Conversation
LinkML Linting ResultsSummary
Problems per Schema/home/runner/work/mixs/mixs/src/mixs/schema/mixs.yamlErrors
|
|
| - marker gene | ||
| - taxonomy | ||
| slot_uri: MIXS:0001337 | ||
| range: boolean |
There was a problem hiding this comment.
maybe I'm just crazy.. I thought boolean was TRUE FALSE but you have yes.. so wouldn't this need to be a yes / no enum?
There was a problem hiding this comment.
Good catch, and no you're not crazy. xsd:boolean only accepts true/false/1/0 — the yes example is technically invalid against the declared range: boolean. This is inherited from x16s_recover and affects 5 of the 6 boolean-range slots in MIxS (all use yes or no examples).
The good news: linkml/linkml#3144 (merged to main 2026-03-25, not yet released) adds schema-aware boolean coercion for CSV/TSV. MIxS could add boolean_truthy/boolean_falsy annotations to accept yes/no during loading while outputting canonical true/false.
For this PR, I'll change the example to true to be correct against the current spec. The broader boolean story is tracked in #1191.
2655164 to
32b4a17
Compare
LinkML Linting ResultsSummary
|
There was a problem hiding this comment.
Pull request overview
This PR extends the MIxS LinkML schema to introduce taxon-neutral “marker gene recovery” slots intended to complement existing 16S-specific recovery terms, and wires those new slots into relevant genome checklist classes.
Changes:
- Add new slots
marker_gene_recovandmarker_gene_recov_swto the schema. - Include the new slots in the
MimagandMisagchecklist slot lists.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
Merge order and requirementsThis PR should merge before #755, which tests for digit-first slot names — a check that currently fails on 20-character structured comment name limitThe INSDC Feature Table Specification (Section 3.1) sets a 20-character maximum for qualifier names. The GSC TWG reaffirmed this constraint in issue #763 (Feb 2024, decision: option 3 — keep the limit, leave existing names as-is, apply the constraint going forward). Both new slot names here are within the limit:
Outstanding Copilot review itemsCopilot raised several actionable points (duplicate Related
|
ID assignment updateNew MIXS IDs claimed for this PR's slots:
Registered on both the MIxS new IDs and MIXS ID Space tabs. Schema updated to use these IDs (previously had a conflicting reuse of MIXS:0001338, which belongs to See #1004 for broader discussion of ID assignment policy. |
marker_gene_recov → MIXS:0001354 marker_gene_recov_sw → MIXS:0001355 Previously reused MIXS:0001337 and MIXS:0001338, which belong to prod_label_claims and isotope respectively. New IDs registered in the MIxS IDs spreadsheet.
LinkML Linting ResultsSummary
|
Mark both slots as deprecated with exact replacements pointing to the taxon-neutral marker_gene_recov (MIXS:0001354) and marker_gene_recov_sw (MIXS:0001355) from PR #974. First release cycle of the two-step deprecation process per src/docs/schema_element_deprecation_guide.md. Closes #1186. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Use deprecated_element_has_exact_replacement for machine-readable replacement pointers (MIXS:0001354, MIXS:0001355) - Clean up deprecated: field to be a human-readable reason string, not a redundant restatement of the replacement CURIE - Add modified_by with ORCID - Add see_also linking to both the tracking issue (#1186) and the incoming replacement terms PR (#974) Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
xsd:boolean only accepts true/false/1/0. The 'yes' example was inherited from x16s_recover but is technically invalid. Broader boolean coercion support is coming via linkml#3144; tracked in #1191. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
LinkML Linting ResultsSummary
|
Does not disassociate, deprecate or remove