How to extract ancestral sequence for Cow from 46 eutherian mammals EPO Ensembl?

Question

[Claire]

Hello everyone,

I am trying to figure out how to extract the ancestral sequence of the bovine species from an Ensembl multiple alignment file: 46 eutherian mammals EPO. In the .emf files, I can see that there is indeed an inferred ancestral sequence for each node of the phylogenetic tree (with a different name for each block if I understand), as well as in the corresponding .maf files. I tried to use the WGAbed pacakge (https://henryjuho.github.io/WGAbed/) to transform the .maf files into .bed files with the cow as the reference species but I only get one matching ancestral sequence for each position in the bed file, and no information about which node it matches. Does anyone know how to extract the ancestral sequences already inferred in these .emf files please?

Thanks in advance for your help. Claire

Answer 1

[minesrebollo]

Hi Claire, I had to do the same thing with rice's EPO. Rice was also the reference species for the alignment. By plotting some trees with icytree.org and activating internal node text to see the name of ancestors, I saw that there was a pattern and that rice's direct ancestor was always the ancestor directly after rice's sequence in the .maf so this is what I did:

1. Converted all .emf files to .maf with https://github.com/Ensembl/ensembl-compara/raw/master/scripts/dumps/emf2maf.pl
2. Saved all rice-chromosome related sequences into one .maf file with

   cat 8_rice.epo.[1-9]*.maf > AllChrsRice8EPO.maf

3. Filter rice and ancestor from the .maf file

   grep '^a' -A 2 -B 3 AllChrsRice8EPO.maf > RiceAncestral.maf

4. compressing the filtered .maf to the transform it to wga bed

I hope this answer doesn't come too late and that you find it useful, I just came across your question.