Problems with mapping files

[Manuel Lafond]

Hi,

I need to know which nodes of the locus trees are duplication/speciation.

For that, I need to look at the .mapsl file.

I activated the option of outputting the ID of each node, 

but in the newick strings, the trees never have an ID for the root (whether it be species/locus/gene tree, see attached files). 

So although the .mapsl tells me which nodes are Dup, I can't tell for the root.  

Is that a bug, or a misinterpretation of my part?

I also have another problem with the mapping - I don't trust the locus-species mapping.

For instance, the 01.mapsl maps node 22 of the locus tree to the node 5 of the species tree.

Node 22 is an internal node, whereas node 5 is a leaf - I can't make sense of this.  

Again, is that a bug?

Thank you

[Diego M.]

Dear Manuel,

Your questions do not directly relate to bugs, but rather to format specifications that may not be clear enough, let me explain them:

a) The root of a locus tree simulated by SimPhy will be a speciation by definition since it is the origin of the simulation. This would not be necessarily true if SimPhy pruned them, but it is not the case.

b) You can trust the mapping, but you have to be careful with the notation. Quoted labels (e.g. '5') represent tip names and do not indicate their node number in post-order as internal-node ids do. In your example, the node 22 of the locus tree is mapping to the node 5 of the species tree, which is not the same as the tip '5'. In fact, you can check in the trees that it corresponds to the locus-tree clade "8_1,6_1,7_1", which maps to the species tree "8,6,7" and corresponds to a speciation in the new locus 1. 

However, checking your questions I did find a bug. The internal-node ids in the Newick species tree were being printed in pre-order, while the mappings, locus and gene trees were in post-order. I have provisionally fixed it in a new github branch (postorderSptree) which I will merge to the master as soon as I double-check it.

I hope this helps,

            Diego M.