Contig Reordering With Multiple Reference Genomes

Question

[Neo]

Hi all, Im not sure if my previous post on the same topic was posted right. so here I go again.

I have tried re-ordering 454 derived contigs from the bacterial genome Im working with, using mauve (with one reference genome). however, I would now like to use multiple reference genomes to reduce gaps and ultimately the number of PCRs that may need to be done to stitch things together.

is anyone aware of a program that would let us do this? I have tried PGA4genomics web server with little success.

any input greatly appreciated. thanks guys.

Answer 1

[Michael]

Possibly the BACCardI tool (homepage) could assist. I don't know if it really does exactly what you want, and if it's still up to date, but at least I have seen it working. From the web-site:

This tool has been applied to numerous genome projects to solve various problems including (a) validation of whole genome shotgun assemblies, (b) support for contig ordering in the finishing phase of a genome project, and (c) intergenome comparison between related strains when only one of the strains has been sequenced and a large insert library is available for the other.

[Lhl]

Hi Michael, do you think it can also be applied to organelle genomes?

[Michael]

If it works for bacterial genomes, then why not for organelle. They are not much different in size. The problem is rather to get the software up and running.

Answer 2

[Val]

We are currently evaluating two tools for quite the same question (Reordering solexa assembly based on reference genome) :

• Abacas, based on Mummer
• Mauve Contig Mover

[Val]

For simply reordering, Mauve Contig Mover was the solution we choose.

Answer 3

[Ketil]

I'm going to question the usefulness of this. Isn't reordering (shuffling) of genomes a typical event in the evolution of genomes - especially bacterial ones? It seems you want to disregard evidence that is based on your reads, and instead use evidence based on homology, which I think is likely to be less reliable.

You don't say what kind of data you had, but I would first have tried to get some paired-end or mate-pair library to see if that helped. But as this is an old thread, is your work published or the result of the reordering otherwise made available?