A robust and stable gene selection algorithm based on graph theory and machine learning

Saha, Subrata; Soliman, Ahmed; Rajasekaran, Sanguthevar

doi:10.1186/s40246-021-00366-9

Human Genomics

Table 3 Performance of multi-class gene selection

From: A robust and stable gene selection algorithm based on graph theory and machine learning

		Average Jaccard indices (\({JI}_{avg}\))						Average informedness (\({I}_{avg}\))
Dataset	Top genes	SU	GR	KLD	RELIEF	SVM-RFE	RSGSA	SU	GR	KLD	RELIEF	SVM-RFE	RSGSA
D6	50	0.34	0.46	0.10	0.30	0.30	0.52	0.93	0.94	0.87	0.94	0.96	0.98
	100	0.32	0.46	0.08	0.27	0.36	0.58	0.94	0.94	0.90	0.95	0.96	0.99
	150	0.31	0.43	0.08	0.27	0.36	0.61	0.94	0.94	0.90	0.96	0.96	0.99
	200	0.30	0.42	0.09	0.27	0.38	0.62	0.94	0.94	0.92	0.95	0.96	0.99
D7	50	0.39	0.49	0.12	0.17	0.25	0.65	0.90	0.90	0.82	0.84	0.92	0.99
	100	0.35	0.48	0.13	0.18	0.28	0.62	0.90	0.91	0.89	0.86	0.94	0.99
	150	0.34	0.45	0.13	0.18	0.31	0.64	0.90	0.91	0.90	0.87	0.95	0.99
	200	0.33	0.44	0.12	0.18	0.32	0.68	0.90	0.91	0.90	0.88	0.94	0.99
D8	50	0.60	0.61	0.19	0.20	0.22	0.50	0.89	0.91	0.89	0.90	0.95	1.00
	100	0.60	0.69	0.17	0.24	0.28	0.54	0.91	0.91	0.91	0.91	0.96	1.00
	150	0.60	0.70	0.17	0.26	0.31	0.57	0.92	0.91	0.93	0.93	0.96	1.00
	200	0.59	0.68	0.16	0.28	0.34	0.61	0.93	0.92	0.93	0.93	0.96	1.00
D9	50	0.34	0.36	0.10	0.24	0.29	0.48	0.92	0.93	0.91	0.93	0.98	0.97
	100	0.36	0.36	0.07	0.27	0.33	0.50	0.94	0.94	0.93	0.94	0.98	0.98
	150	0.36	0.37	0.07	0.27	0.35	0.48	0.95	0.95	0.93	0.94	0.98	0.98
	200	0.36	0.37	0.07	0.27	0.36	0.47	0.96	0.95	0.93	0.94	0.98	0.99
D10	50	0.32	0.50	0.16	0.27	0.42	0.61	0.99	1.00	0.98	0.98	0.99	1.00
	100	0.34	0.49	0.16	0.31	0.47	0.64	1.00	1.00	0.99	1.00	0.99	1.00
	150	0.36	0.49	0.18	0.34	0.48	0.66	1.00	1.00	1.00	1.00	1.00	1.00
	200	0.38	0.49	0.19	0.36	0.51	0.68	1.00	1.00	1.00	1.00	1.00	1.00
Average	50	0.40	0.48	0.13	0.24	0.30	0.55	0.93	0.94	0.89	0.92	0.96	0.99
	100	0.39	0.50	0.12	0.25	0.34	0.58	0.94	0.94	0.92	0.93	0.97	0.99
	150	0.39	0.49	0.13	0.26	0.36	0.59	0.94	0.94	0.93	0.94	0.97	0.99
	200	0.39	0.48	0.13	0.27	0.38	0.61	0.95	0.94	0.94	0.94	0.97	0.99
Average RSGSA Gain	50	0.38	0.15	3.23	1.29	0.83		0.06	0.05	0.11	0.08	0.03
	100	0.49	0.16	3.83	1.32	0.71		0.05	0.05	0.08	0.06	0.02
	150	0.51	0.20	3.54	1.27	0.64		0.05	0.05	0.06	0.05	0.02
	200	0.56	0.27	3.69	1.26	0.61		0.04	0.05	0.05	0.05	0.02

The best performance metric value among the algorithms on each dataset is highlighted in bold typeface

Back to article page

ISSN: 1479-7364

Contact us

General enquiries: journalsubmissions@springernature.com