Evaluation of glycowork

[ReneRanzinger]

There was a talk about glycowork in the GLIC seminar series. Part of the python package is a glycan library with more than 50,000 records. Not all records are complete but they can have the following information:

• glycan
• Species
• Genus
• Family
• Order
• Class
• Phylum
• Kingdom
• Domain
• ref
• glytoucan_id
• predicted_taxonomy
• glycan_type
• immunogenicity
• disease_association
• disease_id
• disease_sample
• disease_direction
• disease_ref
• disease_species
• tissue_sample
• tissue_id
• tissue_ref
• tissue_species
• Composition

Which include taxon annotation, reference, disease and tissue information. A quick count showed:

• ~9000 records Human
• ~2000 records Mouse
• ~ 500 records Rat

Partially from other databases but also manual curation.

Careful! Not all records have GlyTouCan IDs. That is something to work on.

The database can be found here.

[Mindyporterfield]

https://github.com/BojarLab/glycowork/blob/master/05_examples.ipynb here are some examples of what you can do in it, could be a good spark for ideas @ReneRanzinger @kmartinez834 @edwardsnj

[kmartinez834]

Changing assignment to Cyrus, as he's been looking into Glycowork