Fig. 1: Example rules and corresponding evidence chains generated from Healx KG. | Nature Communications

Fig. 1: Example rules and corresponding evidence chains generated from Healx KG.

From: An experimentally validated approach to automated biological evidence generation in drug discovery using knowledge graphs

Fig. 1

a An example path from the graph satisfying the rule (Compound – binds −> Gene <− activates – Compound – in trial for −> Disease) for a given candidate drug prediction “Lumacaftor” in cystic fibrosis. b Example rules learnt by the model and paths generated for each rule. c Example of an uninformative evidence chain containing two ancestor relationships. Yellow symbol indicates the type compound, white indicates disease, green indicates gene, red indicates phenotype and blue indicates pathway.

Back to article page