SEQanswers

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



Similar Threads
Thread Thread Starter Forum Replies Last Post
Truseq cDNA libraries show sharp peak of cDNA BUFtoBOS Sample Prep / Library Generation 2 09-23-2014 08:00 AM
Lexogen has launched TeloPrime Full-Length cDNA Amplification Kit lexogen Vendor Forum 0 09-01-2014 11:36 PM
Variance Estimation KellerMac Bioinformatics 13 02-18-2013 01:07 AM
Amplification of total RNA or cDNA quantities fearnoweevil Sample Prep / Library Generation 0 02-14-2013 08:45 AM

Reply
 
Thread Tools
Old 04-11-2017, 09:23 AM   #1
PrimerBuffalo
Junior Member
 
Location: Chicago

Join Date: Mar 2017
Posts: 6
Default Variance in cDNA amplification

Hello,

I recently generated cDNA from 100 T-cells (from the same source) using the Smart-Seq2 protocol from Sandberg et al. After the pre-amplification step (all wells were pre-amplified for 18 cycles), I ran 1uL of my cDNA on a Bioanalyzer and noticed that the amount of cDNA that I recovered was relatively variable (range from ~450pg/uL to ~1,400pg/uL).

Does anyone know how much variance in cDNA concentration is expected well-to-well? Is this variance biological or an artifact of the protocol?

I would like to get an idea I am worried that the variance that I observe is due to technical error on my part, perhaps in pipetting.

Thanks in advance for the help!
PrimerBuffalo is offline   Reply With Quote
Old 04-12-2017, 02:08 AM   #2
Simone78
Senior Member
 
Location: Bonn (Germany)

Join Date: Oct 2010
Posts: 185
Default

Quote:
Originally Posted by PrimerBuffalo View Post
Hello,

I recently generated cDNA from 100 T-cells (from the same source) using the Smart-Seq2 protocol from Sandberg et al. After the pre-amplification step (all wells were pre-amplified for 18 cycles), I ran 1uL of my cDNA on a Bioanalyzer and noticed that the amount of cDNA that I recovered was relatively variable (range from ~450pg/uL to ~1,400pg/uL).

Does anyone know how much variance in cDNA concentration is expected well-to-well? Is this variance biological or an artifact of the protocol?

I would like to get an idea I am worried that the variance that I observe is due to technical error on my part, perhaps in pipetting.

Thanks in advance for the help!
Hi,
the variation is prefectly within the range we normally see when running SS2. I believe the difference is partly biological (cells have different sizes and hence different amount of RNA, are in different parts of the cell cycle, etc) and partly due to an artifact (PCR bias). If you BA profile looks fine (nice peak around 1.5-2 kb, no or little primer dimers, low "background") I wouldnīt worry!
/Simone
Simone78 is offline   Reply With Quote
Old 04-14-2017, 09:15 AM   #3
PrimerBuffalo
Junior Member
 
Location: Chicago

Join Date: Mar 2017
Posts: 6
Default

Thank you so much! That is very reassuring!
PrimerBuffalo is offline   Reply With Quote
Old 07-20-2017, 08:36 AM   #4
PrimerBuffalo
Junior Member
 
Location: Chicago

Join Date: Mar 2017
Posts: 6
Default

Quote:
Originally Posted by Simone78 View Post
Hi,
the variation is prefectly within the range we normally see when running SS2. I believe the difference is partly biological (cells have different sizes and hence different amount of RNA, are in different parts of the cell cycle, etc) and partly due to an artifact (PCR bias). If you BA profile looks fine (nice peak around 1.5-2 kb, no or little primer dimers, low "background") I wouldnīt worry!
/Simone
Hi Simone78,

I am currently generating indexed libraries from my samples and have run into a problem:

I used 1ng of cDNA and ran the indexing PCR for 7 cycles. I did this because I thought using more starting cDNA would result in more product, but that seems not the case.... By bioanalyzer I ended up with too low a concentration of indexed cDNA (200-1000 pMolar) and my avg peak size was 800-1000 bp.

So I feel that I must lower the amount of starting cDNA, and increase the number of cycles.
In your original protocol you use 0.1 ng cDNA for 12 cycles. My question is why do you use so little cDNA when you are probably generating much more during the preamplification step. Why not use 0.5 ng etc?


Thanks for your help!
PrimerBuffalo is offline   Reply With Quote
Old 07-21-2017, 01:46 AM   #5
Simone78
Senior Member
 
Location: Bonn (Germany)

Join Date: Oct 2010
Posts: 185
Default

Quote:
Originally Posted by PrimerBuffalo View Post
Hi Simone78,

I am currently generating indexed libraries from my samples and have run into a problem:

I used 1ng of cDNA and ran the indexing PCR for 7 cycles. I did this because I thought using more starting cDNA would result in more product, but that seems not the case.... By bioanalyzer I ended up with too low a concentration of indexed cDNA (200-1000 pMolar) and my avg peak size was 800-1000 bp.

So I feel that I must lower the amount of starting cDNA, and increase the number of cycles.
In your original protocol you use 0.1 ng cDNA for 12 cycles. My question is why do you use so little cDNA when you are probably generating much more during the preamplification step. Why not use 0.5 ng etc?


Thanks for your help!
Hi,
yes, probably 1 ng input is too much for the Tn5 you add, especially when using the Nextera XT kit. This leads to under-tagmentation and libraries with long(er) average size. Moreover, I would use 10 PCR cycles when starting from 1 ng.
You are also correct when you say that we should generate more cDNA than what we actually use. The reason why I start from 100 pg is just convenience. As shown on Fig 5D of our Genome Res paper, there is no difference in gene detection and SD when starting from 1 ng, 500 pg or 100 pg. By using less input I can therefore use less Tn5 (usually 0.1 ul of a 12.5 uM solution is sufficient) and make my batch last longer!
I think that, at least when using Nextera XT, you should stay in the range of 100-500 pg. When I use the kit I generally do the reaction in 2 ul and use only 50-150 pg DNA and 0.5 ul Tn5. I hope this clarifies things, let me know otherwise.
Best,
Simone
Simone78 is offline   Reply With Quote
Old 07-21-2017, 07:36 AM   #6
PrimerBuffalo
Junior Member
 
Location: Chicago

Join Date: Mar 2017
Posts: 6
Default

Quote:
Originally Posted by Simone78 View Post
Hi,
yes, probably 1 ng input is too much for the Tn5 you add, especially when using the Nextera XT kit. This leads to under-tagmentation and libraries with long(er) average size. Moreover, I would use 10 PCR cycles when starting from 1 ng.
You are also correct when you say that we should generate more cDNA than what we actually use. The reason why I start from 100 pg is just convenience. As shown on Fig 5D of our Genome Res paper, there is no difference in gene detection and SD when starting from 1 ng, 500 pg or 100 pg. By using less input I can therefore use less Tn5 (usually 0.1 ul of a 12.5 uM solution is sufficient) and make my batch last longer!
I think that, at least when using Nextera XT, you should stay in the range of 100-500 pg. When I use the kit I generally do the reaction in 2 ul and use only 50-150 pg DNA and 0.5 ul Tn5. I hope this clarifies things, let me know otherwise.
Best,
Simone
Yes, that is incredibly helpful thank you!

I will definitely reduce my input concentration and increase my number of cycles. I have a colleague who uses 12 cycles for an input of 0.5 ng and I noticed that you also use 12 cycles but for 0.1ng. Is there much difference in the number of cycles past 10?
PrimerBuffalo is offline   Reply With Quote
Old 07-21-2017, 08:14 AM   #7
Simone78
Senior Member
 
Location: Bonn (Germany)

Join Date: Oct 2010
Posts: 185
Default

Quote:
Originally Posted by PrimerBuffalo View Post
Yes, that is incredibly helpful thank you!

I will definitely reduce my input concentration and increase my number of cycles. I have a colleague who uses 12 cycles for an input of 0.5 ng and I noticed that you also use 12 cycles but for 0.1ng. Is there much difference in the number of cycles past 10?
Hi,
I donīt think it really matters if you do 10 or 12 cycles. The largest bias, where you "lost" your transcripts (due to many reasons that would take too long to explain here) is anyway introduced after the 18, 20 or more cycles you do in the first PCR (after RT). After the tagmentation the library is more homogeneous in terms of size, which should decrease the risk of preferential amplification of the short(er) fragments. Sensitivity and reproducibility is more a matter of input used in the tagmentation. If you use too little DNA you might end up with a sub-population of transcripts that is not representative of your real sample and which has low complexity.
Best,
Simone
Simone78 is offline   Reply With Quote
Reply

Tags
cdna synthesis, smart-seq2, technical, variance

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 12:16 PM.


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