Remove unit and definition fields from schema and terms, get from CV …

…when building markdown docs instead

.github/workflows/scripts/generate_metadata_docs.py

@@ -10,9 +10,12 @@
 
 from jsonschema import validate
 from tomark import Tomark
+import psims
 
 
 def rules_to_markdown(rules):
+    cv = psims.load_psims()  # load a fresh copy of the PSI-MS CV mapping
+
     rule_dict = defaultdict(lambda: defaultdict(list))
     for rule in rules["rules"]:
         rule_dict[rule["path"]][rule["requirement_level"]].append(rule)
@@ -31,13 +34,19 @@ def rules_to_markdown(rules):
             for attr in rule_attrs:
                 field = dict()
 
+                # Attempt to lookup term in CV
+                try:
+                    cv_term = cv[attr["accession"]]
+                except KeyError:
+                    cv_term = None
+
                 # Parse name and accession fields
                 field["Name"] = f"{attr['name']} ({attr['accession']})"
 
                 # Parse definition field
                 info = []
-                if "definition" in attr:
-                    info.append(attr["definition"])
+                if cv_term:
+                    info.append(cv_term.definition)
                 if "notes" in attr:
                     info.append(f"**Notes:** {attr['notes']}")
                 field["Info"] = "<br /><br />".join(info).replace(
@@ -56,9 +65,15 @@ def rules_to_markdown(rules):
                     field["Value"] = "Undefined"
 
                 # Parse units field
-                field["Allowed units"] = (
-                    ", ".join(attr["units"]) if "units" in attr else "/"
-                )
+                if cv_term:
+                    try:
+                        units = [u.comment for u in cv_term.has_units]
+                    except AttributeError:
+                        units = None
+                else:
+                    units = None
+
+                field["Allowed units"] = ", ".join(units) if units else "/"
 
                 # Parse repeatable field
                 field["Repeatable"] = attr["repeatable"]

.github/workflows/scripts/requirements.txt

@@ -1 +1,3 @@
 tomark
+jsonschema
+psims

docs/metadata-rules.md

@@ -123,7 +123,7 @@
 
 | Name | Info | Value | Allowed units | Repeatable |
 |-----|-----|-----|-----|-----|
-| charge state ([MS:1000041](https://www.ebi.ac.uk/ols4/ontologies/ms/classes/http%253A%252F%252Fpurl.obolibrary.org%252Fobo%252FMS_1000041)) | "Number of net charges, positive or negative, on an ion." [PSI:MS]<br /><br />**Notes:** The “charge state” attribute has two different meanings depending on its context. When this attribute is included under <Spectrum=...>, it is used to denote the experimental charge state of the precursor as inferred from the data (e.g. by the isotopic pattern of the precursor peak in the MSn-1 spectrum). If the spectrum is identified to an analyte, the charge state refers to that of the analyte believed to produce this spectrum. In such cases, the charge state attribute should be included in the <Analyte=> section instead. | xsd:integer | / | False |
+| charge state ([MS:1000041](https://www.ebi.ac.uk/ols4/ontologies/ms/classes/http%253A%252F%252Fpurl.obolibrary.org%252Fobo%252FMS_1000041)) | "Number of net charges, positive or negative, on an ion." [PSI:MS]<br /><br />**Notes:** The "charge state" attribute has two different meanings depending on its context. When this attribute is included under <Spectrum=...>, it is used to denote the experimental charge state of the precursor as inferred from the data (e.g. by the isotopic pattern of the precursor peak in the MSn-1 spectrum). If the spectrum is identified to an analyte, the charge state refers to that of the analyte believed to produce this spectrum. In such cases, the charge state attribute should be included in the <Analyte=> section instead. | xsd:integer | / | False |
 | theoretical mass ([MS:1001117](https://www.ebi.ac.uk/ols4/ontologies/ms/classes/http%253A%252F%252Fpurl.obolibrary.org%252Fobo%252FMS_1001117)) | "The theoretical neutral mass of the molecule (e.g. the peptide sequence and its modifications) not including its charge carrier." [PSI:PI]<br /><br />**Notes:** This is the theoretical mass of the neutral molecule (does not include the charge-giving moiety. Use the term “[MS:1003243](https://www.ebi.ac.uk/ols4/ontologies/ms/classes/http%253A%252F%252Fpurl.obolibrary.org%252Fobo%252FMS_1003243)\|adduct ion mass” to refer to the mass of the adduct ion, which includes the charge-giving moiety. | xsd:double | dalton | False |
 | adduct ion mass ([MS:1003243](https://www.ebi.ac.uk/ols4/ontologies/ms/classes/http%253A%252F%252Fpurl.obolibrary.org%252Fobo%252FMS_1003243)) | "The theoretical mass of the adduct ion (e.g. for a singly-charged protonated peptide ion, this value would be the neutral peptide molecule's mass plus the mass of a proton)" [PSI:PI] | xsd:float | / | False |
 | adduct ion formula ([MS:1002813](https://www.ebi.ac.uk/ols4/ontologies/ms/classes/http%253A%252F%252Fpurl.obolibrary.org%252Fobo%252FMS_1002813)) | "Adduct formation formula of the form M+X or M-X, as constrained by the provided regular expression." [PSI:MS]<br /><br />**Notes:** For peptides, absence of this field implies that the adduct ion is [M+nH]n+ where n is the charge state. The regular expression for this field is given in the definition of [MS:1002812](https://www.ebi.ac.uk/ols4/ontologies/ms/classes/http%253A%252F%252Fpurl.obolibrary.org%252Fobo%252FMS_1002812)\|Regular expression for adduct ion formula. | xsd:string | / | False |