gl2fasta methods and constant sites

[Alvaro Cortes]

Hi everyone, 

I'm using the gl2fasta function to perform a phylogenetic analysis in IQTREE. When I select methods one and two (for ML phylogenetic analysis), the output file contains a large number of constant sites (over 260,000). Can someone please explain where these sites come from and if they impact downstream analyses using ML phylogenetic methods?

I know IQTREE can handle a proportion of constant/invariant sites, but this number seems high, and I'm worried it will impact the analysis and branch lengths. 

This issue doesn't occur when I select methods 3 and 4, but I'm running an ML analysis, and 3 and 4 are specific to Maximum parsimony. 

I appreciate any help you can provide. 

Cheers, 

-Alvaro

[Jose Luis Mijangos]

Hi Alvaro,

Please have a look at a previous discussion about this topic using the link below:

https://groups.google.com/g/dartr/c/2Ol1Giv3IIo/m/bVSIXWXMAQAJ 

Cheers,

Luis