WGS in one region, Amplicons in another

[nickloman]

Hi there

Does anyone have experience running mixed applications on a single PicoTiterPlate? We wish to run WGS libraries on region 1 and an amplicon library on region 2 (both Titanium). Is this possible? Are any particular settings required on the software when doing image processing?

Many thanks,

Nick.

[kmcarr]

Yes, you can do this so long as they are both Titanium. You will need to run separate signal processing on the two regions; runAnalysisPipe on region 1 and runAnalysisPipeAmplicons on region 2.

[nickloman]

Fantastic, thanks very much!

[flxlex]

How about mixing amplicons and WGS in the same lane? Using the same type of tags, of course (e.g. 454/Roche MIDs on all). I guess you would run both runAnalysisPipe and runAnalysisPipeAmplicons on the run, and use the WGA reads from the first, and amplicon reads from the second?

[nickloman]

I've no idea if that would work personally. Give it a try and report back

[454Sequencing]

Hi flxlex,

One certainly "could" do that - the world won't end, assuredly! However, it's not a recommended/supported strategy. Amplicons and shotgun reads have different analysis pipelines as they require different signal processing as well as quality filtering, given the goals of each type of sequencing. Depending upon the relative mix of the two and the characteristics of each population things might turn out fine, but it's tricky to predict and therefore physical separation via gasket or run is certainly the safest course.

Best regards,
Jason

[greigite]

Hi flxlex,

One certainly "could" do that - the world won't end, assuredly! However, it's not a recommended/supported strategy. Amplicons and shotgun reads have different analysis pipelines as they require different signal processing as well as quality filtering, given the goals of each type of sequencing. Depending upon the relative mix of the two and the characteristics of each population things might turn out fine, but it's tricky to predict and therefore physical separation via gasket or run is certainly the safest course.

Best regards,
Jason

We just ran a plate with this "not-recommended" strategy for logistical reasons. It was a mix of shotgun gDNA libraries and one amplicon library separated with MIDs. The amplicon library showed an abundance of short reads (see attached image), while the whole genome libraries had a more normal distribution. Bioanalyzer traces for all libraries prior to sequencing were quite similar. Could the signal processing be responsible? If so, is there anything I could do to reprocess the raw data to improve the read lengths for the amplicons?

[flxlex]

greigite,

Interesting! Did you include Amplicon Control beads on the plate, so that you can run the amplicon specific image analysis pipeline?

Would you mind sharing the size distribution of the genomic samples? We often see such short reads also for genomic libraries...

[greigite]

greigite,

Interesting! Did you include Amplicon Control beads on the plate, so that you can run the amplicon specific image analysis pipeline?

Would you mind sharing the size distribution of the genomic samples? We often see such short reads also for genomic libraries...

I don't know if the amplicon control beads were included- this was a collaborative project where another lab did the actual sequencing and we are doing data analysis. I can ask. I don't have the read distribution images for the other samples but I know average length for one of the gDNA samples was 289. This sample had a shorter fragment size pre sequencing as well though.

[RCJK]

I was just wondering if anyone else has tried this approach of mixing Lib-A and Lib-L beads within the same region, and if so, how'd it go?

[Mosiep]

Is it possible to combine/mix samples together for a single run in 454 sequencing? is it practical ?if so what are the benefits? I am new to 454 sequencing

[Zaag]

Yes it is; depends on your project; you save some money;

[malachi74]

Bad Idea

Mixing Lib-L and Lib- A is bad idea, the software will only recognize one of the keys, and the reads with the other reads will not keypass.

[Zaag]

And then you force the software to imageProcess for the other key and you have your data.