YACs as a solution to repeat regions

[qnc]

What do people think about using a YAC for targeted enrichment of a linked region and then placing the YAC as a whole on a next gen sequencing platform against a reference sequence

Is this viable?

If so what thinks so I be careful to look out for?

Thanks if you can help

[pmiguel]

What do people think about using a YAC for targeted enrichment of a linked region and then placing the YAC as a whole on a next gen sequencing platform against a reference sequence

Is this viable?

If so what thinks so I be careful to look out for?

Thanks if you can help

I guess we would need to know what brought you to the point where a YAC seemed worth the effort. In principle, it should work fine. But in the hands of most people YACs are monstrously harder to work with than, for example, BACs.

I know that in a single quad of a SOLiD run, you could go 10-20x on at least 8 yeast lines. So even if you didn't want to run a CHEF gel to isolate the YAC, you could just do a genomic DNA prep and still get reasonable coverage of your YAC. So it is plausible, anyway.

--
Phillip

[qnc]

Thanks

I do understand that YACs are hard to handle and BACs are a better solution

What I am really after is an accurate sequence.

single molecule sequencing is what about 2+? years of.

So amplification of source material is necessary.

If short fragments are used to bulk up and prepare DNA then
long repeating elements will never be tackled with any confidence.

So we need to bulk up with confidence.

the 454 has a good potential of having a large read of a single fragment.

If the repeat is less than 300 all good.

Current Pair end approach has good accuracy but is derived from short sequences and so could be placed anywhere in the genome.

So the logic is take the region that is linked/of interest and bulk it up as a whole.

Then take that bulk and fragment consistently.

Then sequence.


Even if you can do that well next gen still has limits. Lucky you know where you are looking so can apply some alignment.

if a repeating element is 600bp long how can you tell if someone has 2 or 3 of them in tandem?

That is what I am looking to solve.

YACs, BACs, Fosmid and long range PCR etc... all deal with reasonable accuracy bulking up source material.

Whatever we use to bulk up there is still the limits of the machines they need longer reads as well.

So its actually 2 things that need sorting out.

Good quality bulking up of source material till single molecule sequence is viable.

And long reads on the machines to ensure that your efforts of preserving the sequence are rewarded.

Thanks for taking an interest in this and I hope we can solve it... any ideas.

[pmiguel]

Seems like you are conflating two unrelated issues:

(1) Until single molecule sequencing (SMS) is possible, methods to amplify single molecules are necessary to sequence them.

(2) Repetitive sequences can confound currently available piecemeal sequencing technologies. This is especially the case when the repeat units are long or the numbers of repeats in a row are large.

SMS does not help you at all with (2). If it did, I expect sales of Heliscopes would be more robust.

However if we turn away from the locally highly repetitive sequence issue, what you state is true and perhaps not widely appreciated. By obtaining and amplifying fairly large segments of a genome and focusing your sequencing efforts on those you can overcome many of the issues involved with dispersed repetitive sequences because in these smaller segments of a genome they are no longer repetitive.

The problem is that making these types of libraries is not cheap. Nor is making 10,000 sub-libraries to sequence (from each BAC in a tiling path across a genome)instead of 1 full genome library to sequence. And you can get a very large percentage of the total information from a genome using full genome shotgun -- and that is comparatively very cheap.

--
Phillip