A novel analysis strategy for integrating methylation and expression data reveals core pathways for thyroid cancer aetiology

Table 2 Rankings of functional enrichment results - Part 1.

	Model 1 (Differential Expression FDR<0.01)				Model 2 (Differential Methylation FDR<0.01)				Model 3 (Both Diff. Meth and Diff. Expressed FDR<0.01)				Model 4 (Meth, Expr. Significances Combined FDR<0.01)
KEGG TERMs	A	B	C	D	A	B	C	D	A	B	C	D	A	B	C	D
MAPK Signalling	6	7	4	1	10	7	7	17	7	-	12	13	17	8	9	8
ECM Receptor	5	5	3	8	26	-	52	8	8	-	13	1	6	1	3	5
ErbB Signalling	13	28	20	45	28	1	6	28	14	-	20	43	20	22	20	7
NF-KB Signalling	32	11	17	85	11	40	18	24	29	-	33	47	71	33	46	17
Wnt-β-Catenin Signalling	51	25	35	73	14	27	16	13	-	-	-	35	86	46	60	28
VEGF Signalling	46	47	47	105	42	13	23	85	36	-	42	-	85	84	82	56
Thyroid Cancer	30	52	34	66	55	42	43	-	12	4	4	65	88	65	74	69
Adherens Junction	34	16	19	24	21	19	20	1	34	5	16	9	21	19	15	9
p53 Signalling	11	18	11	15	68	-	77	59	11	-	18	23	34	16	22	30
TGF-beta Signalling	3	62	12	5	18	6	12	23	-	1	5	5	28	29	29	19
Notch Signalling	60	58	60	57	93	18	42	6	-	-	-	29	59	9	25	13
GnRH Signalling	61	27	42	26	32	24	24	40	31	-	38	26	73	41	54	68
Neurotrophin Signalling	16	9	8	14	8	4	3	30	5	-	10	24	12	5	6	15
Focal Adhesion	8	1	2	7	1	2	1	2	3	-	3	2	7	2	2	3
Transcr. Misregulation	42	21	29	65	13	-	40	-	-	-	-	37	31	23	28	34
Apoptosis	17	19	16	10	12	-	39	20	15	-	21	16	29	35	31	6
Pathways in Cancer	1	2	1	3	2	3	2	16	1	-	1	3	1	2	1	2
Toll-like receptor signalling pathway	14	57	24	32	67	-	76	60	28	-	32	-	102	85	92	43
Pentose-phosphate pathway	91	70	80	88	-	43	99	88	-	-	-	81	112	87	101	85

Rankings of previously identified thyroid-cancer associated pathways in PANOGA functional enrichment results. For each analysis strategy four different results are shown in order to understand differences between different data selection strategies; A) Batch230 Results B) Batch250 Results C) Functional Enrichment Results of different Batches combined D) Batch230+Batch250 (Pooled) results. Model 1 represents, genes with differential expression FDR<0.01, Model 2 represents, genes with differential methylation FDR <0.01, Model 3 represents, genes having both differential expression and differential methylation FDR <0.01, Model 4 represents, genes that have FDR<0.01 after significance values of methylation and expression are combined.

ISSN: 1471-2164