I'm very new to phylogenetics and want to build a cladogram from mammal species using pseudogene fragments. The procedure is as follows:

  • Find several query sequences from Gallus gallus of the genes VIT1/2/3
  • BLAST against 20 mammal species (ncbi reference genomes) from the major Eutherian superorders (Euarchontoglires, Laurasiatheria, Xenarthra, Afrotheria) and several Marsupial and Monotreme mammals.
  • Multiple sequence alignment with clustalw of a ~500bp region of exon 3 and intron 3
  • Run maximum likelihood using the software MEGA under the Tamura Nei model and complete deletion of gaps.

However I also found a ~150bp sequence in the intronic region that repeatedly came up in my BLAST queries. When I searched this sequence against a datatabase of transposable elements, it looks like a potential type 2 LINE element (or at least part of one).

Would it be reasonable to append this sequence to my ~500bp. Especially since the element appears at multiple hits across the genome so I can't be certain it originated from the egg yolk pseudogene. Like I say I'm very new to this so have no reference point, would appreciate any help. Thanks in advance.

  • 1
    $\begingroup$ Thanks for the edit, that helps! But what is the question you are asking? Why do you want to build these trees, what do you hope they will tell you? If you are interested in the evolution of the gene itself, and given that you are comparing with quite distant species, I would ignore the LINE (and most probably the intron as well). If you are trying to see if you can pinpoint when this LINE was introduced, that's different. Alternatively, it might even make more sense to work at the protein level. Basically, the answer depends on the exact question you are trying to answer using your data. $\endgroup$
    – terdon
    Commented Jul 7 at 17:23

1 Answer 1


I will answer this very quickly. LINE are transposons and the proximity to the locus of interest would be unstable across the evolutionary depth you are considering, i.e. because they "jump around the genome", i.e. transpose. This means multiple copies and a process called "concerted evolution", and is a disaster for phylogenetic trees, for complex reasons. What you want is to track the evolution of vertebrates, not the population genetics of LINE.

There are much deeper issues with your data and approach, but this answers your specific question.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.