While sequencing metrics differ between sequencing platforms (e.g. MiSeq versus HiSeq 4000), they are generally consistent across runs on the same platform. We have found, however, that Single Cell 3’ v2 libraries sequenced on NextSeq 500/550 can show significant variability in sequencing metrics (especially for Read 2) across different flow cell lots compared to other sequencers. Our studies suggest that the flow cell surface chemistry may negatively impact sequencing quality depending on the flow cell lot number.
We have tested running the same library on different flow cell lots and observe Read 2 quality scores ranging from the upper 70s for one flow cell lot and between 20-40% for the worst performing flow cell lot. We consider a Q30 quality score less than 50% is to be poor. Overall, the majority of sequenced libraries we tested or our customers tested (55 out of 75) showed Read 2 Q30 scores greater than 50.
Low Q30 quality score may impact application performance. However, key metrics such as the number of “Valid Barcodes”, “Estimated Number of Cells”, “Fraction Reads in Cells”, and “Total Genes Detected” are largely unaffected by variation in Read 2 Q30 scores. Cluster analysis with Loupe Cell Browser indicates that individual PBMC subtypes, for instance, can be identified from even low quality sequencing runs. Application performance however will be sample dependent.
More information can be found in a Technical Note available on our website: https://support.10xgenomics.com/single-cell-gene-expression/index/doc/technical-note-chromium-single-cell-3-v2-libraries-sequencing-performance-on-illumina-nextseq-500-flow-cells