How can I demultiplex two Library preps(Single Index and Dual Index) using bcl2fastq.

Question

[moranr]

How can I demultiplex two Library preps(Single Index and Dual Index) ?

4 samples were used to build two different libraries Dual Index and Single Index. We mixed both libraries and added all the 8 samples to both lanes of an S2 flowcell.

Now we have the BCL files from this run and the problem we are facing is how to extract the 8 samples given that we have 3 adapters and 12 barcodes.

The first library has a single adapter and two indexes per sample. The second library has two adapters and one index per sample.

The first question here is if we can combine this two libraries into one SampleSheet or we should do it using two SampleSheets with bcl2fastq.

Here is part of our RunInfo

<Reads>
<Read Number="1" NumCycles="151" IsIndexedRead="N"/>
<Read Number="2" NumCycles="8" IsIndexedRead="Y"/>
<Read Number="3" NumCycles="151" IsIndexedRead="N"/>
</Reads>
<FlowcellLayout LaneCount="2" SurfaceCount="2" SwathCount="4" TileCount="88" FlowcellSide="2">

This SampleSheet should help to clarify what we are trying to achieve here:

[Settings]                          
Adapter CTGTCTCTTATACACATCT                     
Adapter2    AGATCGGAAGAGCACACGTCTGAACTCCAGTCA                       
AdapterRead3    AGATCGGAAGAGCGTCGTGTAGGGAAAGAGTGT

Thanks,

[GenoMax]

The first library has a single adapter and two indexes per sample.

Are these regular Illumina indexes? According to the RunInfo.xml you have sequenced only one index read.

[moranr]

As far as I am aware, yes ? . Should it look something like this :

<read number="1" numcycles="151" isindexedread="N"/> <read number="2" numcycles="8" isindexedread="Y"/> <read number="3" numcycles="151" isindexedread="N"/> <read number="4" numcycles="8" isindexedread="Y"/>

?

[GenoMax]

Actually it should look like this.

<Read Number="1" NumCycles="151" IsIndexedRead="N"/>
<Read Number="2" NumCycles="8" IsIndexedRead="Y"/>
<Read Number="3" NumCycles="8" IsIndexedRead="Y"/>
<Read Number="4" NumCycles="152" IsIndexedRead="N"/>

Order of reads in Illumina tech is Read 1 --> Index 1 --> Index 2 --> Read 2.

I was worried that you did not sequence the second index becuase the example in main post is for single index reads.

[moranr]

Could it be that this was the XML for one of the library preps and not both (i.e. the single index prep)?

[GenoMax]

RunInfo.xml is for the entire flowcell. If the second index read has not been sequenced then that is bad news. This run would need to be repeated to get those samples which have 2 indexes.

[moranr]

Thanks for the clarification.

[moranr]

If we had something like this :

IsIndexedRead=N Y N

NumCycles=151 8 151

Number=1 2 3

This indicates that the RunInfo.xml was correct ??

Thanks,

[GenoMax]

That file is for a paired-end single index run. Only the first index was sequenced. If you need first and second index to distinguish between samples that have 2 indexes then there is no option but to repeat the run.

If this is an error on sequence provider's part then they should do the re-run at no cost to you.

Answer 1

[kylevoyto]

I create a separate SampleSheet for each type of index. If 4 of the samples have two indexes, then I will put those four samples into their own SampleSheet. I'll put the remaining 4 samples that have one index into their own SampleSheet as well.

[steve]

This is what we do as well. One samplesheet per type of indexes, run bcl2fastq separately for each sample sheet. Just make sure you give each one a different output directory.