# How to make unrooted tree for Likelihood mapping result by using IQ2-tree?

I am a biologist, and I do not fully understand the tree topology of the experimental species. I used four-taxon set (4 sequences) to identify the Four-cluster Likelihood-Mapping by using

iqtree2 -s ***.fa -lmap 2000 -n 0


I have obtained the svg file - triangle (attached - 1), however there were no data mentioned about T1, T2 and T3 unrooted tree based on my input 4 sequences (Haur_2, Shad_6, Equa_oga2 and apa). I used single orthologous copy genes as the supermatrix for this data.

Therefore, I do not understand which topology of tree is reliable depends on the sequence data. I knew that my data set showed similar result topology from Iq2 tree output. There is no species sequence data available for the species that I am looking for a reliable topology of those species. If anyone knows how to understand the assigned sequence on the unrooted tree based on Four-cluster Likelihood-Mapping by using iqtree2? Could you kindly answer this question? I want to know the orange unrooted tree (image 2) for my analyzed data. Thank you very much.

output ***.iqtree Quartet support of regions a1, a2, a3 (mainly for clustered analysis):

     #quartets    a1 (% a1)        a2 (% a2)        a3 (% a3)    name
-----------------------------------------------------------------------------
1   2000     648     (32.40 ) 712     (35.60 ) 640     (32.00 )  Haur_2
2   2000     648     (32.40 ) 712     (35.60 ) 640     (32.00 )  Shad_6
3   2000     648     (32.40 ) 712     (35.60 ) 640     (32.00 )  Equa_oga2
4   2000     648     (32.40 ) 712     (35.60 ) 640     (32.00 )  apa
-----------------------------------------------------------------------------
2000     648     (32.40 ) 712     (35.60 ) 640     (32.00 )


You have provided two different trees. The second was the Arthropoda and labeled "image 2".

The first example the explanation is that it is an exact polytomy. I know this because the values are equally split between the 3 possible alternatives for a 4-taxon tree. Thus there is no bifurcated solution to the data, any bifurcating combination is equally valid therefore it is simply expressed as all 4-taxa sharing the same node (hypothetical common ancestor). In other words there are 3 equally valid solutions.

Thus, the tree is 4-taxa tree with 1-node. Thats the solution. A normal 4-taxon tree will have 2-nodes, this is a star-tree, i.e. one node, technically a polytomy.

The example below concerns Arthropoda and demonstrates the top tree ((Chelicerta, Myriapoda),(Hexapoda, misc.)) is the correct tree. That 90.4 value demonstrates that. What this says is that neither myriapods or Chelicerta share a direct common ancestor with hexapods, moreover "misc" (whatever that is) shares a direct ancestor with the hexapods. "Image 2" is a classical bifurcating tree and represents only one solution.

So to specifically answer the question. The tree for arthropoda is:

and the other tree is,

I should have removed the scale bar in this example, because I didn't have the data, but not when you have genetic quantification of the distances.

• Thank you very much for your answer. I have searched about polytomy. It is right that I used the data from a single ancestral lineage. But then, I do not understand the recent paper Nicolás Mongiardino Koch, et al. eLife (2022) showed a likelihood-mapping analysis in IQ-TREE to visualize the phylogenetic signal for alternative resolutions of the quartet involving these three lineages (Arbacioida, Salenioida, and Stomopneustoida + Camarodonta) and their sister clade (Irregularia). My data was also sister clade. Why my analyzed result won't be like Koch, et al. eLife (2022) Fig2. B?