Merging two bacteria de novo assemblies into one

[gmarco]

Hello,

I'm trying to merge a bacterial genome assembly done by myself with Velvet with another I got from partner.

Both assemblies data come from Illumina MiSeq platform (Paired End).

I've a very good assembly, like 60~ contigs (89% genome) at this moment. I would like to use the other assembly (260~ contigs) to try filling the gaps between my contigs if it's possible so I can achieve an even better assembly.

I've tried CISA but I don't really get any better results. I don't know if there's a better method or software for merging both assemblies.

The only idea I've left is to ask my partner for rawdata files, perform myself different assemblies with his rawdata using different software like Velvet and Spades. Then try to merge them with CISA and my original assembly.

Any suggestions are welcome.

[cliffbeall]

Maybe Mauve Contig Mover, though I couldn't get it to work with large numbers of contigs- my numbers were larger than yours.

[gmarco]

Loading fasta... never ends.

[luc]

This assembly merger looks very promising:
GAM-NGS

https://github.com/vice87/gam-ngs

Please let me know if you can get it too work.

[mberacochea]

Hi,

I'm struggling with a similar issue. I have a draft genome (sequenced with PGM) and I want to get the most of the data. Merging different assemblies (different software) is useful?

Thanks

[Brian Bushnell]

You can merge different assemblies with Dedupe, part of the BBMap package. This will merge them and eliminate duplicate contigs, within some number of edits or percent identity that you can specify.

If you want to actually combine overlapping contigs into larger contigs, you can try a tools like Minimus or an OLC assembler. But bear in mind this can create misassemblies, which is not the case for Dedupe.

[mberacochea]

Thanks, I will try them.

It could be interesting to use a reference genome to merge assemblies, are there programs to do such thing?.