Measuring semantic similarities by combining gene ontology annotations and gene co-function networks

BMC Bioinformatics

Table 1 Comparison of NETSIM ’s performance to other similarity measures

Similarity	Data type used	Median (25th Percentile) LFC score
Measure	Data type used	Yeast	Arabidopsis	human
co-function	Co-function network	0.03(0.00)	0.00(0.00)	0.00(0.00)
Resnik	GO annotation	0.26(0.10)	0.07(0.01)	0.29(0.19)
Yu	GO annotation	0.32(0.10)	0.14(0.02)	0.78(0.48)
Wang	GO annotation	0.26(0.11)	0.16(0.04)	0.69(0.52)
Schlicker	GO annotation	0.33(0.14)	0.24(0.10)	0.71(0.46)
NETSIM	GO annotation,	0.76(0.35)	0.27(0.10)	1.10(0.83)
NETSIM	co-function network	0.76(0.35)	0.27(0.10)	1.10(0.83)

Comparison of NETSIM’s performance to other similarity measures based on the median and the first quartile of Log-transformed Fold Change (LFC) scores. Numbers in bold indicate the best performance for each species.

ISSN: 1471-2105