Inefficient ATAC-seq library prep amplification

lapensee · 11-29-2018, 05:27 AM

Hi all,

I’m having trouble generating an “optimal” library for sequencing.
I am using a human cell line derived from an adrenocortical tumor (from ATCC).

I have tried 12.5, 25, 50, and 100k cells with 2.5ul Tn5 (Illumina).
I've tried lysing with and without Digitonin, no difference.

qPCR determined that I needed ~5-7 extra cycles of amplification.

Qiagen cleanup was not removing the primer, and I had large peaks at ~2000 or bigger, so I switched to AMPure beads (double sided purification, first 0.5X then 1.8X)

I never seem to get those nice big peaks around ~<200, 400 just a blip. See pics below.

Does anyone have any ideas for improving the efficiency of my library amplification?

Thanks
Chris

lapensee · 12-03-2018, 08:41 AM

If not suggestions, would someone be willing to post a pic of a GREAT bioanalyzer profile (for ATAC-Seq)?

lapensee · 12-13-2018, 11:15 AM

Suspecting undertagmentation. Anyone else run into this problem, and if so, what was your remedy?

Meyana · 12-13-2018, 04:38 PM

I have seen the same thing repeatedly with FACS sorted nuclei isolated from frozen tissue, hoping someone has an idea on how to optimize...

lapensee · 03-20-2019, 10:16 AM

FWIW I got this working. I adopted the OMNI-ATAC protocol and get great library amplification from 5k cells using only 2% the amount of Tn5 required for 50k cells. Adding a protease inhibitor during transposition was crucial for my library prep.

Rosmano · 04-03-2019, 02:16 AM

Hey, did you use the new Nextera DNA flex Library Prep kit?
I've been having similar problems to you but nothing I do seems to work.

lapensee · 04-03-2019, 04:45 AM

I used "Nextera DNA Library Prep Kit"
FC-121-1030

Rosmano · 08-10-2019, 09:15 AM

What protease inhibitor did you use in your protocol? I adopted the Omni-ATAC protocol but there's no qPCR amplification, even though I have DNA (7.6ng/uL obtained from 50,000 thymocytes).

I don't even know if my pre-amplification succeeded or not.

lapensee · 08-12-2019, 07:10 AM

Halt protease inhib without EDTA (100x) from thermofisher

Rosmano · 08-12-2019, 09:53 AM

Well, after analyzing my qPCR results again, what I have to say is that the PCR looks awful but there's something there. Unfortunately, by the end of the 20 cycles I didn't hit a plateau stage and the amplification are of very low amounts. The exponential algorithm fails even.
Considering that I tried with the Omni-ATAC buffer with Tween and digitonin, I am considering re-trying only with NP-40 like in the original Buenrostro paper. Does anyone have any other idea of other optimization steps?

Rosmano · 08-14-2019, 02:10 AM

By the way, I am doing the qPCR quantification with Power SYBR Green PCR Master Mix from Thermo Fisher instead of the NEB Next HF Master Mix.

Does anyone know if this might impact the qPCR quantification?

Meyana · 08-14-2019, 02:34 AM

I started out using the PCR mix in the Nextera kit for both PCR and qPCR and got little to no amplification. Switching all steps to NEB (as described in the Omni paper) got everything working smoothly for me and highly reproducible with as low as 500 FACS'ed nuclei.
Based on this, I recommend using NEB products for all steps.
Also, I do not think it is a good idea to "switch" polymerase mix between quantification and amplification step.

Best of luck!

Rosmano · 08-20-2019, 02:47 AM

Yup, I got everything working by using solely the NEBNext master mix.

Now, I'm just wondering about performing bead clean-up... I've decided to do a Fragmet analyzer before doing size selection, and depending on the profile I'll decide if I need to do a size selection or not.

By the way, does anyone have an example of a Fragment Analyzer profile of a successful ATAC-seq library?

Meyana · 08-20-2019, 04:53 PM

I always do both upper and lower size selection using AMPure XP beads, I don't see a reason not to just do it right away (unless you are unfamiliar with bead cleanup and want to have a way to check you process).
I add 0.5x, keep supernatant, add 1.3x (total 1.8), keep beads, elute. Then check BA.

I attached some BA tracks from a test I did, D and E are how I like my libraries to look and these two sequenced nicely.

firsal · 09-03-2019, 07:00 PM

Quote:

Originally Posted by lapensee

FWIW I got this working. I adopted the OMNI-ATAC protocol and get great library amplification from 5k cells using only 2% the amount of Tn5 required for 50k cells. Adding a protease inhibitor during transposition was crucial for my library prep.

Hi, I'm also doing Omni-ATAC-seq but I don't get reproducible results. Did you add 1x protease Inhibitor?

pabanga1 · 10-14-2019, 02:45 AM

Quote:

Originally Posted by lapensee

FWIW I got this working. I adopted the OMNI-ATAC protocol and get great library amplification from 5k cells using only 2% the amount of Tn5 required for 50k cells. Adding a protease inhibitor during transposition was crucial for my library prep.

Hi! I am wondering what does exactly from the Omni-ATAC protocol mean:

"Add 50 ul cold ATAC-Resuspension Buffer (RSB) containing 0.1% NP40, 0.1% Tween-20, and 0.01% Digitonin and pipette up and down 3 times.
Incubate on ice for 3 minutes.
Wash out lysis with 1 ml of cold ATAC-RSB containing 0.1% Tween-20 but NO NP40 or digitonin and invert tube 3 times to mix"

Does this mean I need to removed the previous 50 uL and wash with cold ATAC-RSB?

Rosmano · 12-01-2019, 05:09 AM

Quote:

Originally Posted by pabanga1

Hi! I am wondering what does exactly from the Omni-ATAC protocol mean:

"Add 50 ul cold ATAC-Resuspension Buffer (RSB) containing 0.1% NP40, 0.1% Tween-20, and 0.01% Digitonin and pipette up and down 3 times.
Incubate on ice for 3 minutes.
Wash out lysis with 1 ml of cold ATAC-RSB containing 0.1% Tween-20 but NO NP40 or digitonin and invert tube 3 times to mix"

Does this mean I need to removed the previous 50 uL and wash with cold ATAC-RSB?

This means you add 1mL of cold ATAC-RSB containing 0.1% Tween-20 to the 50uL to wash the sample

Rosmano · 12-01-2019, 05:15 AM

Quote:

Originally Posted by Meyana

I always do both upper and lower size selection using AMPure XP beads, I don't see a reason not to just do it right away (unless you are unfamiliar with bead cleanup and want to have a way to check you process).
I add 0.5x, keep supernatant, add 1.3x (total 1.8), keep beads, elute. Then check BA.

I attached some BA tracks from a test I did, D and E are how I like my libraries to look and these two sequenced nicely.

I did a left side size selection and got the following band in the Fragment Analyzer. I stored my samples at -20ºC, so now I am going to do the right-side size selection as you describe. I was hoping that just doing the left-side size selection would be enough, because I did a mock experiment using unsorted cells to see how it would work.

Do you think that there's an issue in thawing my samples now and do the right size selection?

Meyana · 12-01-2019, 08:10 PM

Quote:

Originally Posted by Rosmano

I did a left side size selection and got the following band in the Fragment Analyzer. I stored my samples at -20ºC, so now I am going to do the right-side size selection as you describe. I was hoping that just doing the left-side size selection would be enough, because I did a mock experiment using unsorted cells to see how it would work.

Do you think that there's an issue in thawing my samples now and do the right size selection?

I can't see why there should be any problem in doing the upper size selection now, I would just try and see how it sequences.

Hope it works.

Rosmano · 10-12-2020, 10:30 AM

Oh, just an update: it worked beautifully!

I prepared a few new libraries. Could you guys provide feedback and tell me what do you think of these new samples?

https://ibb.co/nbWRn1p
https://ibb.co/RNF2vd4
https://ibb.co/xjKBPth

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Rn/cycles: qPCR decided ATAC-seq library amplification	zhaolin	Sample Prep / Library Generation	9	08-14-2019 02:48 AM
inefficient NEB DNA library prep with plenty of starting material	nsmackler	Sample Prep / Library Generation	16	04-21-2016 01:57 PM
Truseq Exome Library Prep: amplification does not work	user 31888	Illumina/Solexa	2	04-05-2016 02:58 PM
ChIPseq Library Prep Amplification: How many PCR cycles do you use?	JIrish	Epigenetics	4	04-24-2014 12:06 PM
Amplification of ChIP'd DNA before library prep	OptimusBrien	Sample Prep / Library Generation	1	07-12-2012 01:26 PM

08-14-2019, 02:34 AM	#12
Meyana Member Location: Japan Join Date: Sep 2017 Posts: 40	I started out using the PCR mix in the Nextera kit for both PCR and qPCR and got little to no amplification. Switching all steps to NEB (as described in the Omni paper) got everything working smoothly for me and highly reproducible with as low as 500 FACS'ed nuclei. Based on this, I recommend using NEB products for all steps. Also, I do not think it is a good idea to "switch" polymerase mix between quantification and amplification step. Best of luck! Last edited by Meyana; 08-14-2019 at 05:18 PM.

08-20-2019, 02:47 AM	#13
Rosmano Member Location: Portugal Join Date: May 2012 Posts: 21	Yup, I got everything working by using solely the NEBNext master mix. Now, I'm just wondering about performing bead clean-up... I've decided to do a Fragmet analyzer before doing size selection, and depending on the profile I'll decide if I need to do a size selection or not. By the way, does anyone have an example of a Fragment Analyzer profile of a successful ATAC-seq library? Last edited by Rosmano; 08-20-2019 at 03:26 AM.

12-03-2018, 08:41 AM	#2
lapensee Junior Member Location: Michigan Join Date: Oct 2018 Posts: 6	If not suggestions, would someone be willing to post a pic of a GREAT bioanalyzer profile (for ATAC-Seq)?

12-13-2018, 04:38 PM	#4
Meyana Member Location: Japan Join Date: Sep 2017 Posts: 40	I have seen the same thing repeatedly with FACS sorted nuclei isolated from frozen tissue, hoping someone has an idea on how to optimize...

03-20-2019, 10:16 AM	#5
lapensee Junior Member Location: Michigan Join Date: Oct 2018 Posts: 6	FWIW I got this working. I adopted the OMNI-ATAC protocol and get great library amplification from 5k cells using only 2% the amount of Tn5 required for 50k cells. Adding a protease inhibitor during transposition was crucial for my library prep.

04-03-2019, 02:16 AM	#6
Rosmano Member Location: Portugal Join Date: May 2012 Posts: 21	Hey, did you use the new Nextera DNA flex Library Prep kit? I've been having similar problems to you but nothing I do seems to work.

04-03-2019, 04:45 AM	#7
lapensee Junior Member Location: Michigan Join Date: Oct 2018 Posts: 6	I used "Nextera DNA Library Prep Kit" FC-121-1030

08-10-2019, 09:15 AM	#8
Rosmano Member Location: Portugal Join Date: May 2012 Posts: 21	What protease inhibitor did you use in your protocol? I adopted the Omni-ATAC protocol but there's no qPCR amplification, even though I have DNA (7.6ng/uL obtained from 50,000 thymocytes). I don't even know if my pre-amplification succeeded or not.

08-12-2019, 07:10 AM	#9
lapensee Junior Member Location: Michigan Join Date: Oct 2018 Posts: 6	Halt protease inhib without EDTA (100x) from thermofisher

08-12-2019, 09:53 AM	#10
Rosmano Member Location: Portugal Join Date: May 2012 Posts: 21	Well, after analyzing my qPCR results again, what I have to say is that the PCR looks awful but there's something there. Unfortunately, by the end of the 20 cycles I didn't hit a plateau stage and the amplification are of very low amounts. The exponential algorithm fails even. Considering that I tried with the Omni-ATAC buffer with Tween and digitonin, I am considering re-trying only with NP-40 like in the original Buenrostro paper. Does anyone have any other idea of other optimization steps?

08-14-2019, 02:10 AM	#11
Rosmano Member Location: Portugal Join Date: May 2012 Posts: 21	By the way, I am doing the qPCR quantification with Power SYBR Green PCR Master Mix from Thermo Fisher instead of the NEB Next HF Master Mix. Does anyone know if this might impact the qPCR quantification?