Raw wells?

[pmiguel]

Yes, we just got a DynaMag-2. Previously we had been using an older magnet of uncertain origin. Just a pet theory of mine that it would impact enrichment percentages. Your results suggest that I am wrong.

It is not the magnet. We used the DynaMag-2 this time and the results were the same.

The extra duds are probably beads that bind to the enrichment beads, but do not contain enough amplicons to produce robust signal.

I just noticed that we have a modified emPCR method. We use a >15 minute incubation of Enrichment Beads + amplified DNA beads [1] on the Labquake instead of 5 minutes.

Anyone else who is experiencing low numbers of raw wells have an extended incubation at this step?

This is the only step I can find that might pull down more weakly amplified (dud) beads in our protocol.


[1] This section of the emPCR October 2009 Method Manual page 8:

3.6.3 Enrichment of the DNA-Carrying Beads
2. Rotate the tubes on the Labquake, at room temperature (+15 to +25ºC) for 5 minutes.

--
Phillip

[pmiguel]

Phillip- Thanks for the reply. Do you have info about the magnet that you ordered? I've been using the one recommended by Roche, the DynaMag-2. I've been thinking my issue might be a centrifuge issue and plan on doing a bead count of the layer 2 supernatant on my next run. I have been getting a high keypass rate though, even though the raw wells are so low, 75-80%.

Yes, we just got a DynaMag-2. Previously we had been using an older magnet of uncertain origin. Just a pet theory of mine that it would impact enrichment percentages. Your results suggest that I am wrong.

In our case, we have done bead counts on all the supernatants pulled off some months ago. They were all consistent with nearly every well having a bead left in it. So the low raw well numbers were a result of "dud" beads not having enough signal to qualify as a raw well, rather than a large percentage of the wells being empty of any beads.

Note that the bead deposition protocol relies on about 40% the beads being duds. That is, there are about 1.8 million wells in a single region of a 2 region run. But you attempt to load 2 million beads. Nearly all the wells get filled. But you only want about 1 million of the wells to be counted as raw wells. (Too much more than that would be "overloaded".) So really all that is happening here is that rather than 40% of the beads being duds, 60% or more are duds.

The extra duds are probably beads that bind to the enrichment beads, but do not contain enough amplicons to produce robust signal. Could be something as trivial as the number of enrichment beads being used. If that were too high (because a lot# got a little overloaded) then some weak beads might make the cut.

Seems like it would be a delicate balance to maintain. Surprising it has worked as well as it has in the past.

--
Phillip

[rbarry]

Phillip- Thanks for the reply. Do you have info about the magnet that you ordered? I've been using the one recommended by Roche, the DynaMag-2. I've been thinking my issue might be a centrifuge issue and plan on doing a bead count of the layer 2 supernatant on my next run. I have been getting a high keypass rate though, even though the raw wells are so low, 75-80%.

[pmiguel]

Not yet. Our RAC asked us to upload a few runs and was specifically interested in the keySignalPerBase metric in the 454RuntimeMetricsAll.txt file, which for Titanium should be 250-600. Also ppi1 should be 300-400 and ppi3 should be about 50% of the ppi1 values. We tended to be low for all of those. So Roche sent and engineer to check the instrument out and replace some valves.

We have done one eight region run subsequently. Spec for each region should have been 80-120 thousand Passed Filter Wells. 3 of the regions had bad libraries. But for the other five we had >50% filter pass. Four of five broke 80,000 Passed Filter Wells.

Still, they were on the lower end. But our keySignalPerBase was >250 for all of them. So I think our instrument was successfully "tuned up".

I still think there is an enrichment stringency issue at play here. I broke down and ordered one of those expensive rare earth magnets. The older magnets seems like they would be much more gentle (slow) and thus may be allowing some weakly-amplified beads to enrich. The new magnet will be used for our next run. We'll see how that goes.

--
Phillip

[rbarry]

Did anyone find a solution to this yet? I've been having the same problem but I've only been getting about 1.3M raw wells.

[pmiguel]

HI All,
We've recently observed a consistent decrease in teh number of raw wells in our Titanium runs - from 2 million to 1.5 million. We have tried all sorts of solutions, but cannot figure out teh reason for the drop.

2 Questions :
1. Has anyone else seen anything like this before and
2. Does anyone know how a "Raw well" is defined. Is it a physical well detected by the camera? Is it a well with a bead in it?

Any help would be apprecaited.
Cheers
Richard Allcock
Perth, Western Australia

Does anyone have an answer, or even an opinion what could be causing this? As I mention above, we are seeing the same thing starting around March or April of 2009. Not quite this severe, however. Maybe 1.6-1.8 million totalRawWells per run.

--
Phillip

[pmiguel]

BTW, you might want to check out an earlier thread on this topic:

http://seqanswers.com/forums/showthread.php?t=3026

--
Phillip

[pmiguel]

HI All,
We've recently observed a consistent decrease in teh number of raw wells in our Titanium runs - from 2 million to 1.5 million. We have tried all sorts of solutions, but cannot figure out teh reason for the drop.

2 Questions :
1. Has anyone else seen anything like this before and
2. Does anyone know how a "Raw well" is defined. Is it a physical well detected by the camera? Is it a well with a bead in it?

Any help would be apprecaited.
Cheers
Richard Allcock
Perth, Western Australia

Hi Richard,
Yes, we have seen exactly the same thing.
From what I gather a raw well has to have a productive bead in it.

Again, this is just what I've pieced together from various comments from Roche people, but here is what I think must be happening:

(1)Roche's enrichment protocol/bead deposition relies on slightly less than ~50% of the beads being non-productive (duds, no sequence signal produced). That ensures that only about 2 million of the 3.6 million wells of the PTP will emit light. More than that, and read length (presumably) will suffer.

(2)Some change has resulted in a higher % of duds being produced by our enrichment. Not sure what it is. You are the first other lab I've heard of that is seeing this.

That said, our recent results have been pretty good despite the lower number of raw wells we see, so we have not complained too loudly. But we will need to deal with it at some point.

The only non-canonical thing I can think of that we do is that we used the older, weaker magnet for enrichments, etc, rather than the new, more powerful magnet. Not sure how that could be the cause, though.

--
Phillip

[RichardAllcock]

HI All,
We've recently observed a consistent decrease in teh number of raw wells in our Titanium runs - from 2 million to 1.5 million. We have tried all sorts of solutions, but cannot figure out teh reason for the drop.

2 Questions :
1. Has anyone else seen anything like this before and
2. Does anyone know how a "Raw well" is defined. Is it a physical well detected by the camera? Is it a well with a bead in it?

Any help would be apprecaited.
Cheers
Richard Allcock
Perth, Western Australia