
there was not one that identified all species below the node labelled by
0 as one species ...

> which would make sense as they are higher levels
> of classification right?

Yes, I would agree.

> so one should really just look at the lower
> nodes for individual species? and the posteriors values show support for
> these nodes containing one species.

Exactly, they indicate how frequently the sequences below a node were
delimited as one single species.

> Okay I think I explained that to myself. Please correct me if I'm
> misunderstanding.

All correct :-)

Alexis

>
> Thank you
>
> On Monday, May 5, 2014 at 11:29:37 AM UTC+2, Jahnavi Joshi wrote:
>
> I have been using a server based bPTP.
>
> Have question regarding the interpretation of the output
>
> Is there any cut-off which we can use to identify the species?
>
> If there is support of 0.95 vs 0.49? how does one interpret 0.49?
> does it mean it has multiple species? or not resolved? can we treat
> support values as posterior probability?
>
> what does zero mean at any node?
>
> Thanks in advance
>
> Jahnavi
>

> --
> You received this message because you are subscribed to the Google

> Groups "PTP species delimitation" group.

> To unsubscribe from this group and stop receiving emails from it, send
> an email to ptp-species-delimi...@googlegroups.com
> <mailto:ptp-species-delimi...@googlegroups.com>.