SEQanswers

Go Back   SEQanswers > Sequencing Technologies/Companies > Illumina/Solexa



Similar Threads
Thread Thread Starter Forum Replies Last Post
TruSeq or Nextera kit for library prep? rooty Sample Prep / Library Generation 12 10-29-2013 06:02 AM
Automation for TruSeq library prep? jlove Sample Prep / Library Generation 10 01-10-2012 08:34 AM
Library prep for 50ng input with TruSeq DNA- or RNAseq adapters? sehrrot Sample Prep / Library Generation 1 11-09-2011 04:03 AM
Why and how are gDNA contaminations in RNAseq library preps bad? lukas1848 Sample Prep / Library Generation 1 11-04-2011 03:51 AM
Lib prep with 10-200bp (45bp peak) gDNA saikumarkv Sample Prep / Library Generation 0 02-18-2011 01:44 PM

Reply
 
Thread Tools
Old 02-21-2017, 12:25 PM   #41
pmiguel
Senior Member
 
Location: Purdue University, West Lafayette, Indiana

Join Date: Aug 2008
Posts: 2,208
Default

Quote:
Originally Posted by Cistron View Post
Hello folks,

Interesting thread! I got a lot of high molecular weight smear in my last run as well (following EMP protocols for soil samples).

I'm trying to figure out the details of the daisy-chain hypothesis:

From what I see the Illumina adapters on either end aren't complementary. So I guess the thought is that a few annealing bps are enough to lead to 3' extension. However, why would this lead to preferential extension to form multimers, thus starting at the adapter sequences? If it is just a few bps duplexing, surely this could happen anywhere within the amplicon.

Is my thinking off. Am I missing a point?
No, I think you are right. I just hadn't thought carefully enough about it previously. It is very likely that the bubble product hypothesis is correct.

Wait, I just re-read your post. The bubble product hypothesis explains the presence of an extra peak with higher apparently molecular weight. Molecules comprising the "high MW" peak are not actually higher MW, they are just largely "di-monoplex" rather than duplex. That is, the ends are duplex, the middle comprises two non-complementary strands.

This doesn't directly explain a "smear" extending up into much higher MWs. Unless one posits that the bubble products are chain-linking together into larger structures. Not "daisy-chaining", but creating loops that have actually interlocked. (I'm sure there is a term in topology for this, but I don't know what it would be.)

To test this I guess you could massively dilute your sample, and subject it to denaturing temperatures. Then cool it diluted and re-concentrate it. If true, one would expect to get just 1 or 2 peaks afterwards, not the smear.

That said, if you just want libraries to sequence, the answer is generally to ignore the issue, use qPCR to titrate and move ahead. Or just construct libraries with fewer cycles of amplification.

--
Phillip
pmiguel is offline   Reply With Quote
Old 02-21-2017, 12:39 PM   #42
Cistron
Junior Member
 
Location: UK

Join Date: Feb 2013
Posts: 3
Default

Quote:
Originally Posted by pmiguel View Post
No, I think you are right. I just hadn't thought carefully enough about it previously. It is very likely that the bubble product hypothesis is correct.

Wait, I just re-read your post. The bubble product hypothesis explains the presence of an extra peak with higher apparently molecular weight. Molecules comprising the "high MW" peak are not actually higher MW, they are just largely "di-monoplex" rather than duplex. That is, the ends are duplex, the middle comprises two non-complementary strands.

This doesn't directly explain a "smear" extending up into much higher MWs. Unless one posits that the bubble products are chain-linking together into larger structures. Not "daisy-chaining", but creating loops that have actually interlocked. (I'm sure there is a term in topology for this, but I don't know what it would be.)

To test this I guess you could massively dilute your sample, and subject it to denaturing temperatures. Then cool it diluted and re-concentrate it. If true, one would expect to get just 1 or 2 peaks afterwards, not the smear.
Hi Philip,

Thanks for your answer!

Right, so "daisy-chains" aren't really extended much, but entangled products. That makes more sense.

The bubble products made sense when I first heard about it, however, the appearance of seemingly multiples of the original product on the gel led me off track.

Quote:
That said, if you just want libraries to sequence, the answer is generally to ignore the issue, use qPCR to titrate and move ahead. Or just construct libraries with fewer cycles of amplification.

--
Phillip
The stuff I do for my students
Cistron is offline   Reply With Quote
Old 02-21-2017, 01:13 PM   #43
pmiguel
Senior Member
 
Location: Purdue University, West Lafayette, Indiana

Join Date: Aug 2008
Posts: 2,208
Default

Quote:
Originally Posted by Cistron View Post
Hi Philip,

Thanks for your answer!

Right, so "daisy-chains" aren't really extended much, but entangled products. That makes more sense.
If you like. I had envisioned daisy-chains as linear arrays of products, each linked by duplex ends.

But I was clearly wrong, as you point out.

--
Phillip
pmiguel is offline   Reply With Quote
Old 02-22-2017, 08:52 AM   #44
pmiguel
Senior Member
 
Location: Purdue University, West Lafayette, Indiana

Join Date: Aug 2008
Posts: 2,208
Default

Quote:
Originally Posted by pmiguel View Post
If you like. I had envisioned daisy-chains as linear arrays of products, each linked by duplex ends.

But I was clearly wrong, as you point out.

--
Phillip
Wait, no!! I was Jedi mind-tricked. No reason my original vision of linear daisy chains wouldn't happen.

Each double stranded library amplicon after it is created by the polymerase from a single-stranded template molecule will comprise 2 strands.

Code:

5e-A--M-B'-3e
   |  | |
3e-A'-M-B--5e
Where "5e" denotes "5 prime end", "A" is the entire top strand of the left (forward) adapter, "A'" is the reverse complement, or bottom strand of the left (forward) adapter. "M" should be considered the insert, either strand. "B" and "B'" are just the right (reverse) adapter strand.

During denaturation the two strand melt apart and during annealing could form, for example:

Code:

5e-A--M-B'-3e
        |
     5e-B-M-A'-3e
            |
         5e-A--M-B'-3e
Except this is a linear diagram and you would need to imagine the 5' end fo B annealing to the 3' endo fo B'.

For any number of links of these type to form, the intermolecular kinetics must be faster than the intramolecular formation of a bubble product. Which would require very high DNA concentration and/or longer inserts.

--
Phillip
pmiguel is offline   Reply With Quote
Old 02-22-2017, 09:04 AM   #45
Cistron
Junior Member
 
Location: UK

Join Date: Feb 2013
Posts: 3
Default

Ah, yes - do you mean like this?

Code:
     3'5'      3'5'
     | |       | |
5'__/   \_____/   \___3'
I guess this could still form a complex bubble.
Cistron is offline   Reply With Quote
Old 02-22-2017, 10:09 AM   #46
pmiguel
Senior Member
 
Location: Purdue University, West Lafayette, Indiana

Join Date: Aug 2008
Posts: 2,208
Default

Quote:
Originally Posted by Cistron View Post
Ah, yes - do you mean like this?

Code:
     3'5'      3'5'
     | |       | |
5'__/   \_____/   \___3'
I guess this could still form a complex bubble.
Yes. That's it.
And yes, large circles could form. As could the chain link topologies discussed upthread.

--
Phillip
pmiguel is offline   Reply With Quote
Reply

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off




All times are GMT -8. The time now is 06:45 AM.


Powered by vBulletin® Version 3.8.9
Copyright ©2000 - 2017, vBulletin Solutions, Inc.
Single Sign On provided by vBSSO