Toronto Dataset

Classification of protein interactions

For a list of protein pairs (PPI) we consider the following attributes:

Column	Description	Feature Type	Data Type	Source
Human PPI	(1 = interaction annotated; 0 = not)	Categorical	integer	[1], [2]
Fly PPI	(1 = interaction annotated; 0 = not)	Categorical	integer	[1], [2]
Worm PPI	(1 = interaction annotated; 0 = not)	Categorical	integer	[1], [2]
Yeast PPI	(1 = interaction annotated; 0 = not)	Categorical	integer	[1], [2]
Domain PPI	(1 = interaction annotated; 0 = not)	Categorical	integer	[3]
Co-expressed	(1 = yes; 0 = no)	Categorical	integer	[4]
GO BP sharing	(1 = share GO biological process, 0 = not)	Categorical	integer	[5]
Target	Functional (1 = true; 0 = false)	Classification	integer	[6]

The list of PPIs is composed of two types: functional and non-functional.

• Functional: PPIs derived from Reactome pathway database^[6]
• Non-functional: PPIs created selecting random pairs of proteins already in participating in pathways.

References

[1] IntAct : Orchard, Sandra, et al. "The MIntAct project—IntAct as a common curation platform for 11 molecular interaction databases." Nucleic acids research 42.D1 (2014): D358-D363.
[2] Biogrid : Oughtred, Rose, et al. "The BioGRID interaction database: 2019 update." Nucleic acids research 47.D1 (2019): D529-D541.
[3] pFam: Oughtred, Rose, et al. "The BioGRID interaction database: 2019 update." Nucleic acids research 47.D1 (2019): D529-D541.
[4] COXPRESdb: Obayashi, Takeshi, et al. "COXPRESdb v7: a gene coexpression database for 11 animal species supported by 23 coexpression platforms for technical evaluation and evolutionary inference." Nucleic acids research 47.D1 (2019): D55-D62.
[5] GO: Ashburner, Michael, et al. "Gene ontology: tool for the unification of biology." Nature genetics 25.1 (2000): 25-29.
[6] Reactome: Jassal, Bijay, et al. "The reactome pathway knowledgebase." Nucleic acids research 48.D1 (2020): D498-D503.