metasoarous
commented
6 years ago I've decided this isn't really too big a deal if we just add extra information to the table view in order to clarify that there might be different seed records per sample in the table. Ultimately, we don't really need to change how we represent things here. Whether semantically it makes sense to is a little more up for discussion, but I don't think a major priority right now, and wouldn't likely change anything in how we interface with the data anyway.
Right now, we create multiple seed entities for what should be a single seed entity, because of how we nest seeds per sample now (what we had before did the same thing to samples and wasn't tenable with unseeded partitions). I think the ticket here is to go the route I initially envisioned of specifying a scope wherein a given seed id must be unique as the tripl.nestly machine does its magic. May require some tweaking there.