GO Bench Supplementary Data

GO Bench Swissprot Dataset
Swissprot download link containing sequences (in fasta format) for each protein included in GO Bench datasets. Sequences are not included directly from GO_Bench, so they must be sourced from here and combined with GO_Bench annotations for modeling. (See GO_Bench_Sample code on github for examples)

Gene Ontology
Link to the go-basic.obo gene ontology file used by GO Bench. Frequently useful for structured modeling of protein annotations, or for reproducing GO Bench annotation propagation.

Gene Ontology Annotation Database
Link to the Gene Ontology Annotation database used to construct GO Bench. Use for reproducing GO Bench, or for additional data.

Negative Benchmark Dataset
Link to data and code for the Negative Benchmarking dataset by Dessimoz Lab. All GO Bench training datasets are built to be compatible with Negative Benchmarking dataset, which can be used for higher confidence testing under some circumstances.