sprint 6: dark taxon

[slint]

Next steps

1. Create a sample dataset with specimens (including barcode sequences) + images
2. Review mapped fields to "standard" metadata and custom fields
3. Convert the sample to the Lycophron CSV input format using a script
4. Upload the sample on Sandbox
5. Go through some of the more real-life scenarios
   • Physical specimen without images
   • Images without sequences or both without sample
   • Metadata updates

---

Notes

• Full list of metadata fields
• Full list of custom metadata fields
• There's a full MIXS terms list as YAML. This can be used to generate labels for the terms
• We need to know which vocabulary each terms is from (mixs, dwc, ac, etc.)
• Types for fields
  • keyword: exact terms
  • text for free-text/flexible
  • float for decimal or intergers (e.g. sample size)
• We would like to display in the Event records, under the "Linked records" section of the page the search results for the referencing Physical specimens and Photos (see e.g. https://zenodo.org/records/15123769)
• Photos should link to the concept DOI of the Physical Specimen
• Interesting talk about storing genomics data in ElasticSearch: https://www.slideshare.net/slideshow/boston-meetupgoingorganic/25204280#1

---

[asrivathsan]

Genera guidelines: https://docs.gbif.org/publishing-dna-derived-data/en/

Barcode specific terms ( https://rs.gbif.org/extension/gbif/1.0/dna_derived_data_2024-07-11.xml )

GGBN: https://terms.tdwg.org/wiki/GGBN_Data_Standard

term	vocabulary	type
sampleSizeValue	dwc	text
sampleSizeUnit	dwc	text
associatedSequences	dwc	hyperlink to doi of sequence dataset, internal
eventID	dwc	keyword
samplingProtocol	dwc	text
locationID	dwc	keyword
habitat	dwc	text
countryCode	dwc	keyword
samplingEffort	dwc	text
geodeticDatum	dwc	keyword
occurrenceID	dwc	keyword
occurrenceStatus	dwc	keyword (actually true false)
occurrenceQuantity	dwc	float
parentEventID	dwc	keyword
datasetID	dwc	keyword
digitizationDate	ac	date
ampliconSize	gbif	float
DNA_sequence	gbif	text
pcr_primer_forward	gbif	keyword
pcr_primer_reverse	gbif	keyword
pcr_primer_name_forward	gbif	text
pcr_primer_name_reverse	gbif	text
sop	mixs	hyperlink to a doi, likely external
target_gene	mixs	keyword
target_subfragment	mixs	keyword
seq_meth	mixs	text
pcr_cond	mixs	text
assembly_software	mixs	text
env_broad_scale	mixs	keyword
env_local_scale	mixs	keyword
env_medium	mixs	keyword

additional features for technical table, no available terms, for barcode data

• Consensus_Coverage
• Ambiguties
• Gaps
• Sequencing Date

Link for the test dataset spreadsheet: https://docs.google.com/spreadsheets/d/1ipBjykoJrWLJN8BzpFnLv50gmRoNCrYBhN6OCwIl-XY/edit?usp=sharing

Link for data: https://drive.google.com/drive/folders/11NFI_mNgKwW8OkSJ953ZV9jw0eiMNfud?usp=sharing

[asrivathsan]

To avoid mixing up biodiversity information and sequence metadata fields: is it possible to list some of these under a different category of data
"DNA derived data"