SNP vs variant+invariant sites ?

[Adrien Rieux]

Hi there,

Fairly simple question I think but I couldn't dinf the answer in the manual. I wonder wether I'm allowed to run PartitionFinder2 on a alignment containing variant sites (SNP) only or if i have to use an alignment where both variant and invariant sites are present ?
I know where every SNP falls into a reference genome and this is how I can annotate each SNP witch a particular scheme (e.g coding 1,2,3 or non coding).

Thanks,

Adrien

[Rob Lanfear]

Hi Adrien,

You're certainly 'allowed' to do this. How you treat the resulting information is another question I think. PF2 doesn't implement the models you need here, which are those with an ascertainment bias correction. Because of this, I think a pragmatic route might be to first run PartitionFinder on your data, and to then use the partitioning scheme it suggest, but ignore the models. For the models it would be more appropriate to use those with an ascertainment bias correction.

An alternative is to put the invariant sites back into your analysis, as you mention. That would be my preference, because the ascertainment bias corrections are rough guesses, whereas the invariant sites contain precisely the information that the standard models need.

Rob

--
You received this message because you are subscribed to the Google Groups "PartitionFinder" group.
To unsubscribe from this group and stop receiving emails from it, send an email to partitionfinder+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

Rob Lanfear
Division of Ecology and Evolution,

Research School of Biology,

The Australian National University,

Canberra

www.robertlanfear.com