Hypernetwork modelling reveals epigenome-proteome interactions in post-surgical progression of non-functioning pituitary adenomas

Hypernetwork modelling reveals epigenome-proteome interactions in post-surgical progression of non-functioning pituitary adenomas | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Hypernetwork modelling reveals epigenome-proteome interactions in post-surgical progression of non-functioning pituitary adenomas Medha Suman, Tobias Hallén, Terence Garner, Annika Thorsell, Adam Stevens, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9534122/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 4 You are reading this latest preprint version Abstract Background Non-functioning pituitary adenomas (NFPAs) present a complex clinical challenge due to their indolent and invasive growth patterns, and critical anatomical location. Post-surgical tumor progression is frequent in patients with NFPAs, which often necessitates additional therapeutic interventions. The molecular mechanisms underlying post-surgical progression remain poorly understood and there are currently no reliable methods to stratify patients according to risk of tumor progression. The aim of this study was to comprehensively characterize the molecular alterations, together with an integrated understanding of their interactions, that could potentially uncover the biological processes driving post-surgical tumor progression of NFPAs. Methods We performed an integrated analysis of genome-wide DNA methylation and proteomics in 25 progressive and 15 indolent NFPAs using hypernetwork modelling linking CpG sites to the differentially expressed proteins to identify functional alterations associated with tumor progression. In addition, we investigated cis -regulatory relationships by examining CpG sites located within or in close proximity to the genes encoding the corresponding proteins, allowing assessment of the direct impact of DNA methylation changes on protein expression levels. Results Hypernetwork analysis uncovered extensive indirect and higher-order associations, capturing coordinated epigenetic influences on protein networks in indolent and progressive NFPAs. Progressive NFPAs were characterized by a compact and highly interconnected hub network with proteins primarily involved in DNA replication and transcription regulation (MCM6 and HDGFL2), chromatin organization (SAFB, HDGFL2, KDM3B, and TAF7), and cytoskeleton organization and cell structure maintenance (AJM1 and SYNE2). In contrast, indolent adenomas exhibited a broader and more diffuse network architecture with hub proteins linked to protein processing and transport (PSMD6, APMAP, B4GAT1, and COPE), extracellular matrix organization (LAMB2), and oxidative stress response (CISD2). Hub proteins in progressive NFPAs were enriched for metabolic pathways including glycolysis and tricarboxylic acid cycle, while enriched pathways for hub proteins in the indolent group were associated with genome maintenance and cellular stress responses. Conclusions Hypernetwork analysis highlighted distinct epigenetic-proteomic regulatory mechanisms linked to tumor behavior that were not detected through cis -acting correlation analysis. Collectively, this integrative approach provides insight beyond direct regulation effects and offers a framework for identifying network-informed candidate markers with mechanistic relevance in tumor progression. Non-functioning pituitary adenoma Tumor progression Hypernetwork DNA methylation Protein expression Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Introduction Pituitary adenomas, also referred to as pituitary neuroendocrine tumors under the current World Health Organization classification, account for approximately 10–15% of all intracranial tumors and arise from hormone-producing cells of the anterior pituitary gland ( 1 ). Non-functioning pituitary adenomas (NFPAs) represent a clinically heterogeneous subgroup of pituitary adenomas characterized by the absence of hormone hypersecretion and a typically slow yet often invasive growth pattern ( 2 – 4 ). They comprise 22–54% of all pituitary adenomas and can cause significant clinical challenges due to compression of the normal pituitary gland and surrounding structures, including the optic chiasm and hypothalamus. Consequently, patients may develop visual impairment, hypopituitarism, and in some cases, neurocognitive dysfunction. These conditions substantially impair quality of life and are associated with increased morbidity and mortality among affected individuals ( 5 , 6 ). Surgical resection remains the primary treatment modality for NFPAs ( 7 ). However, complete tumor removal is often limited by local invasion with residual tumor persisting in more than half of patients. Within this subgroup, over 50% undergo tumor progression requiring reintervention, while the remainder exhibit an indolent clinical course ( 5 , 8 – 10 ). Importantly, patients with progressive disease have been shown to have higher mortality compared to those with stable tumors ( 11 ). This variability in clinical behavior, coupled with the lack of reliable predictors of progression, necessitates long-term radiological surveillance for all patients, placing a considerable burden on both the individual and healthcare systems [7]. This highlights a critical gap in personalized management of NFPAs driven by limited understanding of the molecular mechanisms and highlights the urgent need for biomarkers capable of stratifying patients by risk and guiding therapeutic decisions. Genetic alterations appear to play a limited role in NFPA biology as most known mutations are associated with functioning subtypes ( 12 – 15 ). By contrast, epigenetic dysregulation, particularly DNA methylation, has emerged as a key driver of pituitary tumorigenesis ( 16 ). Genome-wide methylation studies have identified distinct epigenetic subgroups associated with tumor lineage, invasiveness, and size ( 17 , 18 ). Complementary proteomic analyses have demonstrated differential protein expression patterns linked to invasiveness ( 19 ). However, most prior studies have focused either on tumor classification or invasiveness, with limited attention to clinically relevant outcomes such as tumor regrowth after surgery that requires re-intervention. In our previous work, we independently characterized genome-wide DNA methylation ( 20 ) and proteomic landscapes ( 21 ) in NFPAs, identifying distinct molecular signatures associated with post-surgical tumor progression. Findings from these studies not only underscored the complexity of tumor biology but also the limitations of single-layer analyses, as gene regulation is governed by complex, multi-level interactions that extend beyond simple linear relationships. In this study, we aimed to expand on previous findings with a more comprehensive multi-omics approach in order to better elucidate the complex molecular mechanisms driving tumor progression. DNA methylation regulates gene and protein expression through both direct ( cis ) and indirect ( trans ) mechanisms and can also drive global changes in expression through effects on chromatin and epigenetic regulators (Fig. 1 ). Traditional integrative strategies often fail to capture the higher-order network structures inherent in biological systems. In contrast, network-based and graph-based models emphasize uncovering functionally relevant modules and interactions that are not apparent when analyzing each data type independently. Among such approaches, hypernetwork modelling has emerged as a powerful framework for capturing higher-order relationships between biological entities ( 22 , 23 ). By simultaneous association of multiple CpG sites with individual proteins or genes, this approach reflects the complex regulatory architecture of epigenetic and proteomic interactions, thus providing a more holistic view of how widespread epigenetic changes may converge on key functional nodes that potentially drive tumor behavior. In this study, we apply a hypernetwork-based integrative framework to combine genome-wide DNA methylation and proteomic data in NFPAs. By constructing protein-centric networks, we aim to identify coordinated epigenetic-proteomic alterations associated with tumor progression. Material and methods Data retrieval and processing This study included a subset of samples with paired DNA methylation and protein expression data derived from our previously published studies ( 20 , 21 ). Patients were grouped into progressive and indolent groups according to the inclusion and exclusion criteria defined previously ( 20 ). Briefly, patients who required reintervention after surgery due to residual tumor progression were classified as having progressive NFPAs, whereas those with residual tumor showing no progression for ≥ 5 years were classified as having indolent NFPAs. Only NFPAs of gonadotropinoma lineage were included to ensure molecular and clinical homogeneity across cohorts. Tumor progression was assessed through neuroradiological evaluation. As the current analysis focused on a subset of samples with paired datasets, all data were reprocessed and normalized specifically for this study, and previously processed data were not reused. Methylation array data Raw methylation intensity files (IDATs) were imported into the R computing environment and data were pre-processed and normalized using the normalization method “noobBMIQ” using the R package ChAMP ( 24 ). For every CpG position in every sample, detection p -value and bead count were calculated and were used to evaluate data quality. Probes with an average detection p -value of > 0.01 were considered unreliable and removed from further analysis. CpG probes that were aligned to multiple sites and with a bead count of 0.05 were removed from further analysis. Batch correction was performed to correct for possible batch effect using “ComBat” batch correction. After data pre-processing and normalization, a total of 734,065 CpG positions remained for further analysis. Beta-values (ratio of methylated probe intensity to the sum of methylated and unmethylated probe intensity) were calculated and used for all downstream analysis. Protein expression data Raw protein abundances of total quantified proteins ( n = 4075) were obtained for all 40 NFPA samples and log2 transformed to improve the interpretability and comparability of expression values. Proteins that were not detected in > 70% of samples in each group were removed and 3008 proteins remained for further analysis. Missing values that remained even after filtering out the proteins with missing data were imputed using R package missMDA ( 25 ). The imputed data were used only when necessary for visualization. Differential protein expression Differential protein expression analysis was performed using the limma ( 26 ) package in R. A false discovery rate threshold 1 was considered significant. Correlation analysis Genes encoding the 457 differentially expressed proteins (DEPs) were identified through ID matching in UniProt and all CpG sites located within or in the vicinity of those genes were selected. Pearson’s correlation was used to calculate the correlation between protein expression levels and DNA methylation levels of the CpG sites encoding these proteins. Hypernetwork analysis We applied hypernetwork modelling to explore the relationship between DEPs and genome-wide DNA methylation variations. In this framework, proteins were treated as nodes, while CpG sites served as hyperedges connecting proteins that share correlated methylation signatures ( 27 – 30 ). Separate protein-methylation hypernetworks were constructed for progressive and indolent NFPAs by calculating Pearson correlations between DEPs (identified between progressive and indolent groups) and methylation levels across all CpG sites. These correlation matrices were then binarized using a correlation coefficient cut-off threshold, which was equal to the standard deviation of the absolute correlation values, to retain only larger (positive or negative) correlations between the two sets of molecular data. The resulting binary matrix served as the incidence matrix of the hypernetwork, where proteins (DEPs) are nodes and CpGs are hyperedges connecting them. To derive the adjacency matrix of the hypernetwork, we multiplied the binarized incidence matrix by its transpose (M × Mᵗ). Each entry in the adjacency matrix reflects the number of shared CpG correlations between pairs of proteins, thereby identifying groups of proteins whose expression is jointly associated with multiple methylation sites ( 31 , 32 ). This adjacency matrix (M × Mᵗ) captures the degree of shared epigenetic influence among proteins and was used for hierarchical clustering to detect densely connected clusters. The adjacency matrices were visualized as heatmaps, and clusters of interest were identified based on the presence of extensive shared CpG associations (i.e., hyperedges) and distinct branching patterns. Gene enrichment analysis Over-representation analysis was performed using WebGestalt ( https://www.webgestalt.org ) ( 33 ) and Reactome database was used with default settings to find the pathways that were enriched in the set of genes. Assessment of copy number alterations Copy number alteration (CNA) plots were generated based on the methylation array data using the R package conumee ( 34 ). CNA plots were reviewed manually to evaluate the presence of CNAs in NFPAs. Results Clinical characteristics and overview of molecular data This study included a total of 25 progressive and 15 indolent NFPAs. Age at NFPA diagnosis for the entire cohort was 20–82 years (median: 59 years). The adenomas were diagnosed as grade 0 to grade 4 according to Knosp grading. The clinical and pathological characteristics of the samples included in this study are presented in Fig. 2 A. For a complete characterization of the epigenomic and proteomic landscape of the NFPAs, we examined DNA methylation levels at 734,065 CpG sites and protein expression of 3008 proteins in 40 NFPA samples. As the primary aim of this study was to elucidate the biological mechanisms underlying the progressive behavior of the NFPAs, we focused our analysis on proteins that showed significant difference in protein expression (|log2FC|>1, false discovery rate threshold < 0.05) between progressive and indolent NFPAs and examined their correlation with the genome-wide DNA methylation levels (Fig. 2 B). Cis -acting relationship between DNA methylation and protein expression To understand cis -acting dynamics between DNA methylation and protein expression in NFPAs, we conducted a Pearson’s correlation analysis between the 457 DEPs and the CpG sites located within or in the vicinity of their respective genes. We performed the analyses separately for indolent and progressive NFPA and subsequently compared the resulting correlation patterns between the groups. The DEPs mapped to a total of 470 genes (some proteins mapped to more than one gene). Of these, 407 genes had overlapping CpG sites ( n = 13,693), while the remaining 63 genes lacked mapped CpGs and were thus excluded from this analysis. Assessing the distribution of CpG sites in various genomic regions, we found that they were predominantly located in the gene body region (53%): they were also distributed in promoter-associated regions including transcription start site 1500 (TSS1500), TSS200, 5' untranslated region (5'UTR), 3'UTR, and the 1st Exon (Fig. 3 A). In relation to the CpG-island context (GC-rich regions with a high density of CpG dinucleotides, often located in gene promoters), most CpGs were located in intergenic regions (42%) followed by CpG islands (26.9%), shores (N-shore 0–2 kilo base pair (kb) upstream of CpG island and S-shore 0–2 kb downstream of CpG island), and the shelf region (N-shelf 2–4 kb upstream of CpG island and S-shelf 2–4 kb downstream of CpG island) (Fig. 3 A). Indolent NFPAs exhibit stronger cis- regulation compared to progressive NFPAs The strength of the correlation between protein expression and DNA methylation was higher in the indolent group, with correlation coefficients ( r ) ranging between − 0.87 to 0.89 compared to the progressive group ( r =–0.79 to 0.77). In progressive NFPAs, most methylation-protein expression correlations were weak (| r |0.5, 0.01%) (Fig. 3 B). In contrast, indolent NFPAs exhibited a higher proportion of strong correlations (| r |>0.5, 8.6%), while weak and moderate correlations accounted for 66.5% and 24.9% of relationships, respectively (Fig. 3 B). In terms of direction of correlation, there was no overall tendency toward positive or negative correlations in either clinical group (Fig. 3 C). To assess whether there was any genomic location preference with respect to correlation direction or correlation strength in either clinical group, we compared the distribution of correlated sites across different regions of the genome based on correlation direction (positive and negative) and strength (weak, moderate, and strong). Contrary to theoretical expectations, we found that CpGs closest to transcription initiation sites and promoter regions exhibited more frequent positive correlations, whereas downstream gene body regions were more strongly associated with negative correlations (Fig. 3 D). Importantly, this pattern was consistent across the outcome groups, despite modest variation in absolute proportions. With respect to the strength of correlation a relatively clearer difference was observed between the outcome groups: progressive NFPAs exhibited an enrichment for weak correlations and a complete absence of strong correlations across the promoter regions compared to indolent NFPAs. Strong correlations were also absent in the 3'UTR and TSS200 regions of the genome in progressive NFPAs (Fig. 3 E). Concordant direct DNA methylation-protein expression regulation is rare in progressive NFPAs To further investigate the cis -regulation between DNA methylation changes and downstream protein expression in progressive and indolent NFPAs, we identified the overlap between differentially methylated genes and DEPs across the outcome groups. Genes showing concordant directionality, i.e., hypermethylation with reduced protein expression and hypomethylation with increased protein expression were considered as under functional epigenetic control. A total of 23,718 CpGs mapping to 16,472 genes were differentially methylated between the progressive and indolent NFPAs. Of these, 336 genes also exhibited differential protein expressions between the outcome groups. Based on methylation-expression concordance, genes were classified into four groups: hypermethylation with upregulation (hyper-up), hypermethylation with downregulation (hyper-down), hypomethylation with upregulation (hypo-up), and hypomethylation with downregulation (hypo-down) (Fig. 4 A). Among CpGs showing the expected inverse methylation-protein expression relationship (hyper-down and hypo-up), only a limited number of gene-protein pairs demonstrated significant differences at both the methylation and protein expression levels. One CpG (cg05046666 located in the shore region [related to TSS1500, 5'UTR] of the gene SPATA20 ) was significantly hypermethylated in the progressive NFPAs, which resulted in a significant downregulation of the encoded protein (log2FC=–1.5). On the other hand, four CpG sites (cg05155595, cg14361862, cg14229540, and cg09050452 mapping to genes ANXA4 , CDS2 , LTBP4 , and DHRS13 , respectively) were significantly hypomethylated in progressive NFPAs and demonstrated a significant upregulation (log2FC > 1) of the corresponding proteins (Fig. 4 B). Since DNA methylation patterns and, ultimately, protein expression can also be influenced by underlying CNAs, we assessed whether the observed differences in protein expression levels between the outcome groups were driven by CNAs. To address this, we analyzed the copy number alteration profiles of all samples (generated based on the methylation data). The majority of NFPAs exhibited flat copy number profiles, indicating an absence of significant CNAs. Only five out of the 40 NFPAs showed detectable focal copy number alterations. Higher-order integration of DNA methylation and proteomics reveals coordinated regulatory networks unique to progressive and indolent NFPAs To capture the complex and higher order interactions between DNA methylation and protein expression, we built a hypernetwork in which a single CpG site can connect to multiple proteins, allowing us to identify coordinated protein networks regulated by epigenetic changes. For hypernetwork modelling, 457 DEPs were represented as nodes and genome-wide methylation levels (734,065 CpG sites) were represented as hyperedges connecting subsets of these nodes. Hypernetworks were constructed separately for progressive and indolent NFPAs. The workflow of hypernetwork analysis is presented in Fig. 5 . The hypernetwork framework in progressive NFPAs derived several protein pairs where multiple CpG sites served as shared regulatory links (Fig. 6 A). Of the 457 DEPs, 307 (67%) were linked to others in the hypernetwork through shared CpG sites. The strength of coordinated epigenetic regulation and, thus, the functional relevance of a protein pair was determined based on the number of shared CpG sites between the proteins that ranged between 28,284 and 323,172. Proteins with a large number of shared CpG positions were referred to as hub proteins and they showed a clear separation from the remaining proteins and clustered in branch 1 that contained a total of 101 proteins in progressive NFPAs (Fig. 6 A). In progressive NFPAs, the top 20 protein pairs that exhibited the highest number of overlapping CpG connections had 15 unique proteins. The top protein pair with 323,172 overlapping CpG correlations were MCM6 and SAFB (Table 1 ). These proteins shared hypernetwork connections with other proteins (HDGFL2, SUPT5H, AJM1, SYNE2, DCD, ATP6AP1, INPP4A, KDM3B, EIF3CL, ATP8A1, FUCA1, CAMK2B, and TBCB). The top 20 hypernetwork protein pairs with the number of overlapping CpG sites between them, the function of genes encoding these proteins, and their possible associations with tumor is presented in Table 1 . Several of these proteins including MCM6, SAFB, SUPT5H, DCD, AJM1, SYNE2, FUCA1, and HDGFL2 were consistent at higher filtering cut-off of 0.5. Table 1 Top 20 protein pairs with most overlapping CpG sites in progressive NFPAs No. Protein Gene encoding the protein Upregulated in progressive NFPAs Gene function Described in relation to pituitary adenoma or other cancer No. of overlapping CpG sites 1 Q14566 MCM6 Yes, p = 0.0007 DNA replication regulation • MCM2 and MCM7 genes from the same family as MCM6 have been shown to be upregulated in pituitary tumors and linked to progression ( 41 , 45 ) • Overexpressed in several cancers including glioma ( 40 ), breast cancer ( 46 , 47 ), lung cancer ( 48 ), hepatocellular carcinoma ( 39 ), and endometrial adenocarcinoma ( 49 ), and associated with cell proliferation, migration, and invasion 323,172 Q15424 SAFB Yes, p = 2.12e-06 Acts as a scaffold protein involved in organizing chromatin structure and recruiting transcriptional corepressors • Role in breast cancer tumorigenesis ( 50 ) 2 Q15424 SAFB As above As above As above 296,880 Q7Z4V5 HDGFL2 Yes, p = 2.9e-05 Involved in cellular growth control through the regulation of cyclin D1 expression • Overexpressed in several cancer including hepatocellular carcinoma ( 51 , 52 ), pancreatic cancer ( 53 ), and colorectal cancer ( 54 ), and contributes to tumor cell proliferation and invasion 3 O00267 SUPT5H Yes, p = 7.9e-07 mRNA processing • Overexpressed in colorectal cancer associated with distant metastasis ( 55 ) • Role in breast cancer tumorigenesis by regulating the expression levels of genes that control proliferation, migration, and cell cycle ( 56 ) 290,448 Q15424 SAFB As above As above As above 4 C9J069 AJM1 Yes, p = 0.0016 Role in control of adherens junction integrity • Identified as key prognostic marker in pancreatic adenocarcinoma ( 57 ) 287,077 Q15424 SAFB As above As above As above 5 Q14566 MCM6 As above As above As above 286,286 Q8WXH0 SYNE2 Yes, p = 5.6e-05 Crucial role in linking the nuclear envelope to the cytoskeleton • Role in p21 regulation 6 P81605 DCD Yes, p = 0.000136 Survival-promoting peptide • Overexpressed in breast cancer (particularly in aggressive forms), and involved in cell proliferation and resistance to apoptosis ( 58 , 59 ) • Overexpressed in colorectal cancer and biomarker for poor prognosis ( 60 ) 281,673 Q14566 MCM6 As above As above As above 7 P81605 DCD As above As above As above 281,579 Q15424 SAFB As above As above As above 8 Q14566 MCM6 As above As above As above 281,472 Q15904 ATP6AP1 Yes, p = 1.4e-06 Transporter activity and ATPase activity • Overexpressed in breast cancer and promotes proliferation ( 61 , 62 ) 9 Q14566 MCM6 As above As above As above 281,417 Q96PE3 INPP4A Yes, p = 1.2e-05 Negative regulator of the phosphoinositide 3-kinase (PI3K)/AKT signaling pathway NA 10 O00267 SUPT5H As above As above As above 278,181 Q14566 MCM6 As above As above As above 11 Q14566 MCM6 As above As above As above As above Q7LBC6 KDM3B Yes, p = 6.3e-05 Regulates gene expression via histone modification • Acts as oncogene in prostate cancer ( 63 ), hepatocellular carcinoma ( 64 ), renal cell carcinoma ( 65 ), and acute lymphoblastic leukemia ( 66 ) promoting cell proliferation 12 B5ME19 EIF3CL Yes, p = 1.3e-06 Protein synthesis • Overexpressed in diffuse intrinsic pontine glioma and associated with cell proliferation and anchorage-independent growth ( 67 ) 273,273 Q15904 ATP6AP1 As above As above As above 13 Q15424 SAFB As above As above As above 270,136 Q96PE3 INPP4A As above As above As above 14 B5ME19 EIF3CL Yes, p = 1.3e-06 As above As above 269,770 Q14566 MCM6 As above As above As above 15 Q15904 ATP6AP1 As above As above As above 268,036 Q9Y2Q0 ATP8A1 Yes, p = 3.8e-12 Role in maintaining membrane structure and function by proper distribution of phospholipids • Overexpressed in non-small cell lung cancer and associated with increased migration ability ( 37 ) 16 P04066 FUCA1 Yes, p = 0.000204 • p53 target, regulates growth and survival of cancer cells 266,054 Q14566 MCM6 As above As above As above 17 B5ME19 EIF3CL Yes, p = 1.3e-06 As above As above 265,910 Q15424 SAFB As above As above As above 18 Q13554 CAMK2B No, p = 3.1e-09 Calcium signaling with roles in cytoskeletal remodeling • Mediates both microenvironmental remodeling and tumor progression ( 68 ) 265,430 Q99426 TBCB Yes, p = 0.00138 Tubulin folding cofactor involved in cytoskeletal dynamics • Promotes cell proliferation and associated with poor prognosis in acute myeloid leukemia ( 69 ) 19 Q14566 MCM6 As above As above As above 264,997 Q7Z4V5 HDGFL2 As above As above As above 20 Q15424 SAFB As above As above As above 264,464 Q15904 ATP6AP1 As above As above As above NA, not applicable; NFPA, non-functioning pituitary adenoma. In the case of indolent NFPAs, 70% (318/457) DEPs formed paired connections based on shared CpG sites that ranged between 18,337 to 378,731 CpGs. Hub protein pairs in indolent NFPAs clustered in branch 2 that contained a total of 147 proteins (Fig. 6 B). The top 20 protein pairs that exhibited the highest number of overlapping CpG connections had 11 unique proteins (LAMB2, PSMD6, COPE, APMAP, CISD2, ATP6V1D, B4GAT1, GNAQ, NPTN, PPP2R2A, and RUNDC3A]. The top 20 hypernetwork protein pairs with the number of overlapping CpG sites between them, the function of genes encoding these proteins, and their possible associations with tumor are presented in Table 2 . We compared the protein branches that contained the hub protein pairs in progressive (branch 1) and indolent (branch 2) NFPAs to understand if they consist of the same or different proteins. Interestingly, we found limited overlap, with only 28 proteins common to hypernetwork clusters (Fig. 6 C). Table 2 Top 20 protein pairs with most overlapping CpG sites in indolent NFPAs No. Protein Gene encoding protein Upregulated in progressive NFPAs Gene function Described in relation to pituitary adenoma or other cancer No. of overlapping CpG sites 1 P55268 LAMB2 Yes, p = 0.02 Mediate the attachment, migration, and organization of cells into tissues during embryonic development by interacting with other extracellular matrix components • Promotes metastasis in gastric cancer ( 70 ) 376,215 Q15008 PSMD6 Yes, p = 4.1e-06 Involved in the ATP-dependent degradation of ubiquinated proteins • Overexpressed in hepatocellular carcinoma and essential for tumor cell growth ( 71 ) 2 O14579 COPE Yes, p = 1.9e-06 Mediate biosynthetic protein transport NA 363,088 Q15008 PSMD6 As above As above As above 3 P55268 LAMB2 As above As above As above 360,499 Q9HDC9 APMAP Yes, p = 1.2e-08 Involved in biosynthetic processes • Promotes epithelial-mesenchymal transition and metastasis in cervical cancer ( 72 ), prostate cancer ( 73 ) 4 Q15008 PSMD6 As above As above As above 344,893 Q8N5K1 CISD2 Yes, p = 1.4e-06 Involved in the regulation of aging, skeletal muscle maintenance, neurodegeneration, cancer, autophagy, and apoptosis • Overexpressed in liver cancer ( 74 ), early-stage cervical cancer ( 75 ), and gastric cancer ( 76 ), and associated with increased proliferation and enhanced progression in all three tumor types 5 Q15008 PSMD6 As above As above As above 341,296 Q9HDC9 APMAP As above As above As above 6 P55268 LAMB2 As above As above As above 337,050 Q9Y5K8 ATP6V1D Yes, p = 8.8e-08 Proton-transporting ATPase activity • Promotes stemness and progression in hepatocellular carcinoma ( 77 ) 7 O43505 B4GAT1 Yes, p = 2.0e-05 Protein modification; protein glycosylation • Promotes metastasis in hepatocellular carcinoma ( 78 ) 332,998 P50148 GNAQ Yes, p = 1.1e-07 Vital for proper cellular signaling and vascular development • Frequently mutated in uveal melanomas ( 79 ) 8 Q9HDC9 APMAP As above As above As above 332,990 Q9Y639 NPTN Yes, p = 2.6e-07 Cell adhesion molecule binding NA 9 Q15008 PSMD6 As above As above As above 332,825 Q9Y639 NPTN As above As above As above 10 O43505 B4GAT1 As above As above As above 331,665 Q9Y639 NPTN As above As above As above 11 P55268 LAMB2 As above As above As above 327,606 Q8N5K1 CISD2 As above As above As above 12 O14579 COPE As above As above As above 326,809 P63151 PPP2R2A Yes, p = 2.3e-07 Regulation of chromosome segregation, protein phosphatase regulator activity • Promotes cancer via deregulated signaling and DNA repair defects ( 80 ) 13 P50148 GNAQ As above As above As above 326,474 P55268 LAMB2 As above As above As above 14 P50148 GNAQ As above As above As above 326,447 Q15008 PSMD6 As above As above As above 15 O43505 B4GAT1 As above As above As above 325,727 Q59EK9 RUNDC3A Yes, p = 1.4e-05 GTPase regulator activity • Promotes tumor progression and chemoresistance in gastric neuroendocrine carcinoma ( 81 ) 16 P50148 GNAQ As above As above As above 323,536 Q59EK9 RUNDC3A As above As above As above 17 O14579 COPE As above As above As above 323,267 P55268 LAMB2 As above As above As above 18 Q59EK9 RUNDC3A As above As above As above 323,081 Q9Y639 NPTN As above As above As above 19 P55268 LAMB2 As above As above As above 322,715 Q9Y639 NPTN As above As above As above 20 O14579 COPE As above As above As above 321,589 P50148 GNAQ As above As above As above NA, not applicable; NFPA, non-functioning pituitary adenoma. Genomic distribution of hyperedges To further characterize the hypernetworks, we analyzed CpG sites (hyperedges) linked to hub protein clusters. These CpGs represent methylation regions most strongly associated with the identified protein networks. In indolent NFPAs, 540 CpG sites were connected to more than 60% of proteins in the hypernetwork, whereas 9961 CpG sites met this criterion in the progressive group. Notably, only five CpG sites were shared between the two clinical groups among those connected to over 60% of proteins. In progressive NFPAs, we further identified 1248 hyperedges connected to a higher proportion of proteins (> 70%), a pattern that was not observed in indolent NFPAs. Figure 7 shows the genomic distribution of the hyperedges forming connections to > 60% of hypernetwork proteins in the two outcome groups. Hyperedges in the progressive group showed enrichment for promoter-associated regions including CpG island and N-shore regions (corresponding to TSS200), while hyperedges in the indolent NFPAs were enriched in intergenic regions of the genome (Fig. 7 A). We also assessed their associations with regulatory features and enhancer regions across the genome. In terms of the regulatory features, the hyperedges were enriched for unclassified cell-type specific regulatory regions (Fig. 7 B). These regions are mainly involved in the regulatory functions that do not directly involve the gene promoter, suggesting that the hypernetwork proteins expose the functional regulatory networks which are not necessarily located in the promoter regions of the gene. We found that the hyperedges associated with both progressive and indolent NFPAs were enriched for enhancer regions as defined in the 450K version of the methylation array (Fig. 7 C). Hub proteins are upregulated in progressive NFPAs and enriched for biological processes involved in cancer Over-representation analysis involving all 101 proteins in branch 1 of the hypernetwork, in which the hub protein pairs clustered in progressive NFPAs, revealed that most of the proteins were membrane proteins involved in metabolic processes (Biological process category) and protein binding (Molecular function category) (Fig. 8 A). Among the most enriched Reactome pathways were the Citric acid cycle (TCA cycle) (R-HSA-71403), Glucose metabolism (R-HSA-70326), snRNP Assembly (R-HSA-191859), Metabolism of non-coding RNA (R-HSA-194441), and Metabolism of carbohydrates (R-HSA-71387). In case of indolent NFPAs, the majority of the proteins in branch 2 (147 proteins) were membrane proteins involved in metabolic processes (Biological process category) and protein binding (Molecular function category) (Fig. 8 B). Among the top enriched Reactome pathways were Cellular responses to stimuli (R-HSA-8953897), Packaging of telomere ends (R-HSA-171306), Cellular responses to stress (R-HSA-2262752), Recognition and association of DNA glycosylase with site containing an affected purine (R-HSA-110330), and Cleavage of the damaged purine (R-HSA-110331). All top hub proteins identified in both progressive and indolent NFPAs were significantly upregulated in progressive NFPAs, except for CAMK2B, which was downregulated in the progressive group. Discussion Integrated analysis of genome-wide DNA methylation and protein expression data in 25 progressive and 15 indolent NFPAs revealed that cis -acting DNA methylation-protein expression regulations are relatively uncommon in NFPAs. This relationship was particularly weak in progressive NFPAs with only a small fraction of strong associations, suggesting that methylation-driven control of individual protein levels is limited or may be disrupted in more aggressive disease states. In contrast, network-based hypernetwork analysis uncovered a broader and more coordinated association with DNA methylation highlighting indirect and higher-order regulatory effects. This approach captured relationships that are not apparent in direct analyses, indicating that, while methylation may not strongly dictate individual protein expression, it may still play a significant role in shaping protein interaction networks and pathway-level regulation. Overall, methylation-protein cis -regulation was stronger and more prevalent in indolent tumors, whereas progressive NFPAs were mainly dominated by weak correlations. Although correlations between DNA methylation and protein expression were observed in both progressive and indolent NFPAs, cases showing significant concordance where hypermethylation corresponded to a significant downregulation or hypomethylation to a significant upregulation of protein expression was limited. Only a small number of sporadic CpG-protein pairs showed the expected inverse relationship with protein expression (Fig. 4 B). This was consistent with previous reports demonstrating predominantly low-to-moderate direct correlations between DNA methylation and protein abundance ( 35 ). Consistent with the literature, there was limited prevalence of CNAs in NFPAs, which suggests that they are unlikely to play a major role in driving the observed alterations in protein abundance ( 36 ). Instead, these findings indicate that proteomic changes in NFPAs may be influenced less by absolute gains or losses in DNA methylation and more by alternative regulatory mechanisms, such as microRNA activity, chromatin remodeling, or post-translational modifications, mechanisms not captured by direct methylation-protein correlation analysis. Direct correlation analyses may also fail to capture the complexity of underlying network interactions. For example, scenarios, where methylation at gene A influences the expression of another gene that subsequently modulates protein A through network-mediated effects, as illustrated in Fig. 1 . This highlights the importance of considering regulatory dynamics, rather than static methylation levels alone, when interpreting epigenetic alterations in disease progression. By deploying a hypernetwork model that treats each CpG as a hyperedge connecting multiple proteins, we captured thousands of indirect and higher-order associations that pairwise methods would miss, highlighting coordinated epigenetic regulation in both progressive and indolent NFPAs. Such architectures better reflect the reality that single CpGs can act through enhancers, chromatin loops, or trans -acting factors to influence distant proteins. Notably, a substantial proportion of DEPs were incorporated into the formation of hypernetwork in both groups (67% in progressive and 70% in indolent NFPAs), indicating that DNA methylation broadly contributes to structuring proteomic interactions. We observed progressive NFPAs formed a compact network of highly interconnected hub proteins within branch 1, which suggests the presence of tightly coordinated epigenetic regulation affecting a compact and highly interconnected protein network. The large number of shared CpG sites between protein pairs, reaching over 300,000 in some cases, indicates shared epigenetic control and potentially strong co-regulation. This concentrated and interconnected network architecture may reflect a more synchronized regulatory program that enables rapid tumor growth and progression. Key hub proteins in progressive NFPAs, including MCM6, SAFB, and SUPT5H, are associated with core cellular processes such as DNA replication, transcriptional regulation, and chromatin organization. Several of these proteins have been reported in different cancers and have roles in cell proliferation, migration, and tumor growth (Table 1 ). Li et al. (2017) showed that ATP8A1 knock-down showed reduced migration ability in non-small cell lung cancer cells ( 37 ). MCM6, which is the top hub protein node in the hypernetwork analysis of progressive NFPAs, plays a key role in DNA replication regulation and cell cycle progression. It is overexpressed in many different cancers and considerably lower expression has been reported in normal tissue. There are several studies suggesting its role in cell proliferation, potentially promoting cancer progression ( 38 – 40 ). Another gene of the MCM family has previously been identified as a biomarker for post-surgical progression in NFPAs ( 41 ). Among the hub proteins, many have roles in chromatin binding and histone modification, such as SAFB, HDGFL2, KDM3B, and TAF7, which indicates a possible global impact on protein expression. In contrast, indolent NFPAs exhibited a larger hub-containing branch (147 proteins) with an even broader range of CpG sharing, suggesting a more diffuse but extensive epigenetic network. The hub proteins in this group, such as LAMB2, PPP2R2A, and GNAQ, are linked to cellular signaling, structural organization, and regulatory pathways. They are reported in a range of malignancies and were primarily associated with tumor growth and metastasis (Table 2 ). A key finding is the limited overlap between the hub-associated branches of progressive and indolent NFPAs, with only 28 shared proteins. This difference underscores the presence of distinct epigenetic-proteomic regulatory programs underlying different tumor behavior. While both groups exhibit extensive methylation-associated connectivity, the specific proteins and network structures differ markedly. Over-representation analysis of proteins within the hypernetwork branches revealed distinct biological signatures between progressive and indolent NFPAs. In progressive NFPAs, branch 1 proteins were predominantly membrane-associated and enriched for metabolic processes and protein-binding functions. The prominence of pathways such as the tricarboxylic acid cycle, glucose metabolism, and broader carbohydrate metabolism points toward enhanced metabolic activity and bioenergetic reprogramming. This metabolic shift is a well-recognized hallmark of aggressive tumor behavior, supporting increased proliferation, survival, and adaptation to microenvironmental demands ( 42 – 44 ). Additionally, enrichment of RNA-related processes, including small nuclear ribonucleic acid particle (snRNP) assembly and non-coding RNA metabolism, suggests heightened post-transcriptional regulation. In contrast for indolent NFPAs, the enriched pathways involving hub proteins were associated with cellular responses to stimuli and stress as well as DNA damage recognition and repair mechanisms. Enrichment of pathways related to telomere maintenance and base excision repair indicate a greater emphasis on genomic stability and controlled cellular responses, which may contribute to their more stable and less progressive behavior. Hypernetwork hubs represent mechanistically coherent, network-stabilized proteins that are more likely to be robust across cohorts and biologically actionable. Biologically relevant processes are typically reflected by coordinated regulation of multiple functionally related proteins in the same direction, strengthening evidence for true alterations rather than random variation. Given that robustness is a key requirement for biomarker development, the proteins identified in this analysis represent promising candidates for future studies aimed at identifying biomarkers of progression. A major limitation of this study is the lack of external validation, and these findings require confirmation in an independent cohort. Future studies with larger sample size and integrative multi-omics approaches including transcriptomic and chromatin accessibility data will further strengthen these findings by enabling deeper dissection of the regulatory mechanisms linking DNA methylation to protein expression and by validating their functional relevance. Conclusion This study demonstrates that direct regulation of protein expression by DNA methylation is limited in NFPAs, particularly in progressive tumors. Instead, our findings revealed that proteomic landscapes in NFPAs are shaped by complex, multi-layered regulatory mechanisms. Distinct epigenetic-proteomic network architecture observed in progressive and indolent NFPAs suggests the presence of fundamentally different regulatory programs underlying the different tumor behavior. The identification of hypernetwork hub proteins provides a set of biologically meaningful and potentially robust candidates for future biomarker development. Further validation in independent cohorts and integration with additional omics layers will be essential to confirm these findings and to better understand the regulatory landscape driving NFPA progression. Abbreviations CNA Copy number alteration DEP Differentially expressed protein kb kilo base pair NFPA Non–functioning pituitary adenoma TSS Transcription start site UTR Untranslated region Declarations Acknowledgements We are grateful to Peter Todd (Tajut Ltd., Kaiapoi, New Zealand) for third-party writing assistance in drafting of this manuscript, for which he received financial compensation from ALF-funding. Author contributions Medha Suman, Gudmundur Johannsson, Thomas Skoglund, Tobias Hallén, Terence Garner, and Adam Stevens designed the study and methodology. All authors collected data from the original study. Medha Suman, Gudmundur Johannsson, Tobias Hallén, Annika Thorsell, Helena Carén, Oskar Ragnarsson, Linus Köster, and Thomas Skoglund contributed with resources and planning of the study. Gudmundur Johannsson and Thomas Skoglund obtained funding. Gudmundur Johannsson, Tobias Hallén, Thomas Skoglund, Adam Stevens, Oskar Ragnarsson, Annika Thorsell, and Helena Carén supervised the study. Medha Suman, Terence Garner, and Adam Stevens had full access to all data and performed the analyses. Medha Suman, Helena Carén, Tobias Hallén, Thomas Skoglund, Oskar Ragnarsson, Adam Stevens, Annika Thorsell, and Gudmundur Johannsson drafted the manuscript, and all authors revised it and approved the final version. Funding The study was supported by grants from Västra Götalandsregionen through the Swedish state under the agreement between the Swedish government and the County Councils, the ALF-agreement (ALFGBG-772591, ALFGBG -1006371 and ALFGBG-966066), the Swedish Cancer Society (21 1774 Pj & 24 3687 Pj). Data availability All data supporting the findings of this study are available from the corresponding author on reasonable request. Ethics approval and consent to participate The study was approved by the Regional Ethical Review Board in Gothenburg (Dnr: 100-15). The research was conducted in accordance with the Declaration of Helsinki. Consent to Participate declarations: not applicable. Consent for publication Not applicable. Competing interests Gudmundur Johannsson has served as a consultant for Crinetics, Lundbeck, Novo Nordisk, and AstraZeneca, and has received lecture fees from Ascendic Pharma, Crinetics, Novo Nordisk, and Pharmanovia. All other authors declare no conflicts of interest. References Ezzat S, Asa SL, Couldwell WT, Barr CE, Dodge WE, Vance ML, et al. The prevalence of pituitary adenomas: a systematic review. Cancer. 2004;101:613–9. Fernandez A, Karavitaki N, Wass JA. Prevalence of pituitary adenomas: a community-based, cross-sectional study in Banbury (Oxfordshire, UK). Clin Endocrinol (Oxf). 2010;72:377–82. Aflorei, ED, Korbonits M. Epidemiology and etiopathogenesis of pituitary adenomas. J Neurooncol. 2014;117:379-94. Tjörnstrand A, Gunnarsson K, Evert M, Holmberg E, Ragnarsson O, Rosén T, et al. The incidence rate of pituitary adenomas in western Sweden for the period 2001–2011. Eur J Endocrinol. 2014;171:519-26. Melmed S, Kaiser UB, Lopes MB, Bertherat J, Syro LV, Raverot G, et al. Clinical biology of the pituitary adenoma. Endocr Rev. 2022;43:1003–37. Higham CE, Johannsson G, Shalet SM. Hypopituitarism. Lancet. 2016;388:2403–15. Esposito D, Olsson DS, Ragnarsson O, Buchfelder M, Skoglund T, Johannsson G. Non-functioning pituitary adenomas: indications for pituitary surgery and post-surgical management. Pituitary. 2019;22:422–434. Olsson DS, Buchfelder M, Schlaffer S, Bengtsson BA, Jakobsson KE, Johannsson G, et al. Comparing progression of non-functioning pituitary adenomas in hypopituitarism patients with and without long-term GH replacement therapy. Eur J Endocrinol. 2009 Nov;161:663–9. Olsson DS, Buchfelder M, Wiendieck K, Kremenevskaja N, Bengtsson BÅ, Jakobsson KE, et al. Tumour recurrence and enlargement in patients with craniopharyngioma with and without GH replacement therapy during more than 10 years of follow-up. Eur J Endocrinol. 2012;166:1061–8. Chen Y, Wang CD, Su ZP, Chen YX, Cai L, Zhuge QC, et al. Natural history of postoperative nonfunctioning pituitary adenomas: a systematic review and meta-analysis. Neuroendocrinology. 2012;96:333–42. Erfurth EM, Bulow B, Nordström CH, Mikoczy Z, Hagmar L, Strömberg U. Doubled mortality rate in irradiated patients reoperated for regrowth of a macroadenoma of the pituitary gland. Eur J Endocrinol. 2004;150:497–502. Mayr B, Apenberg S, Rothämel T, von zur Mühlen A, Brabant G. Menin mutations in patients with multiple endocrine neoplasia type 1. Eur J Endocrinol. 1997;137:684–7. Beckers A, Aaltonen LA, Daly AF, Karhu A. Familial isolated pituitary adenomas (FIPA) and the pituitary adenoma predisposition due to mutations in the aryl hydrocarbon receptor interacting protein (AIP) gene. Endocr Rev. 2013;34:239–77. Reincke M, Sbiera S, Hayakawa A, Theodoropoulou M, Osswald A, Beuschlein F, et al. Mutations in the deubiquitinase gene USP8 cause Cushing's disease. Nat Genet. 2015;47:31–8. Mantovani G, Lania AG, Spada A. GNAS imprinting and pituitary tumors. Mol Cell Endocrinol. 2010;326:15–8. Peterson R, Cote DJ, Zada G. Current understanding of the role of DNA methylation in pituitary neuroendocrine tumors. Neurooncol Adv. 2025;7 Suppl 1:i2–9. Srirangam Nadhamuni V, Korbonits M. Novel insights into pituitary tumorigenesis: genetic and epigenetic mechanisms. Endocr Rev. 2020;41:821–46. Miyake Y, Adachi JI, Suzuki T, Mishima K, Araki R, Mizuno R, et al. TERT promoter methylation is significantly associated with TERT upregulation and disease progression in pituitary adenomas. J Neurooncol. 2019;141:131–8. Zhang Y, Li L, Ma X, Liu C, Liu G, Bie Z, et al. Quantitative proteomics identified a novel invasion biomarker associated with EMT in pituitary adenomas. Front Endocrinol (Lausanne). 2023;14:1137648. Hallén T, Johannsson G, Dahlén R, Glad CAM, Örndal C, Engvall A, et al. Genome-wide DNA methylation differences in nonfunctioning pituitary adenomas with and without postsurgical progression. J Clin Endocrinol Metab. 2022;107:2318–28. Hallén T, Johannsson G, Thorsell A, Olsson DS, Örndal C, Engvall A, et al. Proteomic profiles associated with postsurgical progression in nonfunctioning pituitary adenomas. J Clin Endocrinol Metab. 2024;109:1485–93. Lugo-Martinez J, Zeiberg D, Gaudelet T, Malod-Dognin N, Przulj N, Radivojac P. Classification in biological networks with hypergraphlet kernels. Bioinformatics. 2021;37:1000–7. Feng L, Gong H, Zhang S, Liu X, Wang Y, Che J, et al. Hypernetwork modeling and topology of high-order interactions for complex systems. Proc Natl Acad Sci U S A. 2024;121:e2412220121. Tian Y, Morris TJ, Webster AP, Yang Z, Beck S, Feber A, et al. ChAMP: updated methylation analysis pipeline for Illumina BeadChips. Bioinformatics. 2017;33:3982–4. Josse J, Husson F. missMDA: a package for handling missing values in multivariate data analysis. Journal of Statistical Software, 2016;70:1–31. Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015;43:e47. Gaudelet T, Malod-Dognin N, Pržulj N. Higher-order molecular organization as a source of biological function. Bioinformatics. 2018;34:i944–53. Johnson J. Hypernetworks in the science of complex systems. In: Complex Sciences. London: Imperial College Press; 2013. https://books.google.co.nz/books?hl=en&lr=&id=jQO4CgAAQBAJ&oi=fnd&pg=PP7&dq=hypernetworks+in+the+science+of+complex+systems+johnson+%222012%22&ots=_Y1yr7dHOw&sig=Gs0x9cY8nqKyH37j8Y2TMlP98pw#v=onepage&q=hypernetworks%20in%20the%20science%20of%20complex%20systems%20jjohnso%20%222011%22&f=false. Accessed 18 Apr 2026. Pearcy N, Chuzhanova N, Crofts JJ. Complexity and robustness in hypernetwork models of metabolism. J Theor Biol. 2016;406:99–104. Ripley DM, Garner T, Stevens A. Developing the 'omic toolkit of comparative physiologists. Comp Biochem Physiol Part D Genomics Proteomics. 2024;52:101287. Contreras-Aso G, Criado R, Vera de Salas G, Yang J. Detecting communities in higher-order networks by using their derivative graphs. Chaos, Solitons & Fractals, 2023;177:114200. Kumar T, Vaidyanathan S, Ananthapadmanabhan H, Parthasarathy S, Ravindran B. Hypergraph clustering by iteratively reweighted modularity maximization. Applied Network Science 2020;5:52. Elizarraras JM, Liao Y, Shi Z, Zhu Q, Pico AR, Zhang B. WebGestalt 2024: faster gene set analysis and new support for metabolomics and multi-omics. Nucleic Acids Res. 2024;52:W415–21. Daenekas B, Pérez E, Boniolo F, Stefan S, Benfatto S, Sill M, et al. Conumee 2.0: enhanced copy-number variation analysis from DNA methylation arrays for humans and mice. Bioinformatics. 2024;40:btae029. Liang WW, Lu RJ, Jayasinghe RG, Foltz SM, Porta-Pardo E, Geffen Y, et al; Clinical Proteomic Tumor Analysis Consortium. Integrative multi-omic cancer profiling reveals DNA methylation patterns associated with therapeutic vulnerability and cell-of-origin. Cancer Cell. 2023;41:1567–85.e7. Bi WL, Horowitz P, Greenwald NF, Abedalthagafi M, Agarwalla PK, Gibson WJ, et al. Landscape of genomic alterations in pituitary adenomas. Clin Cancer Res. 2017;23:1841–51. Li D, Xu T, Wang X, Ma X, Liu T, Wang Y, et al. The role of ATP8A1 in non-small cell lung cancer. Int J Clin Exp Pathol. 2017;10:7760–6. Zeng T, Guan Y, Li YK, Wu Q, Tang XJ, Zeng X, et al. The DNA replication regulator MCM6: an emerging cancer biomarker and target. Clin Chim Acta. 2021;517:92–8. Liu Z, Li J, Chen J, Shan Q, Dai H, Xie H, et al. MCM family in HCC: MCM6 indicates adverse tumor features and poor outcomes and promotes S/G2 cell cycle progression. BMC Cancer. 2018;18:200. Cai HQ, Cheng ZJ, Zhang HP, Wang PF, Zhang Y, Hao JJ, et al. Overexpression of MCM6 predicts poor survival in patients with glioma. Hum Pathol. 2018;78:182–7. Hallén T, Olsson DS, Hammarstrand C, Örndal C, Engvall A, Ragnarsson O, et al. MCM7 as a marker of postsurgical progression in non-functioning pituitary adenomas. Eur J Endocrinol. 2021;184:521–31. Xia L, Oyang L, Lin J, Tan S, Han Y, Wu N, Yi P, Tang L, et al. The cancer metabolic reprogramming and immune response. Mol Cancer. 2021;20:28. Sun L, Suo C, Li ST, Zhang H, Gao P. Metabolic reprogramming for cancer cells and their microenvironment: beyond the Warburg effect. Biochim Biophys Acta Rev Cancer. 2018;1870:51–66. Lue HW, Podolak J, Kolahi K, Cheng L, Rao S, Garg D, et al. Metabolic reprogramming ensures cancer cell survival despite oncogenic signaling blockade. Genes Dev. 2017;31:2067–84. Huang Y, Wang Q, Zhou W, Jiang Y, He K, Huang W, et al. Prenatal p25-activated Cdk5 induces pituitary tumorigenesis through MCM2 phosphorylation-mediated cell proliferation. Neoplasia. 2024;57:101054. Cheng L, Tan Z, Huang Z, Pan Y, Zhang W, Wang J. Expression profile and prognostic values of mini-chromosome maintenance families (MCMs) in breast cancer. Med Sci Monit. 2020;26:e923673. Issac MSM, Yousef E, Tahir MR, Gaboury LA. MCM2, MCM4, and MCM6 in breast cancer: clinical utility in diagnosis and prognosis. Neoplasia. 2019;21:1015–35. Liu YZ, Wang BS, Jiang YY, Cao J, Hao JJ, Zhang Y, et al. MCMs expression in lung cancer: implication of prognostic significance. J Cancer. 2017;8:3641–7. Hotton J, Agopiantz M, Leroux A, Charra-Brunaud C, Marie B, Busby-Venner H, et al. Minichromosome maintenance complex component 6 (MCM6) expression correlates with histological grade and survival in endometrioid endometrial adenocarcinoma. Virchows Arch. 2018;472:623–33. Hammerich-Hille S, Bardout VJ, Hilsenbeck SG, Osborne CK, Oesterreich S. Low SAFB levels are associated with worse outcome in breast cancer patients. Breast Cancer Res Treat. 2010;121:503–9. Enomoto H, Nakamura H, Liu W, Nishiguchi S. Hepatoma-derived growth factor: its possible involvement in the progression of hepatocellular carcinoma. Int J Mol Sci. 2015;16:14086–97. Yoshida K, Nakamura H, Okuda Y, Enomoto H, Kishima Y, Uyama H, et al. Expression of hepatoma-derived growth factor in hepatocarcinogenesis. J Gastroenterol Hepatol. 2003;18:1293–301. Uyama H, Tomita Y, Nakamura H, Nakamori S, Zhang B, Hoshida Y, et al. Hepatoma-derived growth factor is a novel prognostic factor for patients with pancreatic cancer. Clin Cancer Res. 2006;12:6043–8. Liao F, Dong W, Fan L. Apoptosis of human colorectal carcinoma cells is induced by blocking hepatoma-derived growth factor. Med Oncol. 2010;27:1219–26. Chen R, Zhu J, Dong Y, He C, Hu X. Suppressor of Ty homolog-5, a novel tumor-specific human telomerase reverse transcriptase promoter-binding protein and activator in colon cancer cells. Oncotarget. 2015;6:32841–55. Lone BA, Ahmad F, Karna SKL, Pokharel YR. SUPT5H Post-transcriptional silencing modulates PIN1 expression, inhibits tumorigenicity, and induces apoptosis of human breast cancer cells. Cell Physiol Biochem. 2020;54:928–46. Ma T, Gongye X, Dongzhi C, Chai Y, Wang Q, Tian M. Machine learning-based WGCNA approach for developing an immunogenic cell death-related hub gene signature and identification of AJM1 as a prognostic biomarker in pancreatic adenocarcinoma. Int J Med Sci. 2025;22:4493–508. Brauer HA, D'Arcy M, Libby TE, Thompson HJ, Yasui YY, Hamajima N, et al. Dermcidin expression is associated with disease progression and survival among breast cancer patients. Breast Cancer Res Treat. 2014;144:299–306. Bancovik J, Moreira DF, Carrasco D, Yao J, Porter D, Moura R, et al. Dermcidin exerts its oncogenic effects in breast cancer via modulation of ERBB signaling. BMC Cancer. 2015;15:70. Zhang S, Wang Y, Zhang X, Wang M, Wu H, Tao Y, et al. ATP6AP1 as a potential prognostic biomarker in CRC by comprehensive analysis and verification. Sci Rep. 2024;14:4018. Yan Z, Huang A, Ma D, Hong C, Zhang S, He L, et al. ATP6AP1 promotes cell proliferation and tamoxifen resistance in luminal breast cancer by inducing autophagy. Cell Death Dis. 2025;16:201. Wang J, Liu Y, Zhang S. Prognostic and immunological value of ATP6AP1 in breast cancer: implications for SARS-CoV-2. Aging (Albany NY). 2021;13:16904–21. Björkman M, Östling P, Härmä V, Virtanen J, Mpindi JP, Rantala J, et al. Systematic knockdown of epigenetic enzymes identifies a novel histone demethylase PHF8 overexpressed in prostate cancer with an impact on cell proliferation, migration and invasion. Oncogene. 2012;31:3444–56. An MJ, Kim DH, Kim CH, Kim M, Rhee S, Seo SB, et al. Histone demethylase KDM3B regulates the transcriptional network of cell-cycle genes in hepatocarcinoma HepG2 cells. Biochem Biophys Res Commun. 2019;508:576–82. Mahamdallie S, Yost S, Poyastro-Pearson E, Holt E, Zachariou A, Seal S, et al. Identification of new Wilms tumour predisposition genes: an exome sequencing study. Lancet Child Adolesc Health. 2019;3:322–31. Kim JY, Kim KB, Eom GH, Choe N, Kee HJ, Son HJ, et al. KDM3B is the H3K9 demethylase involved in transcriptional activation of lmo2 in leukemia. Mol Cell Biol. 2012;32:2917–33. Chen J, Lin Z, Barrett L, Dai L, Qin Z. Identification of new therapeutic targets and natural compounds against diffuse intrinsic pontine glioma (DIPG). Bioorg Chem. 2020;99:103847. Jia Q, Liao X, Zhang Y, Xu B, Song Y, Bian G, et al. Anti-tumor role of CAMK2B in remodeling the stromal microenvironment and inhibiting proliferation in papillary renal cell carcinoma. Front Oncol. 2022;12:740051. Wang B, Wang W, Li Q, Guo T, Yang S, Shi J, et al. High expression of microtubule-associated protein TBCB predicts adverse outcome and immunosuppression in acute myeloid leukemia. J Cancer. 2023;14:1707–24. Li D, Jin Y, He X, Deng J, Lu W, Yang Z, et al. Hypoxia-induced LAMB2-enriched extracellular vesicles promote peritoneal metastasis in gastric cancer via the ROCK1-CAV1-Rab11 axis. Oncogene. 2024;43:2768–80. Zhou SS, Ye YP, Chen Y, Zeng DT, Zheng GC, He RQ, et al. Overexpression pattern, function, and clinical value of proteasome 26S subunit non-ATPase 6 in hepatocellular carcinoma. World J Clin Oncol. 2025;16:99839. Zhu X, Xiang Z, Zou L, Chen X, Peng X, Xu D. APMAP promotes epithelial-mesenchymal transition and metastasis of cervical cancer cells by activating the Wnt/β-catenin pathway. J Cancer. 2021;12:6265–73. Jiang S, Wang X, Song D, Liu X, Gu Y, Xu Z, et al. Cholesterol induces epithelial-to-mesenchymal transition of prostate cancer cells by suppressing degradation of EGFR through APMAP. Cancer Res. 2019;79:3063–75. Chen B, Shen S, Wu J, Hua Y, Kuang M, Li S, et al. CISD2 associated with proliferation indicates negative prognosis in patients with hepatocellular carcinoma. Int J Clin Exp Pathol. 2015;8:13725–38. Liu L, Xia M, Wang J, Zhang W, Zhang Y, He M. CISD2 expression is a novel marker correlating with pelvic lymph node metastasis and prognosis in patients with early-stage cervical cancer. Med Oncol. 2014;31:183. Wang L, Ouyang F, Liu X, Wu S, Wu HM, Xu Y, et al. Overexpressed CISD2 has prognostic value in human gastric cancer and promotes gastric cancer cell proliferation and tumorigenesis via AKT signaling pathway. Oncotarget. 2016;7:3791–805. Xu Z, Liu R, Ke H, Xu F, Yang P, Zhang W, et al. ATP6V1D drives hepatocellular carcinoma stemness and progression via both lysosome acidification-dependent and -independent mechanisms. Autophagy. 2025;21:513–29. Chen PD, Liao YY, Cheng YC, Wu HY, Wu YM, Huang MC. Decreased B4GALT1 promotes hepatocellular carcinoma cell invasiveness by regulating the laminin-integrin pathway. Oncogenesis. 2023;12:49. Silva-Rodríguez P, Fernández-Díaz D, Bande M, Pardo M, Loidi L, Blanco-Teijeiro MJ. GNAQ and GNA11 genes: a comprehensive review on oncogenesis, prognosis and therapeutic opportunities in uveal melanoma. Cancers (Basel). 2022;14:3066. Cheng Y, Liu W, Kim ST, Sun J, Lu L, Sun J, et al. Evaluation of PPP2R2A as a prostate cancer susceptibility gene: a comprehensive germline and somatic study. Cancer Genet. 2011;204:375–81. Lin Z, Kwok HF. RUNDC3A/SNAP25/Akt signaling mediates tumor progression and chemoresistance in gastric neuroendocrine carcinoma. Cell Death Dis. 2022;13:840. Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 29 Apr, 2026 Reviewers invited by journal 29 Apr, 2026 Editor assigned by journal 28 Apr, 2026 First submitted to journal 27 Apr, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9534122","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":631838467,"identity":"f6b4e184-901e-45a4-ab90-8e83ce38e1d3","order_by":0,"name":"Medha Suman","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAzElEQVRIiWNgGAWjYHACMxiD8QHJWpgNSNbCJkGUet325m0PfjDUJa5tP3us8msbgz0/QSvOHCs37GE4nLjtTF7abdk2hsSZDYS03Mgxk+BhOJC47UCO2W3JNoYEgwOEtNx/Yyb5B+iwbeffmBUDtdjbE9Ryg8dMmoeBOXEb0DrGj20MjBsI+yWtTFrG4LDxthtvjKUZzkkkziBoy/HD2yTfVNTJbjufY/jxR5mNPX8DIWvAABqHzDwMxEUNAjD+IFHDKBgFo2AUjAwAAJ8YQCi7DA84AAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0001-8432-7227","institution":"University of Gothenburg: Goteborgs Universitet","correspondingAuthor":true,"prefix":"","firstName":"Medha","middleName":"","lastName":"Suman","suffix":""},{"id":631838468,"identity":"fd524157-3ede-4e82-8bd7-b7cda1b6f11e","order_by":1,"name":"Tobias Hallén","email":"","orcid":"","institution":"University of Gothenburg Institute of Neuroscience and Physiology: Goteborgs universitet Institutionen for neurovetenskap och fysiologi","correspondingAuthor":false,"prefix":"","firstName":"Tobias","middleName":"","lastName":"Hallén","suffix":""},{"id":631838469,"identity":"6273cd6f-3d06-4514-9063-d447088531b8","order_by":2,"name":"Terence Garner","email":"","orcid":"","institution":"The University of Manchester Faculty of Biology Medicine and Health","correspondingAuthor":false,"prefix":"","firstName":"Terence","middleName":"","lastName":"Garner","suffix":""},{"id":631838470,"identity":"81ccc8da-404e-473b-b598-c9c7ebdcc107","order_by":3,"name":"Annika Thorsell","email":"","orcid":"","institution":"University of Gothenburg: Goteborgs Universitet","correspondingAuthor":false,"prefix":"","firstName":"Annika","middleName":"","lastName":"Thorsell","suffix":""},{"id":631838471,"identity":"463c05e2-3914-4ebc-810f-d7b318f53c30","order_by":4,"name":"Adam Stevens","email":"","orcid":"","institution":"University of Manchester School of Medicine: The University of Manchester Faculty of Biology Medicine and Health","correspondingAuthor":false,"prefix":"","firstName":"Adam","middleName":"","lastName":"Stevens","suffix":""},{"id":631838472,"identity":"6fc807cf-9bef-4af0-80e3-f5038cea33fc","order_by":5,"name":"Linus Köster","email":"","orcid":"","institution":"University of Gothenburg Institute of Neuroscience and Physiology: Goteborgs universitet Institutionen for neurovetenskap och fysiologi","correspondingAuthor":false,"prefix":"","firstName":"Linus","middleName":"","lastName":"Köster","suffix":""},{"id":631838473,"identity":"82f6953e-d2a9-4643-b1fc-a1d3287e51c5","order_by":6,"name":"Helena Carén","email":"","orcid":"","institution":"University of Gothenburg Institute of Biomedicine: Goteborgs Universitet Institutionen for biomedicin","correspondingAuthor":false,"prefix":"","firstName":"Helena","middleName":"","lastName":"Carén","suffix":""},{"id":631838474,"identity":"1e58aea0-d74b-4022-8f96-63a621b2d7af","order_by":7,"name":"Oskar Ragnarsson","email":"","orcid":"","institution":"University of Gothenburg Institute of Medicine: Goteborgs universitet Institutionen for medicin","correspondingAuthor":false,"prefix":"","firstName":"Oskar","middleName":"","lastName":"Ragnarsson","suffix":""},{"id":631838475,"identity":"b672c4c2-0b45-42b0-9ed5-0b67557899e8","order_by":8,"name":"Thomas Skoglund","email":"","orcid":"","institution":"University of Gothenburg Institute of Neuroscience and Physiology: Goteborgs universitet Institutionen for neurovetenskap och fysiologi","correspondingAuthor":false,"prefix":"","firstName":"Thomas","middleName":"","lastName":"Skoglund","suffix":""},{"id":631838476,"identity":"cf0430f3-fe70-4668-b465-a3d23c99b2ba","order_by":9,"name":"Gudmundur Johannsson","email":"","orcid":"","institution":"University of Gothenburg Institute of Medicine: Goteborgs universitet Institutionen for medicin","correspondingAuthor":false,"prefix":"","firstName":"Gudmundur","middleName":"","lastName":"Johannsson","suffix":""}],"badges":[],"createdAt":"2026-04-26 19:12:41","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9534122/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9534122/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108947487,"identity":"4ed44e71-fd05-4f08-9c80-8ebec51a0527","added_by":"auto","created_at":"2026-05-11 06:29:25","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":109487,"visible":true,"origin":"","legend":"\u003cp\u003eEpigenetic regulation of gene expression. Schematics show multiple mechanisms of gene and protein expression regulation by\u003cstrong\u003e \u003c/strong\u003eDNA methylation alterations. In \u003cem\u003ecis\u003c/em\u003e-effects, methylation at the promoter directly regulates the expression of the same gene, leading to changes in its corresponding protein levels. In \u003cem\u003etrans\u003c/em\u003e-effects, methylation at one genomic locus affects the expression of other genes indirectly via transcription factors, regulatory long non-coding RNAs, or microRNAs, which in turn regulate downstream target genes and their protein levels. In indirect or network effects, methylation alterations influence chromatin-modifying enzymes (e.g., histone deacetylase [HDACs], DNA methyltransferases [DNMTs]), resulting in global chromatin remodeling and widespread changes in gene and protein expression networks.\u003c/p\u003e","description":"","filename":"OnlineFig1.1.png","url":"https://assets-eu.researchsquare.com/files/rs-9534122/v1/27c1a8efe9f1ff90f844513b.png"},{"id":108947495,"identity":"1d4b8ad1-d487-4754-ab92-ea2afcc1e90b","added_by":"auto","created_at":"2026-05-11 06:29:27","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":88218,"visible":true,"origin":"","legend":"\u003cp\u003eOverview of clinical characteristics and multi-omics analysis strategy.\u003cstrong\u003e (A) \u003c/strong\u003eClinical characteristics of the samples in progressive and indolent non-functioning pituitary adenomas. \u003cstrong\u003e(B)\u003c/strong\u003e Summary of datasets and samples used for integrative analysis.\u003c/p\u003e","description":"","filename":"OnlineFig2.png","url":"https://assets-eu.researchsquare.com/files/rs-9534122/v1/14569c1dfd5cf64334524462.png"},{"id":108947576,"identity":"80ed1f25-c990-4722-92a7-c1416b8d93b2","added_by":"auto","created_at":"2026-05-11 06:29:50","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":279972,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003ecis\u003c/em\u003e-Acting DNA methylation regulation of protein expression in progressive and indolent non-functioning pituitary adenomas(NFPAs) and genomic context of the associated methylation sites.\u003cstrong\u003e (A) \u003c/strong\u003ePie chart shows distribution of CpG sites that showed correlation with their corresponding protein expression across genomic regions (gene-related regions: transcription start site 1500 [TSS1500], TSS200, 5' untranslated region (5'UTR), 1st Exon, and 3'UTR), and CpG island-related regions (islands, shores, shelves, and intergenic regions).\u003cstrong\u003e (B)\u003c/strong\u003e Proportion of weak, moderate, and strong correlations identified between DNA methylation and protein expression in progressive and indolent NFPA groups. \u003cstrong\u003e(C)\u003c/strong\u003e Density plots of Pearson correlation coefficients for methylation-protein associations in progressive and indolent NFPAs. Positive (red) and negative (blue) correlations are shown with the proportion of each indicated, highlighting distinct correlation patterns between the groups.\u003cstrong\u003e (D) \u003c/strong\u003eDistribution of positive and negative correlations across genomic regions relative to CpG islands (top) and gene features (bottom) in progressive and indolent NFPAs, illustrating how correlation direction varies by genomic context. \u003cstrong\u003e(E)\u003c/strong\u003e Distribution of correlation strength (weak, moderate, strong) across CpG island-related regions (top) and gene-related regions (bottom) for both progressive and indolent NFPAs, demonstrating differences in the magnitude of methylation-protein correlations across genomic features.\u003c/p\u003e","description":"","filename":"OnlineFig3.png","url":"https://assets-eu.researchsquare.com/files/rs-9534122/v1/33079b00d7b7ee789bb7b2df.png"},{"id":108947551,"identity":"9684919c-d8e6-4991-9fe9-4e962774cd80","added_by":"auto","created_at":"2026-05-11 06:29:35","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":164133,"visible":true,"origin":"","legend":"\u003cp\u003eIntegrated analysis of DNA methylation and protein expression.\u003cstrong\u003e (A)\u003c/strong\u003e Venn diagram shows the overlap between differentially expressed proteins (\u003cem\u003en\u003c/em\u003e=457) and differentially methylated genes (\u003cem\u003en\u003c/em\u003e=16,472). A total of 336 genes exhibit both differential methylation and corresponding changes in protein expression. The bar plot (in the bottom panel) summarizes the classification of these overlapping genes into four regulatory categories based on methylation and protein expression patterns: Hyper-Down (hypermethylated with decreased protein expression), Hyper-Up (hypermethylated with increased protein expression), Hypo-Down (hypomethylated with decreased protein expression), and Hypo-Up (hypomethylated with increased protein expression). \u003cstrong\u003e(B)\u003c/strong\u003e Scatter plot illustrates the relationship between mean DNA methylation difference (Δβ) and log2-fold change in protein expression for the overlapping genes. Vertical dashed lines indicate methylation thresholds and horizontal dashed lines indicate protein expression thresholds. Points are colored according to regulatory pattern as shown in the legend.\u003c/p\u003e","description":"","filename":"OnlineFig4.png","url":"https://assets-eu.researchsquare.com/files/rs-9534122/v1/3e069d5c4722497de2544518.png"},{"id":108947488,"identity":"36110270-988d-462c-ab9c-674e68b94c9b","added_by":"auto","created_at":"2026-05-11 06:29:25","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":479034,"visible":true,"origin":"","legend":"\u003cp\u003eHypernetwork modelling workflow. Schematic shows the overview of the hypernetwork construction integrating protein expression and DNA methylation data. Pairwise correlations are computed between each differentially expressed proteins (DEPs) (\u003cem\u003en\u003c/em\u003e=457) and genome-wide CpG sites (\u003cem\u003en\u003c/em\u003e=734,065), generating a correlation matrix that reflects the strength of association between protein expression and DNA methylation. The correlation matrix is filtered and binarized based on a standard deviation (SD) threshold. Correlations greater than or equal to the SD threshold are assigned a value of 1, while correlations below the threshold are assigned a value of 0, resulting in a binary incidence matrix. Matrix multiplication of the binary incidence matrix is performed to generate a protein-protein adjacency matrix, representing shared methylation associations and defining the structure of the hypernetwork.\u003c/p\u003e","description":"","filename":"OnlineFig5.png","url":"https://assets-eu.researchsquare.com/files/rs-9534122/v1/9f7d4e76376bbd24ca045405.png"},{"id":108947497,"identity":"210fa143-3cc9-4539-9aed-ea1fb57639a6","added_by":"auto","created_at":"2026-05-11 06:29:28","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":921060,"visible":true,"origin":"","legend":"\u003cp\u003eHeatmaps of the protein-protein adjacency matrices derived from the hypernetwork analysis. \u003cstrong\u003e(A)\u003c/strong\u003e Progressive and \u003cstrong\u003e(B)\u003c/strong\u003eindolent non-functioning pituitary adenomas (NFPAs). The color scale represents the number of shared CpG-associated hyperedges between protein pairs as represented in the legend. Black boxes highlight densely connected protein networks within each branch. \u003cstrong\u003e(C)\u003c/strong\u003e Venn diagram showing the overlap between proteins from hypernetwork-derived branches in progressive and indolent NFPAs.\u003c/p\u003e","description":"","filename":"OnlineFig6.png","url":"https://assets-eu.researchsquare.com/files/rs-9534122/v1/a1ae74368ab90ddae4c6eedb.png"},{"id":108978044,"identity":"acc782ce-fb19-4be6-bd04-425f4b5e9fa7","added_by":"auto","created_at":"2026-05-11 11:33:49","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":114264,"visible":true,"origin":"","legend":"\u003cp\u003eGenomic distribution of CpG sites (hyperedges) linked to \u0026gt;60% of proteins in the hypernetwork.\u003cstrong\u003e \u003c/strong\u003eDistribution of hyperedges across \u003cstrong\u003e(A)\u003c/strong\u003egenomic regions defined in relation to the CpG island including Intergenic, Island, N_Shore, N_Shelf, S_Shelf, and S_Shore regions, \u003cstrong\u003e(B)\u003c/strong\u003e promoter-associated regions, and \u003cstrong\u003e(C)\u003c/strong\u003e enhancer regions in comparison to EPICv2 probes.\u003c/p\u003e","description":"","filename":"OnlineFig7.png","url":"https://assets-eu.researchsquare.com/files/rs-9534122/v1/06b24eff1d1e11a33046d2a9.png"},{"id":108947568,"identity":"e24d4f37-f7ad-4f84-a7c5-a797b2cbe03c","added_by":"auto","created_at":"2026-05-11 06:29:42","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":236877,"visible":true,"origin":"","legend":"\u003cp\u003eFunctional enrichment and comparison of hub proteins in progressive and indolent non-functioning pituitary adenomas (NFPAs). Gene ontology (GO) enrichment analysis of hypernetwork-derived hub proteins in progressive (top panels) and indolent (bottom panels) NFPAs categorized by Biological Process (red), Cellular Component (blue), and Molecular Function (green) terms. Bar plots represent the most significantly enriched GO categories.\u003c/p\u003e","description":"","filename":"OnlineFig8.png","url":"https://assets-eu.researchsquare.com/files/rs-9534122/v1/8e8e7146f0ff2fbc2d9fa844.png"},{"id":108979911,"identity":"a5364d48-98a8-44ca-a950-0a77b4287a7e","added_by":"auto","created_at":"2026-05-11 12:02:12","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4462793,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9534122/v1/6be7ad40-f294-4008-9699-722526790e13.pdf"}],"financialInterests":"","formattedTitle":"Hypernetwork modelling reveals epigenome-proteome interactions in post-surgical progression of non-functioning pituitary adenomas","fulltext":[{"header":"Introduction","content":"\u003cp\u003ePituitary adenomas, also referred to as pituitary neuroendocrine tumors under the current World Health Organization classification, account for approximately 10\u0026ndash;15% of all intracranial tumors and arise from hormone-producing cells of the anterior pituitary gland (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e). Non-functioning pituitary adenomas (NFPAs) represent a clinically heterogeneous subgroup of pituitary adenomas characterized by the absence of hormone hypersecretion and a typically slow yet often invasive growth pattern (\u003cspan additionalcitationids=\"CR3\" citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e). They comprise 22\u0026ndash;54% of all pituitary adenomas and can cause significant clinical challenges due to compression of the normal pituitary gland and surrounding structures, including the optic chiasm and hypothalamus. Consequently, patients may develop visual impairment, hypopituitarism, and in some cases, neurocognitive dysfunction. These conditions substantially impair quality of life and are associated with increased morbidity and mortality among affected individuals (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eSurgical resection remains the primary treatment modality for NFPAs (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e). However, complete tumor removal is often limited by local invasion with residual tumor persisting in more than half of patients. Within this subgroup, over 50% undergo tumor progression requiring reintervention, while the remainder exhibit an indolent clinical course (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan additionalcitationids=\"CR9\" citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e). Importantly, patients with progressive disease have been shown to have higher mortality compared to those with stable tumors (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e). This variability in clinical behavior, coupled with the lack of reliable predictors of progression, necessitates long-term radiological surveillance for all patients, placing a considerable burden on both the individual and healthcare systems [7]. This highlights a critical gap in personalized management of NFPAs driven by limited understanding of the molecular mechanisms and highlights the urgent need for biomarkers capable of stratifying patients by risk and guiding therapeutic decisions.\u003c/p\u003e \u003cp\u003eGenetic alterations appear to play a limited role in NFPA biology as most known mutations are associated with functioning subtypes (\u003cspan additionalcitationids=\"CR13 CR14\" citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e). By contrast, epigenetic dysregulation, particularly DNA methylation, has emerged as a key driver of pituitary tumorigenesis (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e). Genome-wide methylation studies have identified distinct epigenetic subgroups associated with tumor lineage, invasiveness, and size (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e). Complementary proteomic analyses have demonstrated differential protein expression patterns linked to invasiveness (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e). However, most prior studies have focused either on tumor classification or invasiveness, with limited attention to clinically relevant outcomes such as tumor regrowth after surgery that requires re-intervention. In our previous work, we independently characterized genome-wide DNA methylation (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e) and proteomic landscapes (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e) in NFPAs, identifying distinct molecular signatures associated with post-surgical tumor progression. Findings from these studies not only underscored the complexity of tumor biology but also the limitations of single-layer analyses, as gene regulation is governed by complex, multi-level interactions that extend beyond simple linear relationships. In this study, we aimed to expand on previous findings with a more comprehensive multi-omics approach in order to better elucidate the complex molecular mechanisms driving tumor progression.\u003c/p\u003e \u003cp\u003eDNA methylation regulates gene and protein expression through both direct (\u003cem\u003ecis\u003c/em\u003e) and indirect (\u003cem\u003etrans\u003c/em\u003e) mechanisms and can also drive global changes in expression through effects on chromatin and epigenetic regulators (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Traditional integrative strategies often fail to capture the higher-order network structures inherent in biological systems. In contrast, network-based and graph-based models emphasize uncovering functionally relevant modules and interactions that are not apparent when analyzing each data type independently. Among such approaches, hypernetwork modelling has emerged as a powerful framework for capturing higher-order relationships between biological entities (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e). By simultaneous association of multiple CpG sites with individual proteins or genes, this approach reflects the complex regulatory architecture of epigenetic and proteomic interactions, thus providing a more holistic view of how widespread epigenetic changes may converge on key functional nodes that potentially drive tumor behavior.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eIn this study, we apply a hypernetwork-based integrative framework to combine genome-wide DNA methylation and proteomic data in NFPAs. By constructing protein-centric networks, we aim to identify coordinated epigenetic-proteomic alterations associated with tumor progression.\u003c/p\u003e"},{"header":"Material and methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eData retrieval and processing\u003c/h2\u003e \u003cp\u003eThis study included a subset of samples with paired DNA methylation and protein expression data derived from our previously published studies (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e). Patients were grouped into progressive and indolent groups according to the inclusion and exclusion criteria defined previously (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e). Briefly, patients who required reintervention after surgery due to residual tumor progression were classified as having progressive NFPAs, whereas those with residual tumor showing no progression for \u0026ge;\u0026thinsp;5 years were classified as having indolent NFPAs. Only NFPAs of gonadotropinoma lineage were included to ensure molecular and clinical homogeneity across cohorts. Tumor progression was assessed through neuroradiological evaluation. As the current analysis focused on a subset of samples with paired datasets, all data were reprocessed and normalized specifically for this study, and previously processed data were not reused.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eMethylation array data\u003c/h3\u003e\n\u003cp\u003eRaw methylation intensity files (IDATs) were imported into the R computing environment and data were pre-processed and normalized using the normalization method \u0026ldquo;noobBMIQ\u0026rdquo; using the R package \u003cem\u003eChAMP\u003c/em\u003e (\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e). For every CpG position in every sample, detection \u003cem\u003ep\u003c/em\u003e-value and bead count were calculated and were used to evaluate data quality. Probes with an average detection \u003cem\u003ep\u003c/em\u003e-value of \u0026gt;\u0026thinsp;0.01 were considered unreliable and removed from further analysis. CpG probes that were aligned to multiple sites and with a bead count of \u0026lt;\u0026thinsp;3 were also removed. Samples with a \u003cem\u003ep\u003c/em\u003e-value\u0026thinsp;\u0026gt;\u0026thinsp;0.05 were removed from further analysis. Batch correction was performed to correct for possible batch effect using \u0026ldquo;ComBat\u0026rdquo; batch correction. After data pre-processing and normalization, a total of 734,065 CpG positions remained for further analysis. Beta-values (ratio of methylated probe intensity to the sum of methylated and unmethylated probe intensity) were calculated and used for all downstream analysis.\u003c/p\u003e\n\u003ch3\u003eProtein expression data\u003c/h3\u003e\n\u003cp\u003eRaw protein abundances of total quantified proteins (\u003cem\u003en\u003c/em\u003e\u0026thinsp;=\u0026thinsp;4075) were obtained for all 40 NFPA samples and log2 transformed to improve the interpretability and comparability of expression values. Proteins that were not detected in \u0026gt;\u0026thinsp;70% of samples in each group were removed and 3008 proteins remained for further analysis. Missing values that remained even after filtering out the proteins with missing data were imputed using R package \u003cem\u003emissMDA\u003c/em\u003e (\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e). The imputed data were used only when necessary for visualization.\u003c/p\u003e\n\u003ch3\u003eDifferential protein expression\u003c/h3\u003e\n\u003cp\u003eDifferential protein expression analysis was performed using the \u003cem\u003elimma\u003c/em\u003e (\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e) package in R. A false discovery rate threshold\u0026thinsp;\u0026lt;\u0026thinsp;0.05 and an absolute log2-fold change (|log2FC|)\u0026thinsp;\u0026gt;\u0026thinsp;1 was considered significant.\u003c/p\u003e\n\u003ch3\u003eCorrelation analysis\u003c/h3\u003e\n\u003cp\u003eGenes encoding the 457 differentially expressed proteins (DEPs) were identified through ID matching in UniProt and all CpG sites located within or in the vicinity of those genes were selected. Pearson\u0026rsquo;s correlation was used to calculate the correlation between protein expression levels and DNA methylation levels of the CpG sites encoding these proteins.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eHypernetwork analysis\u003c/h2\u003e \u003cp\u003eWe applied hypernetwork modelling to explore the relationship between DEPs and genome-wide DNA methylation variations. In this framework, proteins were treated as nodes, while CpG sites served as hyperedges connecting proteins that share correlated methylation signatures (\u003cspan additionalcitationids=\"CR28 CR29\" citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e). Separate protein-methylation hypernetworks were constructed for progressive and indolent NFPAs by calculating Pearson correlations between DEPs (identified between progressive and indolent groups) and methylation levels across all CpG sites. These correlation matrices were then binarized using a correlation coefficient cut-off threshold, which was equal to the standard deviation of the absolute correlation values, to retain only larger (positive or negative) correlations between the two sets of molecular data. The resulting binary matrix served as the incidence matrix of the hypernetwork, where proteins (DEPs) are nodes and CpGs are hyperedges connecting them.\u003c/p\u003e \u003cp\u003eTo derive the adjacency matrix of the hypernetwork, we multiplied the binarized incidence matrix by its transpose (M \u0026times; Mᵗ). Each entry in the adjacency matrix reflects the number of shared CpG correlations between pairs of proteins, thereby identifying groups of proteins whose expression is jointly associated with multiple methylation sites (\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e). This adjacency matrix (M \u0026times; Mᵗ) captures the degree of shared epigenetic influence among proteins and was used for hierarchical clustering to detect densely connected clusters. The adjacency matrices were visualized as heatmaps, and clusters of interest were identified based on the presence of extensive shared CpG associations (i.e., hyperedges) and distinct branching patterns.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eGene enrichment analysis\u003c/h3\u003e\n\u003cp\u003eOver-representation analysis was performed using WebGestalt (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.webgestalt.org\u003c/span\u003e\u003cspan address=\"https://www.webgestalt.org\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) (\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e) and Reactome database was used with default settings to find the pathways that were enriched in the set of genes.\u003c/p\u003e\n\u003ch3\u003eAssessment of copy number alterations\u003c/h3\u003e\n\u003cp\u003eCopy number alteration (CNA) plots were generated based on the methylation array data using the R package \u003cem\u003econumee\u003c/em\u003e (\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e). CNA plots were reviewed manually to evaluate the presence of CNAs in NFPAs.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eClinical characteristics and overview of molecular data\u003c/h2\u003e \u003cp\u003eThis study included a total of 25 progressive and 15 indolent NFPAs. Age at NFPA diagnosis for the entire cohort was 20\u0026ndash;82 years (median: 59 years). The adenomas were diagnosed as grade 0 to grade 4 according to Knosp grading. The clinical and pathological characteristics of the samples included in this study are presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFor a complete characterization of the epigenomic and proteomic landscape of the NFPAs, we examined DNA methylation levels at 734,065 CpG sites and protein expression of 3008 proteins in 40 NFPA samples. As the primary aim of this study was to elucidate the biological mechanisms underlying the progressive behavior of the NFPAs, we focused our analysis on proteins that showed significant difference in protein expression (|log2FC|\u0026gt;1, false discovery rate threshold\u0026thinsp;\u0026lt;\u0026thinsp;0.05) between progressive and indolent NFPAs and examined their correlation with the genome-wide DNA methylation levels (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB).\u003c/p\u003e \u003cp\u003e \u003cb\u003eCis\u003c/b\u003e \u003cb\u003e-acting relationship between DNA methylation and protein expression\u003c/b\u003e \u003c/p\u003e \u003cp\u003eTo understand \u003cem\u003ecis\u003c/em\u003e-acting dynamics between DNA methylation and protein expression in NFPAs, we conducted a Pearson\u0026rsquo;s correlation analysis between the 457 DEPs and the CpG sites located within or in the vicinity of their respective genes. We performed the analyses separately for indolent and progressive NFPA and subsequently compared the resulting correlation patterns between the groups.\u003c/p\u003e \u003cp\u003eThe DEPs mapped to a total of 470 genes (some proteins mapped to more than one gene). Of these, 407 genes had overlapping CpG sites (\u003cem\u003en\u003c/em\u003e\u0026thinsp;=\u0026thinsp;13,693), while the remaining 63 genes lacked mapped CpGs and were thus excluded from this analysis. Assessing the distribution of CpG sites in various genomic regions, we found that they were predominantly located in the gene body region (53%): they were also distributed in promoter-associated regions including transcription start site 1500 (TSS1500), TSS200, 5' untranslated region (5'UTR), 3'UTR, and the 1st Exon (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA). In relation to the CpG-island context (GC-rich regions with a high density of CpG dinucleotides, often located in gene promoters), most CpGs were located in intergenic regions (42%) followed by CpG islands (26.9%), shores (N-shore 0\u0026ndash;2 kilo base pair (kb) upstream of CpG island and S-shore 0\u0026ndash;2 kb downstream of CpG island), and the shelf region (N-shelf 2\u0026ndash;4 kb upstream of CpG island and S-shelf 2\u0026ndash;4 kb downstream of CpG island) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eIndolent NFPAs exhibit stronger\u003c/b\u003e \u003cb\u003ecis-\u003c/b\u003e\u003cb\u003eregulation compared to progressive NFPAs\u003c/b\u003e\u003c/p\u003e \u003cp\u003eThe strength of the correlation between protein expression and DNA methylation was higher in the indolent group, with correlation coefficients (\u003cem\u003er\u003c/em\u003e) ranging between \u0026minus;\u0026thinsp;0.87 to 0.89 compared to the progressive group (\u003cem\u003er\u003c/em\u003e=\u0026ndash;0.79 to 0.77). In progressive NFPAs, most methylation-protein expression correlations were weak (|\u003cem\u003er\u003c/em\u003e|\u0026lt;0.3, 84%) or moderate (|\u003cem\u003er\u003c/em\u003e|=0.3\u0026ndash;0.5, 14.6%) and only a small proportion were strong (|\u003cem\u003er\u003c/em\u003e|\u0026gt;0.5, 0.01%) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). In contrast, indolent NFPAs exhibited a higher proportion of strong correlations (|\u003cem\u003er\u003c/em\u003e|\u0026gt;0.5, 8.6%), while weak and moderate correlations accounted for 66.5% and 24.9% of relationships, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). In terms of direction of correlation, there was no overall tendency toward positive or negative correlations in either clinical group (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC). To assess whether there was any genomic location preference with respect to correlation direction or correlation strength in either clinical group, we compared the distribution of correlated sites across different regions of the genome based on correlation direction (positive and negative) and strength (weak, moderate, and strong). Contrary to theoretical expectations, we found that CpGs closest to transcription initiation sites and promoter regions exhibited more frequent positive correlations, whereas downstream gene body regions were more strongly associated with negative correlations (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD). Importantly, this pattern was consistent across the outcome groups, despite modest variation in absolute proportions. With respect to the strength of correlation a relatively clearer difference was observed between the outcome groups: progressive NFPAs exhibited an enrichment for weak correlations and a complete absence of strong correlations across the promoter regions compared to indolent NFPAs. Strong correlations were also absent in the 3'UTR and TSS200 regions of the genome in progressive NFPAs (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eConcordant direct DNA methylation-protein expression regulation is rare in progressive NFPAs\u003c/h2\u003e \u003cp\u003eTo further investigate the \u003cem\u003ecis\u003c/em\u003e-regulation between DNA methylation changes and downstream protein expression in progressive and indolent NFPAs, we identified the overlap between differentially methylated genes and DEPs across the outcome groups. Genes showing concordant directionality, i.e., hypermethylation with reduced protein expression and hypomethylation with increased protein expression were considered as under functional epigenetic control.\u003c/p\u003e \u003cp\u003eA total of 23,718 CpGs mapping to 16,472 genes were differentially methylated between the progressive and indolent NFPAs. Of these, 336 genes also exhibited differential protein expressions between the outcome groups. Based on methylation-expression concordance, genes were classified into four groups: hypermethylation with upregulation (hyper-up), hypermethylation with downregulation (hyper-down), hypomethylation with upregulation (hypo-up), and hypomethylation with downregulation (hypo-down) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA). Among CpGs showing the expected inverse methylation-protein expression relationship (hyper-down and hypo-up), only a limited number of gene-protein pairs demonstrated significant differences at both the methylation and protein expression levels. One CpG (cg05046666 located in the shore region [related to TSS1500, 5'UTR] of the gene \u003cem\u003eSPATA20\u003c/em\u003e) was significantly hypermethylated in the progressive NFPAs, which resulted in a significant downregulation of the encoded protein (log2FC=\u0026ndash;1.5). On the other hand, four CpG sites (cg05155595, cg14361862, cg14229540, and cg09050452 mapping to genes \u003cem\u003eANXA4\u003c/em\u003e, \u003cem\u003eCDS2\u003c/em\u003e, \u003cem\u003eLTBP4\u003c/em\u003e, and \u003cem\u003eDHRS13\u003c/em\u003e, respectively) were significantly hypomethylated in progressive NFPAs and demonstrated a significant upregulation (log2FC\u0026thinsp;\u0026gt;\u0026thinsp;1) of the corresponding proteins (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eSince DNA methylation patterns and, ultimately, protein expression can also be influenced by underlying CNAs, we assessed whether the observed differences in protein expression levels between the outcome groups were driven by CNAs. To address this, we analyzed the copy number alteration profiles of all samples (generated based on the methylation data). The majority of NFPAs exhibited flat copy number profiles, indicating an absence of significant CNAs. Only five out of the 40 NFPAs showed detectable focal copy number alterations.\u003c/p\u003e \u003cp\u003e \u003cb\u003eHigher-order integration of DNA methylation and proteomics reveals coordinated regulatory networks unique to progressive and indolent NFPAs\u003c/b\u003e \u003c/p\u003e \u003cp\u003eTo capture the complex and higher order interactions between DNA methylation and protein expression, we built a hypernetwork in which a single CpG site can connect to multiple proteins, allowing us to identify coordinated protein networks regulated by epigenetic changes. For hypernetwork modelling, 457 DEPs were represented as nodes and genome-wide methylation levels (734,065 CpG sites) were represented as hyperedges connecting subsets of these nodes. Hypernetworks were constructed separately for progressive and indolent NFPAs. The workflow of hypernetwork analysis is presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe hypernetwork framework in progressive NFPAs derived several protein pairs where multiple CpG sites served as shared regulatory links (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA). Of the 457 DEPs, 307 (67%) were linked to others in the hypernetwork through shared CpG sites. The strength of coordinated epigenetic regulation and, thus, the functional relevance of a protein pair was determined based on the number of shared CpG sites between the proteins that ranged between 28,284 and 323,172. Proteins with a large number of shared CpG positions were referred to as hub proteins and they showed a clear separation from the remaining proteins and clustered in branch 1 that contained a total of 101 proteins in progressive NFPAs (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA). In progressive NFPAs, the top 20 protein pairs that exhibited the highest number of overlapping CpG connections had 15 unique proteins. The top protein pair with 323,172 overlapping CpG correlations were MCM6 and SAFB (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). These proteins shared hypernetwork connections with other proteins (HDGFL2, SUPT5H, AJM1, SYNE2, DCD, ATP6AP1, INPP4A, KDM3B, EIF3CL, ATP8A1, FUCA1, CAMK2B, and TBCB). The top 20 hypernetwork protein pairs with the number of overlapping CpG sites between them, the function of genes encoding these proteins, and their possible associations with tumor is presented in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. Several of these proteins including MCM6, SAFB, SUPT5H, DCD, AJM1, SYNE2, FUCA1, and HDGFL2 were consistent at higher filtering cut-off of 0.5.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eTop 20 protein pairs with most overlapping CpG sites in progressive NFPAs\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo.\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eProtein\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eGene encoding the protein\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eUpregulated in progressive NFPAs\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eGene function\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eDescribed in relation to pituitary adenoma or other cancer\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eNo. of overlapping CpG sites\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ14566\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eMCM6\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0007\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eDNA replication regulation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026bull; \u003cem\u003eMCM2\u003c/em\u003e and \u003cem\u003eMCM7\u003c/em\u003e genes from the same family as \u003cem\u003eMCM6\u003c/em\u003e have been shown to be upregulated in pituitary tumors and linked to progression (\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e)\u003c/p\u003e \u003cp\u003e\u0026bull; Overexpressed in several cancers including glioma (\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e), breast cancer (\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e, \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e), lung cancer (\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e), hepatocellular carcinoma (\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e), and endometrial adenocarcinoma (\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e), and associated with cell proliferation, migration, and invasion\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e323,172\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ15424\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eSAFB\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;2.12e-06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eActs as a scaffold protein involved in organizing chromatin structure and recruiting transcriptional corepressors\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026bull; Role in breast cancer tumorigenesis (\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ15424\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eSAFB\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e296,880\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ7Z4V5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eHDGFL2\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;2.9e-05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eInvolved in cellular growth control through the regulation of cyclin D1 expression\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026bull; Overexpressed in several cancer including hepatocellular carcinoma (\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e, \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e), pancreatic cancer (\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e), and colorectal cancer (\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e), and contributes to tumor cell proliferation and invasion\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eO00267\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eSUPT5H\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;7.9e-07\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003emRNA processing\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026bull; Overexpressed in colorectal cancer associated with distant metastasis (\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e)\u003c/p\u003e \u003cp\u003e\u0026bull; Role in breast cancer tumorigenesis by regulating the expression levels of genes that control proliferation, migration, and cell cycle (\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e290,448\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ15424\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eSAFB\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eC9J069\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eAJM1\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0016\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eRole in control of adherens junction integrity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026bull; Identified as key prognostic marker in pancreatic adenocarcinoma (\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e287,077\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ15424\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eSAFB\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ14566\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eMCM6\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e286,286\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ8WXH0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eSYNE2\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;5.6e-05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eCrucial role in linking the nuclear envelope to the cytoskeleton\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026bull; Role in p21 regulation\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eP81605\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eDCD\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.000136\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSurvival-promoting peptide\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026bull; Overexpressed in breast cancer (particularly in aggressive forms), and involved in cell proliferation and resistance to apoptosis (\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e, \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e)\u003c/p\u003e \u003cp\u003e\u0026bull; Overexpressed in colorectal cancer and biomarker for poor prognosis (\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e281,673\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ14566\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eMCM6\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eP81605\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eDCD\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e281,579\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ15424\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eSAFB\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ14566\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eMCM6\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e281,472\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ15904\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eATP6AP1\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.4e-06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eTransporter activity and ATPase activity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026bull; Overexpressed in breast cancer and promotes proliferation (\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e, \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ14566\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eMCM6\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e281,417\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ96PE3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eINPP4A\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.2e-05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNegative regulator of the phosphoinositide 3-kinase (PI3K)/AKT signaling pathway\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNA\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eO00267\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eSUPT5H\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e278,181\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ14566\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eMCM6\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ14566\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eMCM6\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ7LBC6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eKDM3B\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;6.3e-05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eRegulates gene expression via histone modification\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026bull; Acts as oncogene in prostate cancer (\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e63\u003c/span\u003e), hepatocellular carcinoma (\u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e64\u003c/span\u003e), renal cell carcinoma (\u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e), and acute lymphoblastic leukemia (\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e) promoting cell proliferation\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eB5ME19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eEIF3CL\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.3e-06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eProtein synthesis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026bull; Overexpressed in diffuse intrinsic pontine glioma and associated with cell proliferation and anchorage-independent growth (\u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e67\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e273,273\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ15904\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eATP6AP1\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ15424\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eSAFB\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e270,136\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ96PE3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eINPP4A\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eB5ME19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eEIF3CL\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.3e-06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e269,770\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ14566\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eMCM6\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ15904\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eATP6AP1\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e268,036\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ9Y2Q0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eATP8A1\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;3.8e-12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eRole in maintaining membrane structure and function by proper distribution of phospholipids\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026bull; Overexpressed in non-small cell lung cancer and associated with increased migration ability (\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eP04066\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eFUCA1\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.000204\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026bull; p53 target, regulates growth and survival of cancer cells\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e266,054\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ14566\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eMCM6\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eB5ME19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eEIF3CL\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.3e-06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e265,910\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ15424\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eSAFB\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ13554\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eCAMK2B\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNo, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;3.1e-09\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eCalcium signaling with roles in cytoskeletal remodeling\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026bull; Mediates both microenvironmental remodeling and tumor progression (\u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e68\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e265,430\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ99426\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eTBCB\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.00138\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eTubulin folding cofactor involved in cytoskeletal dynamics\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026bull; Promotes cell proliferation and associated with poor prognosis in acute myeloid leukemia (\u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e69\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ14566\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eMCM6\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e264,997\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ7Z4V5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eHDGFL2\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ15424\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eSAFB\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e264,464\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ15904\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eATP6AP1\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"7\"\u003eNA, not applicable; NFPA, non-functioning pituitary adenoma.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eIn the case of indolent NFPAs, 70% (318/457) DEPs formed paired connections based on shared CpG sites that ranged between 18,337 to 378,731 CpGs. Hub protein pairs in indolent NFPAs clustered in branch 2 that contained a total of 147 proteins (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB). The top 20 protein pairs that exhibited the highest number of overlapping CpG connections had 11 unique proteins (LAMB2, PSMD6, COPE, APMAP, CISD2, ATP6V1D, B4GAT1, GNAQ, NPTN, PPP2R2A, and RUNDC3A]. The top 20 hypernetwork protein pairs with the number of overlapping CpG sites between them, the function of genes encoding these proteins, and their possible associations with tumor are presented in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. We compared the protein branches that contained the hub protein pairs in progressive (branch 1) and indolent (branch 2) NFPAs to understand if they consist of the same or different proteins. Interestingly, we found limited overlap, with only 28 proteins common to hypernetwork clusters (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eC).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eTop 20 protein pairs with most overlapping CpG sites in indolent NFPAs\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo.\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eProtein\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eGene encoding protein\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eUpregulated in progressive NFPAs\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eGene function\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eDescribed in relation to pituitary adenoma or other cancer\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eNo. of overlapping CpG sites\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eP55268\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eLAMB2\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eMediate the attachment, migration, and organization of cells into tissues during embryonic development by interacting with other extracellular matrix components\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026bull; Promotes metastasis in gastric cancer (\u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e70\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e376,215\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ15008\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003ePSMD6\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;4.1e-06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eInvolved in the ATP-dependent degradation of ubiquinated proteins\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026bull; Overexpressed in hepatocellular carcinoma and essential for tumor cell growth (\u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e71\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eO14579\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eCOPE\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.9e-06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eMediate biosynthetic protein transport\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e363,088\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ15008\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003ePSMD6\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eP55268\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eLAMB2\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e360,499\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ9HDC9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eAPMAP\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.2e-08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eInvolved in biosynthetic processes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026bull; Promotes epithelial-mesenchymal transition and metastasis in cervical cancer (\u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e72\u003c/span\u003e), prostate cancer (\u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e73\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ15008\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003ePSMD6\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e344,893\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ8N5K1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eCISD2\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.4e-06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eInvolved in the regulation of aging, skeletal muscle maintenance, neurodegeneration, cancer, autophagy, and apoptosis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026bull; Overexpressed in liver cancer (\u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e74\u003c/span\u003e), early-stage cervical cancer (\u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e75\u003c/span\u003e), and gastric cancer (\u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e76\u003c/span\u003e), and associated with increased proliferation and enhanced progression in all three tumor types\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ15008\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003ePSMD6\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e341,296\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ9HDC9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eAPMAP\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eP55268\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eLAMB2\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e337,050\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ9Y5K8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eATP6V1D\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;8.8e-08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eProton-transporting ATPase activity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026bull; Promotes stemness and progression in hepatocellular carcinoma (\u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e77\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eO43505\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eB4GAT1\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;2.0e-05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eProtein modification; protein glycosylation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026bull; Promotes metastasis in hepatocellular carcinoma (\u003cspan citationid=\"CR78\" class=\"CitationRef\"\u003e78\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e332,998\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eP50148\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eGNAQ\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.1e-07\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eVital for proper cellular signaling and vascular development\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026bull; Frequently mutated in uveal melanomas (\u003cspan citationid=\"CR79\" class=\"CitationRef\"\u003e79\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ9HDC9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eAPMAP\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e332,990\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ9Y639\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eNPTN\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;2.6e-07\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eCell adhesion molecule binding\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNA\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ15008\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003ePSMD6\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e332,825\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ9Y639\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eNPTN\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eO43505\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eB4GAT1\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e331,665\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ9Y639\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eNPTN\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eP55268\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eLAMB2\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e327,606\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ8N5K1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eCISD2\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eO14579\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eCOPE\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e326,809\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eP63151\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003ePPP2R2A\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;2.3e-07\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eRegulation of chromosome segregation, protein phosphatase regulator activity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026bull; Promotes cancer via deregulated signaling and DNA repair defects (\u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e80\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eP50148\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eGNAQ\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e326,474\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eP55268\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eLAMB2\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eP50148\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eGNAQ\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e326,447\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ15008\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003ePSMD6\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eO43505\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eB4GAT1\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e325,727\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ59EK9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eRUNDC3A\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.4e-05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eGTPase regulator activity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026bull; Promotes tumor progression and chemoresistance in gastric neuroendocrine carcinoma (\u003cspan citationid=\"CR81\" class=\"CitationRef\"\u003e81\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eP50148\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eGNAQ\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e323,536\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ59EK9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eRUNDC3A\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eO14579\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eCOPE\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e323,267\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eP55268\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eLAMB2\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ59EK9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eRUNDC3A\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e323,081\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ9Y639\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eNPTN\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eP55268\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eLAMB2\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e322,715\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ9Y639\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eNPTN\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eO14579\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eCOPE\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e321,589\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eP50148\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eGNAQ\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAs above\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"7\"\u003eNA, not applicable; NFPA, non-functioning pituitary adenoma.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eGenomic distribution of hyperedges\u003c/h2\u003e \u003cp\u003eTo further characterize the hypernetworks, we analyzed CpG sites (hyperedges) linked to hub protein clusters. These CpGs represent methylation regions most strongly associated with the identified protein networks.\u003c/p\u003e \u003cp\u003eIn indolent NFPAs, 540 CpG sites were connected to more than 60% of proteins in the hypernetwork, whereas 9961 CpG sites met this criterion in the progressive group. Notably, only five CpG sites were shared between the two clinical groups among those connected to over 60% of proteins. In progressive NFPAs, we further identified 1248 hyperedges connected to a higher proportion of proteins (\u0026gt;\u0026thinsp;70%), a pattern that was not observed in indolent NFPAs. Figure\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e shows the genomic distribution of the hyperedges forming connections to \u0026gt;\u0026thinsp;60% of hypernetwork proteins in the two outcome groups. Hyperedges in the progressive group showed enrichment for promoter-associated regions including CpG island and N-shore regions (corresponding to TSS200), while hyperedges in the indolent NFPAs were enriched in intergenic regions of the genome (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eA). We also assessed their associations with regulatory features and enhancer regions across the genome. In terms of the regulatory features, the hyperedges were enriched for unclassified cell-type specific regulatory regions (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eB). These regions are mainly involved in the regulatory functions that do not directly involve the gene promoter, suggesting that the hypernetwork proteins expose the functional regulatory networks which are not necessarily located in the promoter regions of the gene. We found that the hyperedges associated with both progressive and indolent NFPAs were enriched for enhancer regions as defined in the 450K version of the methylation array (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eC).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eHub proteins are upregulated in progressive NFPAs and enriched for biological processes involved in cancer\u003c/h2\u003e \u003cp\u003eOver-representation analysis involving all 101 proteins in branch 1 of the hypernetwork, in which the hub protein pairs clustered in progressive NFPAs, revealed that most of the proteins were membrane proteins involved in metabolic processes (Biological process category) and protein binding (Molecular function category) (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eA). Among the most enriched Reactome pathways were the Citric acid cycle (TCA cycle) (R-HSA-71403), Glucose metabolism (R-HSA-70326), snRNP Assembly (R-HSA-191859), Metabolism of non-coding RNA (R-HSA-194441), and Metabolism of carbohydrates (R-HSA-71387). In case of indolent NFPAs, the majority of the proteins in branch 2 (147 proteins) were membrane proteins involved in metabolic processes (Biological process category) and protein binding (Molecular function category) (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eB). Among the top enriched Reactome pathways were Cellular responses to stimuli (R-HSA-8953897), Packaging of telomere ends (R-HSA-171306), Cellular responses to stress (R-HSA-2262752), Recognition and association of DNA glycosylase with site containing an affected purine (R-HSA-110330), and Cleavage of the damaged purine (R-HSA-110331). All top hub proteins identified in both progressive and indolent NFPAs were significantly upregulated in progressive NFPAs, except for CAMK2B, which was downregulated in the progressive group.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eIntegrated analysis of genome-wide DNA methylation and protein expression data in 25 progressive and 15 indolent NFPAs revealed that \u003cem\u003ecis\u003c/em\u003e-acting DNA methylation-protein expression regulations are relatively uncommon in NFPAs. This relationship was particularly weak in progressive NFPAs with only a small fraction of strong associations, suggesting that methylation-driven control of individual protein levels is limited or may be disrupted in more aggressive disease states. In contrast, network-based hypernetwork analysis uncovered a broader and more coordinated association with DNA methylation highlighting indirect and higher-order regulatory effects. This approach captured relationships that are not apparent in direct analyses, indicating that, while methylation may not strongly dictate individual protein expression, it may still play a significant role in shaping protein interaction networks and pathway-level regulation.\u003c/p\u003e \u003cp\u003eOverall, methylation-protein \u003cem\u003ecis\u003c/em\u003e-regulation was stronger and more prevalent in indolent tumors, whereas progressive NFPAs were mainly dominated by weak correlations. Although correlations between DNA methylation and protein expression were observed in both progressive and indolent NFPAs, cases showing significant concordance where hypermethylation corresponded to a significant downregulation or hypomethylation to a significant upregulation of protein expression was limited. Only a small number of sporadic CpG-protein pairs showed the expected inverse relationship with protein expression (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB). This was consistent with previous reports demonstrating predominantly low-to-moderate direct correlations between DNA methylation and protein abundance (\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e). Consistent with the literature, there was limited prevalence of CNAs in NFPAs, which suggests that they are unlikely to play a major role in driving the observed alterations in protein abundance (\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e). Instead, these findings indicate that proteomic changes in NFPAs may be influenced less by absolute gains or losses in DNA methylation and more by alternative regulatory mechanisms, such as microRNA activity, chromatin remodeling, or post-translational modifications, mechanisms not captured by direct methylation-protein correlation analysis. Direct correlation analyses may also fail to capture the complexity of underlying network interactions. For example, scenarios, where methylation at gene A influences the expression of another gene that subsequently modulates protein A through network-mediated effects, as illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. This highlights the importance of considering regulatory dynamics, rather than static methylation levels alone, when interpreting epigenetic alterations in disease progression.\u003c/p\u003e \u003cp\u003eBy deploying a hypernetwork model that treats each CpG as a hyperedge connecting multiple proteins, we captured thousands of indirect and higher-order associations that pairwise methods would miss, highlighting coordinated epigenetic regulation in both progressive and indolent NFPAs. Such architectures better reflect the reality that single CpGs can act through enhancers, chromatin loops, or \u003cem\u003etrans\u003c/em\u003e-acting factors to influence distant proteins. Notably, a substantial proportion of DEPs were incorporated into the formation of hypernetwork in both groups (67% in progressive and 70% in indolent NFPAs), indicating that DNA methylation broadly contributes to structuring proteomic interactions. We observed progressive NFPAs formed a compact network of highly interconnected hub proteins within branch 1, which suggests the presence of tightly coordinated epigenetic regulation affecting a compact and highly interconnected protein network. The large number of shared CpG sites between protein pairs, reaching over 300,000 in some cases, indicates shared epigenetic control and potentially strong co-regulation. This concentrated and interconnected network architecture may reflect a more synchronized regulatory program that enables rapid tumor growth and progression. Key hub proteins in progressive NFPAs, including MCM6, SAFB, and SUPT5H, are associated with core cellular processes such as DNA replication, transcriptional regulation, and chromatin organization. Several of these proteins have been reported in different cancers and have roles in cell proliferation, migration, and tumor growth (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Li et al. (2017) showed that \u003cem\u003eATP8A1\u003c/em\u003e knock-down showed reduced migration ability in non-small cell lung cancer cells (\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e). MCM6, which is the top hub protein node in the hypernetwork analysis of progressive NFPAs, plays a key role in DNA replication regulation and cell cycle progression. It is overexpressed in many different cancers and considerably lower expression has been reported in normal tissue. There are several studies suggesting its role in cell proliferation, potentially promoting cancer progression (\u003cspan additionalcitationids=\"CR39\" citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e). Another gene of the MCM family has previously been identified as a biomarker for post-surgical progression in NFPAs (\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e). Among the hub proteins, many have roles in chromatin binding and histone modification, such as SAFB, HDGFL2, KDM3B, and TAF7, which indicates a possible global impact on protein expression.\u003c/p\u003e \u003cp\u003eIn contrast, indolent NFPAs exhibited a larger hub-containing branch (147 proteins) with an even broader range of CpG sharing, suggesting a more diffuse but extensive epigenetic network. The hub proteins in this group, such as LAMB2, PPP2R2A, and GNAQ, are linked to cellular signaling, structural organization, and regulatory pathways. They are reported in a range of malignancies and were primarily associated with tumor growth and metastasis (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). A key finding is the limited overlap between the hub-associated branches of progressive and indolent NFPAs, with only 28 shared proteins. This difference underscores the presence of distinct epigenetic-proteomic regulatory programs underlying different tumor behavior. While both groups exhibit extensive methylation-associated connectivity, the specific proteins and network structures differ markedly.\u003c/p\u003e \u003cp\u003eOver-representation analysis of proteins within the hypernetwork branches revealed distinct biological signatures between progressive and indolent NFPAs. In progressive NFPAs, branch 1 proteins were predominantly membrane-associated and enriched for metabolic processes and protein-binding functions. The prominence of pathways such as the tricarboxylic acid cycle, glucose metabolism, and broader carbohydrate metabolism points toward enhanced metabolic activity and bioenergetic reprogramming. This metabolic shift is a well-recognized hallmark of aggressive tumor behavior, supporting increased proliferation, survival, and adaptation to microenvironmental demands (\u003cspan additionalcitationids=\"CR43\" citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e). Additionally, enrichment of RNA-related processes, including small nuclear ribonucleic acid particle (snRNP) assembly and non-coding RNA metabolism, suggests heightened post-transcriptional regulation. In contrast for indolent NFPAs, the enriched pathways involving hub proteins were associated with cellular responses to stimuli and stress as well as DNA damage recognition and repair mechanisms. Enrichment of pathways related to telomere maintenance and base excision repair indicate a greater emphasis on genomic stability and controlled cellular responses, which may contribute to their more stable and less progressive behavior.\u003c/p\u003e \u003cp\u003eHypernetwork hubs represent mechanistically coherent, network-stabilized proteins that are more likely to be robust across cohorts and biologically actionable. Biologically relevant processes are typically reflected by coordinated regulation of multiple functionally related proteins in the same direction, strengthening evidence for true alterations rather than random variation. Given that robustness is a key requirement for biomarker development, the proteins identified in this analysis represent promising candidates for future studies aimed at identifying biomarkers of progression. A major limitation of this study is the lack of external validation, and these findings require confirmation in an independent cohort. Future studies with larger sample size and integrative multi-omics approaches including transcriptomic and chromatin accessibility data will further strengthen these findings by enabling deeper dissection of the regulatory mechanisms linking DNA methylation to protein expression and by validating their functional relevance.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThis study demonstrates that direct regulation of protein expression by DNA methylation is limited in NFPAs, particularly in progressive tumors. Instead, our findings revealed that proteomic landscapes in NFPAs are shaped by complex, multi-layered regulatory mechanisms. Distinct epigenetic-proteomic network architecture observed in progressive and indolent NFPAs suggests the presence of fundamentally different regulatory programs underlying the different tumor behavior. The identification of hypernetwork hub proteins provides a set of biologically meaningful and potentially robust candidates for future biomarker development. Further validation in independent cohorts and integration with additional omics layers will be essential to confirm these findings and to better understand the regulatory landscape driving NFPA progression.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cdiv class=\"DefinitionList\"\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eCNA\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eCopy number alteration\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eDEP\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eDifferentially expressed protein\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003ekb\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003ekilo base pair\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eNFPA\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eNon\u0026ndash;functioning pituitary adenoma\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eTSS\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eTranscription start site\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eUTR\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eUntranslated region\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe are grateful to Peter Todd (Tajut Ltd., Kaiapoi, New Zealand) for third-party writing assistance in drafting of this manuscript, for which he received financial compensation from ALF-funding.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMedha Suman, Gudmundur Johannsson, Thomas Skoglund, Tobias Hall\u0026eacute;n, Terence Garner, and Adam Stevens designed the study and methodology. All authors collected data from the original study. Medha Suman, Gudmundur Johannsson, Tobias Hall\u0026eacute;n, Annika Thorsell, Helena Car\u0026eacute;n, Oskar Ragnarsson, Linus K\u0026ouml;ster, and Thomas Skoglund contributed with resources and planning of the study. Gudmundur Johannsson and Thomas Skoglund obtained funding. Gudmundur Johannsson, Tobias Hall\u0026eacute;n, Thomas Skoglund, Adam Stevens, Oskar Ragnarsson, Annika Thorsell, and Helena Car\u0026eacute;n supervised the study. Medha Suman, Terence Garner, and Adam Stevens had full access to all data and performed the analyses. Medha Suman, Helena Car\u0026eacute;n, Tobias Hall\u0026eacute;n, Thomas Skoglund, Oskar Ragnarsson, Adam Stevens, Annika Thorsell, and Gudmundur Johannsson drafted the manuscript, and all authors revised it and approved the final version.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe study was supported by grants from V\u0026auml;stra G\u0026ouml;talandsregionen through the Swedish state under the agreement between the Swedish government and the County Councils, the ALF-agreement (ALFGBG-772591, ALFGBG -1006371 and ALFGBG-966066), the Swedish Cancer Society (21 1774 Pj \u0026amp; 24 3687 Pj).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll data supporting the findings of this study are available from the corresponding author on reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe study was approved by the Regional Ethical Review Board in Gothenburg (Dnr: 100-15). The research was conducted in accordance with the Declaration of Helsinki.\u003c/p\u003e\n\u003cp\u003eConsent to Participate declarations: not applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGudmundur Johannsson has served as a consultant for Crinetics, Lundbeck, Novo Nordisk, and AstraZeneca, and has received lecture fees from Ascendic Pharma, Crinetics, Novo Nordisk, and Pharmanovia. All other authors declare no conflicts of interest.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eEzzat S, Asa SL, Couldwell WT, Barr CE, Dodge WE, Vance ML, et al. The prevalence of pituitary adenomas: a systematic review. Cancer. 2004;101:613\u0026ndash;9.\u003c/li\u003e\n \u003cli\u003eFernandez A, Karavitaki N, Wass JA. Prevalence of pituitary adenomas: a community-based, cross-sectional study in Banbury (Oxfordshire, UK). Clin Endocrinol (Oxf). 2010;72:377\u0026ndash;82.\u003c/li\u003e\n \u003cli\u003eAflorei, ED, Korbonits M. Epidemiology and etiopathogenesis of pituitary adenomas. J Neurooncol. 2014;117:379-94.\u003c/li\u003e\n \u003cli\u003eTj\u0026ouml;rnstrand A, Gunnarsson K, Evert M, Holmberg E, Ragnarsson O, Ros\u0026eacute;n T, et al. The incidence rate of pituitary adenomas in western Sweden for the period 2001\u0026ndash;2011. Eur J Endocrinol. 2014;171:519-26.\u003c/li\u003e\n \u003cli\u003eMelmed S, Kaiser UB, Lopes MB, Bertherat J, Syro LV, Raverot G, et al. Clinical biology of the pituitary adenoma. Endocr Rev. 2022;43:1003\u0026ndash;37.\u003c/li\u003e\n \u003cli\u003eHigham CE, Johannsson G, Shalet SM. Hypopituitarism. Lancet. 2016;388:2403\u0026ndash;15.\u003c/li\u003e\n \u003cli\u003eEsposito D, Olsson DS, Ragnarsson O, Buchfelder M, Skoglund T, Johannsson G. Non-functioning pituitary adenomas: indications for pituitary surgery and post-surgical management. Pituitary. 2019;22:422\u0026ndash;434.\u003c/li\u003e\n \u003cli\u003eOlsson DS, Buchfelder M, Schlaffer S, Bengtsson BA, Jakobsson KE, Johannsson G, et al. Comparing progression of non-functioning pituitary adenomas in hypopituitarism patients with and without long-term GH replacement therapy. Eur J Endocrinol. 2009 Nov;161:663\u0026ndash;9.\u003c/li\u003e\n \u003cli\u003eOlsson DS, Buchfelder M, Wiendieck K, Kremenevskaja N, Bengtsson B\u0026Aring;, Jakobsson KE, et al. Tumour recurrence and enlargement in patients with craniopharyngioma with and without GH replacement therapy during more than 10 years of follow-up. Eur J Endocrinol. 2012;166:1061\u0026ndash;8.\u003c/li\u003e\n \u003cli\u003eChen Y, Wang CD, Su ZP, Chen YX, Cai L, Zhuge QC, et al. Natural history of postoperative nonfunctioning pituitary adenomas: a systematic review and meta-analysis. Neuroendocrinology. 2012;96:333\u0026ndash;42.\u003c/li\u003e\n \u003cli\u003eErfurth EM, Bulow B, Nordstr\u0026ouml;m CH, Mikoczy Z, Hagmar L, Str\u0026ouml;mberg U. Doubled mortality rate in irradiated patients reoperated for regrowth of a macroadenoma of the pituitary gland. Eur J Endocrinol. 2004;150:497\u0026ndash;502.\u003c/li\u003e\n \u003cli\u003eMayr B, Apenberg S, Roth\u0026auml;mel T, von zur M\u0026uuml;hlen A, Brabant G. Menin mutations in patients with multiple endocrine neoplasia type 1. Eur J Endocrinol. 1997;137:684\u0026ndash;7.\u003c/li\u003e\n \u003cli\u003eBeckers A, Aaltonen LA, Daly AF, Karhu A. Familial isolated pituitary adenomas (FIPA) and the pituitary adenoma predisposition due to mutations in the aryl hydrocarbon receptor interacting protein (AIP) gene. Endocr Rev. 2013;34:239\u0026ndash;77.\u003c/li\u003e\n \u003cli\u003eReincke M, Sbiera S, Hayakawa A, Theodoropoulou M, Osswald A, Beuschlein F, et al. Mutations in the deubiquitinase gene USP8 cause Cushing\u0026apos;s disease. Nat Genet. 2015;47:31\u0026ndash;8.\u003c/li\u003e\n \u003cli\u003eMantovani G, Lania AG, Spada A. GNAS imprinting and pituitary tumors. Mol Cell Endocrinol. 2010;326:15\u0026ndash;8.\u003c/li\u003e\n \u003cli\u003ePeterson R, Cote DJ, Zada G. Current understanding of the role of DNA methylation in pituitary neuroendocrine tumors. Neurooncol Adv. 2025;7 Suppl 1:i2\u0026ndash;9.\u003c/li\u003e\n \u003cli\u003eSrirangam Nadhamuni V, Korbonits M. Novel insights into pituitary tumorigenesis: genetic and epigenetic mechanisms. Endocr Rev. 2020;41:821\u0026ndash;46.\u003c/li\u003e\n \u003cli\u003eMiyake Y, Adachi JI, Suzuki T, Mishima K, Araki R, Mizuno R, et al. TERT promoter methylation is significantly associated with TERT upregulation and disease progression in pituitary adenomas. J Neurooncol. 2019;141:131\u0026ndash;8.\u003c/li\u003e\n \u003cli\u003eZhang Y, Li L, Ma X, Liu C, Liu G, Bie Z, et al. Quantitative proteomics identified a novel invasion biomarker associated with EMT in pituitary adenomas. Front Endocrinol (Lausanne). 2023;14:1137648.\u003c/li\u003e\n \u003cli\u003eHall\u0026eacute;n T, Johannsson G, Dahl\u0026eacute;n R, Glad CAM, \u0026Ouml;rndal C, Engvall A, et al. Genome-wide DNA methylation differences in nonfunctioning pituitary adenomas with and without postsurgical progression. J Clin Endocrinol Metab. 2022;107:2318\u0026ndash;28.\u003c/li\u003e\n \u003cli\u003eHall\u0026eacute;n T, Johannsson G, Thorsell A, Olsson DS, \u0026Ouml;rndal C, Engvall A, et al. Proteomic profiles associated with postsurgical progression in nonfunctioning pituitary adenomas. J Clin Endocrinol Metab. 2024;109:1485\u0026ndash;93.\u003c/li\u003e\n \u003cli\u003eLugo-Martinez J, Zeiberg D, Gaudelet T, Malod-Dognin N, Przulj N, Radivojac P. Classification in biological networks with hypergraphlet kernels. Bioinformatics. 2021;37:1000\u0026ndash;7.\u003c/li\u003e\n \u003cli\u003eFeng L, Gong H, Zhang S, Liu X, Wang Y, Che J, et al. Hypernetwork modeling and topology of high-order interactions for complex systems. Proc Natl Acad Sci U S A. 2024;121:e2412220121.\u003c/li\u003e\n \u003cli\u003eTian Y, Morris TJ, Webster AP, Yang Z, Beck S, Feber A, et al. ChAMP: updated methylation analysis pipeline for Illumina BeadChips. Bioinformatics. 2017;33:3982\u0026ndash;4.\u003c/li\u003e\n \u003cli\u003eJosse J, Husson F. missMDA: a package for handling missing values in multivariate data analysis. Journal of Statistical Software, 2016;70:1\u0026ndash;31.\u003c/li\u003e\n \u003cli\u003eRitchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015;43:e47.\u003c/li\u003e\n \u003cli\u003eGaudelet T, Malod-Dognin N, Pržulj N. Higher-order molecular organization as a source of biological function. Bioinformatics. 2018;34:i944\u0026ndash;53.\u003c/li\u003e\n \u003cli\u003eJohnson J. Hypernetworks in the science of complex systems. In: Complex Sciences. London: Imperial College Press; 2013. https://books.google.co.nz/books?hl=en\u0026amp;lr=\u0026amp;id=jQO4CgAAQBAJ\u0026amp;oi=fnd\u0026amp;pg=PP7\u0026amp;dq=hypernetworks+in+the+science+of+complex+systems+johnson+%222012%22\u0026amp;ots=_Y1yr7dHOw\u0026amp;sig=Gs0x9cY8nqKyH37j8Y2TMlP98pw#v=onepage\u0026amp;q=hypernetworks%20in%20the%20science%20of%20complex%20systems%20jjohnso%20%222011%22\u0026amp;f=false. Accessed 18 Apr 2026.\u003c/li\u003e\n \u003cli\u003ePearcy N, Chuzhanova N, Crofts JJ. Complexity and robustness in hypernetwork models of metabolism. J Theor Biol. 2016;406:99\u0026ndash;104.\u003c/li\u003e\n \u003cli\u003eRipley DM, Garner T, Stevens A. Developing the \u0026apos;omic toolkit of comparative physiologists. Comp Biochem Physiol Part D Genomics Proteomics. 2024;52:101287.\u003c/li\u003e\n \u003cli\u003eContreras-Aso G, Criado R, Vera de Salas G, Yang J. Detecting communities in higher-order networks by using their derivative graphs. Chaos, Solitons \u0026amp; Fractals, 2023;177:114200.\u003c/li\u003e\n \u003cli\u003eKumar T, Vaidyanathan S, Ananthapadmanabhan H, Parthasarathy S, Ravindran B. Hypergraph clustering by iteratively reweighted modularity maximization. Applied Network Science 2020;5:52.\u003c/li\u003e\n \u003cli\u003eElizarraras JM, Liao Y, Shi Z, Zhu Q, Pico AR, Zhang B. WebGestalt 2024: faster gene set analysis and new support for metabolomics and multi-omics. Nucleic Acids Res. 2024;52:W415\u0026ndash;21.\u003c/li\u003e\n \u003cli\u003eDaenekas B, P\u0026eacute;rez E, Boniolo F, Stefan S, Benfatto S, Sill M, et al. Conumee 2.0: enhanced copy-number variation analysis from DNA methylation arrays for humans and mice. Bioinformatics. 2024;40:btae029.\u003c/li\u003e\n \u003cli\u003eLiang WW, Lu RJ, Jayasinghe RG, Foltz SM, Porta-Pardo E, Geffen Y, et al; Clinical Proteomic Tumor Analysis Consortium. Integrative multi-omic cancer profiling reveals DNA methylation patterns associated with therapeutic vulnerability and cell-of-origin. Cancer Cell. 2023;41:1567\u0026ndash;85.e7.\u003c/li\u003e\n \u003cli\u003eBi WL, Horowitz P, Greenwald NF, Abedalthagafi M, Agarwalla PK, Gibson WJ, et al. Landscape of genomic alterations in pituitary adenomas. Clin Cancer Res. 2017;23:1841\u0026ndash;51.\u003c/li\u003e\n \u003cli\u003eLi D, Xu T, Wang X, Ma X, Liu T, Wang Y, et al. The role of ATP8A1 in non-small cell lung cancer. Int J Clin Exp Pathol. 2017;10:7760\u0026ndash;6.\u003c/li\u003e\n \u003cli\u003eZeng T, Guan Y, Li YK, Wu Q, Tang XJ, Zeng X, et al. The DNA replication regulator MCM6: an emerging cancer biomarker and target. Clin Chim Acta. 2021;517:92\u0026ndash;8.\u003c/li\u003e\n \u003cli\u003eLiu Z, Li J, Chen J, Shan Q, Dai H, Xie H, et al. MCM family in HCC: MCM6 indicates adverse tumor features and poor outcomes and promotes S/G2 cell cycle progression. BMC Cancer. 2018;18:200.\u003c/li\u003e\n \u003cli\u003eCai HQ, Cheng ZJ, Zhang HP, Wang PF, Zhang Y, Hao JJ, et al. Overexpression of MCM6 predicts poor survival in patients with glioma. Hum Pathol. 2018;78:182\u0026ndash;7.\u003c/li\u003e\n \u003cli\u003eHall\u0026eacute;n T, Olsson DS, Hammarstrand C, \u0026Ouml;rndal C, Engvall A, Ragnarsson O, et al. MCM7 as a marker of postsurgical progression in non-functioning pituitary adenomas. Eur J Endocrinol. 2021;184:521\u0026ndash;31.\u003c/li\u003e\n \u003cli\u003eXia L, Oyang L, Lin J, Tan S, Han Y, Wu N, Yi P, Tang L, et al. The cancer metabolic reprogramming and immune response. Mol Cancer. 2021;20:28.\u003c/li\u003e\n \u003cli\u003eSun L, Suo C, Li ST, Zhang H, Gao P. Metabolic reprogramming for cancer cells and their microenvironment: beyond the Warburg effect. Biochim Biophys Acta Rev Cancer. 2018;1870:51\u0026ndash;66.\u003c/li\u003e\n \u003cli\u003eLue HW, Podolak J, Kolahi K, Cheng L, Rao S, Garg D, et al. Metabolic reprogramming ensures cancer cell survival despite oncogenic signaling blockade. Genes Dev. 2017;31:2067\u0026ndash;84.\u003c/li\u003e\n \u003cli\u003eHuang Y, Wang Q, Zhou W, Jiang Y, He K, Huang W, et al. Prenatal p25-activated Cdk5 induces pituitary tumorigenesis through MCM2 phosphorylation-mediated cell proliferation. Neoplasia. 2024;57:101054.\u003c/li\u003e\n \u003cli\u003eCheng L, Tan Z, Huang Z, Pan Y, Zhang W, Wang J. Expression profile and prognostic values of mini-chromosome maintenance families (MCMs) in breast cancer. Med Sci Monit. 2020;26:e923673.\u003c/li\u003e\n \u003cli\u003eIssac MSM, Yousef E, Tahir MR, Gaboury LA. MCM2, MCM4, and MCM6 in breast cancer: clinical utility in diagnosis and prognosis. Neoplasia. 2019;21:1015\u0026ndash;35.\u003c/li\u003e\n \u003cli\u003eLiu YZ, Wang BS, Jiang YY, Cao J, Hao JJ, Zhang Y, et al. MCMs expression in lung cancer: implication of prognostic significance. J Cancer. 2017;8:3641\u0026ndash;7.\u003c/li\u003e\n \u003cli\u003eHotton J, Agopiantz M, Leroux A, Charra-Brunaud C, Marie B, Busby-Venner H, et al. Minichromosome maintenance complex component 6 (MCM6) expression correlates with histological grade and survival in endometrioid endometrial adenocarcinoma. Virchows Arch. 2018;472:623\u0026ndash;33.\u003c/li\u003e\n \u003cli\u003eHammerich-Hille S, Bardout VJ, Hilsenbeck SG, Osborne CK, Oesterreich S. Low SAFB levels are associated with worse outcome in breast cancer patients. Breast Cancer Res Treat. 2010;121:503\u0026ndash;9.\u003c/li\u003e\n \u003cli\u003eEnomoto H, Nakamura H, Liu W, Nishiguchi S. Hepatoma-derived growth factor: its possible involvement in the progression of hepatocellular carcinoma. Int J Mol Sci. 2015;16:14086\u0026ndash;97.\u003c/li\u003e\n \u003cli\u003eYoshida K, Nakamura H, Okuda Y, Enomoto H, Kishima Y, Uyama H, et al. Expression of hepatoma-derived growth factor in hepatocarcinogenesis. J Gastroenterol Hepatol. 2003;18:1293\u0026ndash;301.\u003c/li\u003e\n \u003cli\u003eUyama H, Tomita Y, Nakamura H, Nakamori S, Zhang B, Hoshida Y, et al. Hepatoma-derived growth factor is a novel prognostic factor for patients with pancreatic cancer. Clin Cancer Res. 2006;12:6043\u0026ndash;8.\u003c/li\u003e\n \u003cli\u003eLiao F, Dong W, Fan L. Apoptosis of human colorectal carcinoma cells is induced by blocking hepatoma-derived growth factor. Med Oncol. 2010;27:1219\u0026ndash;26.\u003c/li\u003e\n \u003cli\u003eChen R, Zhu J, Dong Y, He C, Hu X. Suppressor of Ty homolog-5, a novel tumor-specific human telomerase reverse transcriptase promoter-binding protein and activator in colon cancer cells. Oncotarget. 2015;6:32841\u0026ndash;55.\u003c/li\u003e\n \u003cli\u003eLone BA, Ahmad F, Karna SKL, Pokharel YR. SUPT5H Post-transcriptional silencing modulates PIN1 expression, inhibits tumorigenicity, and induces apoptosis of human breast cancer cells. Cell Physiol Biochem. 2020;54:928\u0026ndash;46.\u003c/li\u003e\n \u003cli\u003eMa T, Gongye X, Dongzhi C, Chai Y, Wang Q, Tian M. Machine learning-based WGCNA approach for developing an immunogenic cell death-related hub gene signature and identification of AJM1 as a prognostic biomarker in pancreatic adenocarcinoma. Int J Med Sci. 2025;22:4493\u0026ndash;508.\u003c/li\u003e\n \u003cli\u003eBrauer HA, D\u0026apos;Arcy M, Libby TE, Thompson HJ, Yasui YY, Hamajima N, et al. Dermcidin expression is associated with disease progression and survival among breast cancer patients. Breast Cancer Res Treat. 2014;144:299\u0026ndash;306.\u003c/li\u003e\n \u003cli\u003eBancovik J, Moreira DF, Carrasco D, Yao J, Porter D, Moura R, et al. Dermcidin exerts its oncogenic effects in breast cancer via modulation of ERBB signaling. BMC Cancer. 2015;15:70.\u003c/li\u003e\n \u003cli\u003eZhang S, Wang Y, Zhang X, Wang M, Wu H, Tao Y, et al. ATP6AP1 as a potential prognostic biomarker in CRC by comprehensive analysis and verification. Sci Rep. 2024;14:4018.\u003c/li\u003e\n \u003cli\u003eYan Z, Huang A, Ma D, Hong C, Zhang S, He L, et al. ATP6AP1 promotes cell proliferation and tamoxifen resistance in luminal breast cancer by inducing autophagy. Cell Death Dis. 2025;16:201.\u003c/li\u003e\n \u003cli\u003eWang J, Liu Y, Zhang S. Prognostic and immunological value of ATP6AP1 in breast cancer: implications for SARS-CoV-2. Aging (Albany NY). 2021;13:16904\u0026ndash;21.\u003c/li\u003e\n \u003cli\u003eBj\u0026ouml;rkman M, \u0026Ouml;stling P, H\u0026auml;rm\u0026auml; V, Virtanen J, Mpindi JP, Rantala J, et al. Systematic knockdown of epigenetic enzymes identifies a novel histone demethylase PHF8 overexpressed in prostate cancer with an impact on cell proliferation, migration and invasion. Oncogene. 2012;31:3444\u0026ndash;56.\u003c/li\u003e\n \u003cli\u003eAn MJ, Kim DH, Kim CH, Kim M, Rhee S, Seo SB, et al. Histone demethylase KDM3B regulates the transcriptional network of cell-cycle genes in hepatocarcinoma HepG2 cells. Biochem Biophys Res Commun. 2019;508:576\u0026ndash;82.\u003c/li\u003e\n \u003cli\u003eMahamdallie S, Yost S, Poyastro-Pearson E, Holt E, Zachariou A, Seal S, et al. Identification of new Wilms tumour predisposition genes: an exome sequencing study. Lancet Child Adolesc Health. 2019;3:322\u0026ndash;31.\u003c/li\u003e\n \u003cli\u003eKim JY, Kim KB, Eom GH, Choe N, Kee HJ, Son HJ, et al. KDM3B is the H3K9 demethylase involved in transcriptional activation of lmo2 in leukemia. Mol Cell Biol. 2012;32:2917\u0026ndash;33.\u003c/li\u003e\n \u003cli\u003eChen J, Lin Z, Barrett L, Dai L, Qin Z. Identification of new therapeutic targets and natural compounds against diffuse intrinsic pontine glioma (DIPG). Bioorg Chem. 2020;99:103847.\u003c/li\u003e\n \u003cli\u003eJia Q, Liao X, Zhang Y, Xu B, Song Y, Bian G, et al. Anti-tumor role of CAMK2B in remodeling the stromal microenvironment and inhibiting proliferation in papillary renal cell carcinoma. Front Oncol. 2022;12:740051.\u003c/li\u003e\n \u003cli\u003eWang B, Wang W, Li Q, Guo T, Yang S, Shi J, et al. High expression of microtubule-associated protein TBCB predicts adverse outcome and immunosuppression in acute myeloid leukemia. J Cancer. 2023;14:1707\u0026ndash;24.\u003c/li\u003e\n \u003cli\u003eLi D, Jin Y, He X, Deng J, Lu W, Yang Z, et al. Hypoxia-induced LAMB2-enriched extracellular vesicles promote peritoneal metastasis in gastric cancer via the ROCK1-CAV1-Rab11 axis. Oncogene. 2024;43:2768\u0026ndash;80.\u003c/li\u003e\n \u003cli\u003eZhou SS, Ye YP, Chen Y, Zeng DT, Zheng GC, He RQ, et al. Overexpression pattern, function, and clinical value of proteasome 26S subunit non-ATPase 6 in hepatocellular carcinoma. World J Clin Oncol. 2025;16:99839.\u003c/li\u003e\n \u003cli\u003eZhu X, Xiang Z, Zou L, Chen X, Peng X, Xu D. APMAP promotes epithelial-mesenchymal transition and metastasis of cervical cancer cells by activating the Wnt/\u0026beta;-catenin pathway. J Cancer. 2021;12:6265\u0026ndash;73.\u003c/li\u003e\n \u003cli\u003eJiang S, Wang X, Song D, Liu X, Gu Y, Xu Z, et al. Cholesterol induces epithelial-to-mesenchymal transition of prostate cancer cells by suppressing degradation of EGFR through APMAP. Cancer Res. 2019;79:3063\u0026ndash;75.\u003c/li\u003e\n \u003cli\u003eChen B, Shen S, Wu J, Hua Y, Kuang M, Li S, et al. CISD2 associated with proliferation indicates negative prognosis in patients with hepatocellular carcinoma. Int J Clin Exp Pathol. 2015;8:13725\u0026ndash;38.\u003c/li\u003e\n \u003cli\u003eLiu L, Xia M, Wang J, Zhang W, Zhang Y, He M. CISD2 expression is a novel marker correlating with pelvic lymph node metastasis and prognosis in patients with early-stage cervical cancer. Med Oncol. 2014;31:183.\u003c/li\u003e\n \u003cli\u003eWang L, Ouyang F, Liu X, Wu S, Wu HM, Xu Y, et al. Overexpressed CISD2 has prognostic value in human gastric cancer and promotes gastric cancer cell proliferation and tumorigenesis via AKT signaling pathway. Oncotarget. 2016;7:3791\u0026ndash;805.\u003c/li\u003e\n \u003cli\u003eXu Z, Liu R, Ke H, Xu F, Yang P, Zhang W, et al. ATP6V1D drives hepatocellular carcinoma stemness and progression via both lysosome acidification-dependent and -independent mechanisms. Autophagy. 2025;21:513\u0026ndash;29.\u003c/li\u003e\n \u003cli\u003eChen PD, Liao YY, Cheng YC, Wu HY, Wu YM, Huang MC. Decreased B4GALT1 promotes hepatocellular carcinoma cell invasiveness by regulating the laminin-integrin pathway. Oncogenesis. 2023;12:49.\u003c/li\u003e\n \u003cli\u003eSilva-Rodr\u0026iacute;guez P, Fern\u0026aacute;ndez-D\u0026iacute;az D, Bande M, Pardo M, Loidi L, Blanco-Teijeiro MJ. GNAQ and GNA11 genes: a comprehensive review on oncogenesis, prognosis and therapeutic opportunities in uveal melanoma. Cancers (Basel). 2022;14:3066.\u003c/li\u003e\n \u003cli\u003eCheng Y, Liu W, Kim ST, Sun J, Lu L, Sun J, et al. Evaluation of PPP2R2A as a prostate cancer susceptibility gene: a comprehensive germline and somatic study. Cancer Genet. 2011;204:375\u0026ndash;81.\u003c/li\u003e\n \u003cli\u003eLin Z, Kwok HF. RUNDC3A/SNAP25/Akt signaling mediates tumor progression and chemoresistance in gastric neuroendocrine carcinoma. Cell Death Dis. 2022;13:840.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"journal-of-translational-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jtrm","sideBox":"Learn more about [Journal of Translational Medicine](http://translational-medicine.biomedcentral.com)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/jtrm/default.aspx","title":"Journal of Translational Medicine","twitterHandle":"@BioMedCentral","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Non-functioning pituitary adenoma, Tumor progression, Hypernetwork, DNA methylation, Protein expression","lastPublishedDoi":"10.21203/rs.3.rs-9534122/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9534122/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eNon-functioning pituitary adenomas (NFPAs) present a complex clinical challenge due to their indolent and invasive growth patterns, and critical anatomical location. Post-surgical tumor progression is frequent in patients with NFPAs, which often necessitates additional therapeutic interventions. The molecular mechanisms underlying post-surgical progression remain poorly understood and there are currently no reliable methods to stratify patients according to risk of tumor progression. The aim of this study was to comprehensively characterize the molecular alterations, together with an integrated understanding of their interactions, that could potentially uncover the biological processes driving post-surgical tumor progression of NFPAs.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eWe performed an integrated analysis of genome-wide DNA methylation and proteomics in 25 progressive and 15 indolent NFPAs using hypernetwork modelling linking CpG sites to the differentially expressed proteins to identify functional alterations associated with tumor progression. In addition, we investigated \u003cem\u003ecis\u003c/em\u003e-regulatory relationships by examining CpG sites located within or in close proximity to the genes encoding the corresponding proteins, allowing assessment of the direct impact of DNA methylation changes on protein expression levels.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eHypernetwork analysis uncovered extensive indirect and higher-order associations, capturing coordinated epigenetic influences on protein networks in indolent and progressive NFPAs. Progressive NFPAs were characterized by a compact and highly interconnected hub network with proteins primarily involved in DNA replication and transcription regulation (MCM6 and HDGFL2), chromatin organization (SAFB, HDGFL2, KDM3B, and TAF7), and cytoskeleton organization and cell structure maintenance (AJM1 and SYNE2). In contrast, indolent adenomas exhibited a broader and more diffuse network architecture with hub proteins linked to protein processing and transport (PSMD6, APMAP, B4GAT1, and COPE), extracellular matrix organization (LAMB2), and oxidative stress response (CISD2). Hub proteins in progressive NFPAs were enriched for metabolic pathways including glycolysis and tricarboxylic acid cycle, while enriched pathways for hub proteins in the indolent group were associated with genome maintenance and cellular stress responses.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eHypernetwork analysis highlighted distinct epigenetic-proteomic regulatory mechanisms linked to tumor behavior that were not detected through \u003cem\u003ecis\u003c/em\u003e-acting correlation analysis. Collectively, this integrative approach provides insight beyond direct regulation effects and offers a framework for identifying network-informed candidate markers with mechanistic relevance in tumor progression.\u003c/p\u003e","manuscriptTitle":"Hypernetwork modelling reveals epigenome-proteome interactions in post-surgical progression of non-functioning pituitary adenomas","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-11 06:26:16","doi":"10.21203/rs.3.rs-9534122/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"","date":"2026-04-29T15:15:01+00:00","index":0,"fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-29T14:12:24+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-28T12:14:12+00:00","index":"","fulltext":""},{"type":"submitted","content":"Journal of Translational Medicine","date":"2026-04-27T10:43:39+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"journal-of-translational-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jtrm","sideBox":"Learn more about [Journal of Translational Medicine](http://translational-medicine.biomedcentral.com)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/jtrm/default.aspx","title":"Journal of Translational Medicine","twitterHandle":"@BioMedCentral","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"4b6b382a-d324-4e8e-84ce-305202f7a04b","owner":[],"postedDate":"May 11th, 2026","published":true,"recentEditorialEvents":[{"type":"reviewerAgreed","content":"","date":"2026-04-29T15:15:01+00:00","index":0,"fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-29T14:12:24+00:00","index":"","fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-05-11T06:26:16+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-11 06:26:16","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9534122","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9534122","identity":"rs-9534122","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}