short read length on XL+ - Page 2

madseq · 04-04-2012, 01:38 AM

Hi FGponce,

what was the accuracy at 600 bp (98% match) of your first control test fragment run? The number of reads above 98% perfect match at 600 bp should be at least 56% for a control run (this can change when using libraries).

Below that number you could have an instrument problem.

You can ask Roche for a validated E coli XL+ library they prepared in Branford recently.

FGponce · 04-10-2012, 04:52 PM

Hi madseq,

So the machine looked fine with avg 62% above 98% match for the control run.

The data for the sample run attached above actually contains reads from 5 different biological samples/organisms. I may expect one or two to be recalcitrant due to long stretches of AAAs etc in the genome but not all five. Also, if the machine is working well why don't the control beads act as controls and produce long reads? Surely a control that doesn't act as a control during a run with bad libraries isn't actually a control?!

I'll try and get a library that has run well on a FLX+ supplied by Roche.

We havent done much FLX seq is it normal to see large sample dependent read length variation from libraries of similar fragment size?

madseq · 04-11-2012, 01:17 AM

Quote:

Originally Posted by FGponce

Hi madseq,

So the machine looked fine with avg 62% above 98% match for the control run.

The data for the sample run attached above actually contains reads from 5 different biological samples/organisms. I may expect one or two to be recalcitrant due to long stretches of AAAs etc in the genome but not all five. Also, if the machine is working well why don't the control beads act as controls and produce long reads? Surely a control that doesn't act as a control during a run with bad libraries isn't actually a control?!

I'll try and get a library that has run well on a FLX+ supplied by Roche.

We havent done much FLX seq is it normal to see large sample dependent read length variation from libraries of similar fragment size?

Did you separate the polyA containing libraries from the others, or did you sequence all of them together?? If you mixed them all, this could be the problem. And is it possible you have small fragments in the library on lane 1 of the bioanalizer?

Good runs in the FLX+ give readlength averages of 650-700. Don´t expect readlenths of 1000-1500.

And yes, the control beads perform differently depending on the sample you´re running, loading density, etc... They help to diagnose problems but they do not behave always exactly the same.

You should try to have your run analized by customer support, they have the tools to tell you if your failed run can be due to the samples used or if you are having issues with the instrument.

MissDNA · 04-11-2012, 11:18 AM

Quote:

Originally Posted by RCJK

MissDNA-I've only done 1 XL+ run with metagenomic samples (the 3rd image I posted) so maybe it was a fluke that it was shorter? I don't recall having any specifically bad XLR70 metagenomic runs in the past.

Hi RCJK, one question: do you follow exactly the XL+ Rapid Library protocol or you guys do any modification?

FAS has recommended us to change the nebulization conditions from 15 psi for 1 min to 20 psi for 1 min. This sort of recommendation made me think the real problem on our preps are the really long fragments rather possibity of having small ones. I repeated the libraries today and we got a larger % of fragments around 650-1200 bp than before. I sent the traces to Roche, and I am waiting on their ok to titrate and do the LV. We are perfoming the protocol for evaluation of fragments size after the first melt during enrichment. Only if we get fragments in the right range, we will sequence.

They did tell me that metagenomic samples can be harder to get to work on Plus.

RCJK · 04-11-2012, 04:41 PM

Quote:

Originally Posted by MissDNA

Hi RCJK, one question: do you follow exactly the XL+ Rapid Library protocol or you guys do any modification?

Hi MissDNA, yes, for the XL+ RL preps I've done so far I've followed the protocol exactly. I haven't heard of any changes to the recommended nebulizing conditions, but if that's what they are recommending now maybe they have realized it results in better libraries.

Actually, I'd like to start using our Covaris to shear the DNA, but am not quite sure what size to go for or what settings to use. I'm sure they'll just say that it's not supported and not to do it. I've used it for XLR70 RL preps just fine, but am a bit hesitant to proceed with it for the XL+ RL preps.

On a side note, I did hear from Roche last week that they are recommending 6-8% enrichment recovery for best XL+ results. They said to still proceed with beads up to 20% enrichment, but that over ~8%, the signal intensities on the beads starts to get too high. I've asked that they put out some sort of bulletin or something with all their tips for a successful XL+ run since they seem to be highly variable.

Good luck and let us know how the run goes!

Jason

MissDNA · 04-12-2012, 07:53 AM

Thanks, Jason. I will most definitely let you guys know.

I haven´t heard anything about enrichment rates. Our bad run it was with beads with very low enrichment. Problem is we used a box of emPCR reagent that expired on nov/11, this way we cannot get replacement for our run. I honestly don´t think that was the problem but I cannot argue with Roche on that, as using of expired reagents are not supposed to happen. Anyway, now we are doing all this tests under their supervision, with non expired reagents. We aslo got rid of the bead recovery reagents lots that have been showing problems.

I agrre theu should release a bulletin with tips, like they did for amplicon sequencing.

Could you do me a favor, and post the Agilent traces from your best and worse run, so I can see how they look like?

Nicole 454 Sequencing · 04-23-2012, 06:43 AM

Hi Jason,

The recommended enrichment percentage for the GS FLX+, GS FLX and GS Junior platforms remains between 5% and 20%. Within this window, there is no correlation between increasing enrichment percentage and signal per base. A higher signal per base is typically indicative of other issues.

Please feel free to contact your local Roche representative with any issues or concerns and visit our website at www.my454.com/my454 for the most up to date guidelines.

Regards,

Nicole
Technical Support Scientist
454 Life Sciences, A Roche Company

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Mapping Short Reads with unequal length using MAQ	TOLEN	Illumina/Solexa	0	12-30-2010 07:57 PM
read length distributions?	greigite	Illumina/Solexa	1	12-01-2009 07:42 AM
De novo fragment assembly with short mate-paired reads, Does the read length matter?	strob	Literature Watch	3	10-23-2009 01:36 PM
BWA Read Length	AnamikaDarwin	Bioinformatics	1	04-10-2009 11:47 PM
MAQ and short read length (DGE)	jms1223	Illumina/Solexa	5	02-23-2009 07:07 AM

04-04-2012, 01:38 AM	#21
madseq Member Location: Spain Join Date: Dec 2010 Posts: 13	Hi FGponce, what was the accuracy at 600 bp (98% match) of your first control test fragment run? The number of reads above 98% perfect match at 600 bp should be at least 56% for a control run (this can change when using libraries). Below that number you could have an instrument problem. You can ask Roche for a validated E coli XL+ library they prepared in Branford recently.

04-10-2012, 04:52 PM	#22
FGponce Junior Member Location: New Zealand Join Date: Dec 2011 Posts: 9	Hi madseq, So the machine looked fine with avg 62% above 98% match for the control run. The data for the sample run attached above actually contains reads from 5 different biological samples/organisms. I may expect one or two to be recalcitrant due to long stretches of AAAs etc in the genome but not all five. Also, if the machine is working well why don't the control beads act as controls and produce long reads? Surely a control that doesn't act as a control during a run with bad libraries isn't actually a control?! I'll try and get a library that has run well on a FLX+ supplied by Roche. We havent done much FLX seq is it normal to see large sample dependent read length variation from libraries of similar fragment size?

04-12-2012, 07:53 AM	#26
MissDNA Senior Member Location: Brazil Join Date: Nov 2010 Posts: 143	Thanks, Jason. I will most definitely let you guys know. I haven´t heard anything about enrichment rates. Our bad run it was with beads with very low enrichment. Problem is we used a box of emPCR reagent that expired on nov/11, this way we cannot get replacement for our run. I honestly don´t think that was the problem but I cannot argue with Roche on that, as using of expired reagents are not supposed to happen. Anyway, now we are doing all this tests under their supervision, with non expired reagents. We aslo got rid of the bead recovery reagents lots that have been showing problems. I agrre theu should release a bulletin with tips, like they did for amplicon sequencing. Could you do me a favor, and post the Agilent traces from your best and worse run, so I can see how they look like?

04-23-2012, 06:43 AM	#27
Nicole 454 Sequencing Member Location: Branford Join Date: Apr 2011 Posts: 17	Hi Jason, The recommended enrichment percentage for the GS FLX+, GS FLX and GS Junior platforms remains between 5% and 20%. Within this window, there is no correlation between increasing enrichment percentage and signal per base. A higher signal per base is typically indicative of other issues. Please feel free to contact your local Roche representative with any issues or concerns and visit our website at www.my454.com/my454 for the most up to date guidelines. Regards, Nicole Technical Support Scientist 454 Life Sciences, A Roche Company