troubles with Radseq library preparation

crom · 08-16-2017, 05:42 PM

Hello all,

I am preparing a library of a selected number of specimens (20 ind) following the ddRad seq protocol (Peterson et al. 2013). I am having difficulties because I start with 500ng of Dna but I loose so much DNA that I end to get very low concentration before the final PCR.
In particular I loose more than 50% of DNA every beads wash or pooling. I using a concentration of 3X beads. Do you know if I am supposed to loose so much?
How many max PCR cycles may I run whitout increasing the PCR errors?

thank you

nucacidhunter · 08-16-2017, 06:35 PM

50% loss is too much; an acceptable loss is 10-15% and 3x bead is also excessive. You are using too much DNA to start with so if you go to PCR (multiple 20 ul reactions) with 2 ng of adapted size selected tags and 5% is amplifiable then up to 12 cycle should produce more than enough library for QC and sequencing.

crom · 08-16-2017, 06:54 PM

Quote:

Originally Posted by nucacidhunter

50% loss is too much; an acceptable loss is 10-15% and 3x bead is also excessive. You are using too much DNA to start with so if you go to PCR (multiple 20 ul reactions) with 2 ng of adapted size selected tags and 5% is amplifiable then up to 12 cycle should produce more than enough library for QC and sequencing.

thank you for answering, I supposed that it was too much loss, but I do not know how to reduce it. Actually I am using more beads (3X) to reduce the loss but it is not working. I beleive it is related to the nature of digested samples because with other type of samples and the same beads I do not get so much loss. Any suggestions? otherwise 12 cycles will be not enough for my samples

nucacidhunter · 08-16-2017, 07:23 PM

If digestable DNA fragments mostly are over 2kb, with standard clean up protocol loss might be higher. The quantity of Pippin size selected DNA is more important than loss and 10-20 ng recovery at the end should be enough. I normally do 12 cycles of PCR but I do not think that 13 cycle would cause any bias.

crom · 08-17-2017, 03:53 PM

After the Pippin site selection I have a concentration of 1.4 ng/ul, do you think I can do two runs of PCR (each of 20 ul and with 13 cycles) and than pool the PCR products 20+20ul?

nucacidhunter · 08-17-2017, 05:34 PM

I Wonder how many ul of 1.4 you have and what was the total quantity of input DNA.

crom · 08-17-2017, 06:10 PM

At the end after Pippin (before PCR) I have 30ul of 1.4 ul solution (for a total 42ng) that is a pool of 4 samples, originally I started with a total of 2300 ng...as you can see I loose so much DNA along the way, is it normal?

nucacidhunter · 08-17-2017, 08:10 PM

It seems OK. Final DNA recovery from Pippin depends on the quantity of fragments in the selected size range which will vary based on input, genome composition and restriction enzymes used among other factors.

I think you can safely set up four 20 ul reactions using 5 ul of size selected tags per reaction and do 12 cycles. After 12 cycle you can check the PCR yield and size by TapeStation if it is available or at least check concentration with Qubit or PicoGreen. If PCR yield is around 1 ug/ul then no extra cycles would be needed. If it is lower then do 1 or 2 more cycles considering that each cycle yield will double.

crom · 08-18-2017, 10:27 AM

Thank you very much nucacidhunter!
just to be clear: do you mean I can do different reactions and than pool the yields?

If the quantity is not enough may I use the PCR product (add again polymerase, NTP and so on) and re-run the PCR for extra run?

thanks again

nucacidhunter · 08-18-2017, 11:37 AM

PCR yield should be enough but it would depend on the amplifiable portion of size selected tags. It is better to do more cycles than re-amplifying PCR product because standard ddRAD does not include UMIs for identifying PCR duplicates and also it will affect allele frequency. 50 ng final library should be enough for QC and sequencing.

Edit: For extra 1-2 cycles there is no need for reagent addition. Just keep the reaction at 4 C while quantifying yield.

cbird · 12-05-2017, 10:41 PM

some suggestions:

Minimize tube transfers

Use the with bead method to reuse beads https://genomebiology.biomedcentral....b-2011-12-1-r1

don't throw out supernatant, then you can see if that's where your DNA is going by running it on a gel or fragment analyzer

if you have to do a transfer, elute the DNA from beads twice with smaller volume rather than once with larger volume

good luck

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
NGS library prep method for RRS/GBS/RADseq of degraded/ancient DNA?	BioGenomics	Sample Prep / Library Generation	6	01-14-2015 09:46 PM
RADseq library preparation restriction enzymes	LoeraI	General	1	02-21-2013 06:42 AM
Library Preparation Available	CarlHSLee	Sample Prep / Library Generation	0	01-16-2013 12:25 PM
Scriptminer Small RNA-seq library troubles	Helleaks	Sample Prep / Library Generation	1	11-27-2012 09:00 AM

08-16-2017, 05:42 PM	#1
crom Junior Member Location: san diego Join Date: Jan 2016 Posts: 5	troubles with Radseq library preparation Hello all, I am preparing a library of a selected number of specimens (20 ind) following the ddRad seq protocol (Peterson et al. 2013). I am having difficulties because I start with 500ng of Dna but I loose so much DNA that I end to get very low concentration before the final PCR. In particular I loose more than 50% of DNA every beads wash or pooling. I using a concentration of 3X beads. Do you know if I am supposed to loose so much? How many max PCR cycles may I run whitout increasing the PCR errors? thank you

08-18-2017, 11:37 AM	#10
nucacidhunter Jafar Jabbari Location: Melbourne Join Date: Jan 2013 Posts: 1,238	PCR yield should be enough but it would depend on the amplifiable portion of size selected tags. It is better to do more cycles than re-amplifying PCR product because standard ddRAD does not include UMIs for identifying PCR duplicates and also it will affect allele frequency. 50 ng final library should be enough for QC and sequencing. Edit: For extra 1-2 cycles there is no need for reagent addition. Just keep the reaction at 4 C while quantifying yield. Last edited by nucacidhunter; 08-18-2017 at 11:41 AM.

08-16-2017, 06:35 PM	#2
nucacidhunter Jafar Jabbari Location: Melbourne Join Date: Jan 2013 Posts: 1,238	50% loss is too much; an acceptable loss is 10-15% and 3x bead is also excessive. You are using too much DNA to start with so if you go to PCR (multiple 20 ul reactions) with 2 ng of adapted size selected tags and 5% is amplifiable then up to 12 cycle should produce more than enough library for QC and sequencing.

08-16-2017, 07:23 PM	#4
nucacidhunter Jafar Jabbari Location: Melbourne Join Date: Jan 2013 Posts: 1,238	If digestable DNA fragments mostly are over 2kb, with standard clean up protocol loss might be higher. The quantity of Pippin size selected DNA is more important than loss and 10-20 ng recovery at the end should be enough. I normally do 12 cycles of PCR but I do not think that 13 cycle would cause any bias.

08-17-2017, 03:53 PM	#5
crom Junior Member Location: san diego Join Date: Jan 2016 Posts: 5	After the Pippin site selection I have a concentration of 1.4 ng/ul, do you think I can do two runs of PCR (each of 20 ul and with 13 cycles) and than pool the PCR products 20+20ul?

08-17-2017, 05:34 PM	#6
nucacidhunter Jafar Jabbari Location: Melbourne Join Date: Jan 2013 Posts: 1,238	I Wonder how many ul of 1.4 you have and what was the total quantity of input DNA.

08-17-2017, 06:10 PM	#7
crom Junior Member Location: san diego Join Date: Jan 2016 Posts: 5	At the end after Pippin (before PCR) I have 30ul of 1.4 ul solution (for a total 42ng) that is a pool of 4 samples, originally I started with a total of 2300 ng...as you can see I loose so much DNA along the way, is it normal?

08-17-2017, 08:10 PM	#8
nucacidhunter Jafar Jabbari Location: Melbourne Join Date: Jan 2013 Posts: 1,238	It seems OK. Final DNA recovery from Pippin depends on the quantity of fragments in the selected size range which will vary based on input, genome composition and restriction enzymes used among other factors. I think you can safely set up four 20 ul reactions using 5 ul of size selected tags per reaction and do 12 cycles. After 12 cycle you can check the PCR yield and size by TapeStation if it is available or at least check concentration with Qubit or PicoGreen. If PCR yield is around 1 ug/ul then no extra cycles would be needed. If it is lower then do 1 or 2 more cycles considering that each cycle yield will double.

08-18-2017, 10:27 AM	#9
crom Junior Member Location: san diego Join Date: Jan 2016 Posts: 5	Thank you very much nucacidhunter! just to be clear: do you mean I can do different reactions and than pool the yields? If the quantity is not enough may I use the PCR product (add again polymerase, NTP and so on) and re-run the PCR for extra run? thanks again

12-05-2017, 10:41 PM	#11
cbird Member Location: Corpus Christi, TX Join Date: May 2012 Posts: 10	some suggestions: Minimize tube transfers Use the with bead method to reuse beads https://genomebiology.biomedcentral....b-2011-12-1-r1 don't throw out supernatant, then you can see if that's where your DNA is going by running it on a gel or fragment analyzer if you have to do a transfer, elute the DNA from beads twice with smaller volume rather than once with larger volume good luck