SEQanswers

Go Back   SEQanswers > Sequencing Technologies/Companies > Illumina/Solexa



Similar Threads
Thread Thread Starter Forum Replies Last Post
Adapter Trimming Illumina Mate Pairs (Nextera Protocol) hrarnc Bioinformatics 20 02-28-2014 10:44 AM
TruSeq or Nextera kit for library prep? rooty Sample Prep / Library Generation 12 10-29-2013 07:02 AM
Nextera Exome Enrichment kit dummyseq Illumina/Solexa 8 10-03-2013 06:46 AM
How to prep BACs for Nextera XT DNA kit? OnUrMark Sample Prep / Library Generation 3 05-20-2013 12:56 PM
cDNA prep for Illumina using the Nextera kit fgoetz Sample Prep / Library Generation 3 07-09-2012 07:34 PM

Reply
 
Thread Tools
Old 10-14-2013, 02:27 AM   #1
dfhdfh
Member
 
Location: Germany

Join Date: Jan 2013
Posts: 46
Default Changing the Nextera XT kit protocol and other questions

Hi!

First of all, if this belongs in the Sample Prep / Library Generation forum, feel free to move it there.

Now on to my questions. Until now, I was only part of the bioinformatics of sequencing. However, I recently switched institutes and now I have to establish sequencing in my new lab. We bought a MiSeq and would like the students to participate in this field, too, which is why we offer a practical course in sequencing.
Given the limited time of such a course (~4 x 6-8 h), I chose the Nextera XT kit and would like to sequence some bacterial genomes for demonstration and a bit of computational analysis. The genomes will be bought (are there any limitations?) or prepped in advance, so that we can immediately start with the tagmentation reaction.
Given the fact that the Nextera XT kit is optimized for plate usage, which is rather unsuitable for a practical course, I'd like to change everything to tubes, except for maybe the magnetic bead parts. I don't think this should affect the protocol (as long as we keep track of the indices). Or are these false conclusions?

Do you have any experience in using PCR purification kits other than the Illumina-recommended AMPure XP? E.g. standard centrifuge kits? This would be much more suitable for the course.

Furthermore, what do you guys think would be best for sequencing the libraries? I only have two sensible options (considering the time needed for sequencing). The first one is the v2 kit with 50 cycles and 1x36 or 2x25 and the second one is the v3 kit with 150 cycles and 2x75. I tend toward the v2 kit with 50 cycles because it takes less time than the v3 and guarantees to finish o/n.

I hope it's not too much text and appreciate every feedback

Last edited by dfhdfh; 10-14-2013 at 03:08 AM.
dfhdfh is offline   Reply With Quote
Old 10-14-2013, 05:29 AM   #2
mcnelson.phd
Senior Member
 
Location: Connecticut

Join Date: Jul 2011
Posts: 162
Default

We've been doing short module courses like what you're describing to train users on our MiSeq for over a year and a half now. We've done two genome sequencing modules so far, one using the standard Nextera kit and the second using the XT and we're planning another course this December during the winter break that will either use XT again or use the Nextera Mate-Pair.

The way we've taught it, we either prepare bulk DNA or have students bring their own, and the first steps we do in the class are quantify by Qubit and then dilute the DNA down to the appropriate starting []. This step usually only takes about an hour if we're teaching 6-8 people and the people who have taken the class said they liked having that part included.

For the XT protocol. We have each student prep their own reaction in 200ul PCR tubes, which can be a bit cumbersome but has worked fine so far. We prepare aliquots of all of the reagents so each student has their own aliquot and there's no issue of contamination for the master stock. For the cleanup step we bought single tube magnet stands so each person cleans their own library with AMPure. For the XT kits, yield can often be an issue and column purifications often seem to loose more DNA than the AMPure does. Saying that, we'll often get >10 ng/ul yields, which is often more than enough for sequencing.

As I've said previous times in this forum, we don't do the bead normalization and denaturing step of the XT protocol and instead treat the cleaned libraries like standard libraries whereby we quantify with Qubit, find the average fragment size on the Bioanalyzer, and then dilute the libraries to 4nm and pool them prior to sequencing. This has been more reliable for us than using the normalization beads, and it also allows the students to go through the same process that they would do for most other library types (e.g. TruSeq, standard Nextera, amplicon, etc.)

The way our module is scheduled, we're able to get all of the wet lab stuff done in a day and a half, often on a Thursday/Friday, and then start the sequencing run and let that go over the weekend. Then on Monday we show the students how to get their data both from BaseSpace and directly from the MiSeq and do two days on genome assembly, annotation, and analysis. This has worked really well and not only lets the students prepare libraries and get that experience but then they also get to analyze their own data which always proves interesting as sometimes people bring samples that pose interesting challenges for assembly or other analysis (e.g. very low/high GC genomes, things with high copy number plasmids, etc.)
mcnelson.phd is offline   Reply With Quote
Old 10-14-2013, 07:23 AM   #3
dfhdfh
Member
 
Location: Germany

Join Date: Jan 2013
Posts: 46
Default

Thanks for your answer. Very helpful.

So, you're saying that you would recommend purifying via AMPure?

Concerning the normalization: I don't know if we'll have our Bioanalyzer in time for the course, so at the moment I'm planning to do the kit normalization via beads. However, I could see us doing QC via qPCR.

How long does QC via Bioanalyzer take? As I said, I have only experience post-sequencing
dfhdfh is offline   Reply With Quote
Old 10-14-2013, 08:36 AM   #4
mcnelson.phd
Senior Member
 
Location: Connecticut

Join Date: Jul 2011
Posts: 162
Default

I'd recommend AMPure simply because it's worked well for us, and when we've done column based cleanups for other things we've had poor yields that leave us with too little library. That may be library type and column dependent, so maybe some of the newer columns produced by Qiagen or Zymo would do a better job for XT? For amplicon libraries we're switching to a Qiagen column for cleanup, but since we have no issues with yield we're not concerned with lower recovery efficiencies. For our genomes, the AMPure has worked reliably and is what Illumina recommends so that's what we've been doing, but given how expensive AMPure is, we may explore column purification in the future if we have leftover reagents to test things with.

If you do your library QC by qPCR, you can determine the nM [] straight from there and wouldn't technically need to quantify by Qubit or run anything on the Bioanalyzer. You could then still do either the standard XT normalization, pooling, and denaturing or treat the libraries like normal and pool based on their nM [] from the qPCR and denature.

In the end I guess it would depend on how you want to train other people to make libraries and use the MiSeq.
mcnelson.phd is offline   Reply With Quote
Old 10-15-2013, 02:08 AM   #5
dfhdfh
Member
 
Location: Germany

Join Date: Jan 2013
Posts: 46
Default

It's a practical course integrated into the Master's program, so I think most of them won't ever do anything sequencing-related again. It's more like a demonstration on what is possible with sequencing and what its applications are. They don't need to be able to do it on their own afterward.

Anyways, thanks again. I think for my test run I'll use both AMPure and column purification to study the respective results. Then I'll decide what to do with the students.
dfhdfh is offline   Reply With Quote
Old 10-16-2013, 04:46 AM   #6
mcnelson.phd
Senior Member
 
Location: Connecticut

Join Date: Jul 2011
Posts: 162
Default

I'd be interested to hear how well the column purification goes, so please post an update after your get a chance to test that out. Especially if you can run the libraries on the Bioanalyzer to show the size distribution.

Also, our courses here are actually administered through one of the masters programs at the university, with the specific goal of training students with the skills that industry groups have told the administration they would like to see graduates with a MS have. I'd be interested in trading notes after you've done your course to compare how they were run and what the student response.
mcnelson.phd is offline   Reply With Quote
Old 10-16-2013, 06:04 AM   #7
mdalessio
Member
 
Location: Waterloo

Join Date: Oct 2013
Posts: 19
Default

Re. column purification for clean ups

I did a column purification for clean up after the tagmentation and then did the PCR. While the PCR did work, I believe my yield was quite affected by the column purification. Too much of the DNA was lost (imo).

Our lab has also tried to do gel extraction of the range of size that we want (after PCR) for 300-600 bp fragments. Again, I think this negatively affects the yield far too much to be viable.
mdalessio is offline   Reply With Quote
Old 10-16-2013, 02:40 PM   #8
mcnelson.phd
Senior Member
 
Location: Connecticut

Join Date: Jul 2011
Posts: 162
Default

Quote:
Originally Posted by mdalessio View Post
Re. column purification for clean ups

I did a column purification for clean up after the tagmentation and then did the PCR. While the PCR did work, I believe my yield was quite affected by the column purification. Too much of the DNA was lost (imo).

Our lab has also tried to do gel extraction of the range of size that we want (after PCR) for 300-600 bp fragments. Again, I think this negatively affects the yield far too much to be viable.
Yeah, that was my impression of columns as well, but Qiagen and Zymo keep making new columns that may work better so if someone else is willing to test it out then I'm all for hearing the results

As for size selecting, I once tried doing a double-sided SPRI size selection on some leftover libraries that we had, and based on the Bioanalyzer it looks like I was able to get a fragment size more in line with what I wanted. We never sequenced those libraries though as we had real libraries to get done and the assemblies we were getting were good enough that the extra time and cost wasn't worth it.
mcnelson.phd is offline   Reply With Quote
Old 10-16-2013, 11:29 PM   #9
dfhdfh
Member
 
Location: Germany

Join Date: Jan 2013
Posts: 46
Default

Quote:
Originally Posted by mcnelson.phd View Post
I'd be interested to hear how well the column purification goes, so please post an update after your get a chance to test that out. Especially if you can run the libraries on the Bioanalyzer to show the size distribution.

Also, our courses here are actually administered through one of the masters programs at the university, with the specific goal of training students with the skills that industry groups have told the administration they would like to see graduates with a MS have. I'd be interested in trading notes after you've done your course to compare how they were run and what the student response.
If I do it, I'll surely post updates here.

I'm also interested in trading notes. The course will be in December, so it'll be a while
dfhdfh is offline   Reply With Quote
Old 10-28-2013, 04:44 PM   #10
han526
Member
 
Location: Boston

Join Date: Feb 2010
Posts: 11
Default

What is your sample input volume? If you could reduce sample volume, the yield should go up and cost go down. alternatively, you can try PCRClean, which is equivalent to Ampure but much less.

As for size selection, do you remember what the yield was after double-sided SPRI selection?

Quote:
Originally Posted by mcnelson.phd View Post
Yeah, that was my impression of columns as well, but Qiagen and Zymo keep making new columns that may work better so if someone else is willing to test it out then I'm all for hearing the results

As for size selecting, I once tried doing a double-sided SPRI size selection on some leftover libraries that we had, and based on the Bioanalyzer it looks like I was able to get a fragment size more in line with what I wanted. We never sequenced those libraries though as we had real libraries to get done and the assemblies we were getting were good enough that the extra time and cost wasn't worth it.
han526 is offline   Reply With Quote
Old 10-28-2013, 05:10 PM   #11
mcnelson.phd
Senior Member
 
Location: Connecticut

Join Date: Jul 2011
Posts: 162
Default

Quote:
Originally Posted by han526 View Post
What is your sample input volume? If you could reduce sample volume, the yield should go up and cost go down. alternatively, you can try PCRClean, which is equivalent to Ampure but much less.

As for size selection, do you remember what the yield was after double-sided SPRI selection?
For cleanup, we're using the full 50ul volume from the PCR step. That requires only 30ul of AMPure, so it's not a lot (esp. compared to what TruSeq required). It is completely ridiculous, however, that AMPure costs ~$18/ml if you buy the 5ml but only $11/ml if you buy 60ml. You'd have to do a ton of Nextera libraries to justify the 60ml which means you're really overpaying for the 5ml bottle.

As for the double-sided SPRI, I certainly lost library, but not enough that I couldn't still do the standard denaturing protocol like we do for XT libraries. As I said though, we never sequenced anything because it wasn't necessary and there were other things I wanted to use spare space for. I did however run the libraries on the Bioanalyzer, and did get a narrower peak. The biggest problem with double-sided SPRI for size selection though is that it isn't really that accurate because it relies on volumes, so it was still a pretty broad peak compared to a gel extracted size selection.
mcnelson.phd is offline   Reply With Quote
Old 11-26-2013, 01:41 AM   #12
dfhdfh
Member
 
Location: Germany

Join Date: Jan 2013
Posts: 46
Default

Alright, so I'm close to performing my first test run. It is planned for Thursday (two days from now). I have some last minute questions, and hope they are not too basic.

As a test, I will sequence an E. coli genome with existing reference annotation using the v2 50 cycle kit and Nextera XT for library prep. Unfortunately, we don't have our Bioanalyzer, yet (but the computer and software package that comes with it ).

Here is my first question: Why can't I chose 26 cycles for both reads in the IEM 1.6.0 sample sheet creator? It always jumps back to 151. That's not too bad, of course, because I can just go ahead and edit the .csv file. However, it is a bit annoying and I want to make sure it is not me making a mistake.

2.: If I want to use a phiX spike-in, do I have to use an Index for my test sample? I didn't find anything concerning this. My gut-feeling is that I do but I'm a bit irritated by the fact that v2 phiX contains an Index and v3 doesn't (I don't know which one we have, as I wasn't the one buying it, I will update later when I found the time to check that).

3.: If I only sequence one sample, does the bead normalization of the Nextera XT protocol have any purpose?

Thanks in advance

Edit: I just checked, we got the phiX v2.
Edit2: Ok, so I found answers for the second and third questions. I still haven't found an answer for the first one, though.

Last edited by dfhdfh; 11-27-2013 at 12:14 AM.
dfhdfh 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 04:59 PM.


Powered by vBulletin® Version 3.8.9
Copyright ©2000 - 2021, vBulletin Solutions, Inc.
Single Sign On provided by vBSSO