GLAD4U: deriving and prioritizing gene lists from PubMed literature

Table 1 Overall quality of the retrieved gene lists

Query	GO/ MIM gene count	GLAD4U gene count	EBIMed gene count		GLAD4U	EBIMed
Apoptosis	1039	6037 (958)	1469 (387)	Precision	0.1587	0.2634
		[195715]	[10000]	Recall	0.9220	0.3725
				F-measure	0.2708	0.3086
Cell adhesion	785	4195 (691)	1725 (305)	Precision	0.1647	0.1769
		[125144]	[10000]	Recall	0.8802	0.3885
				F-measure	0.2775	0.2431
DNA repair	282	2476 (263)	1100 (180)	Precision	0.1062	0.1636
		[60952]	[10000]	Recall	0.9326	0.6383
				F-measure	0.1907	0.2605
Hypertension	87	2046 (77)	135 (27)	Precision	0.0376	0.2000
		[323818]	[10000]	Recall	0.8851	0.3103
				F-measure	0.0721	0.2432
Obesity	111	1778 (110)	350 (59)	Precision	0.0619	0.1686
		[141615]	[10000]	Recall	0.9910	0.5315
				F-measure	0.1165	0.2560
Schizophrenia	94	1725 (90)	382 (44)	Precision	0.0522	0.1152
		[91194]	[10000]	Recall	0.9574	0.4681
				F-measure	0.0990	0.1849

Numbers in parentheses indicate the number of genes overlapping between the GLAD4U or EBIMed lists and the corresponding gold standard, numbers in square brackets indicate the number of publications retrieved by the query.