Property types

[bwalsh]

In the jsonld document, most properties are defined as "@type": "rdfs:Class" ( vs. "@type": "rdf:Property")

I noticed this defining numerics in the code vs in the schema

htan-portal/lib/dataSchemaHelpers.ts

Lines 139 to 167 in f496b61

	const NUMERICAL_SCHEMA_DATA_LOOKUP: { [schemaDataId: string]: boolean } = {
	[SchemaDataId.AgeAtDiagnosis]: true,
	[SchemaDataId.YearOfDiagnosis]: true,
	[SchemaDataId.DaysToLastFollowup]: true,
	[SchemaDataId.DaysToLastKnownDiseaseStatus]: true,
	[SchemaDataId.DaysToDiagnosis]: true,
	[SchemaDataId.PercentTumorInvasion]: true,
	[SchemaDataId.GrossTumorWeight]: true,
	[SchemaDataId.BreslowThickness]: true,
	[SchemaDataId.MitoticCount]: true,
	[SchemaDataId.MarginDistance]: true,
	[SchemaDataId.TumorDepth]: true,
	[SchemaDataId.DaysToBirth]: true,
	[SchemaDataId.CollectionDaysfromIndex]: true,
	[SchemaDataId.ProcessingDaysfromIndex]: true,
	[SchemaDataId.DysplasiaFraction]: true,
	[SchemaDataId.NumberProliferatingCells]: true,
	[SchemaDataId.PercentEosinophilInfiltration]: true,
	[SchemaDataId.PercentGranulocyteInfiltration]: true,
	[SchemaDataId.PercentInflamInfiltration]: true,
	[SchemaDataId.PercentLymphocyteInfiltration]: true,
	[SchemaDataId.PercentMonocyteInfiltration]: true,
	[SchemaDataId.PercentNecrosis]: true,
	[SchemaDataId.PercentNeutrophilInfiltration]: true,
	[SchemaDataId.PercentNormalCells]: true,
	[SchemaDataId.PercentStromalCells]: true,
	[SchemaDataId.PercentTumorCells]: true,
	[SchemaDataId.PercentTumorNuclei]: true,
	};

I also noticed type information embedded in comments. Parsing comments, I noticed the following "type" information in comments:

'"uint16"', "'null'", '(True/False)', '(mL).', '(mm).', '(um)', '(um).', ')', 'Integer', 'Integer.', 'String', 'Text.',   'URL',   '[Number]', '[URL]', '[number]', 'age', 'birthday.', 'centimeters.', 'cm', 'comma-separated', 'day.', 'days', 'days.', 'file', 'file.', 'filenames', 'files', 'grams.', 'kilograms.', 'length', 'list', 'mm', 'null', 'number', 'time', 'timepoint.'

Is there a convention on primitive type location?

[alliecreason]

Bump @milen-sage, we would appreciate your feedback here.