EARN: an ensemble machine learning algorithm to predict driver genes in metastatic breast cancer

BMC Medical Genomics

Table 4 12 driver genes predicted by EARN50 which are confirmed for metastatic cancers in the HCMDB

Symbol	Prediction score	Rank	PSMM (%) [54]	PSMM (%) [55]	NSCGMCH	NSCGMBH	MCMGM	PKGECC	PKGEBC
APEX1	0.900511991	5	0.50	1.70	1	#N/A	5	#N/A	#N/A
ARID1A	0.895213526	11	2.40	5.10	2	#N/A	24	✓	✓
KDM6B	0.894029187	13	1.40	4.60	1	#N/A	16	#N/A	#N/A
TBX3	0.893837209	14	2.80	5.10	1	#N/A	21	✓	✓
KDR^a	0.890079401	17	0.90	1.70	7	2	9	✓	#N/A
SERPINE2	0.889205475	19	0.90	0.80	1	#N/A	4	#N/A	#N/A
TBL1XR1	0.871240171	27	0.90	0.80	2	#N/A	4	✓	✓
KRAS	0.868267682	30	1.40	1.70	20	#N/A	7	✓	✓
NOS3	0.861560093	31	2.40	2.10	1	#N/A	12	#N/A	#N/A
RAPGEF3	0.851947423	42	#N/A	2.50	2	#N/A	6	#N/A	#N/A
SELE^a	0.847865292	49	0.90	1.30	12	1	5	#N/A	#N/A
MME^a	0.847698297	50	0.90	2.50	9	1	9	#N/A	#N/A

Also, the rank number, score, and mutation count for these genes are provided in the table. The confirmed genes as the known genes related to any primary cancers or primary breast tumors in OMIM, CGC, and NCG databases have been marked in the last two columns
PSMM, Percentage of samples with one or more mutations based on initial mutation file, NSCGMCH, Number of studies that have cited genes related to different metastatic cancers in the HCMDB, NSCGMBH, Number of studies that have cited genes related to metastatic breast cancer in HCMDB, MCMGM, Mutation counts for mutated genes across 450 metastasis tumor samples based on the initial mutation file, PKGECC, Predicted known genes by EC associated with different cancers that are confirmed in OMIM, CGC, and NCG, PKGEBC, Predicted known genes by EC associated with breast cancer that are confirmed in OMIM, CGC, and NCG
^aThese genes have been specifically introduced concerning metastatic breast cancer

ISSN: 1755-8794