Off-line Basecaller (OLB) not demultiplexing

[ETHANol]

We (I) screwed up and lost the control lane from a run. The other lanes are methylC-seq libraries so they cannot be used for the control lane so we wanted to run Illumina's Off-line Basecaller (OLB) using the control lane from a previous run.

We did that with the following command ("pathTo" being the absolute path to the directories or files):

/pathTo/bustard.py --no-controls --CIF /PathTo/130830_SNL119_0101_AC21TTACXX/Data/Intensities --phasing=0.001711212 --prephasing=0.002223242 --matrix=/PathTo/Data/Intensities/BaseCalls/Matrix/s_1_matrix.txt --make

It seems to work for the sequencing read (read 1) but not for the index read (read 2). Pretty much all the reads are going into the undetermined indexes directory. The indexes in the fastq files do not match any of the indexes we use in the lab so it can't be contamination.

The generated fastq files map to their genomes as well as would be expected. Some lanes have multiple species in them and we get mapping to the respective species at the expected percentages.

Anybody have any idea of what the problem is with the index read? They indexes used were compatible indexes.

Anybody have a script that will count the indexes, i.e. all 4,096 index possibilities and get their frequencies?

We are in contact with Illumina customer support but I thought I'd ask here to see if anyone else has encountered this problem with OLB.

Thanks.

[sklages]

You could also try to let OLB estimate the (pre)phasing from lane 1, e.g. with just '--control-lane=1'. At least to see if you get any difference ..

I have a script which analyzes the undetermined indices; PM me if you'd like to try.

regards, Sven