Create the best prokaryotic genome assembly: Combining different methods

[JurgenP]

I use 454 titanium reads for my prokaryotic (bacterial) whole genome sequencing. The bacteria I am working on is highly diverse from strain to strain, with the gene content differing up to 15%. Therefore there are arguments to think of my project as a de novo sequence approach or a re-sequencing approach.

Doing a de novo assembly with the Newbler assembler, contig borders are in other parts of the genome then contig borders from the Refmapping software from Roche. So I would like to use the results from one assembly method in region A of my genome, and the results from the other assembly in region B of my genome. I would like to develop a strategy where I can combine these assembly methods.

Does anyone have an idea or an opinion on my approach?

Thanx!

[flxlex]

I always feel refmapping is a bad idea, as you might get artifacts, and you should use the unmapped reads for de novo discovery of regions missing in the reference. So, I would assemble de novo and then use comparative tools to order your contigs, such as Mauve Contig Mover (http://bioinformatics.oxfordjournals...ull/25/16/2071) or Projector2 (not a big fan, but here is the site: http://bioinformatics.biol.rug.nl/we...tor2_start.php).

flxlex

[JurgenP]

Thanx for your reply!

What kind of artefacts would you expect to be produced? As far as I know how the Roche Reference Mapper works, I assume it uses a reference sequence/genome just a "guide" by witch a read is mapped to. The reference will not be involved in any base calls, nor will it be responsible for the erroneously joining of 2 genomic regions that should not be joined. I would expect that if one (set of) read maps to 2 distinct genomic regions in the reference genome, the contig will just stop at that part and this region will be a contig boundary. It is exactly this region that will be correctly assembled in a de novo approach. Hence my suggestion of combining both assembly methods in the assembly of highly divergent genomes.

Mauve is a great tool indeed (use it all the time)... Very candy-like pictures! I will look into the other one you suggested!

Jurgen

[Jinyu Wu]

Hello,

PGA4genomics (http://centre.bioinformatics.zj.cn:8080/pga or http://59.79.168.90:8080/pga) to help users automate gap closing based
on comparative genomic syntenies. Extensive evaluations showed that it significantly outperforms previous methods and can produce highly accurate layout result, especially when assembling genomes that are only moderately related.

[amys]

nothing

Create the best prokaryotic genome assembly: Combining different methods