. 2008 Apr 4;82(4):949–958. doi: 10.1016/j.ajhg.2008.02.013

Table 2.

Performance of Five Candidate-Gene-Prioritization Methods on Seven Recently Identified Monogenic Disease Genes

Family	Gene	Rankings
Family	Gene	Random Walk	ENDEAVOUR	SP	DI	SQ
Nephronophthisis	GLIS2³⁷	100	43	100	100	3^∗
ARVD	JUP³⁸	1^∗	1^∗	1^∗	2	67
RP	TOPORS³⁹	23	69	20^∗	100	56
RP	NR2E3⁴⁰	2	2	18	100	1^∗
Noonan Syndrome	RAF1⁴¹	1^∗	3	4	4	42
Brachydactyly	NOG⁴²	1^∗	5	1^∗	1^∗	34
CMT4H	FGD4⁴³	13	2^∗	27	100	9

Mean Enrichment		25.9^∗	18.4	17.2	12.8	10.9

Results of random walk, two local network algorithms, ENDEAVOUR,¹⁰ and the sequence analysis program PROSPECTR¹² for the prediction of recently published genes causing monogenic diseases within artificial linkage intervals containing 100 genes.

“SP” denotes ranking according to shortest path.

“DI” denotes ranking according to direct interaction with a known disease protein.

“SQ” denotes ranking by sequence analysis with PROSPECTR.

“ARVD” denotes arrhythmogenic right ventricular dysplasia.

“RP” denotes retinitis pigmentosa.

“CMT4H” denotes Charcot-Marie-Tooth type 4H.

^∗

indicates best performance.