Fig. 7

Necessity of using Experimental Factor Ontology and respecting ontology tree-like structure in SNP-based enrichment analysis. This is demonstrated using the disease subgraph of the Experimental Factor Ontology (EFO) and analysing cis-eQTLs induced by 24-h IFN-γ. a Side-by-side bar plots comparing the significant EFO terms between the analysis not using EFO (conventional analysis; EFO (-)) and two ontology-based analyses: the EFO (+) & Tree (-) analysis using EFO but without respecting the ontology tree, and the EFO (+) & Tree (+) analysis using EFO and also respecting the ontology tree. The horizontal dotted lines separate commonly identified terms (top), the terms unique to the ontology-based analyses (middle), and the redundant terms identified by the EFO (+) & Tree (-) analysis (bottom). b DAG plot comparing terms identified by all analyses (coloured in cyan), by two analyses (coloured in light cyan), and only by one analysis (coloured in lightest cyan). The term name (if significant) is prefixed in the form of ‘x1-x2-x3’. In this case, x1 for ‘EFO (-)’, x2 for ‘EFO (+) & Tree (-)’, x3 for ‘EFO (+) & Tree (+)’. The value of x1–3 can be ‘1’ or ‘0’, denoting whether this term is identified (present) or not (absent)