Share samples and mixes between experiments

[jakebeal]

If we upload samples from two different experiments (e.g., JHT6 and JHT10), there will often be some elements that are actually the same for both experiments. Can we either detect this or else use deterministic naming such that we end up combining them rather than duplicating sample designs?