SOLiD: TooFewCalls

[pmiguel]

In out case it was not SAGE.

It appears to be related to (in our case) a number of really bad looking panels when reviewing the raw images: large bead lumping artifacts or something.

Bead lumping, eh? That goes back to my wondering whether the new, low profile, v 3.5 flow cells were subjecting beads to higher shear forces than the pre v 3.5 flow cells.

It could be that the shorter terminal transferase reaction times of recent deposition protocols are not sufficient to anchor beads robustly in the new flow cells.

We had a set of templated beads that seemed to work fine under v. 3, but gave terrible results on a WFA we did under v. 3.5 We noticed that this library had a tendency to "cheetah". The only reason we could think of for it working previously then failing miserably under 3.5 were the flow cell changes. Smaller volume, lower "profile", could mean: higher fluid pressure and higher shear forces during fluidics.

Anyway, control beads, and even other sets of templated beads we had left over from months ago, still looked good. So we decided we were going to return to doing the overnight terminal transferase reaction on all templated bead sets.

--
Phillip

[inijman]

My first guess--since this is related to SOLiD-SAGE--is some sort of adaptor sequence issue. Like short inserts or adaptor dimers.

Has anyone got SOLiD SAGE working?

--
Phillip

In out case it was not SAGE.

It appears to be related to (in our case) a number of really bad looking panels when reviewing the raw images: large bead lumping artifacts or something.

[pmiguel]

I think it now splits into two issues: first the fact that you get the error messages and secondly the fact that you cannot map much of your data.

I'm myself curious to see what causes the error messages, since we have also a complete run without data and only these errors.

It could mean that in your case the remaining data is poor resulting in low mapping percentages. Whether the data is really poor of that the error indicates a problem with the image interpretation and therefore making the data worse than it is remains to be checked.

My first guess--since this is related to SOLiD-SAGE--is some sort of adaptor sequence issue. Like short inserts or adaptor dimers.

Has anyone got SOLiD SAGE working?

--
Phillip

[inijman]

I think it now splits into two issues: first the fact that you get the error messages and secondly the fact that you cannot map much of your data.

I'm myself curious to see what causes the error messages, since we have also a complete run without data and only these errors.

It could mean that in your case the remaining data is poor resulting in low mapping percentages. Whether the data is really poor of that the error indicates a problem with the image interpretation and therefore making the data worse than it is remains to be checked.

[pmiguel]

Sorry for the confusion, but the SOLiD SAGE run I've mentioned was done just BEFORE the update to SOLiD3plus.

We also had some software problems with SOLiD3plus (after the update). But the hardware and the results were fine, in contrast to what was displayed by the GUI during the SOLiD run.

Still, SAGE mapping-to-reference results are still poor. This was also experienced by ABI.

Ah, okay. What species were you sequencing? What did you use as your reference sequence?

--
Phillip

[DNAjunk]

Sorry for the confusion, but the SOLiD SAGE run I've mentioned was done just BEFORE the update to SOLiD3plus.

We also had some software problems with SOLiD3plus (after the update). But the hardware and the results were fine, in contrast to what was displayed by the GUI during the SOLiD run.

Still, SAGE mapping-to-reference results are still poor. This was also experienced by ABI.

[pmiguel]

I don't think it is biological but a hardware/software glitch in translating the images. It was a pretty standard transcriptome/chip set of libraries from various organisms. In general we don't use the secundairy analyses, but use bwa to map the reads on our cluster.

Would be good to get a little more information from you and the original poster. Was it obvious from the cycle plots there was an issue? Like, were the number of usable beads dropping each cycle?

We haven't started our first v3.5 run yet. But for various reasons ended up doing a large number of WFAs. We did notice that we had a bead set that was cheetahing fairly consistently--something we hadn't seen for a long time. That made me think that perhaps the new low profile flow cells resulted in higher shear forces on the beads during flow cell filling/emptying.

But I would need more information about your run to be able to draw any conclusions. It could be useful though. Because I can think of factors that would mitigate the potential issue, it if is occurring.

--
Phillip

[inijman]

Were your libraries also SAGE libraries? What species? What did you use as a reference sequence to map to?

--
Phillip

I don't think it is biological but a hardware/software glitch in translating the images. It was a pretty standard transcriptome/chip set of libraries from various organisms. In general we don't use the secundairy analyses, but use bwa to map the reads on our cluster.

[NextGenSeq]

I'm glad I haven't upgraded our instrument yet. Let us know when the bugs are worked out.

[pmiguel]

Last week we also had a 3plus run producing exactly the same output as you have for all the libraries. Currently, Applied guys are working this out but to no avail yet.

I would be really interested to know whether you could track the problem.

Were your libraries also SAGE libraries? What species? What did you use as a reference sequence to map to?

--
Phillip

[inijman]

Last week we also had a 3plus run producing exactly the same output as you have for all the libraries. Currently, Applied guys are working this out but to no avail yet.

I would be really interested to know whether you could track the problem.

[DNAjunk]

Hi

I did a SOLiD 3plus run (SAGE experiment).
In the stats-file the following lines were found:

...
# File: /data/results/solid0365/solid0365_20091027/Sample4/jobs/postPrimerSetPrimary.818/rawseq/solid0365_
20091027_1301_F3_sequence.csfasta - 0 / 11107 [0.0%] (TooFewCalls: 11107 [100.0%])-- UNUSABLE
# File: /data/results/solid0365/solid0365_20091027/Sample4/jobs/postPrimerSetPrimary.818/rawseq/solid0365_
20091027_1302_F3_sequence.csfasta - 0 / 1451 [0.0%] (TooFewCalls: 1451 [100.0%])-- UNUSABLE
Totals (186 p): 19643215 / 19915063 (98.6%): (TooFewCalls: 271848 [1.4%]
Average: 105608.7 / 107070.2 (97.6%): (TooFewCalls: 1461.5 [2.4%])
Usable (184 p) (98.9%): 19643215 / 19902505 (98.7%): (TooFewCalls: 259290.0 [1.3%])
Usable Average: 106756.6 / 108165.8 (98.7%): (TooFewCalls: 1409.2 [1.3%])


What does this all mean?

Thanks for your kind help!