Four specific biomarkers associated with the progression of glioblastoma multiforme in older adults identified using weighted gene co-expression network analysis

Figures & data

Figure 1. Structuring WGCNA. (a) Sample tree clustering and clinical traits (Survival days; vital: white = alive, red = dead, gray = missing value; KPS) heat map of 38 GBM tissues in older adults. (b) Scale independence of various soft-threshold values. (c) Mean connectivity of various soft-threshold values

Table 1. Detail of hub genes in each significant module

Module	Hub gene
Cyan	PPP1R14A DBNDD2 DNAH17 OPALIN HHATL FCHO1 FAM19A4 CLCA4 KIAA1598 ROGDI CAPN3 PLEKHH1 LANCL1 PTGDS S1PR5 LDB3 CD22 TMEM144 CARNS1 TPPP KCNQ1DN LIPE SRCIN1 SPOCK3 SEC14L5 ERMN SLC5A11 MAL SFTPC ADAP1 BOK MBP TMEM63A SLC45A3 CNTN2 AATK ZFP57 CNDP1 NKX6-2 PEX5L CNTNAP4 HAPLN2 CYTH1 LOC283999 KANK4 PLP1 C9orf122 LGI3 TMCC2 GJB1 NKAIN2 QDPR PLCL1 ST18 GPR62 PLCH2 MOBP PCSK6 ENPP2 SEMA4D EDIL3 TMEM151A MAG HCN2 LOC100128675 LOC150622 GJC2 CNP SH3GL3 TMEM125 CDK18 TUBB4 LRP2 SHISA2 LHPP PPP1R16B FA2H MOG C11orf9 C7orf41 KLK6 TMEM88B BCAS1 GAL3ST1 TF RAB40B ZNF536 ZNF488 FAM107B RAB11FIP4 UGT8 CHADL SOX10 GREM1 HOXD1 CLDN11 SLC7A14
Green	DNAH1 HGFAC TRPV1 TFAP2E ATG16L2 JMJD7-PLA2G4B SULT1A3 AGER FAM193B ACCN3 NKTR SCNN1D SGK494 LOC646471 LOC100128288 KAT2A C17orf56 AGAP5 LENG8 DNASE1 C16orf79 PRICKLE4 APBB3 SEC31B KIAA1875 L3MBTL ARID3B SLC26A1 NPIP MYH3 CRIPAK RHOT2 SDHAP3 CEACAM19 STRC NCRNA00176 OSBPL7 NRBP2 HAUS5 PVRIG ATXN7L2 CUL9 TBC1D3B ANKS3 ULK3 FLJ45340 GDPD3 MZF1 CDK5RAP3 TNRC6A KIFC2 SRRM2 PABPC1L PILRB PDXDC2 FAM13AOS TTLL3 ZNF692 TRO LY6G5B LRDD UCKL1AS LOC100132287 CCDC78 CROCCL2 RPL32P3 CRAMP1L YJEFN3 CDK3 NCRNA00107 DND1 ANKRD23 PRDM15 CCDC84 ?|155,060 FBXL6 KLHL17 CSAD WDR90 ?|645,851 LOC100133331 ANKRD36 CCNL2 MSH5 AGAP6 LOC91316 POGZ GSDMB LOC100272228 UCP3 WASH7P DNHD1 PKD1 LOC200030 TTC21A TAF1C C8ORFK29 SPDYE8P CCDC57 GIGYF1 ERAS NCRNA00201 KIAA0895 L UNC5CL ZNF337 NCRNA00105 GNRH1 ASB16 KIAA0907 AFG3L1 ING5 LOC100240726 CDK10 LOC100128842 ANKZF1 LOC349114 HOOK2 STK36 ZNF767 LOC338799 ZNF471 FAM156A LOC150776 AKR7L INE1 FAM71F2 CLK2 TRIM52
Lightcyan	SP3 NEDD1 SAPS3 NUP107 PRPF40A RBM27 NAA15 TMEM194A POLK PRPF4B CDC27
Orange	RC3H2 PAFAH1B2 GTF2A1 SLC30A4 LMBRD2 SBNO1 CCNT1 CRK ETV3 TAOK1 IL6ST NCOA2 LOC284232 MAN1A2 UHMK1

Figure 2. Module clusters and relationships with clinical traits. (a) Clustering dendrograms of all genes with dissimilarity based on topological overlap, together with assigned module colors. (b) Identification of significant modules associated with clinical traits (survival days, vital, and KPS). Each cell in the heat map contains the corresponding correlation score and P-value. Red indicates a positive correlation, whereas green indicates a negative correlation

Figure 3. Relationship between gene significance (GS) and module membership (MM) in significant modules

Figure 4. GO analysis for the module core genes

Figure 5. Selecting hub genes in GBM. (a) Module core genes used to construct protein-protein interaction network. Nodes indicate genes, lines indicate interactions. (b) Genes with top 10 degree score are shown

Figure 6. Relationship between the expression of hub genes and KPS score in older patients with GBM

Figure 7. Relationship between the expression of hub genes and KPS score in young patients with GBM

Table 2. Detail IHC score of ERMN, MOBP, PLP1 and OPALIN in GBM tissues provided by the elder patients with low KPS score and high KPS score

Target	Group	IHC score
		0	1	2	3	4	5	6
ERMN	KPS low	0	1	1	2	2	4	5
	KPS high	0	12	11	2	1	1	1
MOBP	KPS low	0	1	1	1	0	4	8
	KPS high	0	5	14	5	2	1	1
PLP1	KPS low	0	1	0	0	2	4	8
	KPS high	0	11	13	2	0	1	1
OPALIN	KPS low	0	1	1	0	1	3	9
	KPS high	1	17	6	2	1	1	0

Figure 8. ERMN, MOBP, PLP1 and OPALIN were highly expressed in the GBM tissues provided by the older patients with lower KPS scores. (a) IHC stain determined the expression of ERMN, MOBP, PLP1 and OPALIN in GBM tissues provided by the older patients with low and high KPS scores. (b) ROC analysis was performed to determine the diagnostic value of ERMN, MOBP, PLP1 and OPALIN to distinguish the GBM tissues provided by the older patients with low and high KPS scores