SEQanswers

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



Similar Threads
Thread Thread Starter Forum Replies Last Post
Library quantification: opinions? krobison Sample Prep / Library Generation 41 06-23-2016 07:38 PM
Illumina library quantification deepakpatilp Illumina/Solexa 75 06-24-2013 10:21 AM
Library Quantification Confusion! peromhc Sample Prep / Library Generation 9 10-05-2011 08:18 AM
Nanodrop vs bioanalyzer for library quantification k-gun12 Illumina/Solexa 1 09-24-2010 08:26 AM
3'UTR library or random primed cDNA library for quantification? Rosanne82 Sample Prep / Library Generation 0 06-26-2009 06:27 AM

Reply
 
Thread Tools
Old 08-16-2010, 08:53 AM   #1
suludana
Member
 
Location: Spain

Join Date: May 2008
Posts: 61
Question Library quantification

Hi,
I am using qPCR and Bioanalyzer to quantify my libraries before sequencing in a GAII. Even so I still can not get it right. Each run is a surprise (not always good). Does anyone have a really good method?
Thanks
suludana is offline   Reply With Quote
Old 09-07-2010, 08:52 AM   #2
sbsuser
Junior Member
 
Location: Cambridge

Join Date: Sep 2010
Posts: 7
Smile

if you still cant make it from qPCR conc to get the right density of flowcell. i prefer you to sequence for 6 cyc on GA and make the flowcell with the right conc.
sbsuser is offline   Reply With Quote
Old 09-08-2010, 12:07 AM   #3
suludana
Member
 
Location: Spain

Join Date: May 2008
Posts: 61
Default

Do you mean spend a Flow Cell to determine the exact density?
suludana is offline   Reply With Quote
Old 09-08-2010, 03:11 AM   #4
sbsuser
Junior Member
 
Location: Cambridge

Join Date: Sep 2010
Posts: 7
Default

yes, deblock FC after 6 cyc and store in fridge. You can reuse the flowcell if you want to validate your GA.

Last edited by sbsuser; 09-08-2010 at 03:14 AM.
sbsuser is offline   Reply With Quote
Old 09-09-2010, 12:18 AM   #5
simonandrews
Simon Andrews
 
Location: Babraham Inst, Cambridge, UK

Join Date: May 2009
Posts: 871
Default

If you're running a control lane on your flowcells you can also mix (a best guess) 10% of another library in with PhiX to get a pretty accurate quantitation and also check on the quality of the library.
simonandrews is offline   Reply With Quote
Old 09-09-2010, 12:23 AM   #6
sbsuser
Junior Member
 
Location: Cambridge

Join Date: Sep 2010
Posts: 7
Default

How can you mix another library in to phix to get accurate quantitation? Can you explain.
sbsuser is offline   Reply With Quote
Old 09-09-2010, 12:30 AM   #7
simonandrews
Simon Andrews
 
Location: Babraham Inst, Cambridge, UK

Join Date: May 2009
Posts: 871
Default

Quote:
Originally Posted by sbsuser View Post
How can you mix another library in to phix to get accurate quantitation? Can you explain.
Yes, just mix in the other library prior to putting it on the flowcell. After the run extract any reads from that lane which didn't map to PhiX and those will mostly come from your other library. The PhiX stats don't get messed up (other than %mapping) and you still have enough diversity to do the phasing calculations.

If you get around 10% of the other library from your lane we find that around 90% of the unmapped sequences come from this library. You can filter out most of the other PhiX reads by remapping against PhiX with relaxed constraints if you want, and you can get a pretty good idea of what's going on in your other library.
simonandrews is offline   Reply With Quote
Old 09-09-2010, 05:05 AM   #8
sbsuser
Junior Member
 
Location: Cambridge

Join Date: Sep 2010
Posts: 7
Default

is this abt the in lane Phix experiment?
sbsuser is offline   Reply With Quote
Old 09-09-2010, 05:13 AM   #9
simonandrews
Simon Andrews
 
Location: Babraham Inst, Cambridge, UK

Join Date: May 2009
Posts: 871
Default

Quote:
Originally Posted by sbsuser View Post
is this abt the in lane Phix experiment?
It's more the other way. Illumina now offer an option to spike small amounts of PhiX in with each lane for QC purposes, but these don't help if you need a control lane for calibration purposes. If you're still running a full PhiX lane you can spike in another library to that to get a low level of coverage which won't affect the normal function of the control lane.
simonandrews is offline   Reply With Quote
Old 09-15-2010, 03:41 PM   #10
SeqR&D
Member
 
Location: San Diego

Join Date: Sep 2010
Posts: 26
Default

This is how we did it:

1. We wasted a FC
2. Amplified (~14 cycles) gel purified/size selected library (in the size range that you use to sequence)
3. Purified the PCR products
4. Nanodrop quantified the PCR products
5. Made dilution series to hyb to a FC
6. Pick the "best" dilution that gives desired cluster #
7. From this info, create a qPCR STD which you will use to quantify every subsiquent sample that is going on to sequencing. Make tons of this stuff (alliquoted in strip tubes and left in the -20C), and do not run out without using a set to quantify your next batch of STD!!
Our qPCR STD dilution series is this:
100pM, 30pM, 10pM, 3pM, 1pM, 0.3pM, 0.1pM, 0.03pM

As long as you always cluster the amount that you found in step #6, you should be great.

Let me know if this doesn't make sense.
SeqR&D is offline   Reply With Quote
Old 09-20-2010, 02:27 PM   #11
lhandley
Junior Member
 
Location: Southeast USA

Join Date: Apr 2010
Posts: 5
Default

Am I correct to guess that the titration run on the flowcell would have to be repeated with each software upgrade and/or formulation change since that would affect cluster #?
lhandley is offline   Reply With Quote
Old 09-20-2010, 03:44 PM   #12
SeqR&D
Member
 
Location: San Diego

Join Date: Sep 2010
Posts: 26
Default

I wouldn't think so. The most difficult part here is aligning your qPCR results to your cluster #. After that, your cluster # should not change dramatically enough to have to redo the titration. If you see that your cluster # is going down/up all you have to do is change your concentration.
Example: If you are generally clustering at 8pM and you see that your cluster density has dropped across the board (even PhiX control lane), then you might want to start clustering at 10pM or 14pM (and change PhiX if necessary).....

It is possible to make a mistake when clustering the flowcell, so be careful. I occasionally get flowcells that will fail the first base report with low cluster density, but it was the process or the flowcell or the clustering that was the issue...not the concentration.
SeqR&D is offline   Reply With Quote
Old 03-08-2012, 06:49 PM   #13
Genquest
Member
 
Location: Indo

Join Date: Oct 2011
Posts: 20
Default Library qualification by qPCR

Hello everyone,
I am pretty new to this forum and stumbled on this post as even I am facing problem with qPCR quantification method
1.Has anyone encountered with, a higher qPCR concentration than the estimated concentration by pico green??
2. Also if assumed that qPCR is more accurate(w.r.t pico green) as it would give conc. of ssDNA ligated with adapters,then the cluster generation should be accurate with qPCR conc. but it doesn't happen so...the flow cell (V3,Hiseq 2000) remains under clustered.Can any one think of possible reasons??
3. This trend seems to change with every library.Is absolute quantification the right approach for Sybr green qPCR method?Also I am using the PhiX control provided by Illumina as my standard.
4.Do people take the "exact concentration of libraries" as determined by qPCR for cluster generation or they compare it with pico green to see the relative trend of concentration?

Would appreciate if any of the queries are resolved!!
Thanks in advance
Genquest is offline   Reply With Quote
Old 03-09-2012, 04:52 AM   #14
pmiguel
Senior Member
 
Location: Purdue University, West Lafayette, Indiana

Join Date: Aug 2008
Posts: 2,317
Default

There are factors that might confound qPCR. They include:

(1) Ethidium Bromide in your sample. I did not do a careful test of this. But someone in the lab tried spiking EtBr into a sample at a low level. It shifted the Ct by a cycle or two. Ethidium Bromide will also confound fluorimetry, so purifying it away after some size selection methodologies is important.

(2) Your phiX may not be the concentration and/or length that you think it is. Most current (v3) phiX libraries from Illumina are supposed to be about 500 bp. So far every one that we tested has been around that size. I think v2 (including v2 multiplex) phiX is supposed to be 350 bp. The v2 multiplex phiX has always tested out to 350 bp or so for us. But we saw one case where the non-multiplex v2 was 500 bp, rather than 350 bp. Further, I have seen a 500 bp library that had a very weak, but visible, 350 bp peak. Also one library had a ~70 bp peak. I don't know what the adapter length is for these library types, so I don't know if this is an adapter dimer. If it was, then it was 10% of the molar total. But would produce little signal during SYBR Green qPCR.

Since the phiX standards are likely calibrated primarily to produce a given number of clusters, the actual size of the library molecules may not be a great concern for Illumina. However, SYBR green assays do not directly determine molarity -- instead they rely on you to supply the correct average size of libraries. So:

(3) You must have an accurate determination of both your library's average length and your concentration standard's length.

(4) One more: TruSeq libraries, if they are run no-amp, cluster at least an order of magnitude lower than qPCR would suggest they should. Illumina tells you, that you must do at least one cycle of enrichment PCR to "break the fork" of the adapters. Taken literally, this would suggest that the adapters are covalently linked strand-to-strand by a chemistry that breaks during extended heating but not by NaOH. Alternatively the adapters might lack some critical "magic" present in the PPC -- PCR primer cocktail -- of the TruSeq DNA/RNA prep kits. So, these factors could result in issues, if true.
--
Phillip

Last edited by pmiguel; 03-09-2012 at 05:02 AM. Reason: Added point (4).
pmiguel is offline   Reply With Quote
Old 03-09-2012, 09:28 AM   #15
SeqR&D
Member
 
Location: San Diego

Join Date: Sep 2010
Posts: 26
Default

There is no magic in the PPC. Adapters are not a complete "library element"...which is the beauty behind the Y-adapter. You do not have complimentary ends...they are "sticky", and only contain half of what you need to cluster (only half of your library elements have what it takes to hybridize to the flowcell). They should however qPCR with more accuracy than an order of magnitude...as only the first cycle of qPCR should be affected...but maybe thats enough....I haven't done the math.
SeqR&D is offline   Reply With Quote
Old 03-09-2012, 11:37 AM   #16
pmiguel
Senior Member
 
Location: Purdue University, West Lafayette, Indiana

Join Date: Aug 2008
Posts: 2,317
Default

Quote:
Originally Posted by SeqR&D View Post
There is no magic in the PPC. Adapters are not a complete "library element"...which is the beauty behind the Y-adapter. You do not have complimentary ends...they are "sticky", and only contain half of what you need to cluster (only half of your library elements have what it takes to hybridize to the flowcell). They should however qPCR with more accuracy than an order of magnitude...as only the first cycle of qPCR should be affected...but maybe thats enough....I haven't done the math.
From Illumina's FAQ:

Quote:
Is there a PCR-free TruSeq Sample Prep protocol available?

The new index adapter design enables PCR-free protocols. (A single cycle of synthesis is required to separate the forked adapter.)
Upon querying Illumina tech support, I got the following response:

Quote:
We have found that if you simply take the annealed fragment that has the forked adapters on each end and add it to the clustering reaction that even with NaOH treatment before clustering, there is poor initial hybridization and the resultant clustering is poor. By allowing at least one PCR cycle this separates the strands and the forked adapters, allows a single amplification of the entire strands, resulting in a clean, double-stranded species that gives cleaner clustering results.
This makes no sense unless one or both of the following is true:

(1) The adapter oligos are attached to each other in some way that resists their melting apart during NaOH treatment.

(2) The PPC primers themselves have some modification that causes them to bind especially well to the flow cell.

Or, at least, that is my take on the situation.

--
Phillip
pmiguel is offline   Reply With Quote
Old 03-09-2012, 02:02 PM   #17
SeqR&D
Member
 
Location: San Diego

Join Date: Sep 2010
Posts: 26
Default

Neither of those are true. I'm certain that if you run a denaturing gel on the adapters, that they will run as 2 bands. And, PPC primers are complimentary to the adapter ends...nothing special.

Where you might run into issues are with A-tailing efficiency, ligation efficiency, possibly ligating 2 adapters together, having free adapters present when clustering a flowcell (they will take up space, but will not extend or show clusters). Doing some amplification enriches for library elements that have proper adapter ligation on both ends.
SeqR&D is offline   Reply With Quote
Old 03-10-2012, 05:31 AM   #18
pmiguel
Senior Member
 
Location: Purdue University, West Lafayette, Indiana

Join Date: Aug 2008
Posts: 2,317
Default

Quote:
Originally Posted by SeqR&D View Post
Neither of those are true. I'm certain that if you run a denaturing gel on the adapters, that they will run as 2 bands.
What makes you certain?

Quote:
Originally Posted by SeqR&D View Post
And, PPC primers are complimentary to the adapter ends...nothing special.
Why do they run ~80 nt on a pico RNA chip if they are generic oligos?

Quote:
Originally Posted by SeqR&D View Post
Where you might run into issues are with A-tailing efficiency, ligation efficiency, possibly ligating 2 adapters together,
Okay, I'm with you so far...

Quote:
Originally Posted by SeqR&D View Post
having free adapters present when clustering a flowcell (they will take up space, but will not extend or show clusters).
And PCR will dilute out the free adapters, okay, I guess.

Quote:
Originally Posted by SeqR&D View Post
Doing some amplification enriches for library elements that have proper adapter ligation on both ends.
I see your point. But when we tried no amp libraries we used qPCR to determine the concentration of library elements that had proper adapter ligation on both ends. That is how we determined how much to add to the flow cell. qPCR (SYBR green) should not be confounded by any of the factors you list above.

Yet, when we clustered using the unamplified libraries, we saw much lower cluster yields from those libraries. (Like 1% of what was expected, if I recall correctly.) Our qPCR was not otherwise way off -- all the other libraries we quantified this way clustered at about the expected efficiency. So it seems like the simple explanation is that there is something odd about the TruSeq system that Illumina has not revealed.

--
Phillip
pmiguel is offline   Reply With Quote
Old 03-10-2012, 08:53 PM   #19
SeqR&D
Member
 
Location: San Diego

Join Date: Sep 2010
Posts: 26
Default

If you qPCR a non-ampified sample that has varying amounts of free adapter (with a keenly designed SPRI purification you can greatly minimize this), library elements with one ligated adapter and those with both, you will be quantifying the 2 adapter libraries, but the other stuff is in there as well. I'm not exactly sure what ligation efficiency is, but from what I understand it's not great...60% to 70%? (correct me if I'm wrong). That could leave a lot of non-clusterable DNA fragments floating around that will "block" the flowcell surface from clusterable library. I'm not sure if this can account for a 99% reduction in cluster density from one to the other. Keep in mind, this is only my best guess.
SeqR&D is offline   Reply With Quote
Old 03-11-2012, 09:41 AM   #20
pmiguel
Senior Member
 
Location: Purdue University, West Lafayette, Indiana

Join Date: Aug 2008
Posts: 2,317
Default

Quote:
Originally Posted by SeqR&D View Post
If you qPCR a non-ampified sample that has varying amounts of free adapter (with a keenly designed SPRI purification you can greatly minimize this), library elements with one ligated adapter and those with both, you will be quantifying the 2 adapter libraries, but the other stuff is in there as well.
I'm not exactly sure what ligation efficiency is, but from what I understand it's not great...60% to 70%? (correct me if I'm wrong).
You are right, I think. If anything 60%-70% is on the high side. Although if you presume that 60% means "60% of ends get adapters ligated to them", one would expect only 36% to get both ends adapter ligated.

Actually on Friday we looked at the results from a similar ligation chemistry -- the Roche Rapid Library prep kit for 454 library construction -- and saw 12% of the molecules were amplifiable via qPCR. Not entirely clear this is a result of whether adapters were ligated on or not. The genomic DNA given us was in questionable state -- prepared from old refrigerated leaf tissue or even twigs -- it was degraded to 5kb average length prior to any sonication. So the DNA may just have had a moderately high frequency of damage not circumventable by the qPCR polymerase.

Hmmm, actually the Roche Y adapters have FAM modification. So I guess we could actually assess the number of molecule ends with adapters ligated to them.

Quote:
Originally Posted by SeqR&D View Post
That could leave a lot of non-clusterable DNA fragments floating around that will "block" the flowcell surface from clusterable library. I'm not sure if this can account for a 99% reduction in cluster density from one to the other. Keep in mind, this is only my best guess.
Yeah, I like the way you think. The possibility of a blocking issue had not occurred to me prior to your previous post.

As it turns out the attempts we made with no-amp libraries were made from bacterial genomic DNA. So we were only shooting for 10% of a lane for each. (Even that would have been overkill...) So if there were material in the library blocking access of the flow cell oligos to library molecules we would see an effect on the other libraries sharing the lane commensurate with what we saw with with the no-amp libraries. But the other libraries seemed unaffected.

Also, we had previously ran amplified versions of these libraries and they seemed to produce plenty of clusters -- but they were terribly biased. All of this could be the result of the DNA of the organism (Deinococcus radiodurans) -- who know what modifications it hosts. But since Illumina tech support suggested that very poor clustering efficiency was to be expected for TruSeq libraries that had not undergone at least 1 cycle of amplification, I tend to draw other conclusions.

Obviously, I could have this all wrong. Just a couple of hypotheses...

--
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 07:17 PM.


Powered by vBulletin® Version 3.8.9
Copyright ©2000 - 2020, vBulletin Solutions, Inc.
Single Sign On provided by vBSSO