Impact of phosphorothioation on prediction accuracy

[Ge0rges]

Hi Art,

I was wondering if nanopore has any internal metrics for how/if phosphorothioation impacts predictions of methylated bases (e.g. false positives)?

Thanks!

[marcus1487]

We do not have any benchmarks for this internally. Intuitively it would certainly introduce both false positives and false negatives as this signal is not seen in training. But the extent of these errors has not been investigated internally.