idanefroni / Conservatory

Identification of conserved non-coding sequences in plants
GNU General Public License v3.0
12 stars 3 forks source link

Phylogenetic tree construction #5

Closed Averstic closed 1 year ago

Averstic commented 1 year ago

Hi,

I was wondering if you have any recommendations or guidelines on how to exactly generate a phylogenetic tree for use in conservatory analysis?

idanefroni commented 1 year ago

Sure, but can you be more specific?

In general, you can use ortholog protein alignment and RAxML (or iq-tree2) for the structure and branch lengths of the tree.

The model is then determined by phyloFit using an alignment of the fourfold degenerate 3rd codon - this is performed using the buildModel script which identifies orthologs and extracts the variant sites that are informative (found in the scripts directory). We also tried building the tree using the fourfold degenerate site alignments (produced by buildModel) and it seems to be better than protein tress in term of branch lengths assignments.

Hope that helps