mkfastq scATAC 10X genomics -- index read issue

Question

[bkalish1]

Hello, I am analyzing scATAC-seq that was performed using 10X genomics v1.1. For sequencing, the index (i7) read is supposed to be 8bp, but when the libraries were sequenced, the i7 read was 16bp. Therefore, when I run cell ranger-atac mkfastq, I receive an error that the length of the index read does not match the length of the specified indices for the samples. Is there a way to use --use-bases-mask option to correct this issue so I can properly demultiplex samples? The library was sequenced on a NextSeq 500.

Thanks! Brian

[bkalish1]

Unfortunately that did not work. I tried --use-bases-mask Y50n,I8n,I16n,Y50n, but this yielded this error: "The total number of cycles specified by the use-bases-mask does not match the number found in RunInfo.xml. Masks: y50n,i8n,i16n,y50n'"

The RunInfo.xml file specifies 16 for the i7 index.

[GenoMax]

Can you verify what the reads section looks like in your your RunInfo.xml file? One for 10x scATACseq should look like this.

                 <Reads>
                        <Read Number="1" NumCycles="50" IsIndexedRead="N"/>
                        <Read Number="2" NumCycles="8" IsIndexedRead="Y"/>
                        <Read Number="3" NumCycles="16" IsIndexedRead="Y"/>
                        <Read Number="4" NumCycles="50" IsIndexedRead="N"/>
                </Reads>

If it looks different then please paste a copy here.

Check the change I made to my answer below. I forgot to add the second index spec before.

Answer 1

[GenoMax]

Try --use-bases-mask Y*,I8n*,I16,Y* when you run cellranger-atac mkfastq and see if that fixes the issue.