2.1 Gene Expression Analysis in Normal Tissues

The Human Protein Atlas (HPA) (https://www.proteinatlas.org/) contains protein and mRNA expression profiles of genes in organs, tissues, and cells using various omics technologies, such as mass spectrometry-based proteomics, antibody-based imaging, and transcriptomics. It provides great pan-cancer information through tissue and pathology modules. We made an NLRP1 mRNA expression plot in normal tissues utilizing the HPA database (version: 23.0) [12].

2.2 Gene Expression Analysis in Cancerous Tissues Compared to Normal Tissues

The transcriptional level of NLRP1 was also searched in cancerous tissues compared to normal tissues through GEPIA2, TIMER2.0, UALCAN, and starBase v2.0 databases. The Gene Expression Profiling Interactive Analysis version 2 (GEPIA2); (http://gepia2.cancer-pku.cn/#analysis) is a search tool widely used for gene expression analysis of different tumors and normal samples from the TCGA and the GTEx data. This database provides useful information, such as differential expression analysis profile, survival prognosis analysis, correlation analysis, similar gene detection, expression profile boxplot, and stage plot of the desired gene [13]. The Tumor Immune Estimation Resource version 2 (TIMER2.0) (http://timer.cistrome.org/) is an integrative database with unique features of tumor immunity infiltration analysis and differential gene expression analysis [14]. The University of ALabama at Birmingham CANcer data analysis Portal (UALCAN) (http://ualcan.path.uab.edu/analysis.html) is an interactive search tool that enables the user to access valuable cancer OMICS data, such as TCGA, and Clinical Proteomic Tumor Analysis Consortium (CPTAC). Exploring the interest gene, UALCAN provides a pan-cancer gene expression profile, and survival analysis [15]. starBase v2.0 (http://starbase.sysu.edu.cn/panCancer.php) or the Encyclopedia of RNA Interactomes (ENCORI) is a web service designed for exploring the interaction among different cellular RNA. It also provides a pan-cancer module to analyze the differential expression and survival of the interest gene [16]. We also applied the “Pathological Stage Plot” section of GEPIA2 to assess the association between NLRP1 transcriptional level and cancer stages in all TCGA tumors.

2.3 Protein Expression Analysis

UALCAN was utilized to explore the NLRP1 expression analysis with the source of the CPTAC dataset. We carried out the total protein expression level of the primary tumor in comparison to normal tissue by entering “NLRP1”. The available cancer information for the NLRP1 gene was included from the CPTAC dataset, consisting of breast cancer, Lung adenocarcinoma, Head and neck squamous carcinoma, and Pancreatic adenocarcinoma. The HPA database proved to be a valuable resource in investigating the protein expression patterns in healthy tissues, which was a significant advantage.

2.4 Conservation and Survival Analysis

Utilizing the UCSC genome browser (GRCh38/hg38 human genome assembly; https://genome.ucsc.edu/cgi-bin/hgGateway) [17, 18], we reached the NLRP1 gene conservation among vertebrates. The Gene Set Cancer Analysis (GSCA) platform (http://bioinfo.life.hust.edu.cn/GSCA/#/expression) integrates genomics, pharmacogenomics, and immunogenomic datasets for a comprehensive study of gene sets in cancer [19, 20]. In our investigation, we utilized the GSCA database to assess the impact of NLRP1 expression on overall survival (OS), progression-free survival (PFS), disease-free survival (DFS), and disease-specific survival (DSS) outcomes in patients with various tumors. Additionally, the UCSC Xena Browser (https://xenabrowser.net/) was employed to evaluate the association of NLRP1 with OS, PFS, DFS, and DSS [21]. The GEPIA2 database was also utilized to analyze OS and DFS. The “Survival Map” section of GEPIA2 was implemented to achieve the OS and DFS importance map data associated with NLRP1 among all tumors. Furthermore, the starBase v2.0 database was employed to investigate NLRP1-related OS in all TCGA cancers. A log-rank p-value < 0.05 was employed to denote statistical significance across all databases. Finally, we performed an intersection analysis of data from these databases to evaluate the prognostic potential of NLRP1 transcriptional level across different malignancies.

2.5 Drug Sensitivity Analysis

To assess the potential clinical relevance of abnormal NLRP1 gene expression and its viability as a predictive biomarker for drug screening, we carried out a drug sensitivity analysis using the GSCA database. Our analysis involved collecting the IC50 values of 481 Chemotherapeutic agents or medicines across various cell lines, along with their RNA transcription profiles from the Cancer Therapeutics Response Portal (CTRP) dataset within the GSCA database. Using this database, Pearson correlation analysis was executed to ascertain the degree of association between RNA expression profile and medicine IC50 to determine the relationship between gene expression and drug sensitivity, considering a False Discovery Rate (FDR) adjustment method to ensure the validity of the results. A negative correlation would suggest that increased expression of the gene is associated with reduced drug sensitivity or potential resistance, and conversely.

2.6 Co-Expression Analysis

The Enrichr online tool (https://maayanlab.cloud/Enrichr/) [22-24] was deployed to identify the principal 100 genes co-expressed with NLRP1, utilizing the ARCHS4 RNA-seq gene–gene co-expression matrix (Data S2).

2.7 Functional Enrichment Analysis

Enrichment analysis was performed on a cohort of the mentioned 100 genes to investigate the potential functional implications of genes co-expressed with NLRP1 in different cancer types. This analysis included Gene Ontology (GO) enrichment across the domains of cellular components, molecular function, and biological process. Furthermore, pathway enrichment analysis was conducted, incorporating data from the Kyoto Encyclopedia of Genes and Genomes (KEGG), Wikipathway, and Reactome databases. These enrichment processes were facilitated by the Enrichr tool with a defined significance threshold established at a p-value less than 0.05. To graphically display the results of the GO and pathway enrichment analyses, dot plots were generated utilizing the ggplot2 package [25] within the R programming environment (version R-4.2.1, 64-bit, https://www.r-project.org/) [26] and RStudio Desktop (version 2022.7.0.548) [27].

2.8 Cancer-Associated Fibroblast (CAF) Infiltration Analysis

We utilized the TIMER2.0 database to investigate the correlation between NLRP1 expression and the infiltration of CAF across diverse cancer types. Specifically, we employed the “Immune-Gene” section of TIMER2.0 to assess the association between NLRP1 transcriptional level and CAF infiltration in 32 TCGA neoplasms. To estimate immune infiltration, we utilized the “EPIC” algorithm in conjunction with a partial Spearman's correlation test (purity adjustment). A p-value < 0.05 was considered the statistical significance cutoff.

2.9 Genetic Alteration Analysis

The assessment of NLRP1 genetic alterations was conducted using cBioPortal (https://www.cbioportal.org/) [28]. Specifically, within the “Quick Choose” section, we opted for “TCGA Pan-Cancer Atlas Studies” to scrutinize the genetic alteration characteristics of the NLRP1 gene in all TCGA cancers. Entering “NLRP1” into the “Query” section, we set the molecular profiles to include “mutations” and “copy number alterations”. The “Cancer Types Summary” module provides us with the various NLRP1 genetic alteration types including mutation, amplification, deep deletion, and the existence of multiple alterations in various TCGA cancers. Moreover, we examined the types of NLRP1 gene mutations across various cancers using the “mutations” module. Furthermore, we obtained an overview of cancerous samples with NLRP1 mutations in each tumor type from the “Gene_Mutation” module of the TIMER2.0 database.

2.10 Protein–Protein Interaction Analysis

The STRING database version 12 (https://string-db.org/) is a platform designed for the exploration and analysis of protein–protein interaction (PPI) data, facilitating the identification of direct (physical) and indirect (functional) interactions among proteins [29]. Our objective was to acquire PPI network information for NLRP1 by setting the interaction score cutoff to 0.15 and activating all interaction sources, evidence network edge, and a maximum of 100 interactors as the basic settings. Additionally, we utilized the BioGRID version 4.4.228 (https://thebiogrid.org), a curated biological open-access database that provides information on protein–protein interactions, chemical, and genetic interactions, as well as post-translational modifications, to explore validated physical PPI [30]. The ‘Network’ module of this database was employed, with the layout set to ‘Concentric Circles’. Furthermore, we employed the Venny 2.1.0 tool (https://bioinfogp.cnb.csic.es/tools/venny/) to ascertain the overlap between the STRING and BioGRID databases, thereby identifying the NLRP1 PPIs that were mentioned in the literature [31].

2.11 Methylation Analysis

DNMIVD (http://119.3.41.228/dnmivd/index/) is an interactive database designed for the exploration of DNA methylation patterns sourced from the Gene Expression Omnibus (GEO) and TCGA databases [32, 33]. The database provides the methylation status of the NLRP1 gene across all available TCGA cohorts. To compare methylation levels between cancer and normal samples, an independent Student's t-test was employed, considering cancers with |beta difference| > 0.2 and an independent Student's t-test adjusted p-value < 0.05 as tumors exhibiting significant changes in the NLRP1 gene promoter. Furthermore, in instances where abnormal methylation of the NLRP1 gene promoter was observed, we investigated the relationship between gene expression and promoter methylation in primary tissues using Pearson and Spearman correlation analyses, applying predefined criteria (|R value| > 0.1 and p-value < 0.05).

3.1 Gene Expression Analysis in Normal Tissues

The function annotation of the mammalian genome (FANTOME5), in tandem with the HPA, GTEx project, and a hybridized Consensus dataset, which combines the insights from HPA and GTEx, has been scrutinized within the HPA database framework. This critical examination revealed an elevated expression of the NLRP1 gene in diverse anatomical structures, such as the Choroid plexus, skin, spleen, lymphoid nodes, bone marrow, tonsil, and appendix (Figure 1A and Figure S1). These observations suggest a low degree of tissue specificity of the NLRP1, quantitatively supported by a Tau specificity score of 0.38.

At the granularity of single-cell analyses, NLRP1 manifests a high-expression profile within a range of cell types, encompassing microglia in the central nervous system, natural killer (NK) cells, adaptive immune T-cells, adipose cells, B-cell lineages of the immune system, early spermatids, and monocytic cells (Figure 1B). Parallel genomic assessment using the UCSC Genome Browser has demonstrated a modest level of evolutionary conservation of the NLRP1 gene across the vertebrate taxa (Figure 1C).

3.2 Gene Expression Analysis in Cancerous Tissues Compared to Normal Tissues

For the differential expression analysis of NLRP1 in various cancer types compared to normal tissues, we used the TIMER2.0, starBase, GEPIA2, and UALCAN databases. According to TIMER2.0, NLRP1 mRNA expression exhibited a notable decrease in cancerous tissues across multiple cancer types when compared to their respective healthy counterparts. Specifically, tumor tissues of Breast Invasive Carcinoma (BRCA), Colon Adenocarcinoma (COAD), Kidney Chromophobe (KICH), Kidney Renal Papillary Cell Carcinoma (KIRP), Lung Adenocarcinoma (LUAD), Lung Squamous Cell Carcinoma (LUSC), Prostate Adenocarcinoma (PRAD), Uterine Corpus Endometrial Carcinoma (UCEC) (p-value < 0.001), and Bladder Urothelial Carcinoma (BLCA) (p-value < 0.01), as well as Rectum Adenocarcinoma (READ) (p-value < 0.05), displayed a significant downregulation in NLRP1 expression. Conversely, elevated expression of NLRP1 was observed in Head and Neck Squamous Cell Carcinoma (HNSC), Kidney Renal Clear Cell Carcinoma (KIRC), Liver Hepatocellular Carcinoma (LIHC) (p-value < 0.001), as well as Cholangio Carcinoma (CHOL) (p-value < 0.01) and Esophageal carcinoma (ESCA) (p-value < 0.05) when compared to normal tissues (Figure 2A, Table 1). Findings from the starBase database further validated the downregulated expression of NLRP1 in tumor tissues of BLCA, BRCA, COAD, KICH, LUAD, LUSC, PRAD, and UCEC, while KIRC, LIHC, CHOL, and HNSC exhibited higher expression levels relative to normal tissues, with FDR < 0.05 (Figure 2B, Table 1). The GEPIA2 database analysis revealed downregulation of the NLRP1 gene in tumor tissues of BLCA, BRCA, Cervical Squamous Cell Carcinoma and Endocervical Adenocarcinoma (CESC), COAD, Lymphoid Neoplasm Diffuse Large B-cell Lymphoma (DLBC), KICH, LUAD, LUSC, Ovarian Serous Cystadenocarcinoma (OV), PRAD, READ, Skin Cutaneous Melanoma (SKCM), Testicular Germ Cell Tumors (TGCT), UCEC, and Uterine Carcinosarcoma (UCS). Conversely, it exhibited upregulation in CHOL, HNSC, Pancreatic Adenocarcinoma (PAAD), and Pheochromocytoma and Paraganglioma (PCPG) (p-value < 0.05, and |log₂fold change| > 1) (Figure 2C, Table 1). Additionally, insights derived from the UALCAN database supported the downregulation of NLRP1 in BLCA, BRCA, KICH, LUAD, LUSC, PRAD, and UCEC, while upregulation was observed in CHOL, ESCA, HNSC, KIRC, and LIHC (p-value < 0.05) (Figure 2D, Table 1).

The common tumor types with significantly downregulated or upregulated NLRP1 expression patterns were identified between TIMER2.0, starBase, GEPIA2, and UALCAN databases (Table 1). Our findings indicate a consistent downregulation of the NLRP1 gene in tumor tissues of BLCA, BRCA, KICH, LUAD, LUSC, PRAD, and UCEC, while notable overexpression of NLRP1 was observed specifically in CHOL and HNSC. Furthermore, our statistical analysis examining the significance of the differences observed between NLRP1 transcript levels and the pathological stage of various cancers revealed significant relationships solely in BLCA, LUAD, PAAD, and READ tumor tissues (p-value < 0.05, Figure 3A).

3.3 Protein Expression Analysis

We utilized the HPA database to investigate the protein expression profile of NLRP1 in various normal tissues. Our findings revealed a notably high protein expression of NLRP1 in the cerebral cortex, hippocampus, and skin, as depicted in Figure 3B. Additionally, through the UALCAN database, we examined the protein levels of NLRP1 in various primary tumors compared to normal solid tissues. Our results indicated a significant elevation in NLRP1 protein expression in normal samples relative to primary tumors in lung adenocarcinoma patients. Conversely, in the case of pancreatic adenocarcinoma, we observed a substantial increase in NLRP1 protein level within primary tumors when compared to normal samples (Figure 3C). Understanding the implications of these expression levels is crucial, as they may influence key signaling pathways involved in cancer progression.

3.4 Survival Prognosis Analysis

According to the dysregulation of NLRP1 in various malignancies, we postulate that its expression may be associated with the survival of patients with cancer. Our analysis utilized GSCA, GEPIA2, starBase, and UCSC Xena browser to examine the relationship between NLRP1 expression and diverse survival metrics, such as OS, DFS, PFS, and DSS. Through GSCA, a significant correlation was observed between NLRP1 downregulation and decreased OS in Adrenocortical Carcinoma (ACC) (p = 0.0013, HR = 0.28), HNSC (p = 0.031, HR = 0.74), KICH (p = 0.032, HR = 0.14), LUAD (p = 0.0066, HR = 0.66), PCPG (p = 0.024, HR = 0.12), Sarcoma (SARC) (p = 0.012, HR = 0.60), and SKCM (p = 0.00041, HR = 0.62). Conversely, NLRP1 upregulation was associated with poor OS in Uveal Melanoma (UVM) (p = 0.041, HR = 2.38) and KIRC (p = 0.035, HR = 1.37). Moreover, there was a correlation between NLRP1 downregulation and adverse PFS in ACC (p = 8.8e-06, HR = 0.24), CHOL (p = 0.017, HR = 0.33), HNSC (p = 0.0047, HR = 0.70), LUAD (p = 0.023, HR = 0.75), and SKCM (p = 0.023, HR = 0.77), while its upregulation was linked to worse PFS in Brain Lower Grade Glioma (LGG) (p = 0.0083, HR = 1.43), PRAD (p = 0.02, HR = 1.60), and Stomach Adenocarcinoma (STAD) (p = 0.018, HR = 1.40). DSS analysis revealed a connection between poor DSS and NLRP1 downregulation in ACC (p = 0.0016, HR = 0.27), HNSC (p = 0.047, HR = 0.70), LUAD (p = 0.029, HR = 0.65), PCPG (p = 0.0098, HR = 1.49e-09), SARC (p = 0.032, HR = 0.62), and SKCM (p = 4.6e-05, HR = 0.54). Conversely, the upregulation of NLRP1 was correlated with poor DSS in UVM (p = 0.048, HR = 2.43) and COAD (p = 0.041, HR = 2.01). Furthermore, DFS analysis demonstrated that NLRP1 downregulation had a meaningful relationship with a poor prognosis in KIRC (p = 0.041, HR = 0.23) and CHOL (p = 0.0095, HR = 0.18) (Figure 4A, Table 2).

Utilizing the UCSC Xena Browser, we found that the downregulation of NLRP1 was correlated with diminished OS in ACC, HNSC, KICH, and SARC. Additionally, an elevated expression of NLRP1 was linked to poor OS in LGG and UVM (p-value < 0.05). Furthermore, a notable correlation was established between NLRP1 downregulation and adverse PFS in ACC, CHOL, and HNSC, while NLRP1 upregulation was associated with detrimental PFS in LGG, PRAD, and STAD (p-value < 0.05). Moreover, the expression of NLRP1 exhibited correlations with DSS in various human cancers. Indirectly, upregulated NLRP1 was associated with worse survival in LGG and UVM, while in ACC, PCPG, and SARC, its downregulation was directly linked to an adverse prognosis (p-value < 0.05). Additionally, The DFS analysis indicated that reduced NLRP1 expression is associated with poorer DFS in CHOL, KIRC, and Mesothelioma (MESO). (p-value < 0.05, Figure 4B, Table 2).

In the GEPIA2 database, we stratified the samples into NLRP1 high-expression and low-expression categories based on the NLRP1 median transcription level, subsequently evaluating the association between NLRP1 expression and both OS and DFS. The findings revealed that low NLRP1 expression was associated with poor OS in HNSC (p = 0.047, HR = 0.76), LUAD (p = 0.016, HR = 0.69), PAAD (p = 4e-04, HR = 0.47), and SKCM (p = 0.013, HR = 0.72). Additionally, the analysis of DFS data demonstrated that decreased NLRP1 transcription levels correlated with poor prognosis in various cancers, including CHOL (p = 0.039, HR = 0.37), LUAD (p = 0.033, HR = 0.72), and PAAD (p = 0.00037, HR = 0.45). Furthermore, NLRP1 upregulation was linked to adverse DFS in STAD (p = 0.026, HR = 1.5) (Figure 5A, Table 2). Alternatively, the starBase database highlighted the association between NLRP1 downregulation and diminished OS in PAAD (p = 0.0032, HR = 0.53), SKCM (p = 0.0057, HR = 0.68), ACC (p = 0.024, HR = 0.41), LUAD (p = 0.0056, HR = 0.66), and HNSC (p = 0.025, HR = 0.73), while NLRP1 upregulation was related to adverse OS in KIRC (p = 0.043, HR = 1.37) and DLBC (p = 0.024, HR = 8.12) (Figure 5B, Table 2). Ultimately, an intersection analysis of all survival prognosis data groups revealed that NLRP1 downregulation was correlated with poorer OS in HNSC. Additionally, the decreased NLRP1 expression level was linked to adverse PFS in ACC, CHOL, and HNSC, while increased NLRP1 expression was associated with detrimental PFS in LGG, PRAD, and STAD. Furthermore, ACC, PCPG, and SARC displayed worse DSS prognosis in correlation with NLRP1 downregulation, whereas UVM exhibited poor DSS in correlation with NLRP1 overexpression. CHOL emerged as the only cancer type displaying an association between NLRP1 downregulation and poor DFS (Table 2).

3.5 Drug Sensitivity Analysis

Drug sensitivity analysis revealed that cancer patients with low-expression levels of the NLRP1 gene demonstrated heightened sensitivity to SNX-2112, Dabrafenib, Momelotinib, Dacarbazine, AZD7762, NSC632839, and Merck60 (r value < −0.2, FDR < 0.001) based on the findings from the CTRP dataset (Figure 6A). SNX-2112 is a highly selective inhibitor of Hsp90 that effectively suppresses tumor cell proliferation, angiogenesis, and osteoclast formation in multiple myeloma and other hematologic malignancies [34]. Dabrafenib is a kinase inhibitor used for treating patients with unresectable or metastatic melanoma, metastatic non-small cell lung cancer, locally advanced or metastatic anaplastic thyroid cancer, pediatric low-grade glioma, and other solid tumors harboring specific BRAF mutations [35]. Momelotinib is a small-molecule inhibitor of Janus kinase (JAK) designed for the treatment of primary or secondary myelofibrosis in patients at intermediate or high risk [36]. Dacarbazine is a non-specific cell cycle alkylating agent utilized in the management of metastatic malignant melanoma [37]. NSC632839 is an inhibitor of Ubiquitin C-Terminal Hydrolase L1 that significantly induces protein light chain 3 puncta formation and modulates autophagy [38]. Merck60 is a highly potent and selective inhibitor of HDAC1 and HDAC2 that influences histone deacetylation [39]. For a comprehensive overview of these correlations between NLRP1 expression and drug sensitivity, refer to Figure 6A, which visually represents the complex relationship between NLRP1 and drug responses (r value < −0.15, FDR < 0.05).

3.6 Functional Enrichment Analysis

The results of our analysis elucidate a pronounced participation of the delineated genes in an array of critical cellular mechanisms. These genes are discovered to exert pivotal functions within a range of signaling pathways, such as the T Cell Receptor Signaling Pathway, the receptor-mediated Signaling Pathway, and pathways governing the Negative Regulation of Inflammatory Response and the Regulation of Cytokine Production. They are also integral in modulating pathways like the Regulation of Phosphatidylinositol 3-Kinase Signaling and thus underscore their instrumental roles in modulating inflammatory signaling cascades. In addition, our examination has exposed that a significant fraction of these genes is associated with a spectrum of molecular activities, notably including Non-Membrane Spanning Protein Tyrosine Kinase Activity, GTPase Regulator Activity, and Kinase Binding. In terms of cellular localization, the genes are predominantly localized to the Cytoplasmic Side of the Plasma Membrane and intrinsically within the T Cell Receptor Complex, suggesting their active participation in signal transduction pertinent to immune responses. Furthermore, our analysis delves beyond the scope of functional annotations to incorporate a comprehensive examination of the involvement of the identified genes in distinct biological pathways. Notably, we have determined significant pathways by leveraging comprehensive databases, such as KEGG, Wikipathway, and Reactome. These pathways include the T cell receptor signaling pathway, the chemokine signaling pathway, and other pathways that play essential roles within the broader framework of the immune system. These pathway associations provide valuable insights into the potential implications of the identified genes in shaping immune system functionality (Figure 6).

3.7 Cancer-Associated Fibroblast Infiltration Analysis

Previous studies have demonstrated the involvement of CAF in adjusting the behavior of diverse neoplasm-infiltrating immune cells [40]. The infiltration of CAFs may further modulate the role of NLRP1 in the tumor microenvironment. To investigate the association between CAF infiltration and NLRP1 expression, we employed the TIMER2.0 database. Figure 7A displays the outcomes before purity adjustment, revealing a positive correlation between NLRP1 expression and diverse TCGA cancers. Tumor purity emerges as a significant confounding element in this investigation, as the majority of immune cell types exhibit negative correlations with tumor purity. Consequently, upon implementing the purity-adjusted EPIC algorithm, the analysis revealed a negative correlation between NLRP1 expression and CAFs in ACC, BRCA, CESC, COAD, ESCA, Glioblastoma Multiforme (GBM), HNSC, KICH, KIRC, KIRP, LGG, LIHC, LUAD, LUSC, MESO, OV, PAAD, PCPG, PRAD, READ, SKCM, STAD, TGCT, Thyroid Carcinoma (THCA), Thymoma (THYM), and UCEC. Additionally, a negative association was identified between NLRP1 transcriptional level and CAF in specific subgroups, such as HPV-negative HNSC, BRCA basal cell carcinoma, BRCA HER2 positive, BRCA-lumA, BRCA-lumB, and SKCM metastasis based on the EPIC algorithm. Conversely, a positive correlation has been observed in UVM (Figure 7B).

3.8 Genetic Alteration Analysis

Using cBioPortal, we examined the genetic alterations of NLRP1 across 32 TCGA tumor types. Notably, the highest genetic alteration frequencies in the NLRP1 gene were observed in SKCM, UCEC, and ACC, with alteration frequencies of 14.86%, 10.02%, and 5.49%, respectively. Moreover, SKCM, ACC, and UCS tumor subjects exhibited the highest frequencies of NLRP1 mutations (13.96%), deep deletions (2.2%), and amplifications (3.51%), respectively. Several alterations were detected with frequencies < 1% (Figure 8A, Data S3). A total of 303 mutations were identified in the NLRP1 gene across TCGA tumor subjects, comprising 243 missenses, 44 truncating mutations, 15 splice site alterations, and 1 in-frame mutation. These mutations were distributed throughout the NLRP1 gene, encompassing both domain and non-domain sites (Figure 8B, Data S4). Additionally, we utilized TIMER2.0 to further scrutinize NLRP1 gene mutations across various human malignancies. The NLRP1 mutations were predominantly observed in patients with SKCM (61 out of 468 patients), UCEC (51 out of 531 patients), and COAD (22 out of 406 patients) (Figure 8C).

3.9 Protein–Protein Interaction Analysis

The STRING database was employed to extract the top 100 genes exhibiting both functional and physical interaction with NLRP1, as illustrated in Figure 9A. Subsequently, by employing the BioGRID database, we identified 22 genes that were physically associated with NLRP1. This set comprised 14 genes (WWP2, CDK9, APOA1, TRIM68, EIF6, CUL3, UBA52, CLTC, SMC2, CSNK1E, CUL4B, NEK4, KCTD17, CUL4A) curated from high-throughput (HTP) studies, and seven genes (CASP2, HERC5, NLRC4, TRIM65, CASP9, APAF1, MAPK14) obtained from low throughput (LTP) research, with only one gene (TRIM25) curated from both HTP and LTP sources (Figure 9B). The intersection of the genes from STRING and BioGRID was determined and visualized using Venny 2.1 (Figure 9C), leading to the identification of five genes (APAF1, CASP2, CASP9, NEK4, and NLRC4) with literature-confirmed interactions with NLRP1.

3.10 Methylation

When utilizing the DNMIVD database to explore the methylation pattern of the NLRP1 gene's promoter across 22 distinct cancer types, we determined significant hypomethylation in the promoter region of the NLRP1 gene in ESCA (Beta difference = −0.219818 and adjusted p-value = 0.000117) and PAAD (Beta difference = −0.202757 and adjusted p-value = 0.003035) (Data S5). Moreover, the results of the correlation study revealed a notable inverse association between the level of methylation in the promoter region and the expression level of the NLRP1 gene in ESCA (Pearson p-value = 4.0513e-06, Pearson R value = −0.344137, Spearman p-value = 6.55e-8, Spearman R value = −0.398828) (Data S6).