De novo assembly of PacBio with short Illumina data

[boetsie]

Hi all,

I'm a bit stuck on assemblying a bacterial genome with PacBio and Illumina data.

We have around 15x PacBio CLR data (average of 9kb) and 0.1x CCS data. In addition, we have some older Illumina paired-end 50bp data (100x coverage) from the GAIIx.

My problems so far;
- The coverage of the CCS data is too small to use for error correction with Celera Assembler.
- The length of the Illumina reads is too small for error correction, as a size of 100bp is suggested.
- Assemblers (like MIRA and Celera Assembler) require error corrected PacBio data, but i can't correct my data as explained above.

Does anyone have any suggestions on how to assemble these datasets? Are there tools that do not require error corrected PacBio data, or are there tools that can error correct the PacBio data with the short Illumina dataset?

Kind regards,
Boetsie

[jbingham]

Here are 3 approaches you can try. All require installing SMRT Analysis.

First, with PacBio's Allora assembler, you can likely get a reasonable assembly without using the Illumina. Just feed in the PacBio long reads and CCS and see what comes out. The coverage is a bit low, but you'll make some progress at least.

Or you can go the route of assembling the Illumina data separately, and see if you can get any larger contigs out of it. If yes, use RS_AHA_Scaffolding to join contigs with the PacBio long reads.

Another option, even though 50bp Illumina data probably isn't long enough to reliably map for an assembly: use the RS_Allora_Assembly_EC protocol in PacBio's software, with the Illumina for EC.