[pmiguel]

We got it to work on our HiScanSQ -- which uses the same chemistry as the HiSeq, but only scans the top of the flowcell. Not an identical situation, but we had some SMART cDNAs that we sheared and ligated TruSeq adapters on. So about 1/2 of them had the same 50 nt of SMART primer at the beginning. We mixed them 1:1 with a genomic DNA library. Cluster registration went fine.

We are also considering this, may I ask you why half of the reads had the smart oligo at the beginning (since you radomly sheared them)

Thanks and best

Will have to recall from long term memory.

Okay, the cDNAs all were flanked by these 50 bp SMART adapters. These cDNAs were also fairly short prior to shearing for reasons I won't go into. They were fragmented to roughly 1/2 their initial length. Then "end polished" and ligated to Illumina adapters, probably using the TruSeq DNA prep kit. Hence, random chance put 50% of the reads adjacent to the SMART adapter.

SMART is designed to amplify full-length cDNAs using the ingenious "PCR Suppression" method. So if your post-shearing length is a small enough fraction of the amplified cDNA length, it should not be much of an issue. Possible to further reduce contribution towards your sequence from adapter using a rare-cutting restriction enzyme site in your SMART adapter. Not sure it would be worth it, though.

--
Phillip

[simonandrews]

If you only have a strong initial bias but with diverse sequence afterwards there is a somewhat dramatic solution which does work on the Hi-Seq, but unfortunately has to be applied to the whole run.

Basically you start read 1 as normal, but before starting to do imaging you do 5 chemistry cycles without doing any imaging (so called 'dark cycles'). You then do the run as normal and read through the rest of your sequence. At the end of the first read you strip the templates and then re-anneal primer 1 and do a 5 cycle read 2 to get the biased bases at the start. You therefore end up with all of the sequence you want, but spread out over 2 reads instead of 1.

Having said that, in your case that's not what I'd do. Given that you're only using the first 5bp to confirm that you have the correct sequence I'd be tempted to use a custom primer to prime over the biased positions though. If you have a different sequence at the start the primer shouldn't anneal properly so you should only get signal from clusters with the correct sequence.

[JackieBadger]

we sequenced low complexity/diversity libraries by mixing them in with other experiments of high complexity. this way low complexity clustered can be indentifed in amongst high complexity clusters

[josdegraaf]

[QUOTE=pmiguel;57185]We got it to work on our HiScanSQ -- which uses the same chemistry as the HiSeq, but only scans the top of the flowcell. Not an identical situation, but we had some SMART cDNAs that we sheared and ligated TruSeq adapters on. So about 1/2 of them had the same 50 nt of SMART primer at the beginning. We mixed them 1:1 with a genomic DNA library. Cluster registration went fine.

--


We are also considering this, may I ask you why half of the reads had the smart oligo at the beginning (since you radomly sheared them)

Thanks and best

[MRSeq]

Still haven't gotten this to work well--instrument is being checked out right now by Illumina engineers. They say there always was refocusing on read two (really 3; after indexing read) so I needed randomness at first few bases there, too. PhiX spike in didn't help much. Think I need to redesign primers to stagger the signal.

Hillary,

we have had no success so far, either :-( So your 20% spike-in phiX run with 50% normal cluster density was not a success, either? What specific cluster density did you used? Was it chemistry v3? We will be doing another attempt on HiSeQ with v3 chemistry in a couple of weeks so I am very interested in the technical details. TIA

[HMorrison]

Mixed success

Still haven't gotten this to work well--instrument is being checked out right now by Illumina engineers. They say there always was refocusing on read two (really 3; after indexing read) so I needed randomness at first few bases there, too. PhiX spike in didn't help much. Think I need to redesign primers to stagger the signal.

[MRSeq]

Hello,

I am glad to see a thread with so many helpful informations for us dealing with low complexity samples! I am having the same problems - 5-11 nt label at the beginning of the read. So far the quality has been very disappointing and it worsened dramatically when we switched to HiSeQ in spite of 50% phiX spike-in and lower cluster densities. We were also not very happy with our sequencing facility for other reasons. Those of you who had good experience with sequencing low diversity samples - can you please share which sequencing facilities seem to do a good job of it? In our case some preparation is very time consuming and costly and we do not want to loose any more data...

[pmiguel]

Update--
Quality on first read was low and got an "N" in up to half the reads at positions 12, 20, 21, and 24. Indexing read was fine. The instrument completely lost focus on 2nd read (100 nt read) so I got nothing. That read started reading into the specific amplicon reverse primer which I thought would be okay because clusters had been found on the first read. Apparently not true. Doesn't matter for this library but eventually want to join longer reads. Am trying again next run with 20% spike in of high complexity library and 50% lower cluster density.

Actually you likely got burned by one of the recent updates. Until the v3 chemistry update, the instrument did not refocus after strand turnaround. But now it does. If the instrument had used the focal points it was using during the successful index read, you might have been okay.

Thanks for the info, though.

If this happens again, Tech Support may be able to guide you through a forced refocus of the instrument after the cycle where your adapter ends. (I think it mainly would involve halting the run and re-starting it.)

Hmm, if that works, it could be a general (although drastic) solution to low complexity bases even in the first read. Just let as many cycles as necessary complete (ignoring the possibly terrible results you are getting). Stop the run. Start a completely new run on the same flow cell, from that point. At 4 bases from the end your low complexity start, cluster calling would happen.

--
Phillip

[HMorrison]

Update--
Quality on first read was low and got an "N" in up to half the reads at positions 12, 20, 21, and 24. Indexing read was fine. The instrument completely lost focus on 2nd read (100 nt read) so I got nothing. That read started reading into the specific amplicon reverse primer which I thought would be okay because clusters had been found on the first read. Apparently not true. Doesn't matter for this library but eventually want to join longer reads. Am trying again next run with 20% spike in of high complexity library and 50% lower cluster density.

We've had the solenoid change-out and the s/w upgrades (just before this run) and bubble-related loss of focus should not be entire lane starting with cycle 1 (or so Illumina claims). This was 100% loss, not a few percent.

[csquared]

Harold and Phillip,
To update, it appears that I got a reasonable number of reads surviving up until the HiSeq lost focus partway through read 3. I'll try the lower cluster density and phiX or shotgun library spike in next time.

Thanks, all.

Hilary

The loss of focus during read 3 is likely a bubble from fluidics than a diversity problem. If you got that far with good PF clusters, good base quality and a good, flat FWHM metric, it isn't the diversity of the library that is the problem. It is very likely a fluidics issue and one you should raise with your FAS as it would potentially be eligible for a warranty replacement of the affected lane.

It is also worth checking if your HiSeq (assuming it is a HiSeq instrument, if not, this may not apply) has had the new solenoid valves installed. They help prevent, but not eliminate, the bubble issues. I don't know what Illumina is calling the new valves but your FSE or FAS will know.

[pmiguel]

Yes it helps. But from my limited experience, the catastrophic failures come from focusing issues -- where the instrument sees a blank flow cell surface and de-focuses as it attempts to "find" the clusters it expects.

Again, recent firmware upgrades may have mitigated this particular issue. I am particularly paranoid about it because we only recently got an Illumina sequencer and our particular model is an outlier. So problems probably get solved for the HiSeqs first -- those particular to a HiScanSQ would be noticed and fixed later in most cases.

--
Phillip

[TonyBrooks]

Would running low diversity libraries at a low concentration not help solve the problem?
If you are not looking for large number of reads, then running at a low concentration should mean less chance of overlapping clusters and more reads passing filter.

[HESmith]

Hi Hilary,

Apologies for using the incorrect gender, and sorry to hear about the focusing error. Better luck next time.

Harold

[HMorrison]

Semantically "random" does not mean "high complexity". See.

Point being we presume "random" here means all four bases were a random mix of ACGT and therefore an even mixture of all 256 possible sequences. But we would need to know what the method of generating these was to know.

--
Phillip

Harold and Phillip,
I am the poster ("she", not "he", btw) who used the phrase "First four bases were completely random". The 4 random bases were generated by ordering my oligos with "NNNN" where the read is supposed to begin. I did not generate a single "random" sequence to use. Back when I used to synthesize oligos myself, we achieved randomness by mixing reagents into a single bottle that went on the instrument along with A, C, G, T. Don't know what InVitrogen or IDT do these days. (Anybody else go back to Maxam&Gilbert sequencing days, pre-PCR ?)

To update, it appears that I got a reasonable number of reads surviving up until the HiSeq lost focus partway through read 3. I'll try the lower cluster density and phiX or shotgun library spike in next time.

Thanks, all.

Hilary

[pmiguel]

Hi Harold,
Despite the simplicity of what we are discussing here, I think there are ambiguities. I agree your interpretation is likely the correct one. But randomly choosing 5 bases once and prefixing all the reads in a lane with that random sequence would lead to failure of the cluster calling software. That is all I meant. That might seem ludicrous, but I have seen experiments fail for misunderstandings just as ludicrous.

But, yeah, that might be sufficiently unlikely that my bringing up was just distracting, not illuminating. (Also, it could be the malign influence of xkcd forcing my hand to create that link back to it...)

--
Phillip