0% adapter bases after demultiplexing 10x data and "undetermined" sample

Question

[demoraesdiogo2017]

Hello

I am completely new to single cell pre-processing. I am currently trying to demultiplex a single cell output with BCL convert and I have checked the quality parameters, which I am not familiar with at all.

The adapter metrics look like this

Lane    Sample_ID   index   index2  ReadNumber  AdapterBases    SampleBases % Adapter Bases
1   s1  CGAAGTATAC  CTCCAAGTTC  1   0   5161879100  0
1   s1  CGAAGTATAC  CTCCAAGTTC  2   0   16591754250 0
1   s2  AAGTGGAGAG  GTAACAGGAA  1   0   5776440040  0
1   s2  AAGTGGAGAG  GTAACAGGAA  2   0   18567128700 0
1   s3  CGTGACATGC  TTTAGACCAT  1   0   809669112   0
1   s3  CGTGACATGC  TTTAGACCAT  2   0   2602507860  0
1   s4  TCCCAAGGGT  AAAGGTAGTA  1   0   460913348   0
1   s4  TCCCAAGGGT  AAAGGTAGTA  2   0   1481507190  0
1   s5  GCCTTGTCAA  TGATTGGATC  1   0   319250036   0
1   s5  GCCTTGTCAA  TGATTGGATC  2   0   1026160830  0
1   Undetermined            1   0   1153207664  0
1   Undetermined            2   0   3706738920  0

it seems the percentage of the adapter bases are all 0%. I assumed all of them should have the adapters in the index1 and 2, which should be what defines each sample. Can anyone tell if this is expected? The BCL convert documentation mentions these metrics, but it is not explained.

Additionally, there is supposed to be 5 samples in the lane, but an "undetermined" sample is showing up. What does that mean? Are those reads without the adapters?

Answer 1

[GenoMax]

Read 1 should contain cellbarcodes + UMI. In addition your samples are indexed by using standard Illumina indeses. You can see those indexes in the table above e.g. s1 CGAAGTATAC CTCCAAGTTC.

Adapter bcl-convert is referring to are Illumina adapters. Did you provide adapter sequence in your samplesheet and added option to trim the adapters? If you did not then you will always see 0's for that stat.

[demoraesdiogo2017]

oh, I think I mixed up the concept of indexes and adapters. I did not provide adapters in the samplesheet and did not enable trimming. I assume I will be able to do this in the alignment, is that correct? It all makes sense now. Thanks.

[GenoMax]

If you did not do adapter trimming that should be ok since cellranger or alevin-fry should be able to account for that. But you will need to provide correct Illumina indexes to demultiplex the samples in first step.