Re: [R-sig-phylo] simulate Ne changes on a given phylogeny

[Jacob Berv]

On a related note- does anyone know of a function or package that can parse and 
store all of the metadata associated with nodes and branches from a BEAST 
output file? read.beast from phyloch will parse all of the data associated with 
internal nodes, but not with terminal branches (for example, substitution rate 
information). I emailed the developer of that tool and he indicated this was by 
design (though he may be able to offer a solution).

Jake

> On Oct 18, 2015, at 12:04 PM, Jacob Berv  
> wrote:
> 
> Yes, this has come up in my reading…but there do seem to be situations where 
> ’nearly neutral’ substitutions can be negatively or positively associated 
> with population size:
> 
> From Lanfear 2014 Population size and the rate of evolution:
> 
> "Putting aside variation in the mutation rate, we largely expect the total 
> rate of evolution to be negatively correlated with Ne if slightly deleterious 
> mutations dominate evolution, and to be positively correlated with Ne if 
> advantageous mutations dominate evolution.”
> "On the other hand, any process that leads to an association between Ne and 
> mutation rates will cause a similar association between Ne and neutral and 
> effectively neutral substitution rates. These processes could include effects 
> such as the evolution of mutation rates, and the co-variation of Ne with 
> life-history traits such as generation time”
> 
> I think I need to read more about how to simulate these different kinds of 
> demographic scenarios.
> 
> I suppose the question is, what is the more likely null hypothesis? That 
> mutation rates can change extremely rapidly? Or that demographic fluctuations 
> (pop size, generation time) can induce changes in the detectable substitution 
> rate among lineages. 
> 
> Jake
> 
> 
> 
>> On Oct 17, 2015, at 11:21 PM, Liam J. Revell  wrote:
>> 
>> Hi Jacob.
>> 
>> Can I add the somewhat boring & probably obvious comment that under the 
>> neutral theory of molecular evolution the substitution rate is independent 
>> of the effective population size?
>> 
>> All the best, Liam
>> 
>> Liam J. Revell, Associate Professor of Biology
>> University of Massachusetts Boston
>> web: http://faculty.umb.edu/liam.revell/
>> email: liam.rev...@umb.edu
>> blog: http://blog.phytools.org
>> 
>> On 10/17/2015 11:01 PM, Jacob Berv wrote:
>>> Hmmm that seems somewhat indirect but might work… I’ll look into that.
>>> 
>>> To give you more information - I’m actually trying to come up with a way to 
>>> test the idea that substitution rate shifts detected with a relaxed 
>>> molecular clock (BEAST) may be driven by changes in effective population 
>>> size. Simulating data for particular scenarios, and then running that 
>>> simulated data through BEAST could be a useful way to test some explicit 
>>> hypotheses I’m interested in. But I have to simulate the data first.
>>> 
>>> Jake
>>> 
>>> 
 On Oct 17, 2015, at 10:40 PM, Brian O'Meara  wrote:
 
 Dick Hudson's ms software can simulate gene trees along a species tree or 
 network with migration, changing population size, etc. The package 
 phyclust can call ms. You could then just simulate nucleotides on these 
 gene trees.
 
 Best,
 Brian
 
 ___
 Brian O'Meara
 Associate Professor
 Dept. of Ecology & Evolutionary Biology
 U. of Tennessee, Knoxville
 http://www.brianomeara.info 
 
 Postdoc collaborators wanted: http://nimbios.org/postdocs/ 
 
 Calendar: http://www.brianomeara.info/calendars/omeara 
 
 On Sat, Oct 17, 2015 at 10:12 PM, Jacob Berv 
 > 
 wrote:
 Dear R-sig-phylo,
 
 I have a somewhat general simulation question and I was hoping someone on 
 here might have some insight.
 
 I’m trying to figure out if it’s possible to simulate nucleotide sequence 
 data (an arbitrary number of neutral loci under a multi species coalescent 
 model), on an ultrametric input topology (where tips represent species), 
 with user defined changes in effective population size at the start and 
 end of a particular internal branch. In my searching I’ve come across some 
 software by Deren Eaton (https://github.com/dereneaton/simLoci 
  
 >) that looks like it might do what 
 I want - but I’m not sure. It looks like I can specify migration events 
 between taxa, but perhaps not population size changes on internal 
 branches. There are many other applications for simulating sequence data 
 but I am not familiar with any of them. Any thoughts would 

[R-sig-phylo] Q regarding the analysis of ordinal meristic traits such as lamellae number

[Mozes Blom]

Dear R-sig-phylo,

I am fairly novice to the field of phylogenetic comparative methods and I was 
hoping to gain some advice regarding the study of ordinal (pseudo-continuous) 
morphological data.

I am currently studying morphological convergence associated with habitat 
transitions in Australian lizards. To some extent my data is very similar to 
Mahler et al. (2010, 2013), with several measurements of continuous traits such 
as SVL, limb length, head length etc. However in addition to metric data, I 
also have estimates of feet morphology such as the number of toe lamellae.

Now my question is whether it is statistically appropriate to analyse meristic 
data, such as lamellae number, in exactly the same way as the metric data? In 
particular in reference to for example running a Phylo PCA with size corrected 
residuals, phylo-anova (to test for correlation between habitat and trait) and 
running SURFACE to test for convergence?

Part of the reason why I would like your advice, is because the variation in 
counts is much less than for example in Mahler's study; in my case the 
differences in lamellae number ranges from 12 to 17, across 26 species. Ingram 
(2014) mentioned: "We note that one trait (gill raker number) is meristic 
rather than metric but that it is variable enough that species means were 
effectively continuous and approximately normally distributed...". I would like 
to ask if someone would have a suggestion how to assess whether the meristic 
traits can be considered as 'variable enough', to treat it as 
pseudo-continuous? Is this for example commonly done by traditional tests for 
normality?  And if it's not variable enough, could you recommend an alternative 
approach?

Many thanks, your help is much appreciated!

Best,

Moos



[[alternative HTML version deleted]]

___
R-sig-phylo mailing list - R-sig-phylo@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-phylo
Searchable archive at http://www.mail-archive.com/r-sig-phylo@r-project.org/