Rule based glycan structure dictionary terms

[ReneRanzinger]

The attached file: Rulebased for Nathan.xlsx

Contains the 51 terms that we could not associate with a GlyTouCan ID and a structure/substructure search. We believe they could be solved by a set of rules. For example:

"1.77 Fucosylated asialo-biantennary" can be assigned to a glycan that:

• is a N-glycan AND
• has # branches == 2 AND
• has # NeuAc == 0 AND
• has # Fuc > 0

The file describes the rules:

• Column A: ID of the term from the wiki page and the term name
• Column B: Notes
• Column C: GlyGen Glycan Type
• Column D: GlyGen Glycan SubType
• Column E: Number of N-Glycan branches (aka chield residues of the 3 Man and 6 Man of the N-Glycan core)
• Column F: Number of NeuAc (might change to Number of all sialic acids)
• Column G: Number of Fuc
• Column H: Reducing end is an alditol (yes/no)
• Column I: Number of Sulfate
• Column J: not used
• Column K,L: GlyGen Motifs

If a column is not empty the glycan has to match it. If multiple columns are not empty they are AND connected except for K and L which are AND connected to the rest but OR with each other.

e.g. Row 5: Glycan Type == GAG AND Number of Sulfates > 0 AND ( has motif GGM.001212 OR has motif GGM.001211 )

Please see if we can solve this with the GlycoMotif infrastructure.

[edwardsnj]

I have laid some groundwork for this, will add new "arm"-counting motifs to enable this. I will revisit this after SfG

[edwardsnj]

The next glycan-data reslease will have N-linked biantennary, N-linked triantennary, and N-linked tetraantennary subtypes.

[edwardsnj]

@ReneRanzinger @Mindyporterfield Please take a look at the N-linked biantennary, N-linked triantennary, and N-linked tetraantennary subtypes (in beta) and see whether it provides you sufficient information for these rules - N-linked complex is required, so this is a restriction of the complex subtype.