NovaSeq error rate?

[ATϟGC]

I was wondering if anybody can quote the error rate for a NovaSeq? Even an idea of how it compares to MiSeq and HiSeq would be great as I can not seem to find any comparisons or hard numbers anywhere.

I have seen that Brian Bushnell has had problems with index misassignment and has expressed that the quality scores may be inflated from this post here:

https://www.biostars.org/p/259228/

That was back in 2017 so maybe those issues have been largely resolved.

Thanks for any information you can add.

[GenoMax]

It should basically be very similar to other Illumina sequencers. Quality scores on NovaSeq are binned by default so they would not be directly comparable. BBMap may be able to recalibrate the scores but doing that for billions of reads each time may not add a lot of value but will add time to the analysis.

We have tested 48 RNAseq libraries that were originally sequenced on HiSeq 2500 on NovaSeq. The correlation of values was near perfect.

Where possible you should always use unique dual indexes to avoid possible issues with index hopping. Depending on quality of your libraries (insert size needs to be carefully controlled) you would need to worry about getting optical duplicates. Those can easily be checked using `clumpify.sh` from BBMap suite.

[ATϟGC]

Thanks Genomax. It is good to hear that it is comparable to other Illumina instruments and that dual indexes should mitigate the hopping.