As for #1 in your original post, we recently implemented a "Specimen Study" capability that lets you create a child study (much like an ancillary study) from a specimen request. The filtering is driven by the specimens in that request... the resulting study will include only those specimens plus dataset rows & groups associated with the associated participants. This is currently an experimental feature that needs to be enabled via the admin console.

#2 sounds like a bug... the ancillary study is apparently not resolving the samples query to the right folder; I'll open an issue to track this. Several institutions are continuing to fund improvements in the ancillary study / study publication area, though I don't think any of them have encountered your issue.

Adam