Exome kits designed for long read technology?

[ymc]

Does anyone know if there is an exome capture kit that captures exons, intervening introns and promoter of all genes or under development? If we can then use PacBio to sequence the fragments, we can find trinculeotide repeats, phasing, etc in one go. I think the coverage will probably be around 500Mb for human. Does this kind of kit makes sense with the Sequel machine?

[Bukowski]

http://investor.pacificbiosciences.com/releasedetail.cfm?releaseid=908644

People have been doing this for a while...

Genome scale, probably not - maybe on the Sequel but not on the RSII but for small panels - why not?

[ymc]

Thanks for your reply. Your example is 6k fragment. Anyone did 40k fragment?

It would be nice to have a genome wide product or at least a clincial 3000+ gene version. It will be an exome on steroid!

[Bukowski]

Why would you want to do that?

1) A 40kb fragment would have low read quality as it would be a single pass read

2) 6kb fragments overlap quite nicely

Resolving large repeats is still not a solved issue even on the PacBio. And the error profile is still reduced by depth/high pass reads of insert.

[ymc]

I think 20k library is common for bacteria assembly?

https://github.com/PacificBiosciences/DevNet/wiki/E.-coli-Bacterial-Assembly

If you imply 6k with multiple passes, do you mean CCS reads?

[Bukowski]

I think 20k library is common for bacteria assembly?

https://github.com/PacificBiosciences/DevNet/wiki/E.-coli-Bacterial-Assembly

If you imply 6k with multiple passes, do you mean CCS reads?

PacBio don't use the term CCS anymore and the generation of CCS reads is deprecated in SMRTpipe - so no, I mean Reads of Insert.

If you read the '20kb' library preparation protocol, pay specific attention to the Blue Pippin size selection step. You don't generate a library of 20kb fragments with it

[Magdoll]

ymc, do you mean like capturing specific gene regions using using NimbleGen SeqCap EZ?

http://sequencing.roche.com/products/nimblegen-seqcap-target-enrichment/seqcap-ez-system/seqcap-ez-designs.html

We at PacBio have used the oncology panel on both gDNA and cDNA. I don't think we have a poster for that yet (or I can't find it off the top of my head right now). But if this is what you are thinking, it does work.

[ymc]

ymc, do you mean like capturing specific gene regions using using NimbleGen SeqCap EZ?

http://sequencing.roche.com/products/nimblegen-seqcap-target-enrichment/seqcap-ez-system/seqcap-ez-designs.html

We at PacBio have used the oncology panel on both gDNA and cDNA. I don't think we have a poster for that yet (or I can't find it off the top of my head right now). But if this is what you are thinking, it does work.

Thanks for your input. Of course, custom capture can do what I want. But they are costly and design it for thousands of genes can be PITA. I would like a TruSight One panel or whole exome designed for long reads.

http://www.illumina.com/products/trusight-one-sequencing-panel.html

[Magdoll]

Hi ymc,

After a brief conversation with my colleagues it appears that we haven't used the TruSight panel on PacBio. But we also don't see why theoretically (with some minor tweaks) it would not work on PacBio platform.

The NimbleGen panel was originally designed for short reads. We simply used it on PacBio for gDNA capture (fragment to longer pieces) and cDNA capture (using full-length cDNA).

I would imagine PacBio's 6kb capture protocol would be able to work on this.

Best,
--Liz

[ymc]

Hi ymc,

After a brief conversation with my colleagues it appears that we haven't used the TruSight panel on PacBio. But we also don't see why theoretically (with some minor tweaks) it would not work on PacBio platform.

The NimbleGen panel was originally designed for short reads. We simply used it on PacBio for gDNA capture (fragment to longer pieces) and cDNA capture (using full-length cDNA).

I would imagine PacBio's 6kb capture protocol would be able to work on this.

Best,
--Liz

Of course TruSight One can work to a certain extent on PacBio. What I want is to have a TruSight One that captures exon, intron and promoter such that I can find non-coding variants and have phasing across the whole gene. I know custom capture can do this but it will be ideal if there is a well designed panel that can save people hassle and cost.

[ymc]

Dear all, I am thinking about making a custom panel of 20+ genes and then use PacBio to sequence them. My panel will try to cover each gene from promoter to 3'UTR, so the longer the read the better. Which custome panel kit is the best for this? What are the pros and cons of capture-based method and PCR method?

I found this Nimblegen capture kit poster saying it can generate 6kb fragments that has a few fragments at 17kb

http://www.pacb.com/wp-content/uploads/Poster_TargetedSMRTSequencingPhasing_RocheNimbleGenSeqCapEZ.pdf

Is there a custom kit that can do 20kb fragment?

[ymc]

If I use 6kb capture kit, would it waste the potential of the P6C4 chemistry, ie same number of reads but halved the throughput?

[nucacidhunter]

Dear all, I am thinking about making a custom panel of 20+ genes and then use PacBio to sequence them. My panel will try to cover each gene from promoter to 3'UTR, so the longer the read the better. Which custome panel kit is the best for this? What are the pros and cons of capture-based method and PCR method?

Long range PCR would be the easiest and cheapest option for resequencing 20 genes unless following applies:
1- The region is divergent among samples preventing unique primer design
2- You only have partial sequence and want to capture flanking 3' and 5' region
3- No information is available on paralogs or targeting a gene family or genes sharing a domain

Capture would be more expensive because:
1- library prep cost
2- Custom panel design expenses
3- Sequence capture is a long process
4- Only a subset of captured fragments would be on target wasting sequencing

[ymc]

Long range PCR would be the easiest and cheapest option for resequencing 20 genes unless following applies:
1- The region is divergent among samples preventing unique primer design
2- You only have partial sequence and want to capture flanking 3' and 5' region
3- No information is available on paralogs or targeting a gene family or genes sharing a domain

Capture would be more expensive because:
1- library prep cost
2- Custom panel design expenses
3- Sequence capture is a long process
4- Only a subset of captured fragments would be on target wasting sequencing

I am going to do human samples, so I suppose I pass the three deal breakers you mentioned.

If I use long range PCR, does that mean I can take advantage P6C4 chemistry's avg 14k reads and some occasional 40k reads?

[ymc]

I find this long range PCR Qiagen product that says it can amplify up to 40kb

https://www.qiagen.com/hk/shop/pcr/end-point-pcr-enzymes-and-kits/regular-pcr/qiagen-longrange-pcr-kit

Does that mean if I want to amplify genes longer than 40kb, then I need to use multiple kits for one gene?