SEQanswers

Go Back   SEQanswers > Applications Forums > Sample Prep / Library Generation



Similar Threads
Thread Thread Starter Forum Replies Last Post
ATAC-seq protocol wen yuan Sample Prep / Library Generation 65 08-11-2017 07:52 PM
On-Plate ATAC-seq krutipedali Sample Prep / Library Generation 2 02-28-2017 05:16 AM
ATAC-seq primers Fasteno Sample Prep / Library Generation 7 01-20-2017 04:05 PM
ATAC-seq QC using gel Fasteno Sample Prep / Library Generation 0 03-12-2015 05:13 AM
unable to use Bio::Seq in perl bio_informatics Bioinformatics 5 11-28-2013 12:57 PM

Reply
 
Thread Tools
Old 03-14-2017, 08:45 AM   #1
halllb
Junior Member
 
Location: tampa, fl

Join Date: Mar 2017
Posts: 4
Unhappy Unable to get ATAC-seq to work

good morning. for the past several days, i have been attempting to perform ATAC-seq on some Jurkat cell pellets. i am following the Buenrostro/Greenleaf protocol published in Current Protocols (2016). to-date, i cannot get the tagmentation reaction to work. i have tried several different conditions; e.g., 1) running a titration experiment using 7.4K, 14.8K, 29.5K, 44.3K, 59K, and 88.5K live cells from a cell pellet assuming 59% live cells; 2) eliminating the lysis step, which i read in one of the threads on this site, using 14.8K, 44.3K, and 59K live cells. each time, i've isolated the nuclei and immediately performed the tagmentation reaction. after cleaning up the reaction, the BA traces show only one large (10,380bp) peak, indicating no fragmentation and hence, no tagging. to test the enzyme, i used commercial DNA. the tagmentation reaction didn't work, so i called Illumina. they suggested running the DNA through a column since some of the commercial gDNA preps contain EDTA to which the transposase is very sensitive. once i cleaned up the gDNA, the fragmentation reaction worked well. i was going to put the isolated nuclei over a clean-up column, but my boss informed me that it would relax most of the supercoiled DNA, resulting in fragmentation of all the gDNA, instead of just the accessible chromatin sites. does anyone have any suggestions that i can try? i have read on this site that many of you are getting very good results using the Greenfield protocol. i welcome any advice you can offer. thanks so much for your time. laura hall
halllb is offline   Reply With Quote
Old 03-14-2017, 09:52 AM   #2
NickPantelireis
Junior Member
 
Location: London

Join Date: Mar 2017
Posts: 6
Default

I read somewhere in one of the other posts on here that after the tagmentation you still need to amplify your DNA using PCR if you haven't already. The amount of DNA after tagmentation is quite a low amount so might not be being detected on the BA. The 10,380bp peak might just be the upper marker of the HS DNA assay.
NickPantelireis is offline   Reply With Quote
Old 03-14-2017, 10:26 AM   #3
halllb
Junior Member
 
Location: tampa, fl

Join Date: Mar 2017
Posts: 4
Default

hi nick: thanks for replying. I also read that. I'm not sure the large peak is marker, because it's not a sharp peak...it's a rounded one. also, I can see heavy bands that look like DNA in the gel image. below is a link to a pic of the BA... what do you think?
laura
http://imgur.com/a/LCU0T
halllb is offline   Reply With Quote
Old 03-15-2017, 03:17 AM   #4
NickPantelireis
Junior Member
 
Location: London

Join Date: Mar 2017
Posts: 6
Default

Ah yeah, I see what you mean. It does look like the DNA is underdigested or as you said not digested at all. With PCR you should be able to see if your primers attach to your DNA. If not then the enzyme is definatly not tagmenting them as the primers can't attach to their complementary sequence. Otherwise, are you washing, resuspending cells, incubating @37C during the reaction?
NickPantelireis is offline   Reply With Quote
Old 03-15-2017, 12:26 PM   #5
halllb
Junior Member
 
Location: tampa, fl

Join Date: Mar 2017
Posts: 4
Default

hi nick:

thanks for the reply.

regarding doing PCR, i haven't run these current samples through PCR. on a previous attempt, however, i ran the entire protocol, all the way through sequencing, with some samples that appeared to work. unfortunately, upon sequencing, we discovered that we were getting over-tagmentation and essentially performing whole-genome sequencing. a troubleshooting section in the Current Protocols chapter suggests that tagmentation needs to be optimized to cell number. hence, the current experiments with titrating cell numbers; but no fragmentation.

regarding washing, resuspending the cells, and incubating @37C during the reaction, yes i have performed all these steps with strict adherance to the protocol.

i appreciate your suggestions and questions, and look forward to hearing any additional suggestions you may have.

thanks,
laura
halllb is offline   Reply With Quote
Old 03-29-2017, 09:28 AM   #6
kropak1807
Junior Member
 
Location: Tallahassee, FL

Join Date: Jan 2017
Posts: 4
Default

Quote:
Originally Posted by halllb View Post
hi nick:

thanks for the reply.

regarding doing PCR, i haven't run these current samples through PCR. on a previous attempt, however, i ran the entire protocol, all the way through sequencing, with some samples that appeared to work. unfortunately, upon sequencing, we discovered that we were getting over-tagmentation and essentially performing whole-genome sequencing. a troubleshooting section in the Current Protocols chapter suggests that tagmentation needs to be optimized to cell number. hence, the current experiments with titrating cell numbers; but no fragmentation.

regarding washing, resuspending the cells, and incubating @37C during the reaction, yes i have performed all these steps with strict adherance to the protocol.

i appreciate your suggestions and questions, and look forward to hearing any additional suggestions you may have.

thanks,
laura
Make sure you amplify your tagmented DNA for the correct number of cycles (ie. 1/4 max fluorescence; correct number of cycles for any dilution factor). Run a sample of this Post-PCR product on BA and show us what you get.
kropak1807 is offline   Reply With Quote
Old 03-29-2017, 12:00 PM   #7
achouak
Junior Member
 
Location: baltimore

Join Date: Mar 2017
Posts: 2
Default

hi,
i am doing atac seq also for fresh cells from mouse.
when i do it i have some samples are good but the bad ones i try many things but i can not succeed. i can't see bands after pcr.
wouyld you please help me and tell me if you have any suggestions.
Thank you
achouak is offline   Reply With Quote
Old 03-29-2017, 12:10 PM   #8
halllb
Junior Member
 
Location: tampa, fl

Join Date: Mar 2017
Posts: 4
Default

hi: i did finally amplify the tagmented DNA for 10 cycles. i put it on a bioanalyzer, and once i decreased the scale of the chart, i saw the undulating pattern of tagmented nucleosomes. thank you.
03-17-17 - 100K total cells+lysis.png
halllb is offline   Reply With Quote
Old 04-05-2017, 06:51 PM   #9
lilbert5
Junior Member
 
Location: new york

Join Date: Feb 2017
Posts: 1
Default question about ATAC library bioanalyzer

Hi all,

ATAC_bioanalyzer_SUDHL4_20170405.pdf

I am new to ATAC-seq, and am testing a few different conditions based on the original protocol and a newer protocol from the greenleaf lab that was designed for blood cells (fast-ATAC). I made 4 libraries, using either 10k or 100k cells for each method. I used a 50uL reaction, using 2.5 uL of enzyme and incubating at 37C for 30 minutes. After PCR (with cycle number titration) I ran the libraries on an agarose gel and size selected the fragments between 150 and 800 bps (a post-doc in the neighboring lab has done this and has gotten good data from this).

I have attached the BA trace for 100k and 10k using the original method (IGEPAL) and the new method (fast-ATAC). The post-doc has told me that it's best not to sequence any of these because the 3n nucleosomal fraction is much higher than the sub-nucleosomal fraction (~190 bps), however I am inclined to think that the "10k fast" library would give me good data.

Does anyone have experience with taking ATAC from library prep all the way to data analysis and can comment on whether any of these four libraries is worthy of sequencing?

Thanks!
lilbert5 is offline   Reply With Quote
Old 06-13-2017, 03:59 PM   #10
gpmc
Junior Member
 
Location: California

Join Date: Jun 2017
Posts: 4
Default ATAC-Seq Guidance

Hi all,

I am hoping that this is still an active thread!

For the past week I have been attempting to generate ATAC-Seq data following the standard Greenleaf protocol. Briefly, this has involved:

-Lysing ~50,000 HEK293T cells using NP-40
-Adding 50 uL of Nextera tagemenation reaction mixture (containing 2.5 uL of Enzyme)
-Incubating at 37 C for 30 min
-Dynabead clean-up of sample
-PCR (limited to 5 cycles) to attach sample sequencing adaptors and sample indices
-qPCR to determine number of remaining PCR cycles
-Final round of PCR
-1.8x SPRI

However, my BioAnalyzer traces consistently don't reflect any nucleosome patterns and there is always a large amount of mass at ~2,000 bp (see attached). Can anybody provide guidance on what is going wrong here?
Attached Images
File Type: png BioA.PNG (46.5 KB, 26 views)
gpmc is offline   Reply With Quote
Old 06-14-2017, 02:25 AM   #11
NickPantelireis
Junior Member
 
Location: London

Join Date: Mar 2017
Posts: 6
Default

Hey,

I'm by no means an expert in this but assuming this trace is done at the end of your protocol it looks like the DNA is underdigested. From the published protocol it could be down to cell number so maybe try the exact protocol with fewer cells (maybe half). Also you could look into using the FAST-ATAC protocol, it cuts out the lysis buffer and adds digitonin to the transposition reaction.

I'd also do a bioanalyser before and after SPRI beads to see the effect they are having on the sample.

Best,
Nick


Quote:
Originally Posted by gpmc View Post
Hi all,

I am hoping that this is still an active thread!

For the past week I have been attempting to generate ATAC-Seq data following the standard Greenleaf protocol. Briefly, this has involved:

-Lysing ~50,000 HEK293T cells using NP-40
-Adding 50 uL of Nextera tagemenation reaction mixture (containing 2.5 uL of Enzyme)
-Incubating at 37 C for 30 min
-Dynabead clean-up of sample
-PCR (limited to 5 cycles) to attach sample sequencing adaptors and sample indices
-qPCR to determine number of remaining PCR cycles
-Final round of PCR
-1.8x SPRI

However, my BioAnalyzer traces consistently don't reflect any nucleosome patterns and there is always a large amount of mass at ~2,000 bp (see attached). Can anybody provide guidance on what is going wrong here?
NickPantelireis is offline   Reply With Quote
Old 06-14-2017, 05:55 PM   #12
gpmc
Junior Member
 
Location: California

Join Date: Jun 2017
Posts: 4
Default

Thanks for the reply. I repeated the experiment following the Fast-ATAC approach and two different cell concentrations (50,000 anc 25,000) as you recommended. Briefly:
Cells were pipetted into a strip tube
Tranposase reaction mixture containing 0.1% NP-40 was added
Reaction incubated at 37 C for 30 min
Reaction cleaned-up using 1.8x SPRI
11 cycles of PCR
Reaction cleaned-up using 1.8x SPRI
Libraries run on the BioAnalyzer

This time there is some nucleosome patterning but still a large mass > 1,000 kb. Perhaps I will just try double-sided SPRI and sequence these anyway? Do you have any further thoughts?
Attached Images
File Type: png ATAC-2.PNG (45.8 KB, 28 views)
gpmc is offline   Reply With Quote
Old 06-16-2017, 06:13 AM   #13
NickPantelireis
Junior Member
 
Location: London

Join Date: Mar 2017
Posts: 6
Default

Quote:
Originally Posted by gpmc View Post
Thanks for the reply. I repeated the experiment following the Fast-ATAC approach and two different cell concentrations (50,000 anc 25,000) as you recommended. Briefly:
Cells were pipetted into a strip tube
Tranposase reaction mixture containing 0.1% NP-40 was added
Reaction incubated at 37 C for 30 min
Reaction cleaned-up using 1.8x SPRI
11 cycles of PCR
Reaction cleaned-up using 1.8x SPRI
Libraries run on the BioAnalyzer

This time there is some nucleosome patterning but still a large mass > 1,000 kb. Perhaps I will just try double-sided SPRI and sequence these anyway? Do you have any further thoughts?

It's looking better Is there any reason you're using NP-40 and not Digititonin or even Igepal during this step? I read that NP-40 is very powerful and lyses all membranes in the cell, while the others are milder and lyse the cytoplasm keeping the nucleus intact. This might be effecting the chromatin structure of effectiveness of the transposase.

Also I'm not sure if it would have a big effect but using the SPRI beads after the transposition might be affecting things also as I believe it prefers larger fragments. Although I'm not certain about that.
NickPantelireis is offline   Reply With Quote
Old 06-18-2017, 10:09 AM   #14
gpmc
Junior Member
 
Location: California

Join Date: Jun 2017
Posts: 4
Default

Thanks for the reply.

Not using Digitonin because it wasn't available. However, have since ordered and will screen alongside other lysis agents to see if there is a difference.

The bead clean-up after tagmentation is silica based, not SPRI. However, a 1.8x SPRI should work just fine.

Thanks for the suggestions.
gpmc is offline   Reply With Quote
Old 07-26-2017, 06:11 AM   #15
gemomichi
Junior Member
 
Location: Austria

Join Date: Jul 2017
Posts: 1
Default

https://www.dropbox.com/s/oebuvocxkj...michi.png?dl=0

Hi everyone,

Did you resolve the issue with the big fragments peak?
I am new to ATACseq and having the same problem, and I am using the FAST-ATAC protocol with lysis and tagmentation (1% Digitonin 1:100) in one step, stopping the reaction by MinElute Cleanup.

The BA profiles are after double SPRI cleanup (0.5/1.4) - so I guess one issue is the cleanup, however I wonder if there's additionally any possible issues with the tagmentation?

Any suggestions/ideas would be welcome.
gemomichi is offline   Reply With Quote
Old 07-26-2017, 08:53 AM   #16
gpmc
Junior Member
 
Location: California

Join Date: Jun 2017
Posts: 4
Default

Hi,

I have been pretty consistently generating the same looking BioA traces using the FAST-ATAC protocol. Typically, double-sided SPRI removes a significant portion of the larger peak. Sequencing these libraries yields similar metrics to what has previously appeared in the literature. Unsure what that large peak is though? Perhaps it is mito-gDNA?
gpmc 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 09:49 PM.


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