Technology Roadmap (Figure 1)
Table 1: List of Kidney Renal Clear Cell Carcinoma Information.
|
TCGA-KIRC
|
GSE53757
|
GSE26574
|
Platform
|
TCGA
|
GPL570
|
GPL11433
|
Species
|
Homo sapiens
|
Homo sapiens
|
Homo sapiens
|
Tissue
|
Kidney tissues
|
Kidney tissues
|
Kidney tissues
|
Samples in KIRC group
|
539
|
72
|
8
|
Samples in Normal group
|
72
|
72
|
8
|
Reference
|
/
|
Neuronal pentraxin 2 supports clear cell renal cell carcinoma by activating the AMPA-selective glutamate receptor-4
|
An antioxidant response phenotype shared between hereditary and sporadic type 2 papillary renal cell carcinoma.
|
Development of a prognostic model for m1A RGs and evaluation of clinical relevance
To evaluate the prognostic significance of 10 m1A RGs in the TCGA-KIRC dataset, we constructed a prognostic model for these genes employing Lasso regression analysis (Figure 2A). To visualize the regression results, we also created a Lasso variable trajectory graph (Figure 2B, Table 2). The trajectory graph indicates that our m1A RGs prognostic model was effective. We used a risk factor map to visualize the sample groupings in the prognostic model (Figure 2C). The risk factor plot includes risk grouping, survival outcome, and a heat map. Risk grouping involves grouping risk scores obtained from predicting the model for the dataset samples by the median. The survival outcome presents clinical samples' survival time and outcome in the TCGA-KIRC dataset through point plots. Finally, a heat map is employed to visualize the expression of m1A RGs for the samples in the prognostic model. To further elucidate the disparities among the high- and low-risk groups within the TCGA-KIRC dataset, we conducted a Mann-Whitney U test for the 10 m1A RGs and presented the results in subgroup comparison plots (Figure 2D). The analysis results indicate that the expression levels of the 7 m1A RGs (ALKBH1, ALKBH3, TRMT10C, TRMT6, TRMT61B, YTHDC1, and YTHDF2) exhibit significant variation among diverse groups. Specifically, TRMT6 (P-value < 0.001) expression is notably increased in the high-risk group, while TRMT10C (P-value < 0.001), TRMT61B (P-value < 0.001), ALKBH1 (P-value < 0.001), YTHDF2 (P-value < 0.001), and YTHDC1 (P-value < 0.001) are significantly decreased. The expression level of ALKBH3 (P-value > 0.05) shows no statistically significant variation between the two groups.
Table 2: Expression Genes of m1A between high- and low-risk groups.
GeneSymbol
|
logFC
|
AveExpr
|
t
|
P.Value
|
adj.P.Val
|
B
|
ALKBH1
|
-0.40288
|
9.224662
|
-7.42762
|
4.38E-13
|
7.77E-12
|
19.15272
|
ALKBH3
|
-0.05368
|
9.755665
|
-1.08067
|
0.280329
|
0.357177
|
-6.27237
|
TRMT61B
|
-0.64435
|
9.06618
|
-10.7428
|
1.62E-24
|
1.22E-21
|
44.8807
|
TRMT10C
|
-0.35666
|
9.860732
|
-6.47759
|
2.11E-10
|
1.92E-09
|
13.15655
|
TRMT6
|
0.077716
|
9.606906
|
1.406982
|
0.160011
|
0.221609
|
-5.87139
|
TRMT61A
|
0.070882
|
9.703999
|
1.145688
|
0.252435
|
0.327656
|
-6.20082
|
YTHDC1
|
-0.36582
|
12.19435
|
-6.4668
|
2.25E-10
|
2.04E-09
|
13.0924
|
YTHDF3
|
-0.38438
|
12.23562
|
-6.46175
|
2.32E-10
|
2.10E-09
|
13.06246
|
YTHDF2
|
-0.24952
|
11.85847
|
-5.30019
|
1.69E-07
|
8.32E-07
|
6.711437
|
YTHDF1
|
-0.08443
|
11.51887
|
-1.78782
|
0.074369
|
0.114353
|
-5.27153
|
Functional enrichment analysis(GO) of m1A RGs
To explore the interaction between BP, MF, and CC for the 7 m1A RGs in the TCGA-KIRC dataset, we performed a GO analysis on these genes (Table 3). Statistical significance was considered for entries with an enrichment P-value < 0.05 and FDR q-value < 0.05. The results revealed that the 7 m1A RGs in KIRC patient samples were significantly enriched in various biological processes, such as RNA modification, mRNA methylation, tRNA modification, mRNA modification, and tRNA processing, as well as cellular components such as methyltransferase complex, ribonuclease P complex, nuclear euchromatin, mitochondrial matrix, and endoribonuclease complex. Additionally,the molecular functions enriched by these RGs included catalytic activities acting on tRNA and RNA, as well as tRNA, RNA, and mRNA methyltransferase activities. The results were presented using bubble plots (Figure 3A) and circular network diagrams for the BP pathway, CC pathway, and MF pathway (Figure 3B-D).
Table 3: GO enrichment analysis results of m1A RGs.
ONTOLOGY
|
ID
|
Description
|
GeneRatio
|
BgRatio
|
pvalue
|
p.adjust
|
qvalue
|
BP
|
GO:0009451
|
RNA modification
|
5/7
|
163/18670
|
9.88e-10
|
1.52e-07
|
6.86e-08
|
BP
|
GO:0080009
|
mRNA methylation
|
3/7
|
14/18670
|
1.17e-08
|
7.46e-07
|
3.37e-07
|
BP
|
GO:0006400
|
tRNA modification
|
4/7
|
86/18670
|
1.45e-08
|
7.46e-07
|
3.37e-07
|
BP
|
GO:0016556
|
mRNA modification
|
3/7
|
22/18670
|
4.96e-08
|
1.91e-06
|
8.61e-07
|
BP
|
GO:0008033
|
tRNA processing
|
4/7
|
130/18670
|
7.73e-08
|
2.38e-06
|
1.07e-06
|
CC
|
GO:0034708
|
methyltransferase complex
|
2/7
|
113/19717
|
6.71e-04
|
0.010
|
0.004
|
CC
|
GO:0030677
|
ribonuclease P complex
|
1/7
|
14/19717
|
0.005
|
0.025
|
0.009
|
CC
|
GO:0005719
|
nuclear euchromatin
|
1/7
|
30/19717
|
0.011
|
0.025
|
0.009
|
CC
|
GO:0005759
|
mitochondrial matrix
|
2/7
|
469/19717
|
0.011
|
0.025
|
0.009
|
CC
|
GO:1902555
|
endoribonuclease complex
|
1/7
|
31/19717
|
0.011
|
0.025
|
0.009
|
MF
|
GO:0140101
|
catalytic activity, acting on a tRNA
|
4/7
|
122/17697
|
7.40e-08
|
1.56e-06
|
3.59e-07
|
MF
|
GO:0140098
|
catalytic activity, acting on RNA
|
5/7
|
386/17697
|
9.75e-08
|
1.56e-06
|
3.59e-07
|
MF
|
GO:0008175
|
tRNA methyltransferase activity
|
3/7
|
34/17697
|
2.26e-07
|
2.41e-06
|
5.54e-07
|
MF
|
GO:0008173
|
RNA methyltransferase activity
|
3/7
|
66/17697
|
1.72e-06
|
1.37e-05
|
3.16e-06
|
MF
|
GO:0008174
|
mRNA methyltransferase activity
|
2/7
|
11/17697
|
7.36e-06
|
4.71e-05
|
1.09e-05
|
GSEA of DEGs between high- and low-risk groups
To investigate the effect of gene expression levels on KIRC development in distinct groups of the m1A RGs prognosis model in the TCGA-KIRC dataset, we performed a GSEA on the expression of all genes and their participation in BP, CC, and MF between the two groups. We used a significant enrichment screening criterion of P-value < 0.05 and FDR q-value < 0.05 (Figure 4A). The results reveal that genes in both groups of m1A RGs prognosis model in the TCGA-KIRC dataset were significantly enriched in pathways such as IL-5 pathway (Figure4B), IL-6-7 pathway (Figure 4C), IL-27 pathway (Figure 4D), P130Cas linkage to MAPK signaling for integrins (Figure 4E), and Grb2-Sos provides linkage to MAPK signaling for integrins (Figure 4F) (Figure 4B-F, Table 4).
Table 4: GSEA results of m1A RGs Prognosis Model high- and low-risk groups genes.
Description
|
setSize
|
enrichmentScore
|
NES
|
p.adjust
|
qvalues
|
BIOCARTA_IL5_PATHWAY
|
11
|
0.714998
|
1.971499
|
0.098147
|
0.081327
|
REACTOME_TP53_REGULATES_TRANSCRIPTION_OF_GENES_INVOLVED_IN_G1_CELL_CYCLE_ARREST
|
14
|
0.661845
|
1.925429
|
0.081818
|
0.067796
|
PID_IL6_7_PATHWAY
|
47
|
0.45084
|
1.782049
|
0.106413
|
0.088176
|
PID_IL27_PATHWAY
|
26
|
0.513034
|
1.761335
|
0.115452
|
0.095666
|
PID_WNT_SIGNALING_PATHWAY
|
28
|
0.463274
|
1.628843
|
0.157968
|
0.130896
|
REACTOME_P130CAS_LINKAGE_TO_MAPK_SIGNALING_FOR_INTEGRINS
|
15
|
0.533795
|
1.568837
|
0.222955
|
0.184745
|
REACTOME_WNT_LIGAND_BIOGENESIS_AND_TRAFFICKING
|
26
|
0.45512
|
1.562506
|
0.144618
|
0.119833
|
REACTOME_GRB2_SOS_PROVIDES_LINKAGE_TO_MAPK_SIGNALING_FOR_INTEGRINS_
|
15
|
0.524404
|
1.541237
|
0.23248
|
0.192638
|
WP_LNCRNA_INVOLVEMENT_IN_CANONICAL_WNT_SIGNALING_AND_COLORECTAL_CANCER
|
93
|
0.342734
|
1.527341
|
0.174478
|
0.144576
|
WP_NCRNAS_INVOLVED_IN_WNT_SIGNALING_IN_HEPATOCELLULAR_CARCINOMA
|
85
|
0.333619
|
1.468952
|
0.139577
|
0.115657
|
GSVA of DEGs between high- and low-risk groups
We used GSVA on the DEGs in the TCGA-KIRC dataset to investigate the variation in hallmark gene sets between the two groups (Table 5). Our analysis revealed 32 hallmark gene sets that exhibited significant differences (P-value < 0.05) between the two groups. Of these sets, 17 pathways exhibited significantly higher enrichment scores in the high-risk group (P-value < 0.05), while 15 pathways exhibited significantly higher enrichment scores in the low-risk group (P-value < 0.05). Based on the GSVA results, we analyzed the differential expression of the 32 Hallmark pathways between different groups in the TCGA-KIRC dataset and used the R package "pheatmap" to display the specific differential analysis results in a heatmap (Figure 5A). We also analyzed the degree of group differences of the 32 Hallmark pathways between different groups in the TCGA-KIRC dataset by using the Mann-Whitney U test and presented the results using a group comparison plot (Figure 5B), which demonstrated that the differential expression of the 32 Hallmark pathways between different groups in the TCGA-KIRC dataset had statistically significant differences (P-value < 0.05).
Table 5: GSVA results of m1A RGs Prognosis Model high- and low-risk groups genes.
Description
|
logFC
|
AveExpr
|
t
|
adj.P.Val
|
HALLMARK_HEME_METABOLISM
|
0.159138
|
0.001188
|
8.017341
|
3.12E-13
|
HALLMARK_FATTY_ACID_METABOLISM
|
0.174981
|
-0.00928
|
6.840097
|
5.15E-10
|
HALLMARK_KRAS_SIGNALING_DN
|
-0.14085
|
-0.01224
|
-6.40193
|
5.38E-09
|
HALLMARK_IL6_JAK_STAT3_SIGNALING
|
-0.17055
|
0.011854
|
-6.34273
|
5.59E-09
|
HALLMARK_BILE_ACID_METABOLISM
|
0.137566
|
-0.00314
|
6.311484
|
5.59E-09
|
HALLMARK_UV_RESPONSE_DN
|
0.157282
|
0.0105
|
6.257137
|
6.46E-09
|
HALLMARK_SPERMATOGENESIS
|
-0.11792
|
-0.00884
|
-6.18313
|
8.60E-09
|
HALLMARK_TGF_BETA_SIGNALING
|
0.168719
|
0.00611
|
5.96495
|
2.70E-08
|
HALLMARK_ADIPOGENESIS
|
0.14295
|
-0.00603
|
5.669446
|
1.27E-07
|
HALLMARK_PROTEIN_SECRETION
|
0.168965
|
0.001103
|
5.621664
|
1.49E-07
|
HALLMARK_INFLAMMATORY_RESPONSE
|
-0.13737
|
0.014037
|
-5.46219
|
3.21E-07
|
HALLMARK_ALLOGRAFT_REJECTION
|
-0.15805
|
0.012332
|
-5.36911
|
4.82E-07
|
HALLMARK_EPITHELIAL_MESENCHYMAL_TRANSITION
|
-0.15307
|
-0.0043
|
-5.10985
|
1.70E-06
|
HALLMARK_PEROXISOME
|
0.115832
|
-0.01182
|
5.030625
|
2.35E-06
|
HALLMARK_TNFA_SIGNALING_VIA_NFKB
|
-0.12522
|
-0.00243
|
-4.86737
|
4.90E-06
|
HALLMARK_ANDROGEN_RESPONSE
|
0.116462
|
0.007187
|
4.730457
|
8.85E-06
|
HALLMARK_COAGULATION
|
-0.1087
|
2.84E-05
|
-4.65049
|
1.21E-05
|
HALLMARK_MITOTIC_SPINDLE
|
0.110549
|
0.01093
|
4.343229
|
4.62E-05
|
HALLMARK_ESTROGEN_RESPONSE_LATE
|
-0.08657
|
-0.00671
|
-4.09357
|
0.000123
|
HALLMARK_WNT_BETA_CATENIN_SIGNALING
|
0.107711
|
0.002006
|
4.090927
|
0.000123
|
HALLMARK_MYC_TARGETS_V2
|
-0.12097
|
-0.0213
|
-4.0566
|
0.000135
|
HALLMARK_COMPLEMENT
|
-0.0833
|
0.001327
|
-3.63167
|
0.000698
|
HALLMARK_NOTCH_SIGNALING
|
0.087759
|
0.015653
|
3.603017
|
0.000744
|
HALLMARK_HEDGEHOG_SIGNALING
|
0.087783
|
0.009263
|
3.441928
|
0.001292
|
HALLMARK_INTERFERON_GAMMA_RESPONSE
|
-0.07771
|
0.002714
|
-2.7169
|
0.013551
|
HALLMARK_PI3K_AKT_MTOR_SIGNALING
|
0.061528
|
0.008777
|
2.704519
|
0.013551
|
HALLMARK_MYOGENESIS
|
-0.05745
|
0.003408
|
-2.59069
|
0.018195
|
HALLMARK_UV_RESPONSE_UP
|
-0.05076
|
-0.00316
|
-2.47244
|
0.024486
|
HALLMARK_HYPOXIA
|
-0.05675
|
0.006658
|
-2.45774
|
0.024621
|
HALLMARK_E2F_TARGETS
|
-0.06652
|
-0.01995
|
-2.36835
|
0.030338
|
HALLMARK_OXIDATIVE_PHOSPHORYLATION
|
0.08239
|
-0.02942
|
2.331311
|
0.032399
|
HALLMARK_GLYCOLYSIS
|
-0.0491
|
-0.0133
|
-2.07716
|
0.059747
|
Comparison of subgroups of m1A RGs prognostic model in patients with KIRC, PPI network and clinical analysis
To analyze the expression of 7 m1A RGs (ALKBH1, ALKBH3, TRMT10C, TRMT6, TRMT61B, YTHDC1, YTHDF2) in KIRC patients with KIRC tumor tissue (subgroup: KIRC) and healthy kidney tissue (subgroup: Normal), we utilized the Mann-Whitney U test to examine the expression of these genes in subgroup samples from the TCGA-KIRC dataset, GSE53757 dataset, and GSE26574 dataset. The results were presented using group comparison plots (Figure 6A-C). Among the 7 m1A RGs in FPKM format from the TCGA-KIRC dataset, YCHDC1(P-value > 0.05) expression did not exhibit any statistically significant differences between the KIRC and Normal groups. The KIRC group showed significant upregulation in ALKBH1 (P-value < 0.01) and ALKBH3 (P-value < 0.001) expression while TRMT10C (P-value < 0.001), TRMT6 (P-value < 0.001), TRMT61B (P-value < 0.01), and YTHDF2 (P-value < 0.05) expression was significantly downregulated (Figure 6A). In the GSE53757 dataset, TRMT6 and YTHDF2 expression among the 7 m1A RGs did not show any statistically significant differences among different subgroups (P-value > 0.05). The KIRC group exhibited significant upregulation in ALKBH3 (P-value < 0.001) and YTHDC1 (P-value < 0.001) expression while ALKBH1 (P-value < 0.001), TRMT10C (P-value < 0.001), and TRMT61B (P-value < 0.001) expression was significantly downregulated (Figure 6B). In the GSE26574 dataset, TRMT61B, YTHDC1, and YTHDF2 expression did not exhibit any statistically significant differences (P-value > 0.05), whereas TRMT6 (P-value < 0.05) expression was significantly upregulated in the KIRC group and ALKBH1 (P-value < 0.05) and ALKBH3 (P-value < 0.05) expression was significantly downregulated (Figure 6C).
To further investigate the protein-protein interactions of the 7 m1A RGs in the TCGA-KIRC dataset, we constructed a PPI network using the STRING database and visualized it with Cytoscape software (Figure 6D).
Finally, we plotted ROC curves of the 7 m1A RGs in the Tumor and Normal groups of the TCGA-KIRC dataset and presented the results (Figure 6E-K). The expression of ALBKH1 (AUC = 0.610), ALBKH3 (AUC = 0.627), TRMT61B (AUC = 0.605), YTHDC1 (AUC = 0.518), and YTHDF2 (AUC = 0.592) in the TCGA-KIRC dataset in the Tumor and Normal groups had low accuracy. In contrast, TRMT6 (AUC = 0.857) and TRMT10C (AUC = 0.808) showed relatively high accuracy in distinguishing between Tumor and Normal groups in the TCGA-KIRC dataset.
Prognostic analysis of m1A-related genes
We performed a prognostic analysis of 7 m1A RGs, including ALKBH1, ALKBH3, TRMT10C, TRMT61B, TRMT6, YTHDC1, and YTHDF2, using the constructed Lasso regression model, and considered associated molecules to be statistically significant at P-value <0.05. We plotted prognostic survival KM curves for each of the 7 m1A RGs using P-value <0.05 as the screening criterion, and obtained 5 eligible m1A RGs (ALBKH1, TRMT10C, TRMT61B, YTHDC1, YTHDF2) (Figure 7A-E). The results showed that ALBKH1 (P-value < 0.001), TRMT10C (P-value = 0.008), TRMT61B (P-value < 0.001), YTHDC1 (P-value = 0.002), and YTHDF2 (P-value < 0.031) were all statistically significant.
Subsequently, we also plotted the ROC curves of 5 m1A RGs based on disease-specific survival (DSS) events in the TCGA-KIRC dataset, and presented the results with AUC>0.6 (Figure 7F-I, Table 6). As shown in Figure 5, the expression of m1A RGs ALBKH1 (AUC = 0.641, Figure 7F), YTHDC1 (AUC = 0.684, Figure 7H), and YTHDF2 (AUC = 0.605, Figure 7I) only had low accuracy in predicting survival events based on clinical variables in the TCGA-KIRC dataset. TRMT61B (AUC = 0.803, Figure 7G) had moderate accuracy in predicting survival events based on clinical variables in the TCGA-KIRC dataset.
Table 6: Patient Characteristics of KIRC patients in the TCGA datasets.
Characteristic
|
levels
|
Overall
|
n
|
|
539
|
T stage, n (%)
|
T1
|
278 (51.6%)
|
|
T2
|
71 (13.2%)
|
|
T3
|
179 (33.2%)
|
|
T4
|
11 (2%)
|
N stage, n (%)
|
N0
|
241 (93.8%)
|
|
N1
|
16 (6.2%)
|
M stage, n (%)
|
M0
|
428 (84.6%)
|
|
M1
|
78 (15.4%)
|
Pathologic stage, n (%)
|
Stage I
|
272 (50.7%)
|
|
Stage II
|
59 (11%)
|
|
Stage III
|
123 (22.9%)
|
|
Stage IV
|
82 (15.3%)
|
Gender, n (%)
|
Female
|
186 (34.5%)
|
|
Male
|
353 (65.5%)
|
Age, n (%)
|
<=60
|
269 (49.9%)
|
|
>60
|
270 (50.1%)
|
OS event, n (%)
|
Alive
|
366 (67.9%)
|
|
Dead
|
173 (32.1%)
|
DSS event, n (%)
|
Alive
|
420 (79.5%)
|
|
Dead
|
108 (20.5%)
|
PFI event, n (%)
|
Alive
|
378 (70.1%)
|
|
Dead
|
161 (29.9%)
|
Construction of a prognostic model for DEGs in the high- and low-risk groups of the TCGA-KIRC dataset
We conducted a univariate Cox regression analysis to evaluate the prognostic relationship between each variable and m1A RGs. Next, we included all variables that were significantly associated with m1A RGs in the univariate analysis, and built a prognostic model to distinguish high-risk and low-risk groups using multivariate Cox regression analysis. Additionally, we used a forest plot to display the results of the univariate Cox regression analysis, which helped readers better understand the prognostic relationship between each variable and m1A RGs (Figure 8A,Table 7).
Subsequently, we utilized nomogram analysis to assess the prognostic ability of the risk model and constructed a nomogram (Figure 8B) to facilitate visualization. Based on the chart, we observed that the ALKBH1 and TRMT6B1 genes exhibit higher utility compared to other variables. Additionally, we performed a prognostic calibration analysis for the 5 m1A RGs using the nomogram at 1-year, 3-year, and 5-year (Figure 8C-E) intervals and constructed calibration curves (Figures 8F-H). The x-axis of the calibration curve represents the forecasted probability of survival, while the y-axis corresponds to the actual survival probability. The degree of proximity between the calibration curve and the diagonal line serves as an indicator of the accuracy of the model's predictions.
Finally, we utilized DCA to assess the clinical utility of the prognostic model at 1year, 3years, and 5 years (Figure 8F-H). The threshold probability is depicted on the x-axis of the DCA, while the y-axis represents the net benefit. The stable range of x values above the lines for "all positive" and "all negative" could be used to judge the results, with a more extensive x-value range indicating better model performance. The results show that our constructed multivariate Cox regression model has a clinical predictive effect of 5 years > 3 years > 1 year.
Table 7: COX regression to hub genes associated with OS in TCGA-KIRC.
Characteristics
|
Total(N)
|
Univariate analysis
|
Multivariate analysis
|
Hazard ratio (95% CI)
|
P value
|
Hazard ratio (95% CI)
|
P value
|
ALKBH1
|
539
|
|
|
|
|
Low
|
269
|
Reference
|
|
|
|
High
|
270
|
0.539 (0.396-0.733)
|
<0.001
|
0.636 (0.458-0.883)
|
0.007
|
TRMT10C
|
539
|
|
|
|
|
Low
|
269
|
Reference
|
|
|
|
High
|
270
|
0.663 (0.490-0.896)
|
0.008
|
0.810 (0.585-1.122)
|
0.205
|
TRMT61B
|
539
|
|
|
|
|
Low
|
269
|
Reference
|
|
|
|
High
|
270
|
0.465 (0.338-0.639)
|
<0.001
|
0.556 (0.392-0.788)
|
<0.001
|
YTHDC1
|
539
|
|
|
|
|
Low
|
269
|
Reference
|
|
|
|
High
|
270
|
0.613 (0.451-0.833)
|
0.002
|
0.882 (0.624-1.248)
|
0.479
|
YTHDF2
|
539
|
|
|
|
|
Low
|
269
|
Reference
|
|
|
|
High
|
270
|
0.718 (0.532-0.971)
|
0.031
|
0.981 (0.704-1.367)
|
0.910
|
Immunohistochemical analysis of prognostic m1A RGs
We conducted IHC analysis using gene-specific antibodies against prognostic 5 m1A RGs in normal kidney and KIRC tissues, as provided by the HPA database. The IHC analysis revealed more intense ALKBH1 and TRMT61B staining in normal kidney tissue (Figure 9A, 9E) than in KIRC tumor tissue (Figure 9B, 9F), indicating lower expression levels in KIRC tumor tissue compared to normal kidney tissue. Conversely, YTHDC1 and YTHDF2 exhibited weaker staining in normal kidney tissue (Figure 9G, 9I) and more vital staining in KIRC tumor tissue (Figure 9H, 9J), suggesting significantly higher expression levels of these genes in KIRC tumor tissue than in normal kidney tissue. TRMT10C did not exhibit significant differences in staining intensity between normal kidney tissue (Figure 9C) and KIRC tumor tissue (Figure 9D), indicating that its expression levels did not significantly change in KIRC tumor tissue compared to normal kidney tissue.