Add samplingProtocol to eDNA MBARI dataset

[MathewBiddle]

Dataset: https://obis.org/dataset/62b97724-da17-4ca7-9b26-b2a22aeaab51
Protocols: https://mbari-bog.github.io/MBON-Protocols/

I think these would fit under samplingProtocol but should get confirmation from @albenson-usgs first.

If so, I can pull in the DwC-A from https://ipt-obis.gbif.us/archive.do?r=18s_monterey_bay_time_series_edna&v=1.2 and add samplingProtocol to the occurrence.txt file. Then, resubmit.

[albenson-usgs]

When you say "these" what do you mean in practice? I.e. what exactly are you going to put into samplingProtocol? Would it be "https://mbari-bog.github.io/MBON-Protocols/" I would prefer to see a DOI there to the protocol.

[MathewBiddle]

'these' = the protocols at https://mbari-bog.github.io/MBON-Protocols/

I guess I skipped some steps in the comment above.

My proposal was to add the column below to the occurrence.txt file (of course rows repeated for each occurrence record).

samplingProtocol
https://mbari-bog.github.io/MBON-Protocols/

I would prefer to see a DOI there to the protocol.

Agreed. Those docs are hosted out of https://github.com/MBARI-BOG/MBON-Protocols. If a DOI is what we'd like, we can request the maintainers of that repository to mint a DOI through Zenodo. However, we should keep in mind that the DOI will reference the source files (markdown), not the pretty html webpage.

[MathewBiddle]

@sformel-usgs maybe we could talk through this during an SMBDWG meeting?

[sformel-usgs]

Sorry for being slow to answer this. What gaps do these protocols fill in that aren't in the protocols referenced in the DNA derived data extension? From a quick look, here is what already seems to be covered as it is currently published:

DNA extraction: dx.doi.org/10.17504/protocols.io.xjufknw (Nucl_acid_ext field)
PCR/library prep: dx.doi.org/10.17504/protocols.io.n2vdge6 (Nucl_acid_amp field)
Bioinformatics: https://github.com/MBARI-BOG/BOG-Banzai-Dada2-Pipeline (sop field along with above two)

Anyway, yes, I think that's a good idea to discuss, since we hold this up as an example pretty often. Let's put it on the agenda for our next meeting on Nov 15th.

Also, if we're planning on revising this dataset, I think there is some other low-hanging fruit I've noticed before that we could add to make the data a little more robust to interpretation. Fields like occurrenceStatus, and `min/maximumDepthInMeters' and some metadata, like Project metadata.

[MathewBiddle]

ahhh, I might have been confused with the eDNA extension. Thank you for asking these questions.

I was initially looking at the guidance for the occurrence table https://ioos.github.io/bio_mobilization_workshop/edna-extension/#tables and saw that samplingProtocol wasn't in this dataset's occurrence table. Working with the team at MBARI we were able to find the protocols at the url I linked above. So, I was hoping to add them to the record to align with the guidance.

[MathewBiddle]

I don't know if they supplement/summarize the other links you shared. If it's not necessary to have samplingProtocol than we can close this ticket.

[sformel-usgs]

samplingProtocol can also serve as a space for describing how the water samples were taken, since I don't think that is described anywhere in the DNA protocols, but is descibred in the related paper:

Djurhuus, A, Closek, CJ, Kelly, RP et al. (2020). Environmental DNA reveals seasonal shifts and potential interactions in a marine community. Nat Commun 11, 254. https://doi.org/10.1038/s41467-019-14105-1

This might be a conversation for another ticket, but is there someone at MBARI that we could work with the make some of the improvements I mentioned above?

[MathewBiddle]

@kpitz maybe?