low number of raw wells

[JayaDNA]

Hi all,

We have been experiencing a bad run of things lately with our GS Junior.....

We primarily conduct amplicon sequencing and have had great results. We tend to slightly under load our plate and typically obtain ~150,000-190,000 raw wells , ~140,000-180,000 key pass wells and ~ 70,000-90,000 passed filter wells. Recently (our last 3 runs!) our numbers have dropped, our last run (15% enrichment achieved) for example yielded 40,000 raw wells, 6000 key pass and 43 passed filter!! The second to last (5% enrichment achieved) was marginally better at 69,000 raw wells, 57,000 key pass and 24,000 passed filter.

We are at a loss as to what has changed, we have not changed our amplicon library prep or quant. My feeling is that there is a reagent issue, either with the A and B beads or the enrichment beads or some processing issue.... Does any know how a raw well is detected? Is there any literature out there which describes the different steps in the sequencing chemistry??

Has anyone else encountered a similar problem? And if so have they been able to overcome this issue? We have flushed away 3 emPCR and Seq Kits with the same results!

Any advice would be greatly appreciated! Cheers.

[ajthomas]

My first suggestion would be to look at the lot numbers of the kits used before and after this change. Are these last three runs a new lot number of any of the reagents? If not, then there may be a problem with your instrument.

[MissDNA]

We have a few runs with low raw wells number that Roche admitted it was due to a bead recovery lot. Using the same lot we have some good and some bad runs.

[JayaDNA]

Thank you for your replies.

Yes my gut feeling is that it is something to do with the enrichment stage. The last two run did share lot numbers for the emPCR kits which I think might support this. I have sent all the lot # info etc to Roche but I'm still waiting on a reply.....

Thanks again

[tokikake]

lower raw wells only with RL not with Lib-L amplicons

Hi all,

we have the same experience with the low raw wells since a few month. Our lokal support also suggested the bead recovery kit as the "bad guy" and assured that 454 is working on that issue, but is not able to track back the problem to a specific LOT. They think, during the enrichment process also beads with no dna are enriched, counted and subsequently loaded on the PTP.

BUT we only have this problem with rapid libraries, but I never saw this with Lib-L amplicons. And we sequence a lot of amplicons. Moreover for the amplicons we see a slight overloading on every run...
So for me it makes no sense, that the recovery is the problem, because it is the same for RL and amplicons in that case. What do you think?

By the way, we see a constant increase in short reads for the amplicons, now pending between 40-60%. The support is aware of this problem but do not know the reason...

[JayaDNA]

hmmm... that is strange you had the problem with RL and not amplicons... I suppose we need to know how exactly a raw well is detected? Will an 'empty' bead if enriched and loaded onto the PTP be invisible i.e. does a bead require to have template in order to be detected? I have had two different answers from Roche so I am confused. If an 'empty' bead is not detected as a raw well then perhaps there is an issue with enrichment and we are enriching for empty beads.... we are confident in our amplicon library and our quant so cannot understand why the beads would be empty unless something is going wrong at the emulsion pcr stage?? On the other hand if an 'empty' bead would still be detected and we know we are loading enough then this suggests something is going wrong with the sequencing reagents perhaps and that the reaction in which the raw wells are 'illuminated' is failing somehow??

I wonder if our respective agents have indeed highlighted our lot #s in their system as if no one has then it will be impossible to spot any patterns!

Also, we consistently see 35-45% short reads in our amplicon runs. This has not changed in 23 amplicon runs.

[tokikake]

I wonder if our respective agents have indeed highlighted our lot #s in their system as if no one has then it will be impossible to spot any patterns!

Also, we consistently see 35-45% short reads in our amplicon runs. This has not changed in 23 amplicon runs.

Do you see lower raw wells in your ampliocns (Lib-L) as well?? Or do you sequence only Lib-A amplicons? We switched to Lib-L a few month ago; when I had a glance at our Lib-A Amplicon runs at that time, it seems that some of them had lower raw wells. But not so clear and reproducible, that we got suspicious.

We tracked the LOT #s of the emPCR as well as the recovery kit, but cannot see a pattern.

[Aniki]

We have also had problems with low raw well numbers. The issue that I suspect and our FAS suspects are bad lot numbers for the Bead Recovery Rgt. Since we have changed lot numbers, the raw well situation has improved but still not very good.

we are also having enrichment values. Titrations are coming out very nicely but full scale emPCRs achieve very low enrichment values or very high enrichment values using titration cpb. It has come to a point where I am too scared to do a full scale LV.