Recalibrating phasing/prephasing after run

[thomasvangurp]

I've recently obtained one lane PE100 Hiseq data from a highly biased library due to recstriction enzyme overhangs in reads (GBS protocol). Quality values are low for the first 10 cycles, probably due to the non-balance composition due to in-line barcodes.


The lab that normally runs these libraries for us uses a separate PhiX lane which is set as control in the HCS software as described here:technote_phixcontrolv3.pdf.

Is it possible to reprocess the BCL files using the data from another lane (with a non-biased composition) as a control to correct for phasing errors that occured in the GBS-lane? Will this have the same effect as setting a different lane as control prior to the run?

[pmiguel]

I've recently obtained one lane PE100 Hiseq data from a highly biased library due to recstriction enzyme overhangs in reads (GBS protocol). Quality values are low for the first 10 cycles, probably due to the non-balance composition due to in-line barcodes.


The lab that normally runs these libraries for us uses a separate PhiX lane which is set as control in the HCS software as described here:technote_phixcontrolv3.pdf.

Is it possible to reprocess the BCL files using the data from another lane (with a non-biased composition) as a control to correct for phasing errors that occured in the GBS-lane? Will this have the same effect as setting a different lane as control prior to the run?

No. Too late by the time you have a .bcl file. Unlikely phasing mis-calculations were the cause of lower quality values early in the read if the quality values then increased.

Those quality values don't look all that bad. Standard protocol for GBS has in-line barcodes, right? So the area with lower qualities would be the bar codes, no?

--
Phillip

[GenoMax]

Add my vote to Phillip's. That kind of quality values are to be expected if you had a low nucleotide diverse region at the beginning of your reads.