This is a test version of Biostars. For the public version, visit https://www.biostars.org.
How to Handle Incorrect Clade Grouping of a Reference Genome in a Phylogenetic Tree?

Hi everyone,

I encountered an issue while building a phylogenetic tree and would appreciate some guidance or discussion.

I used a reference genome to annotate my target genome, but during the phylogenetic analysis, I found that the reference genome was incorrectly grouped into the same clade as my target genome. Biologically, this reference genome should be much more distant from my species.

Here are my questions:

Has anyone experienced a similar issue? If I want to build a robust phylogenetic tree, should I remove this reference genome and reconstruct the tree? Alternatively, should I keep the reference genome in the tree and describe this anomaly in my discussion section? Any insights or advice on how to handle this situation would be greatly appreciated!

Thank you!

tree clade incorrect phylogenetic reference genome
• 1,594 views • link
written by oasiswho • oasiswho

1 answer

Has anyone experienced a similar issue?

I am sure many people have gotten the trees they didn't expect.

Assuming you know how to do a tree reconstruction, the obtained tree doesn't lie. Either the two species are more related than you think, or something went wrong upstream of tree reconstruction.

You didn't give us enough information to offer more than this general conclusion. We don't know: 1) is this is a single-gene or concatenated tree? 2) DNA or protein? 3) how did you align the sequences? 4 were the alignments trimmed? 5) what were you expecting in terms of tree distances and what did you get? 6) what are other tree members and how are their distances relative to your species of interest?

• 0 views • link
written by Mensur Dlakic Mensur Dlakic

Dear Mensur Dlakic Thank you very much for your response. I am currently working on constructing a phylogenetic tree based on mitochondrial genomes. I included five existing mitochondrial genomes along with two outgroup species to build the tree.

(1)I used two methods: one is a concatenated multi-gene tree constructed with IQ-TREE2, as Figure 1 shown (a), and the other is a coalescent-based species tree constructed using ASTRAL(Figure 1 (B)). (2)The analysis was performed using DNA sequences. (3)I aligned the sequences using MAFFT. (4)The alignments were not trimmed and were directly used for tree construction after alignment. (5)My primary goal is to compare the placement of species in the mitochondrial tree with that in the chloroplast tree. However, I found discrepancies between the placement of species in the mitochondrial tree and the chloroplast tree. (6)The tree members and the two reference papers for comparison are provided in the attached image. The

• 0 views • link
written by oasiswho • oasiswho

Trees don't lie, so I suspect it is something in the upstream methodology.

I am not sure we are comparing apples to apples here. If yours are concatenated trees, is the number of genes the same as in other trees? Have you looked at your alignments? I suspect something is off in them, and doing a visual inspection may give you an answer. Trimming should definitely be done unless the alignment is continuous and without any suspicious gaps.

I can keep asking questions and you will probably keep providing answers, but I don't have the capacity to turn this into a long back-and-forth. I suggest you backtrack all the way and make sure all the steps are correct. Maybe do this first on a smaller scale, say 3-5 concatenated genes, and see what you get from that. If it looks OK, try to replicate it on a larger number of genes. Try also to replicate the exact methodology of other trees.

Once again: trees reflect the alignments. Assuming you have done the tree reconstruction correctly, your problem is most likely somewhere in the alignment creation.

• 0 views • link
written by Mensur Dlakic Mensur Dlakic

Dear Mensur Dlakic

Thank you for your reply very very much, I apologize for replying late.

Just as you said, trees don't lie. I have decided to accept my results and find reasonable explanations for them. Thank you again!

Best Regards!

• 0 views • link
written by oasiswho • oasiswho

Log in to answer this question.

More posts like this