SEQanswers

SEQanswers (http://seqanswers.com/forums/index.php)
-   Sample Prep / Library Generation (http://seqanswers.com/forums/forumdisplay.php?f=25)
-   -   ATAC-seq protocol (http://seqanswers.com/forums/showthread.php?t=45636)

wen yuan 08-06-2014 09:27 AM

ATAC-seq protocol
 
Hi everyone,

I am planning to use ATAC-seq with my samples. But I can not find a detailed protocol in their nature method paper. Also there is no response in ATAC forum. I can not get permission. I do not know why. Does anyone have the protocol for ATAC-seq with more detail? Thank you so much.

Jubs 08-08-2014 10:30 AM

1 Attachment(s)
Hi wen yuan,
Attached is what I got from the authors, or someone from their lab. I'm also waiting for permission to their forum, let's see...
Hope it helps :)

wen yuan 08-08-2014 10:37 AM

Hi Jubs,

Thank you so much. Thanks.


Quote:

Originally Posted by Jubs (Post 147130)
Hi wen yuan,
Attached is what I got from the authors, or someone from their lab. I'm also waiting for permission to their forum, let's see...
Hope it helps :)


Wonghe 10-01-2014 07:34 PM

Hi,
I just downloaded the protocol and is wondering if anyone here has tested it using the Nextera DNA or XT kit? Does it make a difference?
Since there is no size selection before sequencing how do is the final concentration being estimated?

obischof 10-22-2014 12:15 PM

Duplicate reads in ATACseq
 
We have run ATACseq on HDFs (100.000 cells) and after paired-end seq on HiSeq2000 (fast run) we get loads of duplicate reads; in fact we loose between 50-90% from the raw reads. Anybody experienced this? Moreover, is paired-end seq really neccessary?

Chacal 12-16-2014 06:07 AM

We have now run 3 separate ATAC-seq experiments on HDFs, at 50000 cells per sample, trying two different methods. We get 50-90% duplicate reads from FASTQC. Additionally, we also see high variability in peak intensity between different samples. Anyone experienced something similar?

nsmackler 12-16-2014 11:05 AM

Quote:

Originally Posted by Wonghe (Post 151218)
Hi,
I just downloaded the protocol and is wondering if anyone here has tested it using the Nextera DNA or XT kit? Does it make a difference?
Since there is no size selection before sequencing how do is the final concentration being estimated?

I've only used the regular Nextera DNA kit (FC-121-1030) and it works just fine. I don't think you should use the XT kit since it is more expensive.

You'll need to estimate the concentration using a qPCR. We use the KAPA quantification kit for Illumina libraries.

We didn't see many duplicates in our libraries, but we did have a lot of reads map to the mitochrondrial genome (anywhere from 15-75%, depending on the cell type). There is a new protocol out there where you skip the initial cell lysis step and that seems to decrease the number of reads mapping to mtDNA.

Chacal 12-16-2014 01:42 PM

mt genome
 
Can you tell me where this protocol is available?

Thanks,

nsmackler 12-16-2014 02:42 PM

DISCLAIMER: The protocol hasn't been vetted, but in our hands it looks like it decreases the number of reads mapping to mtDNA. Moreover, we don't know if skipping the lysis step alters which regions are 'tagmented'. It could lead to bias in the library.

That being said, it is a pretty simple modification. All you do is skip steps I.4 and I.5 in the protocol in this thread. So you wash the cells once with 50ul PBS, spin down, and then go right into the transposition reaction.

Wonghe 12-29-2014 11:05 PM

I have been trying the ATAC-seq protocol but fail to get the nucleosome pattern. Can anyone help?

I collected my fresh cells from the culture plate by scraping them off in PBS where I split them into 20K, 50K and 100K cells per tube for lysis. The cells were resuspended in 50ul lysis buffer before tagmentation at 37C for 1hr. So far when i ran the tagmented DNA on the Bioanalyzer, I either see a DNA smear or very small fragments of 200bp.

nsmackler 12-30-2014 03:44 AM

Quote:

Originally Posted by Wonghe (Post 157236)
I have been trying the ATAC-seq protocol but fail to get the nucleosome pattern. Can anyone help?

I collected my fresh cells from the culture plate by scraping them off in PBS where I split them into 20K, 50K and 100K cells per tube for lysis. The cells were resuspended in 50ul lysis buffer before tagmentation at 37C for 1hr. So far when i ran the tagmented DNA on the Bioanalyzer, I either see a DNA smear or very small fragments of 200bp.

Here are a few things that might help:
1) Did you do PCR amplification after the tagmentation and before running the bioanalyzer? Prior to PCR the amount of DNA will be much too small to see on a bioanalyzer, which is why you need to do 8-16 cycles of PCR.

2) You only need to tagment for 30min at 37C.

Lastly, could you also please post a picture of the two examples of bioanalyzer traces that you see? The "DNA smear" and "very small fragments of 200bp"?

Wonghe 12-30-2014 06:37 PM

1 Attachment(s)
A correction to my previous post, tagmentation was done at 37C for 30min in the below conditions:
1) 20K cells, 10ul Tag reaction vol (1ul enzyme) no clean up before PCR (10cycles) clean up using 1:1 Ampure beads and ran on Bioanalyzer (~200-400bp)

2) 20K cells, 25ul Tag reaction vol (2ul enzyme) clean up in Qiagen MinElute col elute in 10ul and use all the transpose DNA for 10 cycles PCR, ran 10ul of amplified product on a 6% TBE gel (observed smear).

So far no luck in the nucleosome pattern.
Attachment 3499

nsmackler 12-31-2014 04:17 AM

Quote:

Originally Posted by Wonghe (Post 157307)
A correction to my previous post, tagmentation was done at 37C for 30min in the below conditions:
1) 20K cells, 10ul Tag reaction vol (1ul enzyme) no clean up before PCR (10cycles) clean up using 1:1 Ampure beads and ran on Bioanalyzer (~200-400bp)

2) 20K cells, 25ul Tag reaction vol (2ul enzyme) clean up in Qiagen MinElute col elute in 10ul and use all the transpose DNA for 10 cycles PCR, ran 10ul of amplified product on a 6% TBE gel (observed smear).

So far no luck in the nucleosome pattern.
Attachment 3499

Thanks for the update. That bioanalyzer trace looks like a really good truseq library, not so much a good ATAC-seq library. I wonder what is happening in that your DNA is only amplifying in that really small range. Here are some thoughts:

1) What kind of cells did you use? Also - maybe you could try more than 50K cells? And you can skip the lysis step and go straight to the transposition if you want.

2) Smaller fragments will preferentially amplify, so if you have a lot of small fragments, those will take over the library.

Good luck.

whytcs 02-13-2015 07:08 AM

I think maybe it is because you added relatively more enzyme compared to your cell amount, and the open chromatin was well or even over digested, so you got the main peak from bioanalyzer which is around 200-300bp. It also happened to me actually, I also have no clue. However, this is a quite new technique, it is not surprising that funny things happen.

rbd 04-01-2015 10:06 AM

Quote:

Originally Posted by nsmackler (Post 156630)
DISCLAIMER: The protocol hasn't been vetted, but in our hands it looks like it decreases the number of reads mapping to mtDNA. Moreover, we don't know if skipping the lysis step alters which regions are 'tagmented'. It could lead to bias in the library.

That being said, it is a pretty simple modification. All you do is skip steps I.4 and I.5 in the protocol in this thread. So you wash the cells once with 50ul PBS, spin down, and then go right into the transposition reaction.

If this works, does that mean that both the plasma membrane and the nuclear envelope are permeable to the transposase?

rbd 04-01-2015 10:08 AM

Has anybody tried increasing the cell lysis time?

nsmackler 04-01-2015 10:10 AM

Quote:

Originally Posted by rbd (Post 163680)
If this works, does that mean that both the plasma membrane and the nuclear envelope are permeable to the transposase?

It means that enough of the cells are osmotically lysed in the TD buffer.

Wonghe 05-19-2015 01:37 AM

frozen tissue
 
1 Attachment(s)
Hi All, I've been trying to do ATAC-seq on nuclei isolated from frozen tissues with sucrose gradient. Although I am able to obtain a nucleosome-like pattern (see above attachment) but after sequencing I get many background peaks which are very noisy and the peaks at the TSS were not distinct. Anyone has any idea to improve the method and reduce the background peaks?

ksingh 06-21-2015 05:59 AM

I am looking for methods for alternate to CHIP-Seq for Mouse hematopoietic stem cells. I can't get large number of cells for this protocol. I would appreciate help, if any body aware of protocol (reference). Thanks. KPS

Zaag 06-30-2015 05:56 AM

Quote:

Originally Posted by Jubs (Post 147130)
Hi wen yuan,
Attached is what I got from the authors, or someone from their lab. I'm also waiting for permission to their forum, let's see...
Hope it helps :)

I used this protocol and got this after adapter-PCR:

https://dl.dropboxusercontent.com/s/...06-30.png?dl=0
https://dl.dropboxusercontent.com/s/...06-30.png?dl=0


Can anyone comment? Is it completely crap or not?

To me sample 2 looks better than sample 1, but I have no real clue as in how it should look.


All times are GMT -8. The time now is 01:14 AM.

Powered by vBulletin® Version 3.8.9
Copyright ©2000 - 2018, vBulletin Solutions, Inc.