Comparing a draft genome with its neighbour species genomes

Question

[pbigbig]

Hi,

I just need some suggestions regarding a current project.

I have de novo assembled a draft genome (~1Gbp in size) of a species (non reference), to check the quality of this assembly, should I compare it to a published genome of a neighbour species?

Additionally, which tools are currently best suited for that purpose? (I looked for MUMmer, Mauve, but have no prior experience in using them)

Thank you very much in advance!

 

Answer 1

[Philipp Bayer]

It's a bit hard to check the quality of your assembly by comparing it to a neighboring genome for two reasons:

• you usually don't know how well-assembled the other genome is
• you usually don't know how different your genome is (are non-overlapping regions truly different, or were they just not assembled?)

You can still make a whole genome comparison, mummer is a good idea, I also like to use symap (you can zoom into regions there, it's in itself a wrapper around mummer).

For other ways to compare genome assemblies, have a look at the Assemblathon 1 and 2 papers:

• Assemblathon 1
• Assemblathon 2

[pbigbig]

Thanks a lot.

In fact, I have already checked this de novo assembly with assemblathon_stat.pl; BUSCO eukaryote gene set and REAPR for assembly errors. I will try your suggested program, I am still very fuzzy here but I think it would be nice if we can have some comparison of assemblies from similar species? (however, it still troubles me about which type of comparison and which kind of differences should be expected)

Best regards.

[Philipp Bayer]

I think it's a good figure in a paper, depending on what you're going for in the paper - if you focus on a comparison of both species, then I'd definitely put in a whole genome dotplot via mummer. If you focus on your species only then the whole genome alignment may not be that fitting.

[pbigbig]

Thank you very much for your suggestion!