SEQanswers (
-   Sample Prep / Library Generation (
-   -   Newbie question: low cluster density on MiSeq (Chip-Seq pool) (

HelenaSC 07-05-2013 05:39 AM

Newbie question: low cluster density on MiSeq (Chip-Seq pool)
1 Attachment(s)
Hi all,

I'm new on this and it seems that's we have some problem with cluster density on MiSeq. We have just run a pool of 2 libraries (from Chip-seq ) , 56 cycles and we are getting very bad results:

* Q-score all cycles: 0.1G, 82%
* Cluster density: 471K/mm2
*Clustering passing filter: 20%

I think that i made some mistake in the calculations preparing the pool of quantified libraries at the time of adding NaOH, but I'm not sure...

I have attached a file with the basic calulations. Anyone could have a look and let me know if I shouldn't have diluted the samples so much?We choose 11pM for final concentration of the pool...was it correct or should we have choosed 11pM for each sample?

Thanks a lot in advance!

Calluna 07-05-2013 08:31 AM

I may have done my math incorrectly, but I came up with a final NaOH concentration of 0.002N in your final mixture. If so, that would probably explain the low cluster density/low passing filter. Illumina documents typically say to stay below 0.001M (1 mM) in the final solution after diluting with HT1.

I would recommend reprepping your denature/dilution. Since your libraries are fairly low concentration, you will probably want to denature, but neutralize your final denature with HCl. See this post for more help:

Happy sequencing :)

Calluna 07-05-2013 08:33 AM

I forgot to mention, if your libraries are not very diverse, I would also recommend spiking in some PhiX. If you have the updated Miseq software, I believe you can get by with 5% for very low diversity samples; otherwise, try spiking 25% PhiX and see what you get.

HelenaSC 07-07-2013 11:23 PM

Hi Calluna!

Thanks for your help, we will keep in mind NaOH concentration for next round :)

HelenaSC 07-08-2013 12:26 AM

After reading all these posts, I wonder, can't I just use 0.1N NaOH (instead of 0.2N) to denature my libraries ,with no further neutralization step?
The issue is that I don't want to dilute my samples as much, so , maybe I can start with lower concentration of NaOH??

Thanks a lot!

Calluna 07-10-2013 08:07 AM

Call Illumina and ask if 0.1N NaOH will be strong enough.

If not, you can use a higher concentration NaOH, but less of it... and then neutralize.

I think you can do the following (but double check my math!)
6.5uL total of your 1.1nm normalized/mixed libraries
2uL of 0.5N NaOH
1.5uL H2O
Total denature volume = 10uL

Add 1uL of 1N HCl
Total neutralized denature = 11 uL

Add 639uL HT1

Total final volume = 650uL

If I did the math correctly, I think that gives you an 11pM final concentration, and because you neutralized the NaOH, you shouldn't have to worry about pH.

HelenaSC 07-16-2013 03:39 AM

Hi Calluna!

Thanks a lot, I checked your maths and they were perfect :P They gave me a 10.95pM final concentraction :)

We will perform the run this week, I hope it will work!

Thanks a lot for your help.

HelenaSC 07-17-2013 07:24 AM

My last question before starting (sorry!!! :) ):

I have two samples (1.1nM) that I want to run together on the MiSeq. Following your instructions, do you think this will be ok ?(to end up with a 11pM final concentration):

pool 2 libraries (1.1nM each one) :

5ul lib1 + 5 ul lib2 = 10ul pool 1.1nM
+ 2ul 0.5N NaOH
1ul water
(13ul of denatured samples at 0.84nM)

Then, neutralize with 1ul of 1N HCl (denatured and neutralized samples ; 14ul at 0.78nM)
Finally, dilute samples with 992ul HT1 buffer = samples at final concentration= 11pM (load 600ul).

Thanks a lot!!!

Calluna 07-17-2013 09:13 AM

I'm not sure how crucial it is to have your denature NaOH concentration at at least 0.1N, so I wouldn't mess with it if you don't have to.

This is what I would suggest instead:

5uL of 1.1nM sample A
5uL of 1.1nM sample B
2.5uL of 0.5N NaOH (or 1.25uL H2O + 1.25uL 1N NaOH...either way it is the same thing. This way, you maintain a final effective NaOH concentration in the denature if 0.1N)
Denature volume = 12.5uL

add 1.25uL of 1N HCl
Neutralized volume = 13.75uL

Hyb mix:
add 986.25uL HT1
Total final volume of 1000uL.

Don't forget to use some PhiX to help with any diversity issues.

HelenaSC 07-18-2013 02:08 AM

Thanks a lot!!!
IN a few hours I will let you know if it worked!
I tried to answer your private message but I think I don't know how to do it, since I tried twice and no messages appear on my "Send items" folder!!! :O

HelenaSC 07-18-2013 06:28 AM are the results :(

It seems that changind the final NaOH concentration has improved, somehow, the quality of the run, but not the cluster density...Now, we have more outputh, higher cluster pasing filter value but less cluster density... Here is the comparative between the " old run" and the "new run" from this morning:

“Old run”
Final concentration = 11pM
[NaOH] final = 0.002N
No neutralization
Qscore all cycles= 0.1G, 82%
Cluster density= 471K/mm2
Clustering passing filter= 20%

“New run”
Final concentration = 11pM
[NaOH] final = 0.001N
Neutralized with 1N HCl
Qscore all cycles= 0.3G, 98.6%
Cluster density= 363K/mm2
Clustering passing filter= 93.6%

Any idea? May it be something about the quantification?Should I load more sample?

Thanks a lot!

Calluna 07-18-2013 07:20 AM

I'm glad your Q scores and PF improved. The first thing that comes to mind is to ask how you determined concentration of your libraries. If it was based off High Sensitivity BA, I get quite a bit of variability in accuracy. If you haven't already, I would suggest trying a qPCR to quant the libraries. Second, I would probably suggest giving Illumina tech support a call to see if they can go through your run data with you.

As far as your other thread goes,
I'm not entirely sure what the extra hump would be from. It doesn't look broad enough to be bead carryover...maybe it is a bubble product, although your BA looks more like a shoulder than a bubble to me

Maybe someone else may have some insight.

HelenaSC 07-18-2013 07:28 AM

Hi again,

We quantify libraries using qPCR (KAPA Kit) and correcting by size with Bioanalyzer profile. Having a look at the results it seems that Chip is more "represented" than Input, and we think that the reason is the "strange " profile of chip samples (the correction by its size on bioanalyzer may not be ok, since the profile is more difuse or wide). So , maybe we should try these 2 things:

a) Correct by size again, in orde to have a more equitative pool.
b) Go for 16-17pM final concentration instead of 11pM and see if cluster density grows up, since the run seems to be "perfect" regarding the other metrics.

Thanks for your comments :)

All times are GMT -8. The time now is 09:10 AM.

Powered by vBulletin® Version 3.8.9
Copyright ©2000 - 2020, vBulletin Solutions, Inc.