Library Generation

smadbro · 03-22-2012, 12:11 PM

High guys I'm writing a discussion on the importance of obtaining high-molecular-weight DNA for fragment library generation before sequencing. I'm finding it hard to find any conclusive literature on this as it seems it is just common knowledge, and i was wondering if anybody had any good links to specific papers?
Also, i've been trying to find if there's a 'washing' step before the fragments get the specific adaptor sequences ligated to them to remove low-quality DNA from the sample, or even to remove fragments slightly too small for the sequencing platform?
Any help on this subject would be greatly appreciated, as it seems I have spent a large portion of the last week searching endlessly on the internet for references to make my discussion at least more than 'you need high-quality DNA cause if you dont have it you can create libraries and then cant sequence'.

Thanks!!!

smadbro · 03-22-2012, 12:11 PM

High guys? *Hi
As you can see my brain is destroyed from too many hours staring at a screen!

smadbro · 03-22-2012, 03:59 PM

also was supposed to be *can't in the last line

Thanks!

shawpa · 03-26-2012, 11:16 AM

In the Illumina TruSeq sample protocol it says that you need high quality DNA. Not sure if you count that as reference. It is the manufacturers that generally say things like that. I routinely make libraries from what would be considered "low quality". You maybe don't get as great of sequencing quality but you always get something.

Have you taken a look at a library prep protocol (like illumina TruSeq)? That should answer your questions regarding the washing. Any size fragment will amplify, as long as it gets ligated but usually the preparations steps take out size below 100bp through bead washing.

pmiguel · 03-28-2012, 06:21 AM

We started doing an Ampure clean up after sonication after seeing this recommended in another thread. Should get rid of very short DNA and RNA molecules. Also provides an additional purification step prior to the first enzymatic steps in the TruSeq protocol.

Most genomic DNA preps are actually degraded RNA preps with a little genomic DNA mixed in for good measure.

As far as degraded DNA goes -- as long as it is not degraded below the insert size you want for your library, the only effect the degradation would have, in and of itself, is any site specificity conferred by the agent degrading the DNA will result in a sequencing bias. However, to the extent that degradation might be an indication of general mistreatment of the DNA -- well it might suggest other problems, like DNA damage that will hinder the production of viable (amplifiable) library molecules. That is, if you have lots of pyrimidine dimers then the chances a polymerase is going to be able to replicate the strand are low.

--
Phillip

NextGenSeq · 03-29-2012, 02:30 PM

I think you just need clean DNA. It doesn't matter if it's high molecular weight. You can make RNA-Seq libraries from paraffin embedded RNA which is about 100 bp.

Simone78 · 03-30-2012, 08:53 AM

I wouldn´t care about removing small fragments before adaptor ligation. Adaptors are usually added in such an excess (10 times more than your DNA) that this shouldn´t be a problem. It´s rather more important to get completely rid of them after ligation.

pmiguel · 03-30-2012, 09:50 AM

Quote:

Originally Posted by Simone78

I wouldn´t care about removing small fragments before adaptor ligation. Adaptors are usually added in such an excess (10 times more than your DNA) that this shouldn´t be a problem. It´s rather more important to get completely rid of them after ligation.

Oh, they are? Have you run those calculations on a molar level? What if 10% of your DNA by mass is in the form of fragments 20 bp or less? Sure the first Ampure will remove most of it, but will your end repair enzymes be capable of repairing all those end?

--
Phillip

Simone78 · 03-30-2012, 10:09 AM

Quote:

Originally Posted by pmiguel

Oh, they are? Have you run those calculations on a molar level? What if 10% of your DNA by mass is in the form of fragments 20 bp or less? Sure the first Ampure will remove most of it, but will your end repair enzymes be capable of repairing all those end?

--
Phillip

I made libraries for methylome seq, therefore I started from gDNA (PE 100 bp seq). Running a Bioanalyzer DNA chip after Covaris shearing shows a wide distribution of fragments but with a tail towards long rather than short fragments (with my Covaris settings, of course!). I believe the short fragments shouldn´t be a problem. At least they were not a problem in our case (but we started from 3-5 ug of gDNA). The 10 times molar ratio is just what reported by Illumina when using their methylated adapters.

anurag.gautam · 03-30-2012, 09:25 PM

Hi ,
CAn anybody tell me what should be the appropriate read length for chip seq analyses.
What should be it for Illumina and SOLiD??
Anurag

SarahNGS · 04-02-2012, 05:11 AM

Quote:

Originally Posted by NextGenSeq

I think you just need clean DNA. It doesn't matter if it's high molecular weight. You can make RNA-Seq libraries from paraffin embedded RNA which is about 100 bp.

ditto, its about clean DNA, which most DNA clean-up kits wil give you.

hawaii454-0 · 04-12-2012, 06:42 AM

Has anybody tried making a TruSeq library from DNA at low concentration and around 150bp? I used different ratios of ampure beads in the clean up steps to make sure the smaller fragments aren't lost in the supernatants. Has anyone made any adjustments to the protocol for cases like this?

shawpa · 04-12-2012, 06:50 AM

Quote:

Originally Posted by hawaii454-0

Has anybody tried making a TruSeq library from DNA at low concentration and around 150bp? I used different ratios of ampure beads in the clean up steps to make sure the smaller fragments aren't lost in the supernatants. Has anyone made any adjustments to the protocol for cases like this?

I do this regularly. I make no adjustments except for skipping the gel purification step and upping the pcr cycles in the end. A lot of people will say that you need to adjust the adaptor concentration to make it as close to 10:1 as possible and reduce adaptor dimers. I don't do this at all. Honestly I get some adaptor dimers (will see a small peak at 126bp), but it's not enough to make a difference for me. I've tried diluting the adaptors but still got mixed results. I actually got more adaptor dimers. It's a guess any way on dilution cause you aren't given the concentration. Recently illumina added a library prep without gel purification protocol. I still use the original.

hawaii454-0 · 04-12-2012, 07:03 AM

I have managed to get some okay-looking TruSeq libraries from some tests on E coli which were sheared at around 150bp and the starting material was 50ng, 25ng, and 10ng. The 50ng library (obviously) looks the best. The adjustments I made to the protocol were in the ratios of ampure beads to sample in the clean ups through the protocol. Prior to the adapters being added, the ratio was 1.6:1 and afterwards it was 1.2:1. This seemed to work well for the size of the fragments I sheared. The next step will be checking the Agilent concentration with the ones on the Qubit and after qPCR.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
RRBS library generation-meth dCTP?	shawpa	Sample Prep / Library Generation	1	03-22-2012 06:00 AM
ChIP-Seq problems with library generation	elly	Sample Prep / Library Generation	18	08-02-2011 10:21 AM
PubMed: Large scale library generation for high throughput sequencing.	Newsbot!	Literature Watch	0	05-19-2011 11:30 PM
Repeat Sequence Library Generation	edge	Sample Prep / Library Generation	3	10-12-2010 05:01 PM
Quantify Library before Cluster Generation	kwebb	Illumina/Solexa	7	06-22-2009 05:01 PM

03-22-2012, 12:11 PM	#1
smadbro Junior Member Location: uk Join Date: Mar 2012 Posts: 4	Library Generation High guys I'm writing a discussion on the importance of obtaining high-molecular-weight DNA for fragment library generation before sequencing. I'm finding it hard to find any conclusive literature on this as it seems it is just common knowledge, and i was wondering if anybody had any good links to specific papers? Also, i've been trying to find if there's a 'washing' step before the fragments get the specific adaptor sequences ligated to them to remove low-quality DNA from the sample, or even to remove fragments slightly too small for the sequencing platform? Any help on this subject would be greatly appreciated, as it seems I have spent a large portion of the last week searching endlessly on the internet for references to make my discussion at least more than 'you need high-quality DNA cause if you dont have it you can create libraries and then cant sequence'. Thanks!!!

03-22-2012, 03:59 PM	#3
smadbro Junior Member Location: uk Join Date: Mar 2012 Posts: 4	also was supposed to be *can't in the last line Thanks!

03-26-2012, 11:16 AM	#4
shawpa Member Location: Pittsburgh Join Date: Aug 2011 Posts: 72	In the Illumina TruSeq sample protocol it says that you need high quality DNA. Not sure if you count that as reference. It is the manufacturers that generally say things like that. I routinely make libraries from what would be considered "low quality". You maybe don't get as great of sequencing quality but you always get something. Have you taken a look at a library prep protocol (like illumina TruSeq)? That should answer your questions regarding the washing. Any size fragment will amplify, as long as it gets ligated but usually the preparations steps take out size below 100bp through bead washing.

03-28-2012, 06:21 AM	#5
pmiguel Senior Member Location: Purdue University, West Lafayette, Indiana Join Date: Aug 2008 Posts: 2,315	We started doing an Ampure clean up after sonication after seeing this recommended in another thread. Should get rid of very short DNA and RNA molecules. Also provides an additional purification step prior to the first enzymatic steps in the TruSeq protocol. Most genomic DNA preps are actually degraded RNA preps with a little genomic DNA mixed in for good measure. As far as degraded DNA goes -- as long as it is not degraded below the insert size you want for your library, the only effect the degradation would have, in and of itself, is any site specificity conferred by the agent degrading the DNA will result in a sequencing bias. However, to the extent that degradation might be an indication of general mistreatment of the DNA -- well it might suggest other problems, like DNA damage that will hinder the production of viable (amplifiable) library molecules. That is, if you have lots of pyrimidine dimers then the chances a polymerase is going to be able to replicate the strand are low. -- Phillip

03-29-2012, 02:30 PM	#6
NextGenSeq Senior Member Location: USA Join Date: Apr 2009 Posts: 482	I think you just need clean DNA. It doesn't matter if it's high molecular weight. You can make RNA-Seq libraries from paraffin embedded RNA which is about 100 bp.

03-30-2012, 08:53 AM	#7
Simone78 Senior Member Location: Basel (Switzerland) Join Date: Oct 2010 Posts: 208	I wouldn´t care about removing small fragments before adaptor ligation. Adaptors are usually added in such an excess (10 times more than your DNA) that this shouldn´t be a problem. It´s rather more important to get completely rid of them after ligation.

03-30-2012, 09:25 PM	#10
anurag.gautam Member Location: India Join Date: Oct 2010 Posts: 15	Hi , CAn anybody tell me what should be the appropriate read length for chip seq analyses. What should be it for Illumina and SOLiD?? Anurag

04-12-2012, 06:42 AM	#12
hawaii454-0 Member Location: UK Join Date: Dec 2010 Posts: 23	Has anybody tried making a TruSeq library from DNA at low concentration and around 150bp? I used different ratios of ampure beads in the clean up steps to make sure the smaller fragments aren't lost in the supernatants. Has anyone made any adjustments to the protocol for cases like this?

04-12-2012, 07:03 AM	#14
hawaii454-0 Member Location: UK Join Date: Dec 2010 Posts: 23	I have managed to get some okay-looking TruSeq libraries from some tests on E coli which were sheared at around 150bp and the starting material was 50ng, 25ng, and 10ng. The 50ng library (obviously) looks the best. The adjustments I made to the protocol were in the ratios of ampure beads to sample in the clean ups through the protocol. Prior to the adapters being added, the ratio was 1.6:1 and afterwards it was 1.2:1. This seemed to work well for the size of the fragments I sheared. The next step will be checking the Agilent concentration with the ones on the Qubit and after qPCR.