Strange Miseq Quality Issue, All "#" (Q2) After 100Bp On Read1, Read2 And Barcode Fine

Question

[Nickengland]

I had a sequence back recently with rather strange quality. For the forward read the fastQC graph shows after 100bp (150bp read) the quality drops to 2 for all samples. For the reverse read and the barcode the quality is fine. Looking at the sequences in the fastq shows they all end in a run of "#" from 90-110bp in to the end. Apparently the machine logs show a drop in total signal intensity for this region of bases.

I don't understand how this could happen, as the sample should be OK since the reverse read was perfect. Does the miseq cartridge have a number of sealed compartments containing reagents which could fail to open for some set of bases or contain a dodgy batch?

The forward FastQC plot:

[Dan D]

Have you contacted Illumina tech support? This sounds like the result of a clog, but I don't know enough about the internals of the instrument to give a precise conjecture. Illumina tech support is generally excellent in my experience if you have instrument information at the ready.

[Nickengland]

I haven't directly, but the person who runs the sequencing facility has so we'll have to see what they say.

[Torst]

Phred quality 2 (which is '#' in Sanger encoding and 'B' in the previous Illumina encoding) was described in previous Illumina manuals as a "read quality indicator" and essentially means it couldn't measure anything properly.