why I can't get edge -m changed

[huangl07]

dear,

I have already used the orientagraph software for my data

but one of my data is ambiguous,

my command is "orientagraph -i pop.tmix.gz -m 5 -o pop.5.5 -k 1000 -bootstrap -global"

but the resule llik file is like this:

"Starting ln(likelihood) with 0 migration events: 71.0333 Exiting ln(likelihood) with 1 migration events: 81.6075"

I wander know why?

As I understood ,the -m parameter is equal to the migration events?

[ekmolloy]

You are correct: the -m flag sets the number of migration edges to add. However, a migration edge is only added if this improves the likelihood of the model. My guess is that you have too few populations to add five migration edges. If you send me the file, I can double check what is going on. Also, it looks like you haven't included the -mlno flag, which is necessary for doing maximum likelihood network orientation (MLNO). If you don't want to use MLNO, you can use the original TreeMix method. Thanks, Erin

[huangl07]

pop.tmix.gz

you can try the file

[huangl07]

not only the -m 5, -m2 is also no change

[ekmolloy]

Your file pop.tmix.gz has allele frequencies for 4 populations. It seems reasonable to me that you would be able to explain the data by adding a single admixture edge (and thus no more admixture edges are added). I recommend reading Mark Lipson's 2020 review (download here: https://onlinelibrary.wiley.com/doi/10.1111/1755-0998.13230), especially checking out Figure 3. Overall, if your model has 4 populations and 5 admixture edges, you probably do not want to use methods based on f-statistics.

[huangl07]

thank you，I will read it