Integrated analysis of recurrent properties of cancer genes to identify novel drivers

Table 1 Dataset of known, candidates and mutated cancer genes

Cancer type	Known cancer genes	Candidate cancer genes	Rest of mutated genes	Tumour samples	References
Bladder	24	9	380	97	[8]
Breast	165	280	1,520	196	[2, 11, 40, 52]
Colorectal	42	124	654	11	[2]
Gastric	203	14	5,160	109	[16]
Glioblastoma	127	87	1,679	197	[5, 12, 39]
HNSCC	221	67	5,901	194	[15, 35]
Kidney	71	15	655	517	[7, 9, 55]
Leukaemia	58	169	877	393	[18, 41, 44, 46, 50, 51]
Liver	26	66	370	140	[45, 53]
Lung	218	143	1,012	324	[4, 11, 43, 49]
Lymphoma	46	47	574	133	[14, 47]
Medulloblastoma	16	1	194	88	[13]
Melanoma	96	159	2,845	16	[17, 48, 54]
Myelodysplasia	22	4	206	29	[57]
Myeloma	69	6	1,260	37	[6]
Oligodendroglioma	12	0	149	34	[38]
Ovarian	60	80	109	58	[11]
Pancreas	77	194	1,070	205	[10, 11, 42, 56]
Prostate	54	161	81	67	[11, 37]
Sarcoma	11	10	0	207	[36]
Total	444	608	9,629	3,052	NA

Starting from 10,681 human genes with at least one cancer-specific non-synonymous mutation, three groups were identified: known cancer genes that are part of the cancer gene census [58]; candidate genes that are likely to play a role in cancer because they are recurrently mutated; and the rest of mutated genes that carry at least one non-synonymous mutation. The full description of mutated genes in each of the 39 sequencing screenings is reported in Additional file 2, Table S1. HNSCC; head and neck squamous cell carcinoma.

ISSN: 1474-760X