Table 5.

Performance using gene2pubmed as the gene-literature data source

Scoring method	Novel MEDLINE validation AUC (02/2007-01/2009)	Novel MEDLINE validation AUC (02/2007-04/2010)	Pre-existing CTD validation AUC (11/2008)	Novel CTD validation AUC (11/2008-04/2010)	Pre-existing MEDLINE validation AUC (02/2007)	Mean AUC	Rank
Cosine distance of term frequency-inverse document frequency	0.92	0.91	0.95	0.93	0.98	0.94	2
Cosine distance of P-values	0.53	0.51	0.65	0.63	0.53	0.57	16
Cosine distance of term fractions	0.90	0.89	0.93	0.91	0.96	0.92	5
Sum of the log of combined P-values	0.91	0.89	0.94	0.94	0.94	0.92	3
Sum of the differences of log P-values	0.91	0.91	0.77	0.83	0.93	0.87	7
L2 of log-p of overlapping terms only	0.96	0.95	0.92	0.94	0.99	0.95	1
L2 of term fractions of overlapping terms only	0.64	0.62	0.57	0.60	0.53	0.59	15
L2 of log of P-values	0.90	0.90	0.76	0.83	0.93	0.86	10
L2 of P-values	0.89	0.89	0.75	0.81	0.92	0.86	12
L2 of term fractions	0.92	0.90	0.91	0.92	0.95	0.92	4
L2 of term frequency	0.90	0.90	0.76	0.82	0.93	0.86	11
Term coverage	0.90	0.91	0.77	0.83	0.93	0.87	8
Term overlap	0.91	0.89	0.90	0.92	0.90	0.90	6
Number of gene MeSH terms	0.85	0.82	0.85	0.88	0.83	0.85	13
Number of disease MeSH terms	0.90	0.90	0.76	0.83	0.93	0.86	9
Gene ID	0.75	0.73	0.78	0.79	0.74	0.76	14

AUC of the described scoring methods were compared and tested on the validation sets. CTD, Comparative Toxicogenomics Database.