vinoo-igem
commented
2 years ago Just for memory/documentation sake. In SBOL, an implementation would reflect an aliquot/sample of plasmid DNA. So, at any one point this could be a sample received by FreeGenes directly from TWIST, a sample after some manipulation by FreeGenes, a sample within a 384w distribution plate, etc. So in this particular case, for synthesis status we really just care about the first, right? That it exists at FreeGenes after a synthesis run. I'm assuming there's some method of provenance for implementation so that those latter examples can come after.
To throw out some things we'd want to track of pre-synthesis (decided to), not exhaustive.
- choosing not to synthesize: ex. not Type IIS compatible (#17), curational reasons (redundant, not useful, newer version preferred)
- inability to synthesize: doesn't meet TWIST's requirements (this currently would be done during a synthesis run, but in the future we'd want to tap into the API to determine this before a synthesis run)
during a synthesis run (placed at Twist - to succeeded/failed/abandoned/etc)
- order placed
- failure modes of synthesis for a construct (or more likely, at what stage this failed)
- success
We need to represent the workflow of synthesis orders progressing from decided to placed at Twist to succeeded/failed/abandoned/etc. to actually available at FreeGenes.
I think this can be done with a combination of PROV-O, Implementation, and Collection but this needs to be carefully thought through and a specification document written up.