Construction of genetic classification model for coronary atherosclerosis heart disease using three machine learning methods

Table 3 The result information of 12 optimal feature genes in limma package analysis

Gene symbol	GSE12288		GSE7638		GSE66360		Integrated dataset
Gene symbol	Foldchange	Adjusted p	Foldchange	Adjusted p	Foldchange	Adjusted p	Foldchange	Adjusted p
BCL6	1.07	2.36E−01	1.20	7.07E−03	2.86	2.93E−09	1.27	1.34E−06
CA12	1.05	4.18E−04	1.06	1.99E−03	1.00	9.94E−01	1.08	6.80E−05
CAMK2B	0.88	1.02E−22	0.95	4.57E−02	1.02	8.29E−01	0.90	3.94E−12
CNGB1	0.89	2.72E−08	0.93	1.39E−02	0.83	4.21E−02	0.88	1.22E−09
DDC	0.90	1.03E−17	0.92	6.26E−03	1.00	9.87E−01	0.88	7.23E−12
DERL1	1.03	1.91E−01	1.29	4.81E−19	0.88	4.74E−01	1.17	3.29E−06
HCK	1.07	1.67E−01	1.16	5.43E−06	2.09	1.10E−03	1.28	9.15E−07
HTR4	1.11	4.11E−10	1.00	9.68E−01	1.09	4.12E−01	1.10	8.22E−06
KISS1	0.78	3.49E−22	0.90	3.42E−02	0.92	2.26E−01	0.84	1.05E−15
KLK2	1.08	5.39E−09	1.06	3.58E−02	0.88	1.69E−01	1.09	3.93E−05
LILRA2	1.04	3.88E−01	1.19	1.74E−06	2.22	7.91E−06	1.25	2.60E−07
MTF2	1.07	1.17E−04	1.27	1.02E−08	0.81	1.22E−01	1.15	3.07E−06

Integrated dataset was the combination of GSE12288, GSE7638 and GSE66360; foldchange, the fold change of the average gene expressional level going from control group to case group; adjusted P, the P value adjusted by Benjamini–Hochberg in comparing the gene expressional level between case group and control group

ISSN: 1471-2261