Efficient and biologically relevant consensus strategy for Parkinson’s disease gene prioritization

BMC Medical Genomics

Table 5 Connectivity, differential expression and machine learning data used as criteria for module prioritization

Healthy Control (HC) Modules
Module	n	<k>	<kintra>	<logPD-logHC>	nML	Merit_ML	nLimma	Merit_Limma	nML-Limma	Merit_ML-Limma
HC_01	123	12.04	1.38	−0.021	3	1.23	1	0.51	1	1.23
HC_02	349	34.57	7.29	−0.061	6	0.87	13	2.36	4	1.73
HC_03	1057	19.04	8.85	0.011	4	0.19	2	0.12	2	0.29
HC_04	169	17.02	2.59	−0.002	1	0.30	0	0.00	0	0.00
HC_05	347	9.23	2.59	0.165	2	0.29	1	0.18	1	0.44
HC_06	74	8.26	0.73	0.005	0	0.00	0	0.00	0	0.00
HC_07	290	14.81	5.19	0.073	4	0.70	6	1.31	1	0.52
HC_08	251	10.94	2.05	0.030	11	2.21	10	2.52	5	3.02
HC_09	2	1.15	0.00	0.022	0	0.00	0	0.00	0	0.00
HC_10	37	15.32	1.48	0.043	1	1.36	0	0.00	0	0.00
HC_11	91	10.95	1.23	0.048	3	1.66	0	0.00	0	0.00
HC_12	61	23.65	3.85	0.028	2	1.65	0	0.00	0	0.00
HC_13	164	10.23	1.79	0.007	3	0.92	1	0.39	1	0.92
HC_14	71	8.33	0.81	−0.001	0	0.00	0	0.00	0	0.00
HC_15	2120	49.53	36.69	−0.062	82	1.95	97	2.89	40	2.86
HC_16	3271	22.06	14.66	−0.064	46	0.71	3	0.06	1	0.05
Parkinson’s Disease (PD) Modules
PD_01	603	286.30	70.52	0.022	6	0.50	1	0.10	1	0.25
PD_02	1437	262.21	150.85	−0.126	69	2.42	103	4.53	42	4.42
PD_03	133	210.12	13.36	0.035	1	0.38	0	0.00	0	0.00
PD_04	161	284.83	22.96	0.089	4	1.25	3	1.18	2	1.88
PD_05	789	231.70	62.45	−0.025	5	0.32	1	0.08	0	0.00
PD_06	468	238.37	38.64	0.132	3	0.32	0	0.00	0	0.00
PD_07	494	316.82	58.43	0.103	24	2.45	19	2.43	8	2.45
PD_08	213	218.15	28.17	−0.033	4	0.95	2	0.59	1	0.71
PD_09	4179	333.39	247.08	−0.047	52	0.63	5	0.08	2	0.07

n: number of genes in the module; <k>: average node degree; <k _intra >: intra-modular average node degree; <logPD-logHC>: module average differential of the log transformed average expression of a gene i across PD samples and healthy control samples; n _ML: number of genes identified by ML analysis included in the module; n _Limma: number of genes identified by Limma analysis included in the module; n _ML-Limma: number of common genes identified by both ML and Limma analyses included in the module; Merit_ML = (n _ML/168)/(N/8477): merit assigned to the module based on n _ML, the total number of genes identified by ML analysis (168), N, and the total number of background genes (8477); Merit_Limma = (n _Limma/134)/(N/8477): merit assigned to the module based on n _Limma, the total number of genes identified by Limma analysis (134), N, and the total number of background genes (8477); Merit_ML-Limma = (n _ML-Limma/56)/(N/8477): merit assigned to the module based on n _ML-Limma, the total number of common genes identified by both ML and Limma analyses (56), N, and the total number of background genes (8477)

ISSN: 1755-8794