Interesting Nava, thanks a lot for this. I've always been puzzled why ONT isn't wiping out competitors if their accuracy is as good as the impression conveyed, in addition to their low capex, portability etc; I guess your analysis explains it.
How much improvement would ONT's duplex reads make here? IIRC they claim it brings them to Q29 modal.
Way easier said than done, but it'd be great to have SBS reads aligned to the same reference for comparison...
So, error rate is likely high. The error rate quoted here may not represent all runs, maybe there's run-to-run variation or different filtering would work. But I can't see how they can claim it's Q20. Illumina of course is closer to a Q30 average (high Q20s). Similarly PacBio.
Duplex reads are another question but they are claiming Q30... my guess is it gets closer to low Q20s (again average).
Interesting Nava, thanks a lot for this. I've always been puzzled why ONT isn't wiping out competitors if their accuracy is as good as the impression conveyed, in addition to their low capex, portability etc; I guess your analysis explains it.
How much improvement would ONT's duplex reads make here? IIRC they claim it brings them to Q29 modal.
Way easier said than done, but it'd be great to have SBS reads aligned to the same reference for comparison...
So, error rate is likely high. The error rate quoted here may not represent all runs, maybe there's run-to-run variation or different filtering would work. But I can't see how they can claim it's Q20. Illumina of course is closer to a Q30 average (high Q20s). Similarly PacBio.
Duplex reads are another question but they are claiming Q30... my guess is it gets closer to low Q20s (again average).