Table 1.

The data sets analyzed in this study

Data sets	Total (N)	Accession	Platform	Sex males (%)	Stage I (%)	Histology AC (%)	Molecular TRU (%)	Case arm histology	Case arm molecular
Sato et al. [29]	263	GSE41271	Illumina	54	50	70	38	Train	Train
Der et al. [30]	170	GSE50081	Affymetrix	53	70	75	39	Train	Train
Botling et al. [12]	172	GSE37745	Affymetrix	53	64	62	35	Train	Test
Hou et al. [31]	72	GSE19188	Affymetrix	65	NA	62	31	Train
Clinical Lung Cancer Genome Projecta [14]	191	CLCGP	Illumina	63	47	51	34	Train
Djureinovic et al. [32]	183	GSE81089	RNAseq	47	58	63	50	Train
Karlsson et al. [33]	99	GSE60644	Illumina	46	90	78	40	Train
Lee et al. [34]	138	GSE8894	Affymetrix	75	NA	46	38	Test
Bhattacharjee et al. [19]	211	GSE83227	Affymetrix	36	40	90	37	Test	Test
Tarca et al. [35]	150	GSE43580	Affymetrix	80	50	51	42	Test	Test
Rousseaux et al. [21]	146	GSE30219	Affymetrix	84	90	58	34	Test	Test
Zhu et al.b [16]	123	GSE14814	Affymetrix	67	54	58	35	Test
Wilkerson et al. [10]	116	GSE26939	Agilent	46	53	100	41		Train
Cancer Genome Atlas Research Networkc [11]	230	TCGA LUAD	RNAseq	NA	NA	100	39		Train
Shedden et al. [15]	444	Shedden	Affymetrix	50	62	100	38		Train
Fouret et al. [36]	103	E_MTAB_923d	Affymetrix	16	58	100	42		Train
Okayama et al. [37]	226	GSE31210	Affymetrix	46	74	100	43		Train
Tomida et al. [38]	117	GSE13213	Agilent	51	68	100	40		Test
Chitale et al.e [39]	102	Chitale U133 2plus	Affymetrix	41	69	100	41		Test

^aCLCGP: The Clinical Lung Cancer Genome Project (http://www.uni-koeln.de/med-fak/clcgp/).

^bPresent data set overlaps with Shedden et al. [15] (43 samples).

^cThe Cancer Genome Atlas Network (TCGA).

^dData obtained from the ‘ArrayExpress’ database (https://www.ebi.ac.uk/arrayexpress/experiments/E-MTAB-923/).

^eSamples were divided into two cohorts based on the different Affymetrix platforms, U133A and U133 2plus. Only the latter subset was included in the analysis.