Picard IlluminaBasecallsToSam: Read Structure

[user 31888]

Hi,

I am trying to convert MiSeq .bcl files into uSAM with Picard tools 'CheckIlluminaDirectory', 'ExtractIlluminaBarcodes' and 'IlluminaBasecallsToSam'.

These commands require to provide the READ_STRUCTURE option (described here)

Do you have any idea what the READ_STRUCTURE would be, knowing that I made a library with a TruSeq Exome kit (6 plex) and sequenced it on a MiSeq (2 x 75 bp - 76 cycles)?
Can we get this info from some of the sequencer run log files maybe?

[GenoMax]

This information is in the RunInfo.xml file included in the flowcell_ID directory.

Looking at the help link for Picard above the code would be: 75T1S6B75T1S (Illumina's software generally drops the last base from READ1/2). If you need all 76 basepairs then 76T6B76T.

[user 31888]

Thanks GenoMax !

75T1S6B75T1S (Illumina's software generally drops the last base from READ1/2)

If Illumina's software drops the last base, it should not be sequenced, right?
So, why requesting Picard to skip 1 base then?
I read several times that people apply 1 extra cycle during the run.
Since I am interested to get 75bp long reads, does setting up MiSeq to 76 cycles really necessary?

[GenoMax]

The last base (depending on the sequencing length) is always sequenced. As that last base does not have supporting phasing information it has been generally dropped by illumina's processing pipelines (CASAVA, bcl2fastq). There are options available to have the pipelines report that base, if you want all n+1 bp.

Some facilities supply n-1 bases, some will automatically setup n+1 runs (like in your case) or some may report all n bases.

Setting up an n+1 bp run ensures that resulting "n" bases have valid phasing information.