I noticed that Eland provides an option "QCAL_SOURCE" to choose a quality calibration table from another lane, or a file.
Has anyone else used this option? It is not clear to me (from reading the Illumina Pipeline Manual) how it should be used, or what it's purpose is.
Most of the libraries we are sequencing do not align directly to a known reference genome, but we are running a Phi-X control lane each time. Currently we are using only the raw scores, as produced by Bustard for these samples. Would it improve the accuracy of the quality scores to choose a calibration table from the control lane (or another lane with a reference genome), or to use the "upstream" option, as opposed to just using the raw scores? Is the upstream option identical to the raw scores?
Apparently to choose the control lane for the calibration table, you would use an option like this (if the control is lane 4):
1235678:QCAL_SOURCE auto4
And to choose the upstream option:
1235678:QCAL_SOURCE upstream
Has anyone else used this option? It is not clear to me (from reading the Illumina Pipeline Manual) how it should be used, or what it's purpose is.
Most of the libraries we are sequencing do not align directly to a known reference genome, but we are running a Phi-X control lane each time. Currently we are using only the raw scores, as produced by Bustard for these samples. Would it improve the accuracy of the quality scores to choose a calibration table from the control lane (or another lane with a reference genome), or to use the "upstream" option, as opposed to just using the raw scores? Is the upstream option identical to the raw scores?
Apparently to choose the control lane for the calibration table, you would use an option like this (if the control is lane 4):
1235678:QCAL_SOURCE auto4
And to choose the upstream option:
1235678:QCAL_SOURCE upstream