Right now the seqneut-pipeline is taking many hours to run mostly because the notebooks that plot all the neutralization curves take >5 hours to run. This needs to probably be somehow addressed by fixing how this is handled in neutcurve or seqneut-pipeline. There are also then downstream errors in the final plotting.
@jbloom will work on getting this fixed.