I am trying to use the function gtBroad with my own custom gene set collection created with GSEABase and an expression set but gtBroad is still asking for annotation. But when given it cannot annotate (because it is already there I think):

> gs
GeneSetCollection
  names: KEGG_GLYCOLYSIS_GLUCONEOGENESIS, KEGG_CITRATE_CYCLE_TCA_CYCLE, ..., Top_Positive_FC (903 total)
  unique identifiers: ACSS2, GCK, ..., ESRG (17543 total)
  types in collection:
    geneIdType: SymbolIdentifier (1 total)
    collectionType: NullCollection (1 total)
> > ALD2
ExpressionSet (storageMode: lockedEnvironment)
assayData: 13366 features, 50 samples
  element names: exprs
protocolData: none
phenoData
  rowNames: A01 A02 ... I98 (50 total)
  varLabels: Progression Edad ... AKI (25 total)
  varMetadata: labelDescription
featureData: none
experimentData: use 'experimentData(object)'
Annotation: org.Hs.eg.db

> gtBroad(ALD2, pData(ALD2), id = "Top_Positive_FC", collection = gs)
Error in gtBroad(ALD2, pData(ALD2), id = "Top_Positive_FC", collection = gs) :
  argument "annotation" is missing with no default.
> gtBroad(ALD2, pData(ALD2), id = "Top_Positive_FC", collection = gs, annotation="org.Hs.eg.db")
Error in .mapIdentifiers_isMappable(from, to) :
  unable to map from 'Symbol (org.Hs.eg.db)' to 'Annotation (org.Hs.eg)'
    'from' and 'to' annotations differ

I thought I could use subset to do the same as internally it uses a call to gt with subset, but I didn't manage to make it work:

gt(~Progression, exprs(ALD2), data = pData(ALD2), subsets = sets)
Error in eval(attr(terms(response, data = data), "variables"), data, environment(response))[[attr(terms(response,  :
  attempt to select less than one element in integerOneIndex

Maybe someone can point me to the right direction about how to test a geneSetCollection (or individual sets). Thanks