Integrated multi-omics analysis reveals key genetic, metabolic, and microbial drivers in bladder cancer insights into molecular subtyping and therapeutic approaches: A tumor marker prognostic study

Integrated multi-omics analysis reveals key genetic, metabolic, and microbial drivers in bladder cancer insights into molecular subtyping and therapeutic approaches: A tumor marker prognostic study | 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 Integrated multi-omics analysis reveals key genetic, metabolic, and microbial drivers in bladder cancer insights into molecular subtyping and therapeutic approaches: A tumor marker prognostic study Zhiyong Tan, Xiaorong Chen, Yinglong Huang, Shi Fu, Chen Gong, and 6 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5898970/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background: Bladder cancer (BLCA) is a common malignancy with significant impact on patient health. The aim of this study was to explore the potential mechanisms of BLCA through a combination of multi-omics and single-cell analyses. Methods: In this study, samples from BLCA and paracancerous tissues were collected for transcriptome, whole-exome sequencing, metabolome and intratumoural microbiome sequencing. These data were then co-analyzed with publicly available datasets to identify and analyze key genes, metabolites and microbiomes as well as their regulatory mechanisms in the pathogenesis of BLCA. Different BLCA clusters were then identified on the basis of key genes. Differences among the clusters were then investigated in terms of biological pathways, immunological microenvironment, genetic alterations, immunotherapy and drug susceptibility. The prognostic value of the key genes was then analyzed using publicly available data, and their molecular regulatory mechanisms were further investigated. Finally, the expression patterns of the key genes were observed at the single cell level and key cells were identified. Results: In this paper, three key genes (AHNAK, CSPG4, and NCAM1), 90 key metabolites and two key microorganisms (Sphingomonas koreensis and Rhodospirillaceae) were identified in a multi-omics analysis. Of these, key genes and key metabolites were negatively correlated. The BLCA samples from transcriptome sequencing were then divided into cluster 1 and cluster 2 based on key genes. Single-cell analysis identified nine cell types, with fibroblasts exhibiting the highest expression of key genes, thus establishing fibroblasts as the key cell in this study. Notably, AHNAK expression was higher in fibroblast subtypes. Conclusion: The combined multi-omics analysis revealed a significant correlation between three key genes (AHNAK, CSPG4, and NCAM1) and multiple key metabolites and key microorganisms, which offering a new reference and theoretical support for the treatment and research of BLCA. Bladder cancer Multi-omics Key genes Key metabolites Key microorganisms Single-cell analysis Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 Figure 13 1. Introduction Bladder cancer (BLCA) is a malignant tumor primarily originating from the uroepithelium, exhibiting significant incidence variations by race and gender, and ranks as the 4th most prevalent cancer in men in America [ 1 ] . Non-muscle invasive bladder cancer (NMIBC) and muscle invasive bladder cancer (MIBC) are pathologically categorized by tumor infiltration depth, guiding clinical diagnosis and treatment, with respective 5-year survival rates of approximately 90% and 60% [ 2 ] . Individualizing treatment based solely on pathological grading is challenging, prompting extensive research into BLCA's molecular mechanisms, including gene mutations [ 3 ] , signaling pathways [ 4 ] , tumor microenvironment [ 5 ] , and metabolic pathways [ 6 ] , to enhance diagnostic efficiency and patient survival. However, BLCA's heterogeneity contributes to drug resistance and recurrence post-treatment, leading to suboptimal survival rates [ 7 ] . Thus, comprehensive multi-omics analysis to elucidate multilevel heterogeneity is crucial for improving patient prognosis and is an urgent challenge to address. Single-cell combined transcriptomics is a popular combination in developing BLCA predictive models for predicting patient survival and immune microenvironment [ 8 – 10 ] . While this approach has uncovered some gene expression heterogeneity, it has not elucidated inter-patient and intra-tumor heterogeneity. Following the innovation of science and technology, whole-exome sequencing and metabolomics have gradually been widely used in tumor research. Whole-exome data indicate that normal uroepithelial cells can harbor over 500 to 2,000 mutations by ages 50 to 65, alongside a higher tumor mutational burden (TMB) in BLCA driven mainly by activation of APOBEC3 cytidine deaminase [ 11 ] . In addition, distinct grades of BLCA show differences in glucose, lipid, and carnitine metabolism, which have all been associated with tumor invasion and metastasis [ 12 ] , suggesting that genetic variants and metabolic pathways may be etiological factors contributing to the development of BLCA. Recent research suggested the microbiome may be a causative or cofactor in genitourinary malignancies [ 13 ] ; for instance, Sun et al. found greater microbial diversity in NMIBC compared to MIBC tissues, with distinct dominant species between groups by 2bRAD sequencing for Microbiome (2bRAD-M) [ 14 ] . Increased microbial abundance was noted in tumor patients at higher recurrence or progression risk versus healthy controls [ 15 ] . Therefore, a deeper understanding of the microbiome may yield new prognostic insights for BLCA. Integrating multi-omics data can illuminate how various molecular characterizations synergistically influence disease onset, progression, and therapeutic response, providing a holistic understanding of biological systems. In this research, we incorporated transcriptomics, whole exome sequencing, metabolomics, the analysis of intratumoral microbial populations, and single-cell profiling to effectively identify the key genes, metabolites, and microbes associated with BLCA. We conducted an intensive investigation into their functions and regulatory interactions across various biological hierarchies. Concurrently, we categorized BLCA using molecular profiling of the key genes and thoroughly examined its underlying molecular mechanisms. Ultimately, we illuminated the expression patterns of these crucial genes within pivotal cellular contexts at the single-cell resolution. These findings will provide an insightful understanding of the molecular mechanism of BLCA and the formulation of new therapeutic strategies, with a perspective of contributing to the enhancement of personalized treatment for BLCA patients. 2. Materials and methods 2.1 Patient Selection and Sample Collection A total of 35 cases of cancer and paracancerous (normal) tissues were collected from patients diagnosed with bladder cancer (BLCA) for comprehensive sequencing of the transcriptome, Whole-Exome Sequencing (WES), metabolome and intratumoural microbiome sequencing. Considering the possibility of endogenous contamination during surgery, all samples were collected aseptically in the operating room and clean tissue blocks were selected as much as possible, and then cryopreserved and stored at -80°C for preservation. All patients were from The Second Affiliated Hospital of Kunming Medical University. The study was approved by the Research Ethics Committee of The Second Affiliated Hospital of Kunming Medical University, and written informed consent was obtained from each patient. 2.2 Data Source TCGA-BLCA dataset was sourced from the University of California Santa Cruz (UCSC) Xena database ( https://xena.ucsc.edu/ ) as a transcriptome training set [ 8 ] . The dataset comprised RNA sequencing data from 411 BLCA tumour tissue samples and 9 control tissue samples, along with corresponding clinical information and survival data. Additionally, the GSE13507 dataset (sequencing platform: GPL6102) was obtained from the Gene Expression Omnibus (GEO, https://www.ncbi.nlm.nih.gov/geo/ ) as an external validation set for transcriptome analysis [16] , which encompassed 188 BLCA patient tumour tissue samples and 68 control bladder mucosa tissue samples. Furthermore, two BLCA-associated single-cell datasets were included and integrated into the single-cell dataset for this study, namely, the GSE135337 dataset with GPL24676 platform and the GSE222315 dataset with GPL24676 platform, both of which were obtained from the GEO database, were incorporated into this study. The GSE135337 comprised seven BLCA patient cancers and one paraneoplastic tissue sample. The GSE222315 dataset comprised nine BLCA patients with cancer and four paraneoplastic tissue samples. 2.3 Analysis of Transcriptome Sequencing Data To improve data quality and standardize the data, raw reads from transcriptome sequencing were quality controlled for clean data. The clean data was then compared to the human genome (GRCh38) using TopHat 2.0. Next, the Transcripts Per Kilobase of exon model per Million mapped reads (TPM) values were calculated for each gene to assess the degree of bias in mRNA distribution. Principal component analysis (PCA) was then performed on both groups of samples using the R package ‘ropls’ (version 1.34.0) [ 17 ] to estimate the variability of the data. In addition, the correlation between the two groups of samples was assessed by Spearman analysis (correlation coefficient |Cor| > 0.3, p < 0.05). The objective of differential expression analysis was to identify genes that exhibited significant differences in expression between the various sample groups. In this study, the R package ‘DESeq2’ (version 1.42.0) [ 18 ] was employed to ascertain the disparities in gene expression levels and to acquire differentially expressed genes (DEGs) between cancer and normal samples with adjusted p value 0.5. The Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) were databases that were frequently utilized for the purpose of pathway research. In order to gain insight into the relevant biological functions and pathways of DEGs, GO and KEGG enrichment analyses of DEGs were conducted using the ‘clusterProfiler’ package (version 4.7.1.3) [ 19 ] in R language with adjusted p value < 0.05. 2.4 Analysis of WES Data For the identification of significant mutant genes (SMGs) in both sample groups, tumour mutation information from 35 BLCA tumour samples from the WES data was analyzed using ‘maftools’ (version 2.17.10) [ 20 ] . To explore the biological pathways associated with SMGs, GO and KEGG enrichment analyses of SMGs were also performed using the same methods with p < 0.05.The Oncobox Pathway Database (OncoboxPD, https://open.oncobox.com ) comprised 51,627 human molecular pathways that was processed in a uniform manner. KEGG enrichment was performed according to the above SMGs using OncoboxPD with adjusted p value < 0.05, and pathway activation level (PAL) values were calculated for all KEGG pathways to observe the activation-inhibition status of the relevant pathways enriched by SMGs. 2.5 Analysis of Metabolomic To assess the degree of variation in the metabolomic data of the two groups of samples, PCA analysis was performed on the metabolomic sequencing data of the two groups of samples. Similarly, the correlation between the two groups of samples in the metabolomic sequencing data was assessed by Spearman analysis with |Cor| > 0.3 and p < 0.05. Next, differentially expressed metabolites (DEMs) in cancer and normal samples were screened based on the importance in projection (VIP) values of variables calculated from the orthogonal partial least squares discriminant analysis (OPLS-DA) model and the p values of Student's t-test in paired samples, with thresholds of VIP > 1 and p < 0.05. To investigate the biological processes associated with the DEMs, a metabolic profile was performed using MetaboAnalyst to convert the compound names of the DEMs to KEGG IDs, and the pathways involved in the DEMs were analyzed in the KEGG database at a significance level of p < 0.05. 2.6 Analysis of Intratumoural Microbial Data To ensure the accuracy of subsequent analyses, unprocessed intratumoural microbiome sequencing data were used for quality control and chimeras were removed using UCHIME. Sequences with > 97% similarity were clustered into the same operational taxonomic unit (OTU). For clustered OTUs, taxonomic information was annotated and evaluated using the Mothur algorithm using the Silva database. Furthermore, the alpha and beta diversity of intratumoural microorganisms in different groups of samples were analyzed and compared to assess differences in the composition of intratumoural microorganisms between the cancer and normal groups. In order to analyze the abundance of microorganisms in different groups of samples, bar charts were plotted to demonstrate the relative abundance of intratumoral microorganisms at the phylum, family and genus levels for both groups of samples. Next, specific microbial taxa were identified between the two groups of samples using Linear Discriminant Analysis (LDA) Effect Size (LEfSe) at the genus level (LDA > 2, p < 0.05). To investigate the biological processes associated with these differential microorganisms, KEGG pathway enrichment analysis of differential microorganisms between the two groups at the genus level was performed using PICRUSt2 software. Microbes never exist in isolation from each other and are highly interconnected in response to various environmental changes. Therefore, this research further analyzed the genus-level colonies of both groups of samples and derived the associated genus-level driver colonies. 2.7 Co-analysis of Transcriptome and WES For the purpose of exploring the somatic mutations of BLCA patients, the somatic mutations of BLCA patients were analyzed in all the samples of the TCGA-BLCA and WES data. The amplification and deletion of CNV were also analyzed by Gistic2.0. In order to identify genes with significant differences shared at the transcriptome level and at the WES level, DEGs obtained based on transcriptome analyses were merged with SMGs from WES data, and the resulting genes were defined as candidate genes. The candidate genes were further analyzed for GO and KEGG enrichment analyses with an adjusted p value < 0.05. Patients with BLCA were divided into high and low expression sets according to the optimal expression threshold of each candidate gene. The data were then analyzed using the ‘survminer’ software package (version 0.4.9) [ 21 ] , and Kaplan-Meier (KM) survival curves were constructed for the overall survival (OS) of patients in the two sets. Candidate genes with OS differences at p < 0.05 between sets were considered to have prognostic value and were defined as candidate prognostic genes. The expression of candidate prognostic genes between cancer and normal groups was also analyzed and validated in transcriptome level, TCGA-BLCA dataset, and GSE13507 datasets, respectively, by Wilcoxon test (p < 0.05). In addition, the R package ‘pROC’ was used to plot receiver operating characteristic (ROC) curves and calculate area under the curve (AUC) to assess the ability of candidate prognostic genes to distinguish between cancer and normal samples. Genes with AUC > 0.7 in three sets of transcriptome data were defined as biomarkers for subsequent studies. 2.8 Combined Analysis of Transcriptome, Metabolome and Intratumoural Microbiome To obtain relevant genes and metabolites at the transcriptome, WES and metabolome levels, correlations between biomarkers and DEMs were assessed by Spearman analysis in all BLCA sequencing samples. Biomarkers and DEMs with |Cor| > 0.3 and p < 0.05 were defined as key gene 1 and key metabolite 1, respectively, and their interactions were investigated. Similarly, key gene 2 and key microbiome 1 associated at transcriptome, WES and intratumour microbiome levels were also obtained under the above parameters. In addition, the correlation between DEM and differential microbes was also assessed by Spearman analysis under the same parameters to derive the key metabolite 2 and key microbiome 2 associated at the metabolome as well as intratumour microbiome level. The aforementioned key genes, key metabolites and key microorganisms were integrated to yield key genes, key metabolites and key microorganisms that were meaningful at multi-omics levels. These were subsequently assessed for their regulatory relationships. 2.9 Analysis of BLCA Clusters In order to identify the molecular clusters of BLCA associated with key genes, a consensus clustering approach was employed to categorize BLCA samples into clusters based on the expression of key genes in transcriptome sequencing data. Furthermore, the distribution differences between the various clusters were observed and evaluated using PCA and t-distributed stochastic neighbour embedding (tSNE). Then, based on the clusters just achieved, different cluster samples were analyzed for differences, screened by |log 2 FC| > 0.5 and p < 0.05, to obtain inter-cluster DEGs. Later, GO and KEGG enrichment analyses were performed on the inter-cluster DEGs with adjusted p value < 0.05. To further observe the association between microbial communities in different clusters, the association networks between microbial communities (genus level) in different clusters were analyzed in the intratumoral microbiome sequencing data. Functional prediction of different clusters was also performed to explore the functional differences involved in different clusters. Similarly, intercluster DEMs were screened by VIP > 1.00 and p < 0.05 based on metabolome sequencing data. metabolic pathways involved in transformed intercluster DEMs were subsequently analyzed in the KEGG database (p < 0.05). Furthermore, a series of subsequent in-depth analyses of different BLCA clusters based on transcriptome sequencing data were performed. Genome enrichment analysis (GSEA) with HALLMARK gene set: h.all.v2023.1.Hs.symbols.gmt as the reference set of genes and genome variation analysis (GSVA) with metabolism-related pathways [ 22 ] as the reference gene set were used to investigate significant biological pathways and metabolic pathways between clusters. Differences in TIME, with the TIME signature as the reference gene set, were also investigated. Inter-cluster differences in the abundance of 22 immune-infiltrating cells, 66 immune checkpoint genes [ 23 ] , activity of immune-circulatory pathways, patient response to immunotherapy, somatic mutations, mutations in typical oncogenic pathways, and drugs were also studied. All of the above analyses were considered statistically significant with a p value < 0.05. For the purpose of analyzing the possible potential roles of key microorganisms in immune-related processes, the correlation between key microorganisms and MHC molecules, immune activators, and immunosuppressants in different clusters was assessed by Spearman analysis in intratumoural microbiome sequencing data, respectively. To observe the expression levels of key genes and key metabolites among different clusters, the differences in the expression of key genes, key metabolites and key microorganisms among different clusters were compared by Wilcoxon test in the transcriptome, metabolome and intratumoural microbiome sequencing data, respectively (p < 0.05). The key genes that were significantly different in BLCA clusters were used for subsequent analyses. 2.10 Association Analysis of Clinical Characteristics of Key Genes To examine the expression of key genes in relation to clinical characteristics, BLCA patients were classified into distinct clinical subgroups based on varying clinical characteristics in the TCGA dataset. The differential expression of key genes between subgroups was then evaluated using the Wilcoxon test, with a significance level of p < 0.05. Furthermore, correlations between key genes and individual clinical characteristics were evaluated through the use of a Spearman analysis. Furthermore, BLCA patients were categorized into two groups based on the median critical value of the expression of each key gene. KM survival curves were constructed for the overall survival (OS) of patients with high and low expression of each key gene in different clinical subgroups. 2.11 Enrichment Analysis of Key Genes To explore the biological pathways involved in the key genes in BLCA, single-gene GSEA enrichment analysis of the key genes was performed based on the correlation of 'c2.cp.kegg.v7.0.symbols.gmt' as the reference gene set in the transcriptome sequencing data (adjusted p value < 0.05). In the TCGA dataset, BLCA patients with survival information were divided into high and low expression subgroups based on the median threshold of key gene expression. Pathways or functions that were significantly enriched between the high and low expression subgroups for each key gene were explored using 'HALLMARKgenesets: h.all.v2023.1.Hs.symbols.gmt' as the reference geneset and GSEA based on the multiplicity of differences (adjusted p value < 0.05). 2.12 Drug Prediction and Molecular Docking To further explore potential therapeutic agents for BLCA, potential agents or molecular compounds interacting with the key genes were predicted by L1000 FWD ( https://maayanlab.cloud/l1000fwd/ ). Based on the results of agent analysis in the above, the most relevant drugs to each key gene were selected for molecular docking to explore the binding ability of the key genes to the relevant agents. Binding energy < -5kcal/mol meant high binding activity. 2.13 Information, Function, and Expression Analysis of Proteins Encoded by Key Genes The National Center for Biotechnology Information (NCBI) was conducted further research into the information regarding the key genes, proteins and related functions that they encode. To observe the expression of the key genes in pan-cancer, the expression of each key gene was compared in pan-cancer tumours and control samples based on pan-cancer data from the TCGA database. After that, the expression levels of key genes in different tissues and expression at protein level in tumour and normal samples were analyzed by The Human Protein Atlas (HPA, https://www.proteinatlas.org/ ). 2.14 Analysis of Single-cell Data To ensure the accuracy of the analysis, low-quality cells were first filtered out, and then the top 2,000 highly variable genes among the remaining cells were extracted. The cells were further analyzed by clustering to generate cell clusters. And the cell types obtained from the above clustering were determined by cell annotation. Furthermore, the distribution of each key gene in each cell type was demonstrated in the single-cell dataset, and their expression levels in different samples (cancer and control) were compared. All cell types were also analyzed for KEGG functional enrichment (p < 0.05). In light of the differential expression of the key genes in each cell type and the enrichment results for all cell types, as well as the findings of relevant literature reports, fibroblasts were designated as the key cells for subsequent analyses. Next, ligand-receptor interactions between key cells and other cell types were analyzed in order to reveal the communication of each cell. To further explore the heterogeneity of the key cells, the key cells were analyzed by dimensionality reduction clustering, and the key cells were reclustered into different subclusters. And the subclusters were annotated according to the references [24–26] to obtain the key cell clusters. The potential differentiation trajectories of key cell clusters were further observed by mimetic temporal sequencing analysis, while the expression of key genes in different temporal sequences was visualized. In addition, upstream regulators of key genes in various clusters of key cells were probed to speculate on the potential mechanism of action of the key genes. In addition, to further explore the degree of malignancy of the cell types, the epithelial cells obtained from the above analyses were reclustered into different subclusters and the clusters of the cell subclusters were determined. Based on the above results of epithelial cell descending clustering, the proposed time series of epithelial cell clusters were analyzed and the differentiation trajectories of epithelial cell clusters with different degrees of malignancy were analyzed. The expression of key genes during different time series was also visualized. And their expression in different malignancy degree cell clusters differentiation was also visualized and compared. 2.15 Statistical Analysis Bioinformatic analyses were performed using the R programming language (version 4.2.2). The Wilcoxon test was used to compare differences between two groups. Fisher's exact test was used to calculate the p value < 0.05 for significance. 2.16 Remark Criteria Statement Our research adheres to the Reporting Recommendations for Tumor Marker Prognostic Studies (REMARK) criteria, ensuring transparency and comprehensive reporting of our study findings [ 27 ] . 3. Results 3.1 Quality Assessment of Sequencing Data To assess transcriptome sequencing data quality, we first examined gene expression profiles. Box plot analysis identified four outlier samples (36, 37, 38, and 39), which were excluded. The remaining 70 samples, balanced at 35 tumour and 35 paraneoplastic samples, were retained ( Figure S1 a ). PCA of these samples showed distinct clustering within each group ( Figure S1 b ). Correlation analysis revealed a generally positive correlation between samples from the two groups ( Figure S1 c ). Metabolome sequencing data effectively distinguished tumour from normal samples ( Figure S1 d ). Intratumoural microbiome sequencing data were analyzed to identify cluster OTUs. Sparse curves reached a plateau, indicating sufficient sequencing depth ( Figure S1 e ). Rank-abundance curves illustrated the richness and evenness of the samples ( Figure S1 f ). The species cumulative box plot showed that species diversity increased with sample size until stabilizing ( Figure S1 g ). The aforementioned results collectively demonstrated that the sequencing quality in this study was high and that the samples were effectively differentiated. 3.2 Identification and Function of DEGs, SMGs and DEMs in BLCA Differential expression analysis identified 7,302 DEGs between cancer and normal groups (Fig. 1a, b ). GO enrichment analysis revealed significant terms such as sarcolemma, muscle tissue development, embryonic organ development, contractile fibre and collagen-containing extracellular matrix (Fig. 1c). KEGG enrichment analysis highlighted pathways including cytoskeleton in muscle cells, hypertrophic cardiomyopathy, focal adhesion, axon guidance, dilated cardiomyopathy and other pathways (Fig. 1d). WES of 35 BLCA tumour samples identified 78 SMGs, with missense mutations being predominant (Fig. 1e). GO and KEGG enrichment analyses of SMGs revealed 139 GO entries and five KEGG pathways, including basal transcription factors, spinocerebellar ataxia, human immunodeficiency virus 1 infection, FoxO signaling pathway, and osteoclast differentiation (Fig. 1f, g ). KEGG enrichment using OncoboxPB showed activated pathways like leishmaniasis (PAL = 5, p = 0.0034) and inhibited pathways like p53 signaling (PAL = -43, p < 0.0001) (Fig. 1h). Our findings indicated that the DEGs and SMGs were co-enriched for the FoxO signaling pathway in KEGG, suggesting that this pathway might be a potential recipient of the both the DEGs and SMGs in BLCA. Metabolome sequencing identified 212 DEMs, enriched in 14 KEGG pathways related to various amino acid metabolic processes (Fig. 1i-k). 3.3 Identification and Function of BLCA-associated Differential Microbes To examine the distinctions in species richness and composition of intratumoural microorganisms, alpha and beta diversity were evaluated across diverse groups of sample species. With regard to alpha diversity, the Simpson, Shannon, InvSimpon and Pielou indices demonstrated no statistically significant differences between cancer and normal samples, indicating species diversity and even distribution ( Figure S2a ). In the case of beta diversity, the stress value was 0.11, indicating the reliability of the results ( Figure S2b, c ). Intratumoral microbiome composition was assessed at the phylum, family, and genus levels. Proteobacteria were most abundant in both cancer and normal samples, but Cyanobacteria was notably higher in normal samples and lower in cancer samples (Fig. 2a). Rhodocyclaceae and Comamonadaceae were the most prevalent families in both groups (Fig. 2b). At the genus level, Phaeospirillum and Methyloversatilis were the most abundant (Fig. 2c). Five differential microbiomes were identified between cancer and normal samples: Phaeospirillum, Acinetobacter, Rubrivivax, Staphylococcus, and Dialister (Fig. 2d-f). KEGG enrichment analysis of these microbiomes revealed significant pathways: mycolylarabinogalactan- peptidoglycan complex biosynthesis, gluconeogenesis I, and peptidoglycan biosynthesis II (staphylococci) (Fig. 2g). Analysis of driving species indicated differences in community networks, with greater microbiome density observed in normal samples compared to tumour samples (Fig. 2h; Table S1 ; Figure S3 ). 3.4 Recognition of Biomarkers Analysis of somatic mutations in BLCA patients using TCGA-BLCA and WES data revealed that sense mutations had the highest frequency and missense mutations were most common (Fig. 3). Intersecting DEGs and SMGs identified 26 candidate genes (Fig. 4a), which were associated with clathrin-dependent endocytosis, WW domain binding, lysosomal lumen, protein serine kinase activity, focal adhesion, and other GO entries (Fig. 4b). KEGG enrichment analysis highlighted significant enrichment in Virion-Ebolavirus, Lyssavirus, Morbillivirus, and African trypanosomiasis pathways (Fig. 4c). Prognostic evaluation of these genes classified patients into high- and low-expression groups. KM survival curves indicated significant survival differences for 15 genes ( Figure S4 ). Expression validation in transcriptome, TCGA-BLCA, and GSE13507 datasets further confirmed that APOL1, DHX34, and TNK2 were up-regulated, while AHNAK, CSPG4, NCAM1, and PCDHB4 were down-regulated in the cancer group (Fig. 4d). ROC analysis identified AHNAK, CSPG4, DHX34, NCAM1, and PCDHB4 as biomarkers with AUCs > 0.7 across all datasets (Fig. 4e). 3.5 Identification of Key Genes, Metabolites and Microbiomes A combined analysis of biomarkers, DEMs, and differential microbiomes identified three key genes (AHNAK, CSPG4, NCAM1), 90 key metabolites, and two key microbiomes (Sphingomonas koreensis and Rhodospirillaceae) from multi-omics data ( Figure S5; Fig. 5a-c ). Correlation and network analysis revealed positive correlations among the key genes, no correlation between key microbiomes and key genes or metabolites, and negative correlations between key genes and key metabolites (Fig. 5d-e). 3.6 Two Clusters for BLCA BLCA samples were classified into two clusters based on key genes (Fig. 6a, b ). Analysis revealed 3,464 inter-cluster DEGs, which were enriched in 1,665 GO terms, with the top categories including collagen-containing extracellular matrix and extracellular matrix organization (Fig. 6c-e). A study of a prognostic model of BLCA suggested that PD-L1 expression could predict the prognosis of patients with BLCA, possibly related to extracellular matrix passage of collagen [ 28 ] . KEGG enrichment analysis identified 81 pathways, notably cytoskeleton in muscle cells and focal adhesion (Fig. 6f). Microbial community analysis showed most microbiomes were positively correlated within clusters (Fig. 6g). Additionally, 17 inter-cluster differential metabolites were found, enriched in amino sugar and nucleotide sugar metabolism, and ascorbate and aldarate metabolism (Fig. 6h-j). GSEA identified significant pathways across clusters, including interferon gamma response and epithelial mesenchymal transition ( Figure S6 ). GSVA further identified 10 differential metabolic pathways (Fig. 6k ). 3.7 Tumor Immune Microenvironment (TIME) in BLCA In BLCA clusters, the TIME analysis revealed significant differences between clusters in inhibitory molecules, IFNG response, tumor cell recognition, Priming and activation, and inhibitory cells (MDSCs) (Fig. 7a, b ). Differential immune cells included M0/M1 macrophages, resting mast cells, resting natural killer cells, follicular helper T cells, and gamma delta T cells (Fig. 7c, d ). NCAM1 was positively correlated with gamma delta T cells (Cor = 0.54) and negatively with M0 macrophages (Cor = -0.48) (Fig. 7e). Among 66 immune checkpoint genes, 25 differed between clusters, such as BTLA, CD274, and CTLA4 (Fig. 7f). Differences were observed in cancer immune cycle steps 1, 4, and 7 (Fig. 7g). TIDE scores showed notable divergence, indicating different immune profiles between clusters (Fig. 7h). Key genes correlated positively with stromal, immunity, and ESTIMATE scores, suggesting their relevance to immunotherapy (Fig. 7i). These scores might interact with key metabolites (Fig. 7j). Key microbes with a lot of immune factors showed negative correlations between BLCA clusters (Fig. 7k). 3.8 Gene Mutations in the BLCA The genomic mutation landscape of the top 20 most commonly mutated genes in BLCA revealed detailed information on CNV amplification and deletion, mutation types, and base mutations. MUC4 showed the highest mutation frequency, with a notable percentage of missense mutations and C-to-T base mutations (Fig. 8a, b ). Differences in 10 classical oncogenic pathways were observed between clusters (Fig. 8c). Chemotherapy response analysis identified 236 differential agents in the CTRP database, 43 in GDSC, and 785 in PRISM Repurposing (Fig. 8d; Figure S7a-c ). By comparing the 50% inhibiting concentration (IC50) of common chemotherapeutic agents, FGFR inhibitors and EGFR inhibitors between different clusters, a total of 63 agents were obtained that differed between clusters, and most of them had significant correlation with the gene NCAM1 (Fig. 8e; Figure S7d-j ). At the same time, key genes were also able to act on classical therapeutic pathways as well as pathways through corresponding targeted agents (Fig. 8f). Key genes, including AHNAK, CSPG4, and NCAM1, differed in expression between clusters, and 40 metabolites were differentially expressed, with no microbial differences detected (Fig. 8g; Figure S8 ). 3.9 Exploration of Key Gene-related Mechanisms To analyze key gene expression in BLCA, patients were categorized based on clinical features from TCGA-BLCA. AHNAK and NCAM1 both exhibited differential expression in T and N stage as well as stage, while CSPG4 displayed a more limited differential expression in stage ( Figure S9 ). Meanwhile, AHNAK showed a positive correlation with N stage, T stage, and overall stage, while CSPG4 negatively correlated with M stage. NCAM1 positively correlated with stage and N stage (Fig. 9a). Survival rates differed significantly between high and low expression levels of key genes ( Figure S10-12 ). Single-gene GSEA enrichment analyses highlighted pathways such as focal adhesion, ribosome, and ECM receptor interaction (Fig. 9b-d). AHNAK, CSPG4, and NCAM1 were enriched in 35, 59, and 31 pathways, respectively, related to olfactory transduction, NKCC, and CAMs ( Table S2-4 ). Maltotriose and other compounds showed potential as therapeutic agents, with excellent docking interactions of -6.1 kcal/mol for AHNAK, -8.5 kcal/mol for CSPG4, and − 6 kcal/mol for NCAM1 (Fig. 9e-f; Table S5 ). Also, three key genes varied in most cancers (Fig. 9g; Figure S13 ). Key genes were further analyzed through NCBI and HPA database, revealing high expression in skin, colon, and cerebral cortex ( Table S6; Fig. 9h ). At the protein level, the three key genes did not differ between cancer and normal samples (Fig. 9i). 3.10 Expression of Key Genes at the Single-cell Level Following quality control, single-cell sequencing data included 130,431 cells and 36,137 genes ( Figure S14a, b ). From this, 2,000 highly variable genes were selected, and 32 cell clusters were identified via PCA ( Figure S14c-e ). Nine cell types were annotated: CD4 + T cells, endothelial cells, fibroblasts, myeloid cells, smooth muscle cells, epithelial cells, B cells, CD8 + natural killer cells, and mast cells (Fig. 10; Figure S14f ). The distribution of all nine cell types was presented in samples (Fig. 11a). AHNAK was broadly expressed, CSPG4 was mainly in fibroblasts and smooth muscle cells, and NCAM1 showed minimal expression (Fig. 11b). Thus fibroblasts were used as key cells (Fig. 11c-d). Key cell types focused on the cytoskeleton in muscle cells, the focal adhesion, the PI3K-Akt signaling pathway, and protein processing in the endoplasmic reticulum (Fig. 11e). Significant differences in intercellular communication were observed between normal and tumor cells ( Supplementary Fig. 15 ). In light of these findings and the available literature, we identified fibroblasts as the key cells for further investigation. Fibroblasts were selected for further analysis and reclustered into iCAFs, matCAFs, myCAFs, tCAFs, and vCAFs (Fig. 11f-g). AHNAK was more expressed across fibroblast subtypes (Fig. 11h). Pseudo-time trajectory analysis showed AHNAK expression decreased, CSPG4 varied, and NCAM1 increased then decreased (Fig. 11i-j). Transcription factors and binding motifs in key cell subtypes were identified (Fig. 11k). Furthermore, causal relationships between transcription factors and key genes were inferred and demonstrated by means of directed networks (Fig. 12). BLCA typically originates from epithelial cells. To further investigate malignancy, epithelial cells were categorized into high, intermediate and low malignancy groups based on copy number variation scores of cell subclusters (Fig. 13a-d). Analysis of differentiation trajectories showed that AHNAK expression initially decreased and then increased over time, whereas CSPG4 remained stable and NCAM1 decreased to a point and then remained constant ( Figure S13e-f ). Notably, AHNAK gene expression varied across cell subpo- pulations with different levels of malignancy, whereas CSPG4 and NCAM1 showed no changes only in the low and moderate malignancy groups ( Figure S13g ). 4. Discussion BLCA, recognized as one of the most prevalent malignancies of the urinary tract, predominantly affects individuals over the age of 55 and shows a higher incidence in men [ 29 ] . As our understanding of the pathogenesis of BLCA has expanded, numerous studies have provided a more refined molecular typing based on genes and protein expression and established corresponding prognostic models [ 30 – 32 ] . However, tumorigenesis remains a multifaceted process influenced by various biological programs, so we seek to provide more comprehensive and accurate molecular subtyping for BLCA patients according to the combination of WES and intratumoral microbiome data on this basis. In the present study, three key genes, 90 key metabolites, and two key microbiota were identified by DEGs, SMGs, and DEMs, and the gene-metabolite-microbe regulatory network was constructed accordingly. Meanwhile, the BLCA samples were categorized into 2 subgroups depending on the key genes, which differed significantly in immunity, gene mutation, drug sensitivity, and clinicopathological features. The 3 key genes we selected including AHNAK, CSPG4, and NCAM1, were significantly under-expressed in BLCA patients, with their expression levels negatively correlating with overall survival, but their roles in BLCA have rarely been reported in previous studies [ 33 , 34 ] . Ankyrin Repeat Domain-Containing Protein 1(AHNAK), a cytoskeletal protein, plays a crucial role in calcium homeostasis, muscle formation, and various biological processes like cell proliferation and signaling [ 35 ] . Also, our KEGG enrichment indicated that the association between AHNAK and cell adhesion and the cytoskeleton in BLCA. In BLCA, AHNAK mainly behaves as an inhibitory oncogene, with significantly down-regulated expression in tumor patients, and the level of AHNAK in urine can be used as a biomarker to distinguish between uroepithelial carcinoma and normal uroepithelial cells [ 36 ] , which is in keeping with our results. Previous studies concluded that AHNAK overexpression contributed to DNA damage repair (DDR) and cisplatin resistance, with NAT10 facilitating this process by recruiting LIG4 and XRCC4[34]. Chondroitin sulfate proteoglycan 4(CSPG4), a transmembrane protein, is aberrantly expressed in a variety of tumors and has been involved in tumor invasion, lymphovascular infiltration [ 37 ] , and underexpressed in BLCA, impacting EMT, the immune microenvironment, and energy metabolism, and correlating negatively with patient prognosis [ 38 ] . We also obtained similar conclusions, hinting at a possible engagement in cell migration and invasive processes in BLCA and a possible role in regulating key metabolites and microbes. Neural Cell Adhesion Molecule 1(NCAM1), belongs to the immunoglobulin superfamily of adhesion molecules and correlates with neurogenesis, neuronal synapse growth, proliferation, and cell migration, as described in several tumor entities such as gliomas, ovarian carcinomas, and small-cell lung carcinomas [ 39 ] . It has also been characterized as a potential prognostic and targeted therapeutic cell-cycle related marker in BLCA [ 33 ] . Here, we uncovered a significant correlation between NCAM1 and amino acid metabolism and immune infiltration and hypothesized that NCAM1 may modulate these processes to influence patient prognosis. We synthesized key genes, metabolites, and microbes to construct an interaction network, revealing that key genes were all negatively correlated with metabolites but did not appear to have a strong correlation with microbes. This suggested they may drive BLCA progression via metabolic pathways, while microbes were likely not gene-regulated. Consistent with previous studies [ 6 ] , we elaborated key metabolites enriched in amino acid and nucleotide metabolic pathways, like purine, arginine, glycerophospholipid, and pyrimidine metabolism based on KEGG results. High extracellular adenosine deaminase activity in bladder cancer cell lines, as shown by Hesse et al. [ 40 ] , interfered with antitumor immune responses by altering purine metabolism and impacting adenosine levels. Arginine metabolism also exercises anti-tumor immunity mainly through influencing components of the tumor microenvironment such as macrophages and T cells [ 41 ] . The pyrimidine metabolism- associated enzyme UPP1 is a critical enzyme for the maintenance of uridine homeostasis, promoted BLCA cell proliferation and gemcitabine resistance via the activation of the AKT signaling pathway [ 42 ] . Meanwhile, the functional enrichment analysis of differential microbes demonstrated connections to arginine and pyrimidine metabolism pathways. The key microbes, Sphingomonas koreensis, and Rhodospirillaceae, are environmental bacteria with potential roles in health and disease. Sphingomonas koreensis was first reported as a meningitis patient pathogen in 2015 [ 43 ] . Reduced levels of Rhodospirillaceae correlate with high pancreatic cancer metastatic potential, though its exact mechanisms in cancer progression remain unclear, potentially involving host immune system interactions and bacterial metabolite production [ 44 ] . We categorized BLCA patients into two clusters with respect to the 3 key genes and analyzed the functional differences between them. This analysis indicated that the differential pathways enriched in the subtypes primarily involved the extracellular matrix (ECM), cytoskeleton, adhesion mechanisms, and several inflammatory and tumor-related biological pathways, aligning with the functional enrichment results of the key genes. The ECM, a critical component of the tumor microenvironment, has an altered collagen composition linked to epithelial-mesenchymal transition (EMT), facilitating cell adhesion, migration, and BLCA progression [ 28 ] . Additionally, our findings highlighted a positive correlation between BLCA subtypes and a broad range of microbes. Earlier efforts have documented microbial interactions with the ECM in the tumor microenvironment, where specific bacterial strains disrupt tight junctions within ECM components, enabling tissue colonization and inflammation, thus promoting ECM remodeling and generating reactive oxygen species that can lead to DNA damage and cancer recurrence [ 45 ] . Consequently, the BLCA subtypes identified may influence tumor progression via microbial interactions with the ECM. Our GSEA outcomes revealed clustering around common tumor-related pathways such as EMT, interferon gamma response, TNFA signaling via NFkB, IL6 JAK STAT3 signaling, and the p53 pathway, all of which are crucial in BLCA growth, progression, DNA repair, and response to immunotherapy [ 46 – 48 ] . Accordingly, we examined immune infiltration and mutation profiles across subtypes, observing notable differences in immune microenvironments and oncogenic mutation patterns. Cluster 2 displayed higher immune infiltration and checkpoint expression levels while exhibiting reduced responsiveness to chemotherapeutic agents like FGFR and EGFR inhibitors. Tumor cell growth and development are largely determined by their interactions with the surrounding microenvironment; a high tumor mutational burden (TMB) leads to the production of aberrant proteins recognizable by immune cells, eliciting an effective anti-tumor immune response [ 49 ] . Concurrently, immune cell infiltration modulates this response, correlating directly with tumor progression and therapeutic outcomes [ 50 ] . In single cell analysis, we detected that key genes were overexpressed in CAF, speculating that CAF was the key cell type for the occurrence of BLCA. CAF could moderate tumor proliferation and drug resistance by promoting EMT, ECM remodeling, angiogenesis, and inhibiting anti-tumor immunity [ 51 ] . This is in some congruence with our pathway enrichment results. In particular, iCAF was associated with cell adhesion and intercellular tight junction regulation, potentially enhancing tumor progression and metastasis. Chen et al. also noticed interactions between iCAFs and tumor cells, with secreted cytokines inducing an inflammatory tumor microenvironment with pro-tumorigenic properties, which could reduce the sensitivity of BLCA patients to chemotherapeutic agents and strengthen the tumor invasive and metastatic properties [ 52 ] . Secondly, epithelial cells, as the predominant cell type in BLCA tissues, can exhibit a malignant phenotype to a certain extent [ 53 ] . A subset of epithelial cells expressing N-calmodulin 2 (CDH12) showed specific invasive traits, such as chemoresistance and poor prognosis [ 54 ] . Therefore, we also graded the degree of malignancy of the epithelial cells, and the expression level of AHNAK showed a negative correlation, which may be related to the property that AHNAK itself is a cytoskeletal protein. To summarize, we constructed a network of interactions among genes, metabolites, and microbes by combinatorial analysis of multi-omics data, including transcriptomics, whole-exome, metabolomics, and intratumoral microbiomics, which breaks the limitation of single-omics and contributes to a more profound understanding of the molecular mechanism of BLCA. The subtype analysis of BLCA based on key genes illustrated the discrepancies in biological pathways, immune microenvironment, genomic variants, immunotherapy, and drug sensitivity across subtypes. Nevertheless, there are some limitations that restrict our study. One is that the functions and regulatory mechanisms of key genes were mainly grounded in functional enrichment, which requires more experiments for further validation. The other is that the efficacy of the molecular subtypes we identified in real-world BLCA patients needs to be confirmed by more samples. 5. Conclusion This study successfully employed an integrated multi-omics approach to identify significant correlations between three pivotal genes—AHNAK, CSPG4, and NCAM1—and key metabolites and microorganisms in bladder cancer. The findings offer novel insights into the molecular mechanisms of BLCA and provide a comprehensive reference for developing targeted therapeutic strategies. These biomarkers and their interplay could enhance diagnostic accuracy, personalize treatments, and deepen our understanding of bladder cancer, potentially improving patient outcomes. Declarations Acknowledgements We would like to thank the researchers and staff of the above software and databases. Author contributions H.W. and M.D. : Supervision, Project administration, Funding acquisition. Z.T. : Writing original draft, Investigation, Funding acquisition. X.C. : Bioinformatics analysis. Y.H. and S.F. : Interpretation of the data. H.L. and C.G. : Experimental design and execution. D.L. and C.Y. : Literature review. J.W. : Manuscript revision. All authors have read and approved the final manuscript. Funding This research is supported by the National Natural Science Foundation of China (grant No. 82060464, grant No. 82260609, grant No.82360603). Yunnan Fundamental Research Projects (grant No. 202001AY070001-163, grant No. 202201AU070220, grant No. 202201AY070001-113, grant No. 202401AU070010). Yunnan Provincial Department of Education Project (grant No. 2024J0225). Data availability Derived data supporting the methodology of this study are available from the corresponding author upon reasonable request. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Human Ethics and Consent to Participate declarations Human Ethics Approval All procedures involving human participants were conducted in accordance with the ethical standards of the Ethical Committee of The Second Affiliated Hospital of Kunming Medical University and with the 1964 Helsinki Declaration and its later amendments. The study was approved by the Ethical Committee of The Second Affiliated Hospital of Kunming Medical University (FEY-BG-39-2.0). Consent to Participate Informed consent was obtained from all individual participants included in the study. References Dyrskjøt L, Hansel DE, Efstathiou JA, Knowles MA, Galsky MD, Teoh J, Theodorescu D. Bladder cancer. Nat Rev Dis Primers. 2023 Oct 26;9(1):58. Groeneveld CS, Sanchez-Quiles V, Dufour F, Shi M, Dingli F, Nicolle R, Chapeaublanc E, Poullet P, Jeffery D, Krucker C, Maillé P, Vacherot F, Vordos D, Benhamou S, Lebret T, Micheau O, Zinovyev A, Loew D, Allory Y, de Reyniès A, Bernard-Pierrot I, Radvanyi F. Proteogenomic Characterization of Bladder Cancer Reveals Sensitivity to Apoptosis Induced by Tumor Necrosis Factor-related Apoptosis-inducing Ligand in FGFR3-mutated Tumors. Eur Urol. 2024 May;85(5):483-494. Li Q, Damish AW, Frazier Z, Liu D, Reznichenko E, Kamburov A, Bell A, Zhao H, Jordan EJ, Gao SP, Ma J, Abbosh PH, Bellmunt J, Plimack ER, Lazaro JB, Solit DB, Bajorin D, Rosenberg JE, D'Andrea AD, Riaz N, Van Allen EM, Iyer G, Mouw KW. ERCC2 Helicase Domain Mutations Confer Nucleotide Excision Repair Deficiency and Drive Cisplatin Sensitivity in Muscle-Invasive Bladder Cancer. Clin Cancer Res. 2019 Feb 1;25(3):977-988. Wu G, Weng W, Xia P, Yan S, Zhong C, Xie L, Xie Y, Fan G. Wnt signalling pathway in bladder cancer. Cell Signal. 2021 Mar;79:109886. Komura K, Hirosuna K, Tokushige S, Tsujino T, Nishimura K, Ishida M, Hayashi T, Ura A, Ohno T, Yamazaki S, Nakamori K, Kinoshita S, Maenosono R, Ajiro M, Yoshikawa Y, Takai T, Tsutsumi T, Taniguchi K, Tanaka T, Takahara K, Konuma T, Inamoto T, Hirose Y, Ono F, Shiraishi Y, Yoshimi A, Azuma H. The Impact of FGFR3 Alterations on the Tumor Microenvironment and the Efficacy of Immune Checkpoint Inhibitors in Bladder Cancer. Mol Cancer. 2023 Nov 18;22(1):185. Vantaku V, Donepudi SR, Piyarathna DWB, Amara CS, Ambati CR, Tang W, Putluri V, Chandrashekar DS, Varambally S, Terris MK, Davies K, Ambs S, Bollag R, Apolo AB, Sreekumar A, Putluri N. Large-scale profiling of serum metabolites in African American and European American patients with bladder cancer reveals metabolic pathways associated with patient survival. Cancer. 2019 Mar 15;125(6):921-932. Meeks JJ, Al-Ahmadie H, Faltas BM, Taylor JA 3rd, Flaig TW, DeGraff DJ, Christensen E, Woolbright BL, McConkey DJ, Dyrskjøt L. Genomic heterogeneity in bladder cancer: challenges and possible solutions to improve outcomes. Nat Rev Urol. 2020 May;17(5):259-270. Tan Z, Chen X, Zuo J, Fu S, Wang H, Wang J. Comprehensive analysis of scRNA-Seq and bulk RNA-Seq reveals dynamic changes in the tumor immune microenvironment of bladder cancer and establishes a prognostic model. J Transl Med. 2023 Mar 27;21(1):223. Liang Y, Tan Y, Guan B, Guo B, Xia M, Li J, Shi Y, Yu Z, Zhang Q, Liu D, Yang X, Hao J, Gong Y, Shakeel M, Zhou L, Ci W, Li X. Single-cell atlases link macrophages and CD8 + T-cell subpopulations to disease progression and immunotherapy response in urothelial carcinoma. Theranostics. 2022 Nov 14;12(18):7745-7759. Li J, Kong Z, Qi Y, Wang W, Su Q, Huang W, Zhang Z, Li S, Du E. Single-cell and bulk RNA-sequence identified fibroblasts signature and CD8 + T-cell - fibroblast subtype predicting prognosis and immune therapeutic response of bladder cancer, based on machine learning: bioinformatics multi-omics study. Int J Surg. 2024 Aug 1;110(8):4911-4931. Lawson ARJ, Abascal F, Coorens THH, Hooks Y, O'Neill L, Latimer C, Raine K, Sanders MA, Warren AY, Mahbubani KTA, Bareham B, Butler TM, Harvey LMR, Cagan A, Menzies A, Moore L, Colquhoun AJ, Turner W, Thomas B, Gnanapragasam V, Williams N, Rassl DM, Vöhringer H, Zumalave S, Nangalia J, Tubío JMC, Gerstung M, Saeb-Parsy K, Stratton MR, Campbell PJ, Mitchell TJ, Martincorena I. Extensive heterogeneity in somatic mutation and selection in the human bladder. Science. 2020 Oct 2;370(6512):75-82. Lawson ARJ, Abascal F, Coorens THH, Hooks Y, O'Neill L, Latimer C, Raine K, Sanders MA, Warren AY, Mahbubani KTA, Bareham B, Butler TM, Harvey LMR, Cagan A, Menzies A, Moore L, Colquhoun AJ, Turner W, Thomas B, Gnanapragasam V, Williams N, Rassl DM, Vöhringer H, Zumalave S, Nangalia J, Tubío JMC, Gerstung M, Saeb-Parsy K, Stratton MR, Campbell PJ, Mitchell TJ, Martincorena I. Extensive heterogeneity in somatic mutation and selection in the human bladder. Science. 2020 Oct 2;370(6512):75-82. Markowski MC, Boorjian SA, Burton JP, Hahn NM, Ingersoll MA, Maleki Vareki S, Pal SK, Sfanos KS. The Microbiome and Genitourinary Cancer: A Collaborative Review. Eur Urol. 2019 Apr;75(4):637-646. Sun JX, Xia QD, Zhong XY, Liu Z, Wang SG. The bladder microbiome of NMIBC and MIBC patients revealed by 2bRAD-M. Front Cell Infect Microbiol. 2023 Jun 7;13:1182322. Wu P, Zhang G, Zhao J, Chen J, Chen Y, Huang W, Zhong J, Zeng J. Profiling the Urinary Microbiota in Male Patients With Bladder Cancer in China. Front Cell Infect Microbiol. 2018 May 31;8:167. Erratum in: Front Cell Infect Microbiol. 2018 Dec 11;8:429. Liu J, Wang J, Wu M, Zhang W, Meng L, Wang J, Lv Z, Xia H, Zhang Y, Wang J. Comprehensive Analysis of N6-methyladenosine Modification Patterns Associated With Multiomic Characteristics of Bladder Cancer. Front Med (Lausanne). 2021 Dec 23;8:757432. Thévenot EA, Roux A, Xu Y, Ezan E, Junot C. Analysis of the Human Adult Urinary Metabolome Variations with Age, Body Mass Index, and Gender by Implementing a Comprehensive Workflow for Univariate and OPLS Statistical Analyses. J Proteome Res. 2015 Aug 7;14(8):3322-35. Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15(12):550. Wu T, Hu E, Xu S, Chen M, Guo P, Dai Z, Feng T, Zhou L, Tang W, Zhan L, Fu X, Liu S, Bo X, Yu G. clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. Innovation (Camb). 2021 Jul 1;2(3):100141. Mayakonda A, Lin DC, Assenov Y, Plass C, Koeffler HP. Maftools: efficient and comprehensive analysis of somatic variants in cancer. Genome Res. 2018 Nov;28(11):1747-1756. Liu TT, Li R, Huo C, Li JP, Yao J, Ji XL, Qu YQ. Identification of CDK2-Related Immune Forecast Model and ceRNA in Lung Adenocarcinoma, a Pan-Cancer Analysis. Front Cell Dev Biol. 2021 Jul 30;9:682002. Chen D, Cao Y, Tang H, Zang L, Yao N, Zhu Y, Jiang Y, Zhai S, Liu Y, Shi M, Zhao S, Wang W, Wen C, Peng C, Chen H, Deng X, Jiang L, Shen B. Comprehensive machine learning-generated classifier identifies pro-metastatic characteristics and predicts individual treatment in pancreatic cancer: A multicenter cohort study based on super-enhancer profiling. Theranostics. 2023 May 27;13(10):3290-3309. Hu FF, Liu CJ, Liu LL, Zhang Q, Guo AY. Expression profile of immune checkpoint genes and their roles in predicting immunotherapy response. Brief Bioinform. 2021 May 20;22(3):bbaa176. Chhabra Y, Weeraratna AT. Fibroblasts in cancer: Unity in heterogeneity. Cell. 2023 Apr 13;186(8):1580-1609. Elyada E, Bolisetty M, Laise P, Flynn WF, Courtois ET, Burkhart RA, Teinor JA, Belleau P, Biffi G, Lucito MS, Sivajothi S, Armstrong TD, Engle DD, Yu KH, Hao Y, Wolfgang CL, Park Y, Preall J, Jaffee EM, Califano A, Robson P, Tuveson DA. Cross-Species Single-Cell Analysis of Pancreatic Ductal Adenocarcinoma Reveals Antigen-Presenting Cancer-Associated Fibroblasts. Cancer Discov. 2019 Aug;9(8):1102-1123. Ma Z, Li X, Mao Y, Wei C, Huang Z, Li G, Yin J, Liang X, Liu Z. Interferon-dependent SLC14A1 + cancer-associated fibroblasts promote cancer stemness via WNT5A in bladder cancer. Cancer Cell. 2022 Dec 12;40(12):1550-1565.e7. Altman DG, McShane LM, Sauerbrei W, Taube SE. Reporting recommendations for tumor marker prognostic studies (REMARK): explanation and elaboration. BMC Med. 2012 May 29;10:51. Wang J, Zhu J, Hu J, Wang Z, Wang X, Pan J, Chu Y, Li Z, Jiang W, Liang C, Hou J, Guo J, Dang Y, Jiang S. A novel in vitro prognostic model of bladder cancer based on urine-derived living tumor cells. Genes Dis. 2022 Nov 26;10(6):2586-2596. Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, Bray F. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin. 2021 May;71(3):209-249. Huang X, Du G, Yang Y, Su P, Chen S, Cai C, Huang T, Zeng Y, Tao Y, Tian D, Zhang N. Advancing bladder cancer management: development of a prognostic model and personalized therapy. Front Immunol. 2024 Jul 22;15:1430792. Safder I, Valentine H, Uzzo N, Sfakianos J, Uzzo R, Gupta S, Brown J, Ranti D, Plimack E, Haber G, Weight C, Kutikov A, Abbosh P, Bukavina L. Identification and Validation of Prognostic Model for Tumor Microenvironment-Associated Genes in Bladder Cancer Based on Single-Cell RNA Sequencing Data Sets. JCO Precis Oncol. 2024 Aug;8:e2300661. Feng ZH, Liang YP, Cen JJ, Yao HH, Lin HS, Li JY, Liang H, Wang Z, Deng Q, Cao JZ, Huang Y, Wei JH, Luo JH, Chen W, Chen ZH. m6A-immune-related lncRNA prognostic signature for predicting immune landscape and prognosis of bladder cancer. J Transl Med. 2022 Oct 29;20(1):492. Zeng X, Su H, Liu Z, Wang Y, Lu Z, Cheng S. Integrative analysis of the bladder cancer from a cell cycle NCAM1 perspective at both single cell and bulk resolution. Environ Toxicol. 2024 Apr 6. Xie R, Cheng L, Huang M, Huang L, Chen Z, Zhang Q, Li H, Lu J, Wang H, Zhou Q, Huang J, Chen X, Lin T. NAT10 Drives Cisplatin Chemoresistance by Enhancing ac4C-Associated DNA Repair in Bladder Cancer. Cancer Res. 2023 May 15;83(10):1666-1683. Lee JM, Lim TY, Oh SB, Lee SJ, Bae YS, Jung HS. Ahnak is required to balance calcium ion homeostasis and smooth muscle development in the urinary system. Cell Biosci. 2023 Jun 12;13(1):108. Lee H, Kim K, Woo J, Park J, Kim H, Lee KE, Kim H, Kim Y, Moon KC, Kim JY, Park IA, Shim BB, Moon JH, Han D, Ryu HS. Quantitative Proteomic Analysis Identifies AHNAK (Neuroblast Differentiation-associated Protein AHNAK) as a Novel Candidate Biomarker for Bladder Urothelial Carcinoma Diagnosis by Liquid-based Cytology. Mol Cell Proteomics. 2018 Sep;17(9):1788-1802. Egan CE, Stefanova D, Ahmed A, Raja VJ, Thiesmeyer JW, Chen KJ, Greenberg JA, Zhang T, He B, Finnerty BM, Zarnegar R, Jin MM, Scognamiglio T, Dephoure N, Fahey T 3rd, Min IM. CSPG4 Is a Potential Therapeutic Target in Anaplastic Thyroid Cancer. Thyroid. 2021 Oct;31(10):1481-1493. Qiu Y, Ye W, Wang C, Zang J. Prognostic significance and immunoinfiltration analysis of genes associated with epithelial-mesenchymal transition and energy metabolism in bladder urothelial carcinoma. Aging (Albany NY). 2023 Nov 24;15(22):13312-13328. Sasca D, Szybinski J, Schüler A, Shah V, Heidelberger J, Haehnel PS, Dolnik A, Kriege O, Fehr EM, Gebhardt WH, Reid G, Scholl C, Theobald M, Bullinger L, Beli P, Kindler T. NCAM1 (CD56) promotes leukemogenesis and confers drug resistance in AML. Blood. 2019 May 23;133(21):2305-2319. Hesse J, Steckel B, Dieterich P, Aydin S, Deussen A, Schrader J. Intercellular crosstalk shapes purinergic metabolism and signaling in cancer cells. Cell Rep. 2024 Jan 23;43(1):113643. Nowak M, Klink M. The Role of Tumor-Associated Macrophages in the Progression and Chemoresistance of Ovarian Cancer. Cells. 2020 May 22;9(5):1299. Du W, Tu S, Zhang W, Zhang Y, Liu W, Xiong K, Zhou F, Li N, Zhang R, Yu J, Li M, Xiang W, Qian K, Wang G, Xiao Y, Wang X, Ju L. UPP1 enhances bladder cancer progression and gemcitabine resistance through AKT. Int J Biol Sci. 2024 Jan 27;20(4):1389-1409. Marbjerg LH, Gaini S, Justesen US. First report of Sphingomonas koreensis as a human pathogen in a patient with meningitis. J Clin Microbiol. 2015 Mar;53(3):1028-30. Villani A, Fontana A, Panebianco C, Ferro C, Copetti M, Pavlovic R, Drago D, Fiorentini C, Terracciano F, Bazzocchi F, Canistro G, Pisati F, Maiello E, Latiano TP, Perri F, Pazienza V. A powerful machine learning approach to identify interactions of differentially abundant gut microbial subsets in patients with metastatic and non-metastatic pancreatic cancer. Gut Microbes. 2024 Jan-Dec;16(1):2375483. Alfano M, Canducci F, Nebuloni M, Clementi M, Montorsi F, Salonia A. The interplay of extracellular matrix and microbiome in urothelial bladder cancer. Nat Rev Urol. 2016 Feb;13(2):77-90. Baker SC, Mason AS, Slip RG, Eriksson P, Sjödahl G, Trejdosiewicz LK, Southgate J. The Urothelial Transcriptomic Response to Interferon Gamma: Implications for Bladder Cancer Prognosis and Immunotherapy. Cancers (Basel). 2022 Oct 27;14(21):5295. Li L, Jiang P, Hu W, Zou F, Li M, Rao T, Ruan Y, Yu W, Ning J, Cheng F. AURKB promotes bladder cancer progression by deregulating the p53 DNA damage response pathway via MAD2L2. J Transl Med. 2024 Mar 21;22(1):295. Amara CS, Kami Reddy KR, Yuntao Y, Chan YS, Piyarathna DWB, Dobrolecki LE, Shih DJH, Shi Z, Xu J, Huang S, Ellis MJ, Apolo AB, Ballester LY, Gao J, Hansel DE, Lotan Y, Hodges HC, Lerner SP, Creighton CJ, Sreekumar A, Zheng WJ, Msaouel P, Kavuri SM, Putluri N. The IL6/JAK/STAT3 signaling axis is a therapeutic vulnerability in SMARCB1- deficient bladder cancer. Nat Commun. 2024 Feb 14;15(1):1373. van Dorp J, van der Heijden MS. The bladder cancer immune micro-environment in the context of response to immune checkpoint inhibition. Front Immunol. 2023 Aug 31;14:1235884. Low V, Li Z, Blenis J. Metabolite activation of tumorigenic signaling pathways in the tumor microenvironment. Sci Signal. 2022 Nov 8;15(759):eabj4220. Caramelo B, Zagorac S, Corral S, Marqués M, Real FX. Cancer-associated Fibroblasts in Bladder Cancer: Origin, Biology, and Therapeutic Opportunities. Eur Urol Oncol. 2023 Aug;6(4):366-375. Chen H, Yang W, Xue X, Li Y, Jin Z, Ji Z. Integrated Analysis Revealed an Inflammatory Cancer-Associated Fibroblast-Based Subtypes with Promising Implications in Predicting the Prognosis and Immunotherapeutic Response of Bladder Cancer Patients. Int J Mol Sci. 2022 Dec 15;23(24):15970. Wang J, Zuo Z, Yu Z, Chen Z, Tran LJ, Zhang J, Ao J, Ye F, Sun Z. Collaborating single-cell and bulk RNA sequencing for comprehensive characterization of the intratumor heterogeneity and prognostic model development for bladder cancer. Aging (Albany NY). 2023 Nov 6;15(21):12104-12119. Gouin KH 3rd, Ing N, Plummer JT, Rosser CJ, Ben Cheikh B, Oh C, Chen SS, Chan KS, Furuya H, Tourtellotte WG, Knott SRV, Theodorescu D. An N-Cadherin 2 expressing epithelial cell subpopulation predicts response to surgery, chemotherapy and immunotherapy in bladder cancer. Nat Commun. 2021 Aug 12;12(1):4906. Additional Declarations No competing interests reported. Supplementary Files SupplementaryTables.xlsx FigureLegend.docx FigureS1.tif FigureS2.tif FigureS3.tif FigureS4.tif FigureS5.tif FigureS6.tif FigureS7.tif FigureS8.tif FigureS9.tif FigureS10.tif FigureS11.tif FigureS12.tif FigureS13.tif FigureS14.tif FigureS15.tif FigureS16.tif Cite Share Download PDF Status: Posted Version 1 posted 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-5898970","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":406907702,"identity":"01006ed4-417c-4406-a4da-b7add925fe85","order_by":0,"name":"Zhiyong Tan","email":"","orcid":"","institution":"The Second Affiliated Hospital of Kunming Medical University","correspondingAuthor":false,"prefix":"","firstName":"Zhiyong","middleName":"","lastName":"Tan","suffix":""},{"id":406907703,"identity":"048183b7-d2ed-4992-af90-9265752dc771","order_by":1,"name":"Xiaorong Chen","email":"","orcid":"","institution":"The Third Hospital of Sun Yat-Sen University","correspondingAuthor":false,"prefix":"","firstName":"Xiaorong","middleName":"","lastName":"Chen","suffix":""},{"id":406907706,"identity":"2e64fc8e-f11f-4ede-b1de-d6b613f9a78c","order_by":2,"name":"Yinglong Huang","email":"","orcid":"","institution":"The Second Affiliated Hospital of Kunming Medical University","correspondingAuthor":false,"prefix":"","firstName":"Yinglong","middleName":"","lastName":"Huang","suffix":""},{"id":406907709,"identity":"d73b93f6-b5fa-4a81-91a9-8dfd2c04a578","order_by":3,"name":"Shi Fu","email":"","orcid":"","institution":"The Second Affiliated Hospital of Kunming Medical University","correspondingAuthor":false,"prefix":"","firstName":"Shi","middleName":"","lastName":"Fu","suffix":""},{"id":406907710,"identity":"f3932eb5-575e-4d9d-b395-095812998595","order_by":4,"name":"Chen Gong","email":"","orcid":"","institution":"The Second Affiliated Hospital of Kunming Medical University","correspondingAuthor":false,"prefix":"","firstName":"Chen","middleName":"","lastName":"Gong","suffix":""},{"id":406907711,"identity":"4489740a-81d6-485c-98b0-c84c75c85128","order_by":5,"name":"Dihao Lv","email":"","orcid":"","institution":"The Second Affiliated Hospital of Kunming Medical University","correspondingAuthor":false,"prefix":"","firstName":"Dihao","middleName":"","lastName":"Lv","suffix":""},{"id":406907713,"identity":"75d8dc2c-b447-4242-9025-5e332e8e14ea","order_by":6,"name":"Chadanfeng Yang","email":"","orcid":"","institution":"The Second Affiliated Hospital of Kunming Medical University","correspondingAuthor":false,"prefix":"","firstName":"Chadanfeng","middleName":"","lastName":"Yang","suffix":""},{"id":406907714,"identity":"0b78d0f0-9c7d-46c2-afe1-5105e7048d80","order_by":7,"name":"Haihao Li","email":"","orcid":"","institution":"The Second Affiliated Hospital of Kunming Medical University","correspondingAuthor":false,"prefix":"","firstName":"Haihao","middleName":"","lastName":"Li","suffix":""},{"id":406907718,"identity":"9d35a718-0fbd-47af-816d-eafd0fe81519","order_by":8,"name":"Jiansong Wang","email":"","orcid":"","institution":"The Second Affiliated Hospital of Kunming Medical University","correspondingAuthor":false,"prefix":"","firstName":"Jiansong","middleName":"","lastName":"Wang","suffix":""},{"id":406907719,"identity":"b40414ac-0d3c-4548-bc54-cad8ed52ecda","order_by":9,"name":"Mingxia Ding","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA80lEQVRIiWNgGAWjYDCCwyDCAIiZGRgfNkDEDIjWwmxInJYDCCabJFFa+I7zHn51o+CO3YbjzM8qZ+ZsS2xgb94mwVBzB6cWycN8adY5Bs+SZzazmd3cuO12YgPPsTIJhmPPcGoxOMxjZpxjcDiZn5nB7OZDkBaJHDMJxobDhLWwMbN/KwRrkX9DUIvxY6AWO35mHjNGsMMkePBrkQTawgzUkiDZzFMsOXPbbeM2nrRii4RjuLXwnT9j/Dnnz2F7g/PHN37s3XZbtp/98MYbH2pwawECNgkgkdgA54KIBHwagNH+AUjY41czCkbBKBgFIxoAAJnXWH5rjiFbAAAAAElFTkSuQmCC","orcid":"","institution":"The Second Affiliated Hospital of Kunming Medical University","correspondingAuthor":true,"prefix":"","firstName":"Mingxia","middleName":"","lastName":"Ding","suffix":""},{"id":406907720,"identity":"6eeaf23a-59d5-4fa3-a9d5-fb298f291e36","order_by":10,"name":"Haifeng Wang","email":"","orcid":"","institution":"The Second Affiliated Hospital of Kunming Medical University","correspondingAuthor":false,"prefix":"","firstName":"Haifeng","middleName":"","lastName":"Wang","suffix":""}],"badges":[],"createdAt":"2025-01-25 03:08:16","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5898970/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5898970/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":74960763,"identity":"11dd6263-f909-4d11-a33d-08063471e1b1","added_by":"auto","created_at":"2025-01-28 18:52:41","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1591019,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eIdentification of DEGs, SMGs and DEMs.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea\u003c/strong\u003eVolcano map of DEGs. \u003cstrong\u003eb\u003c/strong\u003e Heat map of DEGs. \u003cstrong\u003ec\u003c/strong\u003e GO enrichment analysis showing the CC, BP, and MF involved in DEGs. \u003cstrong\u003ed\u003c/strong\u003e KEGG enrichment analysis showing the signaling pathways involved in DEGs. \u003cstrong\u003ee\u003c/strong\u003e Pie chart showing the distribution of seven mutation types as well as mutation frequencies in 78 SMGs. \u003cstrong\u003ef\u003c/strong\u003e GO enrichment analysis showing the CC, BP, and MF involved in SMGs. \u003cstrong\u003eg\u003c/strong\u003e KEGG enrichment analysis showing the signaling pathways involved in SMGs. \u003cstrong\u003eh\u003c/strong\u003e Levels of KEGG pathway activation in SMGs analyzed by OncoboxPD. \u003cstrong\u003ei\u003c/strong\u003e Volcano map of DEMs. \u003cstrong\u003ej\u003c/strong\u003e Heat map of DEMs. \u003cstrong\u003ek\u003c/strong\u003eKEGG enrichment analysis showing the signaling pathways involved in DEMs.\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-5898970/v1/11fec40f70ee6e9871c0a55a.png"},{"id":74961673,"identity":"3dfc825b-d180-440c-819f-581000b1502e","added_by":"auto","created_at":"2025-01-28 19:00:46","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":601900,"visible":true,"origin":"","legend":"","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-5898970/v1/5dcd788ea7724a2ecafd9519.png"},{"id":74960771,"identity":"72869df3-c663-4b00-be69-cf18660833d9","added_by":"auto","created_at":"2025-01-28 18:52:44","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":380900,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSomatic mutation analysis. a \u003c/strong\u003eSomatic mutations in BLCA patients in the TCGA-BLCA dataset and WES sequencing data. \u003cstrong\u003eb\u003c/strong\u003e Mutation frequency and type in BLCA patients in WES sequencing data (upper) and TCGA-BLCA (lower). \u003cstrong\u003ec\u003c/strong\u003eAmplification (upper) and deletion (lower) of CNVs.\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-5898970/v1/5f88825960f939969d1fdff5.png"},{"id":74960765,"identity":"e5bd7287-fcc5-43d7-b286-fc24523790ec","added_by":"auto","created_at":"2025-01-28 18:52:43","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":394681,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eIdentification and function analysis of candidate genes.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea \u003c/strong\u003eIdentification of candidate genes.\u003cstrong\u003e b \u003c/strong\u003eGO enrichment analysis showing the CC, BP, and MF involved in candidate genes. \u003cstrong\u003ec\u003c/strong\u003e KEGG enrichment analysis showing the signaling pathways involved in candidate genes. \u003cstrong\u003ed\u003c/strong\u003e Expression analysis of seven differentially expressed candidate genes with prognostic value between groups in transcriptome sequencing data, TCGA-BLCA, and GSE13507. \u003cstrong\u003ee\u003c/strong\u003e ROC curve for the seven genes screened for five biomarkers with AUCs \u0026gt; 0.7.\u003c/p\u003e","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-5898970/v1/9d5613521f24430856385b3f.png"},{"id":74960807,"identity":"b8d701aa-9f60-4332-937e-4fca5b1a5416","added_by":"auto","created_at":"2025-01-28 18:52:45","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":747505,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eIdentification of key genes, metabolites and microbiomes. a \u003c/strong\u003eThree key genes from key gene 1 and key gene 2. \u003cstrong\u003eb\u003c/strong\u003e A total of 90 key metabolites from key metabolite 1 and key metabolite 2. \u003cstrong\u003ec\u003c/strong\u003e Two key microbiomes from key microbiome 1 and key microbiome 2. \u003cstrong\u003ed\u003c/strong\u003e Heatmap showing relevance of key genes, metabolites, and key microbiomes. \u003cstrong\u003ee\u003c/strong\u003e Interactive network diagram of key genes, metabolites, and key microbiomes.\u003c/p\u003e","description":"","filename":"Figure5.png","url":"https://assets-eu.researchsquare.com/files/rs-5898970/v1/236793970bff8ee02c9747db.png"},{"id":74960815,"identity":"9287e1a1-c1d8-44ea-a6d4-d4005cfa9311","added_by":"auto","created_at":"2025-01-28 18:52:46","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":1721345,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCluster analysis.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea, b \u003c/strong\u003eConsensus clustering to two BLCA clusters.\u003cstrong\u003e \u003c/strong\u003ek =2 (\u003cstrong\u003ea\u003c/strong\u003e) and cutoff = 0.44 (\u003cstrong\u003eb\u003c/strong\u003e).\u003cstrong\u003ec\u003c/strong\u003e Volcano map of inter-cluster DEGs. \u003cstrong\u003ed\u003c/strong\u003e Heat map of inter-cluster DEGs. \u003cstrong\u003ee\u003c/strong\u003e GO enrichment analysis showing the CC, BP, and MF involved in inter-cluster DEGs. \u003cstrong\u003ef\u003c/strong\u003e KEGG enrichment analysis showing the signaling pathways involved in inter-cluster DEGs. \u003cstrong\u003eg\u003c/strong\u003e Association networks between microbial communities in different clusters. \u003cstrong\u003eh\u003c/strong\u003e Volcano map of inter-cluster differential metabolites. \u003cstrong\u003ei\u003c/strong\u003e Heat map of inter-cluster differential metabolites. \u003cstrong\u003ej\u003c/strong\u003e KEGG enrichment analysis showing the signaling pathways involved in inter-cluster differential metabolites. \u003cstrong\u003ek\u003c/strong\u003eGSVA results for two clusters.\u003c/p\u003e","description":"","filename":"Figure6.png","url":"https://assets-eu.researchsquare.com/files/rs-5898970/v1/8dd4c6a40f2eecf3d80b8584.png"},{"id":74960858,"identity":"39552158-0265-48c4-8c11-ba67d1de67ac","added_by":"auto","created_at":"2025-01-28 18:52:48","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":1089897,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTumour immune microenvironment in different clusters.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea \u003c/strong\u003eDifferences of pathways in the tumour immune microenvironment between cluster 1 and cluster 2. \u003cstrong\u003eb\u003c/strong\u003e Pathway scores between cluster 1 and cluster 2. \u003cstrong\u003ec\u003c/strong\u003e Abundance of immune cells. \u003cstrong\u003ed\u003c/strong\u003e Differences in immune cell infiltration between cluster 1 and cluster 2. \u003cstrong\u003ee\u003c/strong\u003e Correlation of key genes with immune cells. \u003cstrong\u003ef\u003c/strong\u003eDifferences in the expression of immune checkpoint genes between samples of different clusters. \u003cstrong\u003eg\u003c/strong\u003e Differences in the activity of circulatory pathways of cancer immunity between clusters. \u003cstrong\u003eh\u003c/strong\u003e TIDE score of cluster 1 and cluster 2. \u003cstrong\u003ei\u003c/strong\u003e Correlations showing the relationship of key genes to stromal score, immune score and ESTIMATE score. \u003cstrong\u003ej\u003c/strong\u003e Association of immunotherapy, key genes and key metabolites. \u003cstrong\u003ek\u003c/strong\u003e Correlation network of key microorganisms with MHC molecules, immunostimulators, and immunoinhibitors.\u003c/p\u003e","description":"","filename":"Figure7.png","url":"https://assets-eu.researchsquare.com/files/rs-5898970/v1/1929f9d5af23636693dbf71a.png"},{"id":74961671,"identity":"7770960c-b843-4551-b7f0-2ea74ee52f2f","added_by":"auto","created_at":"2025-01-28 19:00:46","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":752181,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eGene mutations in clusters of BLCA.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea \u003c/strong\u003eMutational landscape of top 20 most commonly mutated genes in different clusters. \u003cstrong\u003eb\u003c/strong\u003eVariant classification, variant type, and base mutations among different clusters. \u003cstrong\u003ec\u003c/strong\u003e Mutational differences in typical oncogenic pathways between clusters. \u003cstrong\u003ed\u003c/strong\u003e Agents were available in the CTRP, GDSC, and PRISM datasets. \u003cstrong\u003ee\u003c/strong\u003e Correlation between key genes and 63 chemotherapeutic agents. \u003cstrong\u003ef\u003c/strong\u003eCorrelation analysis among key genes, targeted agents, classical therapeutic pathways, and pathways. \u003cstrong\u003eg\u003c/strong\u003e Differences in the expression of three key genes between clusters.\u003c/p\u003e","description":"","filename":"Figure8.png","url":"https://assets-eu.researchsquare.com/files/rs-5898970/v1/28075e7b2fd6721f0081173a.png"},{"id":74960826,"identity":"2625629a-bcf5-4468-a599-e6ad39f6c186","added_by":"auto","created_at":"2025-01-28 18:52:47","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":3167285,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eExpression of key genes.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea \u003c/strong\u003eCorrelation of key genes with clinical characteristics. \u003cstrong\u003eb-d\u003c/strong\u003e KEGG enrichment analysis showing the signaling pathways involved in three key genes. Three key genes were AHNAK (\u003cstrong\u003eb\u003c/strong\u003e), CSPG4 (\u003cstrong\u003ec\u003c/strong\u003e), and NCAM1 (\u003cstrong\u003ed\u003c/strong\u003e). \u003cstrong\u003ee\u003c/strong\u003eMolecular structure of maltotriose. \u003cstrong\u003ef\u003c/strong\u003e Molecular docking of key genes (AHNAK, CSPG4, and NCAM1) and maltotriose. \u003cstrong\u003eg\u003c/strong\u003e Expression of individual key genes in pan-cancer data.\u003cstrong\u003e h\u003c/strong\u003e Expression levels of key genes (AHNAK, CSPG4, and NCAM1) in different tissues. \u003cstrong\u003ei\u003c/strong\u003e Expression of key genes (AHNAK, CSPG4, and NCAM1) at the protein level in cancer and normal samples.\u003c/p\u003e","description":"","filename":"Figure9.png","url":"https://assets-eu.researchsquare.com/files/rs-5898970/v1/74b5af13018cf65464314a4d.png"},{"id":74960861,"identity":"747ff1fa-e0b4-4e9c-8b7d-7872e353442e","added_by":"auto","created_at":"2025-01-28 18:52:49","extension":"png","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":1209655,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eNine cell types at single-cell dataset.\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"Figure10.png","url":"https://assets-eu.researchsquare.com/files/rs-5898970/v1/b455eb3e869b0eba7ef17d64.png"},{"id":74960829,"identity":"224eb7ef-9939-40b8-821d-a12f96decbd5","added_by":"auto","created_at":"2025-01-28 18:52:47","extension":"png","order_by":11,"title":"Figure 11","display":"","copyAsset":false,"role":"figure","size":3122081,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSingle-cell analysis.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea \u003c/strong\u003eUMAP of cell types after annotation in all samples, cancer samples and normal samples, respectively. \u003cstrong\u003eb \u003c/strong\u003eDistribution of key genes across cell types.\u003cstrong\u003ec \u003c/strong\u003eUMAP map of fibroblasts.\u003cstrong\u003e d \u003c/strong\u003ePPI network of top 10 specific highly expressed genes in fibroblasts.\u003cstrong\u003e e \u003c/strong\u003eKEGG enrichment analysis showing the signaling pathways involved in cell types. \u003cstrong\u003ef\u003c/strong\u003e Key cells annotated to five subtypes. \u003cstrong\u003eg\u003c/strong\u003e Expression and enrichment pathways of marker genes in various subtypes of key cells. \u003cstrong\u003eh\u003c/strong\u003e Expression of key genes in each key cell subtype. \u003cstrong\u003ei\u003c/strong\u003e Pseudo-time analysis of key cells.\u003cstrong\u003e j\u003c/strong\u003e Expression of key genes in key cell differentiation trajectories in different time stages.\u003cstrong\u003ek\u003c/strong\u003e Coregulator modules and their representative transcription factors and binding motifs for each key cell subtype.\u003c/p\u003e","description":"","filename":"Figure11.png","url":"https://assets-eu.researchsquare.com/files/rs-5898970/v1/b292db957a84d446b35591ce.png"},{"id":74960877,"identity":"bcd1dcbf-309f-46d3-a3a1-1ed4e9120c54","added_by":"auto","created_at":"2025-01-28 18:52:49","extension":"png","order_by":12,"title":"Figure 12","display":"","copyAsset":false,"role":"figure","size":1247829,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDirected network diagrams between transcription factors and key genes.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea \u003c/strong\u003eiCAFs.\u003cstrong\u003e b\u003c/strong\u003e matCAFs.\u003cstrong\u003ec \u003c/strong\u003emyCAFs.\u003cstrong\u003e d \u003c/strong\u003etCAFs.\u003cstrong\u003e e \u003c/strong\u003evCAFs.\u003c/p\u003e","description":"","filename":"Figure12.png","url":"https://assets-eu.researchsquare.com/files/rs-5898970/v1/229b415c2d555b30751013f3.png"},{"id":74960848,"identity":"66449704-11fb-410e-ad48-3ef3060bebef","added_by":"auto","created_at":"2025-01-28 18:52:48","extension":"png","order_by":13,"title":"Figure 13","display":"","copyAsset":false,"role":"figure","size":1955910,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAnalysis of cellular malignancy.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea \u003c/strong\u003eUMAP for downscaling cluster analysis of epithelial cells.\u003cstrong\u003e b \u003c/strong\u003eCNV analysis of various subclusters of epithelial cells.\u003cstrong\u003e c \u003c/strong\u003eDistribution of high malignancy, middle malignancy and low malignancy. \u003cstrong\u003ed\u003c/strong\u003e CNV scores of cell subclusters. \u003cstrong\u003ee\u003c/strong\u003e Pseudo-time analysis of epithelial cells. \u003cstrong\u003ef\u003c/strong\u003eExpression of key genes during different pseudo-time and subcluster differentiation of cells with different degrees of malignancy. \u003cstrong\u003eg\u003c/strong\u003eExpression analysis of key genes in subclusters of cells with different degrees of malignancy.\u003c/p\u003e","description":"","filename":"Figure13.png","url":"https://assets-eu.researchsquare.com/files/rs-5898970/v1/ee2519a94dd0e5b7e0aaa8ba.png"},{"id":78076018,"identity":"53f59480-4374-484e-8464-bae50de71684","added_by":"auto","created_at":"2025-03-09 09:31:58","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":18425895,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5898970/v1/e2be2d64-db6b-457c-8aec-8c9d67df4244.pdf"},{"id":74960764,"identity":"611871c8-16b2-4306-bcf4-ef8ac929fca6","added_by":"auto","created_at":"2025-01-28 18:52:43","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":44342,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTables.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-5898970/v1/03b35031ae69aa4bba2e170a.xlsx"},{"id":74961674,"identity":"5cd62bc0-caa7-450b-abd8-8e919c11851d","added_by":"auto","created_at":"2025-01-28 19:00:46","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":20226,"visible":true,"origin":"","legend":"","description":"","filename":"FigureLegend.docx","url":"https://assets-eu.researchsquare.com/files/rs-5898970/v1/46bc73c70a631947d061554b.docx"},{"id":74960820,"identity":"047c0efb-deca-4872-999a-18453ac0f2ba","added_by":"auto","created_at":"2025-01-28 18:52:46","extension":"tif","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":10478904,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS1.tif","url":"https://assets-eu.researchsquare.com/files/rs-5898970/v1/83df65714a2e7284d06a172d.tif"},{"id":74961666,"identity":"237e7de2-1991-419d-b42e-4641fc6af270","added_by":"auto","created_at":"2025-01-28 19:00:44","extension":"tif","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":2316856,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS2.tif","url":"https://assets-eu.researchsquare.com/files/rs-5898970/v1/4583e09bdb41dc1d6f5e17f4.tif"},{"id":74961670,"identity":"7da13a04-2ec5-4011-bbf3-094c17dbb9f6","added_by":"auto","created_at":"2025-01-28 19:00:45","extension":"tif","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":974440,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS3.tif","url":"https://assets-eu.researchsquare.com/files/rs-5898970/v1/0008c9ef080d98cb0922a3e8.tif"},{"id":74960818,"identity":"422ac5e8-52ae-4bed-9b57-ae87d69df384","added_by":"auto","created_at":"2025-01-28 18:52:46","extension":"tif","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":4028004,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS4.tif","url":"https://assets-eu.researchsquare.com/files/rs-5898970/v1/83f95f225bcc45072ca1fe72.tif"},{"id":74960808,"identity":"4ba1ca9c-fff8-4f62-9d6b-5001e6836e0f","added_by":"auto","created_at":"2025-01-28 18:52:46","extension":"tif","order_by":7,"title":"","display":"","copyAsset":false,"role":"supplement","size":6103720,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS5.tif","url":"https://assets-eu.researchsquare.com/files/rs-5898970/v1/1deb035f9a25bacbb70a00ce.tif"},{"id":74960880,"identity":"a702d999-40a2-47fe-a1ce-25290793e61e","added_by":"auto","created_at":"2025-01-28 18:52:49","extension":"tif","order_by":8,"title":"","display":"","copyAsset":false,"role":"supplement","size":7742384,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS6.tif","url":"https://assets-eu.researchsquare.com/files/rs-5898970/v1/6a8ea4f0e918fe6f0f05de0e.tif"},{"id":74960803,"identity":"74cc573b-db41-422b-ba94-1383254a5e01","added_by":"auto","created_at":"2025-01-28 18:52:45","extension":"tif","order_by":9,"title":"","display":"","copyAsset":false,"role":"supplement","size":3113580,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS7.tif","url":"https://assets-eu.researchsquare.com/files/rs-5898970/v1/93e1a7df7472c806f65872f1.tif"},{"id":74961667,"identity":"fb4acc23-8706-4ef7-b2c0-36a33a74a871","added_by":"auto","created_at":"2025-01-28 19:00:44","extension":"tif","order_by":10,"title":"","display":"","copyAsset":false,"role":"supplement","size":4139716,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS8.tif","url":"https://assets-eu.researchsquare.com/files/rs-5898970/v1/38aaef8a0232ee67a9eb686f.tif"},{"id":74962269,"identity":"84e4484f-f92c-43e9-b085-3917869839c6","added_by":"auto","created_at":"2025-01-28 19:08:47","extension":"tif","order_by":11,"title":"","display":"","copyAsset":false,"role":"supplement","size":3492700,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS9.tif","url":"https://assets-eu.researchsquare.com/files/rs-5898970/v1/5c915c8a7f7cf66dc735e075.tif"},{"id":74960817,"identity":"e0c074f2-ecbc-4af2-ad35-1e5ea37a0a2d","added_by":"auto","created_at":"2025-01-28 18:52:46","extension":"tif","order_by":12,"title":"","display":"","copyAsset":false,"role":"supplement","size":3680972,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS10.tif","url":"https://assets-eu.researchsquare.com/files/rs-5898970/v1/5c0b9a7e220f53d715d4a7ff.tif"},{"id":74960865,"identity":"2a4d1992-b7c3-43f7-b2d8-9518c628c23f","added_by":"auto","created_at":"2025-01-28 18:52:49","extension":"tif","order_by":13,"title":"","display":"","copyAsset":false,"role":"supplement","size":3614620,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS11.tif","url":"https://assets-eu.researchsquare.com/files/rs-5898970/v1/765a5e34e61f29688516f3f1.tif"},{"id":74960841,"identity":"dfdb4c67-bb95-4f98-965d-138bd361603d","added_by":"auto","created_at":"2025-01-28 18:52:48","extension":"tif","order_by":14,"title":"","display":"","copyAsset":false,"role":"supplement","size":3653640,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS12.tif","url":"https://assets-eu.researchsquare.com/files/rs-5898970/v1/b5216cfd5ff601f8b113d72d.tif"},{"id":74960809,"identity":"e189dc76-f3bf-4750-9295-bf76f45d8cd1","added_by":"auto","created_at":"2025-01-28 18:52:46","extension":"tif","order_by":15,"title":"","display":"","copyAsset":false,"role":"supplement","size":11271716,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS13.tif","url":"https://assets-eu.researchsquare.com/files/rs-5898970/v1/44f6945f5008f20b7dfc4948.tif"},{"id":74960821,"identity":"6610f08b-1207-4332-a8d5-ddb459fcbee4","added_by":"auto","created_at":"2025-01-28 18:52:46","extension":"tif","order_by":16,"title":"","display":"","copyAsset":false,"role":"supplement","size":9473344,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS14.tif","url":"https://assets-eu.researchsquare.com/files/rs-5898970/v1/c2d818f4acefb2676d87f617.tif"},{"id":74960840,"identity":"02d3c52e-ddd3-48d3-af8a-7b91b12f07e1","added_by":"auto","created_at":"2025-01-28 18:52:48","extension":"tif","order_by":17,"title":"","display":"","copyAsset":false,"role":"supplement","size":6733456,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS15.tif","url":"https://assets-eu.researchsquare.com/files/rs-5898970/v1/0a50fd937041a1b97cd80879.tif"},{"id":74960802,"identity":"8c9e25dd-d7f2-48b9-9813-652bc2aee40a","added_by":"auto","created_at":"2025-01-28 18:52:45","extension":"tif","order_by":18,"title":"","display":"","copyAsset":false,"role":"supplement","size":5647144,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS16.tif","url":"https://assets-eu.researchsquare.com/files/rs-5898970/v1/52e31c67ffb536746883b43f.tif"}],"financialInterests":"No competing interests reported.","formattedTitle":"Integrated multi-omics analysis reveals key genetic, metabolic, and microbial drivers in bladder cancer insights into molecular subtyping and therapeutic approaches: A tumor marker prognostic study","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eBladder cancer (BLCA) is a malignant tumor primarily originating from the uroepithelium, exhibiting significant incidence variations by race and gender, and ranks as the 4th most prevalent cancer in men in America\u003csup\u003e[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]\u003c/sup\u003e. Non-muscle invasive bladder cancer (NMIBC) and muscle invasive bladder cancer (MIBC) are pathologically categorized by tumor infiltration depth, guiding clinical diagnosis and treatment, with respective 5-year survival rates of approximately 90% and 60%\u003csup\u003e[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]\u003c/sup\u003e. Individualizing treatment based solely on pathological grading is challenging, prompting extensive research into BLCA's molecular mechanisms, including gene mutations\u003csup\u003e[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]\u003c/sup\u003e, signaling pathways\u003csup\u003e[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]\u003c/sup\u003e, tumor microenvironment\u003csup\u003e[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]\u003c/sup\u003e, and metabolic pathways\u003csup\u003e[\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]\u003c/sup\u003e, to enhance diagnostic efficiency and patient survival. However, BLCA's heterogeneity contributes to drug resistance and recurrence post-treatment, leading to suboptimal survival rates\u003csup\u003e[\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]\u003c/sup\u003e. Thus, comprehensive multi-omics analysis to elucidate multilevel heterogeneity is crucial for improving patient prognosis and is an urgent challenge to address.\u003c/p\u003e \u003cp\u003eSingle-cell combined transcriptomics is a popular combination in developing BLCA predictive models for predicting patient survival and immune microenvironment\u003csup\u003e[\u003cspan additionalcitationids=\"CR9\" citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]\u003c/sup\u003e. While this approach has uncovered some gene expression heterogeneity, it has not elucidated inter-patient and intra-tumor heterogeneity. Following the innovation of science and technology, whole-exome sequencing and metabolomics have gradually been widely used in tumor research. Whole-exome data indicate that normal uroepithelial cells can harbor over 500 to 2,000 mutations by ages 50 to 65, alongside a higher tumor mutational burden (TMB) in BLCA driven mainly by activation of APOBEC3 cytidine deaminase\u003csup\u003e[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]\u003c/sup\u003e. In addition, distinct grades of BLCA show differences in glucose, lipid, and carnitine metabolism, which have all been associated with tumor invasion and metastasis\u003csup\u003e[\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]\u003c/sup\u003e, suggesting that genetic variants and metabolic pathways may be etiological factors contributing to the development of BLCA. Recent research suggested the microbiome may be a causative or cofactor in genitourinary malignancies\u003csup\u003e[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]\u003c/sup\u003e; for instance, Sun et al. found greater microbial diversity in NMIBC compared to MIBC tissues, with distinct dominant species between groups by 2bRAD sequencing for Microbiome (2bRAD-M)\u003csup\u003e[\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]\u003c/sup\u003e. Increased microbial abundance was noted in tumor patients at higher recurrence or progression risk versus healthy controls\u003csup\u003e[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]\u003c/sup\u003e. Therefore, a deeper understanding of the microbiome may yield new prognostic insights for BLCA. Integrating multi-omics data can illuminate how various molecular characterizations synergistically influence disease onset, progression, and therapeutic response, providing a holistic understanding of biological systems.\u003c/p\u003e \u003cp\u003eIn this research, we incorporated transcriptomics, whole exome sequencing, metabolomics, the analysis of intratumoral microbial populations, and single-cell profiling to effectively identify the key genes, metabolites, and microbes associated with BLCA. We conducted an intensive investigation into their functions and regulatory interactions across various biological hierarchies. Concurrently, we categorized BLCA using molecular profiling of the key genes and thoroughly examined its underlying molecular mechanisms. Ultimately, we illuminated the expression patterns of these crucial genes within pivotal cellular contexts at the single-cell resolution. These findings will provide an insightful understanding of the molecular mechanism of BLCA and the formulation of new therapeutic strategies, with a perspective of contributing to the enhancement of personalized treatment for BLCA patients.\u003c/p\u003e"},{"header":"2. Materials and methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Patient Selection and Sample Collection\u003c/h2\u003e \u003cp\u003eA total of 35 cases of cancer and paracancerous (normal) tissues were collected from patients diagnosed with bladder cancer (BLCA) for comprehensive sequencing of the transcriptome, Whole-Exome Sequencing (WES), metabolome and intratumoural microbiome sequencing. Considering the possibility of endogenous contamination during surgery, all samples were collected aseptically in the operating room and clean tissue blocks were selected as much as possible, and then cryopreserved and stored at -80\u0026deg;C for preservation. All patients were from The Second Affiliated Hospital of Kunming Medical University. The study was approved by the Research Ethics Committee of The Second Affiliated Hospital of Kunming Medical University, and written informed consent was obtained from each patient.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Data Source\u003c/h2\u003e \u003cp\u003eTCGA-BLCA dataset was sourced from the University of California Santa Cruz (UCSC) Xena database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://xena.ucsc.edu/\u003c/span\u003e\u003cspan address=\"https://xena.ucsc.edu/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) as a transcriptome training set\u003csup\u003e[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]\u003c/sup\u003e. The dataset comprised RNA sequencing data from 411 BLCA tumour tissue samples and 9 control tissue samples, along with corresponding clinical information and survival data. Additionally, the GSE13507 dataset (sequencing platform: GPL6102) was obtained from the Gene Expression Omnibus (GEO, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.ncbi.nlm.nih.gov/geo/\u003c/span\u003e\u003cspan address=\"https://www.ncbi.nlm.nih.gov/geo/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) as an external validation set for transcriptome analysis\u003csup\u003e[16]\u003c/sup\u003e, which encompassed 188 BLCA patient tumour tissue samples and 68 control bladder mucosa tissue samples. Furthermore, two BLCA-associated single-cell datasets were included and integrated into the single-cell dataset for this study, namely, the GSE135337 dataset with GPL24676 platform and the GSE222315 dataset with GPL24676 platform, both of which were obtained from the GEO database, were incorporated into this study. The GSE135337 comprised seven BLCA patient cancers and one paraneoplastic tissue sample. The GSE222315 dataset comprised nine BLCA patients with cancer and four paraneoplastic tissue samples.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Analysis of Transcriptome Sequencing Data\u003c/h2\u003e \u003cp\u003eTo improve data quality and standardize the data, raw reads from transcriptome sequencing were quality controlled for clean data. The clean data was then compared to the human genome (GRCh38) using TopHat 2.0. Next, the Transcripts Per Kilobase of exon model per Million mapped reads (TPM) values were calculated for each gene to assess the degree of bias in mRNA distribution. Principal component analysis (PCA) was then performed on both groups of samples using the R package \u0026lsquo;ropls\u0026rsquo; (version 1.34.0)\u003csup\u003e[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e17\u003c/span\u003e]\u003c/sup\u003e to estimate the variability of the data. In addition, the correlation between the two groups of samples was assessed by Spearman analysis (correlation coefficient |Cor| \u0026gt; 0.3, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05). The objective of differential expression analysis was to identify genes that exhibited significant differences in expression between the various sample groups. In this study, the R package \u0026lsquo;DESeq2\u0026rsquo; (version 1.42.0)\u003csup\u003e[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e18\u003c/span\u003e]\u003c/sup\u003e was employed to ascertain the disparities in gene expression levels and to acquire differentially expressed genes (DEGs) between cancer and normal samples with adjusted p value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 and |log\u003csub\u003e2\u003c/sub\u003eFold Change (FC)| \u0026gt; 0.5. The Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) were databases that were frequently utilized for the purpose of pathway research. In order to gain insight into the relevant biological functions and pathways of DEGs, GO and KEGG enrichment analyses of DEGs were conducted using the \u0026lsquo;clusterProfiler\u0026rsquo; package (version 4.7.1.3)\u003csup\u003e[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e19\u003c/span\u003e]\u003c/sup\u003e in R language with adjusted p value\u0026thinsp;\u0026lt;\u0026thinsp;0.05.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.4 Analysis of WES Data\u003c/h2\u003e \u003cp\u003eFor the identification of significant mutant genes (SMGs) in both sample groups, tumour mutation information from 35 BLCA tumour samples from the WES data was analyzed using \u0026lsquo;maftools\u0026rsquo; (version 2.17.10)\u003csup\u003e[\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e20\u003c/span\u003e]\u003c/sup\u003e. To explore the biological pathways associated with SMGs, GO and KEGG enrichment analyses of SMGs were also performed using the same methods with p\u0026thinsp;\u0026lt;\u0026thinsp;0.05.The Oncobox Pathway Database (OncoboxPD, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://open.oncobox.com\u003c/span\u003e\u003cspan address=\"https://open.oncobox.com\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) comprised 51,627 human molecular pathways that was processed in a uniform manner. KEGG enrichment was performed according to the above SMGs using OncoboxPD with adjusted p value\u0026thinsp;\u0026lt;\u0026thinsp;0.05, and pathway activation level (PAL) values were calculated for all KEGG pathways to observe the activation-inhibition status of the relevant pathways enriched by SMGs.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e2.5 Analysis of Metabolomic\u003c/h2\u003e \u003cp\u003eTo assess the degree of variation in the metabolomic data of the two groups of samples, PCA analysis was performed on the metabolomic sequencing data of the two groups of samples. Similarly, the correlation between the two groups of samples in the metabolomic sequencing data was assessed by Spearman analysis with |Cor| \u0026gt; 0.3 and p\u0026thinsp;\u0026lt;\u0026thinsp;0.05. Next, differentially expressed metabolites (DEMs) in cancer and normal samples were screened based on the importance in projection (VIP) values of variables calculated from the orthogonal partial least squares discriminant analysis (OPLS-DA) model and the p values of Student's t-test in paired samples, with thresholds of VIP\u0026thinsp;\u0026gt;\u0026thinsp;1 and p\u0026thinsp;\u0026lt;\u0026thinsp;0.05. To investigate the biological processes associated with the DEMs, a metabolic profile was performed using MetaboAnalyst to convert the compound names of the DEMs to KEGG IDs, and the pathways involved in the DEMs were analyzed in the KEGG database at a significance level of p\u0026thinsp;\u0026lt;\u0026thinsp;0.05.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e2.6 Analysis of Intratumoural Microbial Data\u003c/h2\u003e \u003cp\u003eTo ensure the accuracy of subsequent analyses, unprocessed intratumoural microbiome sequencing data were used for quality control and chimeras were removed using UCHIME. Sequences with \u0026gt;\u0026thinsp;97% similarity were clustered into the same operational taxonomic unit (OTU). For clustered OTUs, taxonomic information was annotated and evaluated using the Mothur algorithm using the Silva database. Furthermore, the alpha and beta diversity of intratumoural microorganisms in different groups of samples were analyzed and compared to assess differences in the composition of intratumoural microorganisms between the cancer and normal groups. In order to analyze the abundance of microorganisms in different groups of samples, bar charts were plotted to demonstrate the relative abundance of intratumoral microorganisms at the phylum, family and genus levels for both groups of samples. Next, specific microbial taxa were identified between the two groups of samples using Linear Discriminant Analysis (LDA) Effect Size (LEfSe) at the genus level (LDA\u0026thinsp;\u0026gt;\u0026thinsp;2, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05). To investigate the biological processes associated with these differential microorganisms, KEGG pathway enrichment analysis of differential microorganisms between the two groups at the genus level was performed using PICRUSt2 software. Microbes never exist in isolation from each other and are highly interconnected in response to various environmental changes. Therefore, this research further analyzed the genus-level colonies of both groups of samples and derived the associated genus-level driver colonies.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e2.7 Co-analysis of Transcriptome and WES\u003c/h2\u003e \u003cp\u003eFor the purpose of exploring the somatic mutations of BLCA patients, the somatic mutations of BLCA patients were analyzed in all the samples of the TCGA-BLCA and WES data. The amplification and deletion of CNV were also analyzed by Gistic2.0. In order to identify genes with significant differences shared at the transcriptome level and at the WES level, DEGs obtained based on transcriptome analyses were merged with SMGs from WES data, and the resulting genes were defined as candidate genes. The candidate genes were further analyzed for GO and KEGG enrichment analyses with an adjusted p value\u0026thinsp;\u0026lt;\u0026thinsp;0.05. Patients with BLCA were divided into high and low expression sets according to the optimal expression threshold of each candidate gene. The data were then analyzed using the \u0026lsquo;survminer\u0026rsquo; software package (version 0.4.9)\u003csup\u003e[\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e21\u003c/span\u003e]\u003c/sup\u003e, and Kaplan-Meier (KM) survival curves were constructed for the overall survival (OS) of patients in the two sets. Candidate genes with OS differences at p\u0026thinsp;\u0026lt;\u0026thinsp;0.05 between sets were considered to have prognostic value and were defined as candidate prognostic genes. The expression of candidate prognostic genes between cancer and normal groups was also analyzed and validated in transcriptome level, TCGA-BLCA dataset, and GSE13507 datasets, respectively, by Wilcoxon test (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05). In addition, the R package \u0026lsquo;pROC\u0026rsquo; was used to plot receiver operating characteristic (ROC) curves and calculate area under the curve (AUC) to assess the ability of candidate prognostic genes to distinguish between cancer and normal samples. Genes with AUC\u0026thinsp;\u0026gt;\u0026thinsp;0.7 in three sets of transcriptome data were defined as biomarkers for subsequent studies.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e2.8 Combined Analysis of Transcriptome, Metabolome and Intratumoural Microbiome\u003c/h2\u003e \u003cp\u003eTo obtain relevant genes and metabolites at the transcriptome, WES and metabolome levels, correlations between biomarkers and DEMs were assessed by Spearman analysis in all BLCA sequencing samples. Biomarkers and DEMs with |Cor| \u0026gt; 0.3 and p\u0026thinsp;\u0026lt;\u0026thinsp;0.05 were defined as key gene 1 and key metabolite 1, respectively, and their interactions were investigated. Similarly, key gene 2 and key microbiome 1 associated at transcriptome, WES and intratumour microbiome levels were also obtained under the above parameters. In addition, the correlation between DEM and differential microbes was also assessed by Spearman analysis under the same parameters to derive the key metabolite 2 and key microbiome 2 associated at the metabolome as well as intratumour microbiome level. The aforementioned key genes, key metabolites and key microorganisms were integrated to yield key genes, key metabolites and key microorganisms that were meaningful at multi-omics levels. These were subsequently assessed for their regulatory relationships.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e2.9 Analysis of BLCA Clusters\u003c/h2\u003e \u003cp\u003eIn order to identify the molecular clusters of BLCA associated with key genes, a consensus clustering approach was employed to categorize BLCA samples into clusters based on the expression of key genes in transcriptome sequencing data. Furthermore, the distribution differences between the various clusters were observed and evaluated using PCA and t-distributed stochastic neighbour embedding (tSNE). Then, based on the clusters just achieved, different cluster samples were analyzed for differences, screened by |log\u003csub\u003e2\u003c/sub\u003eFC| \u0026gt; 0.5 and p\u0026thinsp;\u0026lt;\u0026thinsp;0.05, to obtain inter-cluster DEGs. Later, GO and KEGG enrichment analyses were performed on the inter-cluster DEGs with adjusted p value\u0026thinsp;\u0026lt;\u0026thinsp;0.05. To further observe the association between microbial communities in different clusters, the association networks between microbial communities (genus level) in different clusters were analyzed in the intratumoral microbiome sequencing data. Functional prediction of different clusters was also performed to explore the functional differences involved in different clusters. Similarly, intercluster DEMs were screened by VIP\u0026thinsp;\u0026gt;\u0026thinsp;1.00 and p\u0026thinsp;\u0026lt;\u0026thinsp;0.05 based on metabolome sequencing data. metabolic pathways involved in transformed intercluster DEMs were subsequently analyzed in the KEGG database (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05). Furthermore, a series of subsequent in-depth analyses of different BLCA clusters based on transcriptome sequencing data were performed. Genome enrichment analysis (GSEA) with HALLMARK gene set: h.all.v2023.1.Hs.symbols.gmt as the reference set of genes and genome variation analysis (GSVA) with metabolism-related pathways\u003csup\u003e[\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e22\u003c/span\u003e]\u003c/sup\u003e as the reference gene set were used to investigate significant biological pathways and metabolic pathways between clusters. Differences in TIME, with the TIME signature as the reference gene set, were also investigated. Inter-cluster differences in the abundance of 22 immune-infiltrating cells, 66 immune checkpoint genes\u003csup\u003e[\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e23\u003c/span\u003e]\u003c/sup\u003e, activity of immune-circulatory pathways, patient response to immunotherapy, somatic mutations, mutations in typical oncogenic pathways, and drugs were also studied. All of the above analyses were considered statistically significant with a p value\u0026thinsp;\u0026lt;\u0026thinsp;0.05. For the purpose of analyzing the possible potential roles of key microorganisms in immune-related processes, the correlation between key microorganisms and MHC molecules, immune activators, and immunosuppressants in different clusters was assessed by Spearman analysis in intratumoural microbiome sequencing data, respectively. To observe the expression levels of key genes and key metabolites among different clusters, the differences in the expression of key genes, key metabolites and key microorganisms among different clusters were compared by Wilcoxon test in the transcriptome, metabolome and intratumoural microbiome sequencing data, respectively (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05). The key genes that were significantly different in BLCA clusters were used for subsequent analyses.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e2.10 Association Analysis of Clinical Characteristics of Key Genes\u003c/h2\u003e \u003cp\u003eTo examine the expression of key genes in relation to clinical characteristics, BLCA patients were classified into distinct clinical subgroups based on varying clinical characteristics in the TCGA dataset. The differential expression of key genes between subgroups was then evaluated using the Wilcoxon test, with a significance level of p\u0026thinsp;\u0026lt;\u0026thinsp;0.05. Furthermore, correlations between key genes and individual clinical characteristics were evaluated through the use of a Spearman analysis. Furthermore, BLCA patients were categorized into two groups based on the median critical value of the expression of each key gene. KM survival curves were constructed for the overall survival (OS) of patients with high and low expression of each key gene in different clinical subgroups.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e2.11 Enrichment Analysis of Key Genes\u003c/h2\u003e \u003cp\u003eTo explore the biological pathways involved in the key genes in BLCA, single-gene GSEA enrichment analysis of the key genes was performed based on the correlation of 'c2.cp.kegg.v7.0.symbols.gmt' as the reference gene set in the transcriptome sequencing data (adjusted p value\u0026thinsp;\u0026lt;\u0026thinsp;0.05). In the TCGA dataset, BLCA patients with survival information were divided into high and low expression subgroups based on the median threshold of key gene expression. Pathways or functions that were significantly enriched between the high and low expression subgroups for each key gene were explored using 'HALLMARKgenesets: h.all.v2023.1.Hs.symbols.gmt' as the reference geneset and GSEA based on the multiplicity of differences (adjusted p value\u0026thinsp;\u0026lt;\u0026thinsp;0.05).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003e2.12 Drug Prediction and Molecular Docking\u003c/h2\u003e \u003cp\u003eTo further explore potential therapeutic agents for BLCA, potential agents or molecular compounds interacting with the key genes were predicted by L1000 FWD (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://maayanlab.cloud/l1000fwd/\u003c/span\u003e\u003cspan address=\"https://maayanlab.cloud/l1000fwd/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Based on the results of agent analysis in the above, the most relevant drugs to each key gene were selected for molecular docking to explore the binding ability of the key genes to the relevant agents. Binding energy \u0026lt; -5kcal/mol meant high binding activity.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003e2.13 Information, Function, and Expression Analysis of Proteins Encoded by Key Genes\u003c/h2\u003e \u003cp\u003eThe National Center for Biotechnology Information (NCBI) was conducted further research into the information regarding the key genes, proteins and related functions that they encode. To observe the expression of the key genes in pan-cancer, the expression of each key gene was compared in pan-cancer tumours and control samples based on pan-cancer data from the TCGA database. After that, the expression levels of key genes in different tissues and expression at protein level in tumour and normal samples were analyzed by The Human Protein Atlas (HPA, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.proteinatlas.org/\u003c/span\u003e\u003cspan address=\"https://www.proteinatlas.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e2.14 Analysis of Single-cell Data\u003c/h2\u003e \u003cp\u003eTo ensure the accuracy of the analysis, low-quality cells were first filtered out, and then the top 2,000 highly variable genes among the remaining cells were extracted. The cells were further analyzed by clustering to generate cell clusters. And the cell types obtained from the above clustering were determined by cell annotation. Furthermore, the distribution of each key gene in each cell type was demonstrated in the single-cell dataset, and their expression levels in different samples (cancer and control) were compared. All cell types were also analyzed for KEGG functional enrichment (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05). In light of the differential expression of the key genes in each cell type and the enrichment results for all cell types, as well as the findings of relevant literature reports, fibroblasts were designated as the key cells for subsequent analyses. Next, ligand-receptor interactions between key cells and other cell types were analyzed in order to reveal the communication of each cell. To further explore the heterogeneity of the key cells, the key cells were analyzed by dimensionality reduction clustering, and the key cells were reclustered into different subclusters. And the subclusters were annotated according to the references\u003csup\u003e[24\u0026ndash;26]\u003c/sup\u003e to obtain the key cell clusters. The potential differentiation trajectories of key cell clusters were further observed by mimetic temporal sequencing analysis, while the expression of key genes in different temporal sequences was visualized. In addition, upstream regulators of key genes in various clusters of key cells were probed to speculate on the potential mechanism of action of the key genes.\u003c/p\u003e \u003cp\u003eIn addition, to further explore the degree of malignancy of the cell types, the epithelial cells obtained from the above analyses were reclustered into different subclusters and the clusters of the cell subclusters were determined. Based on the above results of epithelial cell descending clustering, the proposed time series of epithelial cell clusters were analyzed and the differentiation trajectories of epithelial cell clusters with different degrees of malignancy were analyzed. The expression of key genes during different time series was also visualized. And their expression in different malignancy degree cell clusters differentiation was also visualized and compared.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003e\u003cb\u003e2.15 Statistical Analysis\u003c/b\u003e\u003c/h2\u003e \u003cp\u003eBioinformatic analyses were performed using the R programming language (version 4.2.2). The Wilcoxon test was used to compare differences between two groups. Fisher's exact test was used to calculate the p value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 for significance.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003e2.16 Remark Criteria Statement\u003c/h2\u003e \u003cp\u003eOur research adheres to the Reporting Recommendations for Tumor Marker Prognostic Studies (REMARK) criteria, ensuring transparency and comprehensive reporting of our study findings\u003csup\u003e[\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e27\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Quality Assessment of Sequencing Data\u003c/h2\u003e \u003cp\u003eTo assess transcriptome sequencing data quality, we first examined gene expression profiles. Box plot analysis identified four outlier samples (36, 37, 38, and 39), which were excluded. The remaining 70 samples, balanced at 35 tumour and 35 paraneoplastic samples, were retained (\u003cb\u003eFigure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003ea\u003c/b\u003e). PCA of these samples showed distinct clustering within each group (\u003cb\u003eFigure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eb\u003c/b\u003e). Correlation analysis revealed a generally positive correlation between samples from the two groups (\u003cb\u003eFigure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003ec\u003c/b\u003e). Metabolome sequencing data effectively distinguished tumour from normal samples (\u003cb\u003eFigure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003ed\u003c/b\u003e). Intratumoural microbiome sequencing data were analyzed to identify cluster OTUs. Sparse curves reached a plateau, indicating sufficient sequencing depth (\u003cb\u003eFigure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003ee\u003c/b\u003e). Rank-abundance curves illustrated the richness and evenness of the samples (\u003cb\u003eFigure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003ef\u003c/b\u003e). The species cumulative box plot showed that species diversity increased with sample size until stabilizing (\u003cb\u003eFigure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eg\u003c/b\u003e). The aforementioned results collectively demonstrated that the sequencing quality in this study was high and that the samples were effectively differentiated.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003e\u003cb\u003e3.2 Identification and Function of DEGs, SMGs and DEMs in BLCA\u003c/b\u003e\u003c/h2\u003e \u003cp\u003eDifferential expression analysis identified 7,302 DEGs between cancer and normal groups (Fig.\u0026nbsp;1a, \u003cb\u003eb\u003c/b\u003e). GO enrichment analysis revealed significant terms such as sarcolemma, muscle tissue development, embryonic organ development, contractile fibre and collagen-containing extracellular matrix (Fig.\u0026nbsp;1c). KEGG enrichment analysis highlighted pathways including cytoskeleton in muscle cells, hypertrophic cardiomyopathy, focal adhesion, axon guidance, dilated cardiomyopathy and other pathways (Fig.\u0026nbsp;1d). WES of 35 BLCA tumour samples identified 78 SMGs, with missense mutations being predominant (Fig.\u0026nbsp;1e). GO and KEGG enrichment analyses of SMGs revealed 139 GO entries and five KEGG pathways, including basal transcription factors, spinocerebellar ataxia, human immunodeficiency virus 1 infection, FoxO signaling pathway, and osteoclast differentiation (Fig.\u0026nbsp;1f, \u003cb\u003eg\u003c/b\u003e). KEGG enrichment using OncoboxPB showed activated pathways like leishmaniasis (PAL\u0026thinsp;=\u0026thinsp;5, p\u0026thinsp;=\u0026thinsp;0.0034) and inhibited pathways like p53 signaling (PAL = -43, p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001) (Fig.\u0026nbsp;1h). Our findings indicated that the DEGs and SMGs were co-enriched for the FoxO signaling pathway in KEGG, suggesting that this pathway might be a potential recipient of the both the DEGs and SMGs in BLCA. Metabolome sequencing identified 212 DEMs, enriched in 14 KEGG pathways related to various amino acid metabolic processes (Fig.\u0026nbsp;1i-k).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Identification and Function of BLCA-associated Differential Microbes\u003c/h2\u003e \u003cp\u003eTo examine the distinctions in species richness and composition of intratumoural microorganisms, alpha and beta diversity were evaluated across diverse groups of sample species. With regard to alpha diversity, the Simpson, Shannon, InvSimpon and Pielou indices demonstrated no statistically significant differences between cancer and normal samples, indicating species diversity and even distribution (\u003cb\u003eFigure S2a\u003c/b\u003e). In the case of beta diversity, the stress value was 0.11, indicating the reliability of the results (\u003cb\u003eFigure S2b, c\u003c/b\u003e). Intratumoral microbiome composition was assessed at the phylum, family, and genus levels. Proteobacteria were most abundant in both cancer and normal samples, but Cyanobacteria was notably higher in normal samples and lower in cancer samples (Fig.\u0026nbsp;2a). Rhodocyclaceae and Comamonadaceae were the most prevalent families in both groups (Fig.\u0026nbsp;2b). At the genus level, \u003cem\u003ePhaeospirillum\u003c/em\u003e and \u003cem\u003eMethyloversatilis\u003c/em\u003e were the most abundant (Fig.\u0026nbsp;2c). Five differential microbiomes were identified between cancer and normal samples: Phaeospirillum, Acinetobacter, Rubrivivax, Staphylococcus, and Dialister (Fig.\u0026nbsp;2d-f). KEGG enrichment analysis of these microbiomes revealed significant pathways: mycolylarabinogalactan- peptidoglycan complex biosynthesis, gluconeogenesis I, and peptidoglycan biosynthesis II (staphylococci) (Fig.\u0026nbsp;2g). Analysis of driving species indicated differences in community networks, with greater microbiome density observed in normal samples compared to tumour samples (Fig.\u0026nbsp;2h; \u003cb\u003eTable \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e; Figure S3\u003c/b\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec23\" class=\"Section2\"\u003e \u003ch2\u003e3.4 Recognition of Biomarkers\u003c/h2\u003e \u003cp\u003eAnalysis of somatic mutations in BLCA patients using TCGA-BLCA and WES data revealed that sense mutations had the highest frequency and missense mutations were most common (Fig.\u0026nbsp;3). Intersecting DEGs and SMGs identified 26 candidate genes (Fig.\u0026nbsp;4a), which were associated with clathrin-dependent endocytosis, WW domain binding, lysosomal lumen, protein serine kinase activity, focal adhesion, and other GO entries (Fig.\u0026nbsp;4b). KEGG enrichment analysis highlighted significant enrichment in Virion-Ebolavirus, Lyssavirus, Morbillivirus, and African trypanosomiasis pathways (Fig.\u0026nbsp;4c). Prognostic evaluation of these genes classified patients into high- and low-expression groups. KM survival curves indicated significant survival differences for 15 genes (\u003cb\u003eFigure S4\u003c/b\u003e). Expression validation in transcriptome, TCGA-BLCA, and GSE13507 datasets further confirmed that APOL1, DHX34, and TNK2 were up-regulated, while AHNAK, CSPG4, NCAM1, and PCDHB4 were down-regulated in the cancer group (Fig.\u0026nbsp;4d). ROC analysis identified AHNAK, CSPG4, DHX34, NCAM1, and PCDHB4 as biomarkers with AUCs\u0026thinsp;\u0026gt;\u0026thinsp;0.7 across all datasets (Fig.\u0026nbsp;4e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec24\" class=\"Section2\"\u003e \u003ch2\u003e3.5 Identification of Key Genes, Metabolites and Microbiomes\u003c/h2\u003e \u003cp\u003eA combined analysis of biomarkers, DEMs, and differential microbiomes identified three key genes (AHNAK, CSPG4, NCAM1), 90 key metabolites, and two key microbiomes (Sphingomonas koreensis and Rhodospirillaceae) from multi-omics data (\u003cb\u003eFigure S5; Fig.\u0026nbsp;5a-c\u003c/b\u003e). Correlation and network analysis revealed positive correlations among the key genes, no correlation between key microbiomes and key genes or metabolites, and negative correlations between key genes and key metabolites (Fig.\u0026nbsp;5d-e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec25\" class=\"Section2\"\u003e \u003ch2\u003e3.6 Two Clusters for BLCA\u003c/h2\u003e \u003cp\u003eBLCA samples were classified into two clusters based on key genes (Fig.\u0026nbsp;6a, \u003cb\u003eb\u003c/b\u003e). Analysis revealed 3,464 inter-cluster DEGs, which were enriched in 1,665 GO terms, with the top categories including collagen-containing extracellular matrix and extracellular matrix organization (Fig.\u0026nbsp;6c-e). A study of a prognostic model of BLCA suggested that PD-L1 expression could predict the prognosis of patients with BLCA, possibly related to extracellular matrix passage of collagen\u003csup\u003e[\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e28\u003c/span\u003e]\u003c/sup\u003e. KEGG enrichment analysis identified 81 pathways, notably cytoskeleton in muscle cells and focal adhesion (Fig.\u0026nbsp;6f). Microbial community analysis showed most microbiomes were positively correlated within clusters (Fig.\u0026nbsp;6g). Additionally, 17 inter-cluster differential metabolites were found, enriched in amino sugar and nucleotide sugar metabolism, and ascorbate and aldarate metabolism (Fig.\u0026nbsp;6h-j). GSEA identified significant pathways across clusters, including interferon gamma response and epithelial mesenchymal transition (\u003cb\u003eFigure S6\u003c/b\u003e). GSVA further identified 10 differential metabolic pathways \u003cb\u003e(Fig.\u0026nbsp;6k\u003c/b\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec26\" class=\"Section2\"\u003e \u003ch2\u003e3.7 Tumor Immune Microenvironment (TIME) in BLCA\u003c/h2\u003e \u003cp\u003eIn BLCA clusters, the TIME analysis revealed significant differences between clusters in inhibitory molecules, IFNG response, tumor cell recognition, Priming and activation, and inhibitory cells (MDSCs) (Fig.\u0026nbsp;7a, \u003cb\u003eb\u003c/b\u003e). Differential immune cells included M0/M1 macrophages, resting mast cells, resting natural killer cells, follicular helper T cells, and gamma delta T cells (Fig.\u0026nbsp;7c, \u003cb\u003ed\u003c/b\u003e). NCAM1 was positively correlated with gamma delta T cells (Cor\u0026thinsp;=\u0026thinsp;0.54) and negatively with M0 macrophages (Cor = -0.48) (Fig.\u0026nbsp;7e). Among 66 immune checkpoint genes, 25 differed between clusters, such as BTLA, CD274, and CTLA4 (Fig.\u0026nbsp;7f). Differences were observed in cancer immune cycle steps 1, 4, and 7 (Fig.\u0026nbsp;7g). TIDE scores showed notable divergence, indicating different immune profiles between clusters (Fig.\u0026nbsp;7h). Key genes correlated positively with stromal, immunity, and ESTIMATE scores, suggesting their relevance to immunotherapy (Fig.\u0026nbsp;7i). These scores might interact with key metabolites (Fig.\u0026nbsp;7j). Key microbes with a lot of immune factors showed negative correlations between BLCA clusters (Fig.\u0026nbsp;7k).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec27\" class=\"Section2\"\u003e \u003ch2\u003e3.8 Gene Mutations in the BLCA\u003c/h2\u003e \u003cp\u003eThe genomic mutation landscape of the top 20 most commonly mutated genes in BLCA revealed detailed information on CNV amplification and deletion, mutation types, and base mutations. MUC4 showed the highest mutation frequency, with a notable percentage of missense mutations and C-to-T base mutations (Fig.\u0026nbsp;8a, \u003cb\u003eb\u003c/b\u003e). Differences in 10 classical oncogenic pathways were observed between clusters (Fig.\u0026nbsp;8c). Chemotherapy response analysis identified 236 differential agents in the CTRP database, 43 in GDSC, and 785 in PRISM Repurposing (Fig.\u0026nbsp;8d; \u003cb\u003eFigure S7a-c\u003c/b\u003e). By comparing the 50% inhibiting concentration (IC50) of common chemotherapeutic agents, FGFR inhibitors and EGFR inhibitors between different clusters, a total of 63 agents were obtained that differed between clusters, and most of them had significant correlation with the gene NCAM1 (Fig.\u0026nbsp;8e; \u003cb\u003eFigure S7d-j\u003c/b\u003e). At the same time, key genes were also able to act on classical therapeutic pathways as well as pathways through corresponding targeted agents (Fig.\u0026nbsp;8f). Key genes, including AHNAK, CSPG4, and NCAM1, differed in expression between clusters, and 40 metabolites were differentially expressed, with no microbial differences detected (Fig.\u0026nbsp;8g; \u003cb\u003eFigure S8\u003c/b\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec28\" class=\"Section2\"\u003e \u003ch2\u003e3.9 Exploration of Key Gene-related Mechanisms\u003c/h2\u003e \u003cp\u003eTo analyze key gene expression in BLCA, patients were categorized based on clinical features from TCGA-BLCA. AHNAK and NCAM1 both exhibited differential expression in T and N stage as well as stage, while CSPG4 displayed a more limited differential expression in stage (\u003cb\u003eFigure S9\u003c/b\u003e). Meanwhile, AHNAK showed a positive correlation with N stage, T stage, and overall stage, while CSPG4 negatively correlated with M stage. NCAM1 positively correlated with stage and N stage (Fig.\u0026nbsp;9a). Survival rates differed significantly between high and low expression levels of key genes (\u003cb\u003eFigure S10-12\u003c/b\u003e). Single-gene GSEA enrichment analyses highlighted pathways such as focal adhesion, ribosome, and ECM receptor interaction (Fig.\u0026nbsp;9b-d). AHNAK, CSPG4, and NCAM1 were enriched in 35, 59, and 31 pathways, respectively, related to olfactory transduction, NKCC, and CAMs (\u003cb\u003eTable S2-4\u003c/b\u003e). Maltotriose and other compounds showed potential as therapeutic agents, with excellent docking interactions of -6.1 kcal/mol for AHNAK, -8.5 kcal/mol for CSPG4, and \u0026minus;\u0026thinsp;6 kcal/mol for NCAM1 (Fig.\u0026nbsp;9e-f; \u003cb\u003eTable S5\u003c/b\u003e). Also, three key genes varied in most cancers (Fig.\u0026nbsp;9g; \u003cb\u003eFigure S13\u003c/b\u003e). Key genes were further analyzed through NCBI and HPA database, revealing high expression in skin, colon, and cerebral cortex (\u003cb\u003eTable S6; Fig.\u0026nbsp;9h\u003c/b\u003e). At the protein level, the three key genes did not differ between cancer and normal samples (Fig.\u0026nbsp;9i).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec29\" class=\"Section2\"\u003e \u003ch2\u003e3.10 Expression of Key Genes at the Single-cell Level\u003c/h2\u003e \u003cp\u003eFollowing quality control, single-cell sequencing data included 130,431 cells and 36,137 genes (\u003cb\u003eFigure S14a, b\u003c/b\u003e). From this, 2,000 highly variable genes were selected, and 32 cell clusters were identified via PCA (\u003cb\u003eFigure S14c-e\u003c/b\u003e). Nine cell types were annotated: CD4\u003csup\u003e+\u003c/sup\u003e T cells, endothelial cells, fibroblasts, myeloid cells, smooth muscle cells, epithelial cells, B cells, CD8\u003csup\u003e+\u003c/sup\u003e natural killer cells, and mast cells (Fig.\u0026nbsp;10; \u003cb\u003eFigure S14f\u003c/b\u003e). The distribution of all nine cell types was presented in samples (Fig.\u0026nbsp;11a). AHNAK was broadly expressed, CSPG4 was mainly in fibroblasts and smooth muscle cells, and NCAM1 showed minimal expression (Fig.\u0026nbsp;11b). Thus fibroblasts were used as key cells (Fig.\u0026nbsp;11c-d). Key cell types focused on the cytoskeleton in muscle cells, the focal adhesion, the PI3K-Akt signaling pathway, and protein processing in the endoplasmic reticulum (Fig.\u0026nbsp;11e). Significant differences in intercellular communication were observed between normal and tumor cells (\u003cb\u003eSupplementary Fig.\u0026nbsp;15\u003c/b\u003e). In light of these findings and the available literature, we identified fibroblasts as the key cells for further investigation. Fibroblasts were selected for further analysis and reclustered into iCAFs, matCAFs, myCAFs, tCAFs, and vCAFs (Fig.\u0026nbsp;11f-g). AHNAK was more expressed across fibroblast subtypes (Fig.\u0026nbsp;11h). Pseudo-time trajectory analysis showed AHNAK expression decreased, CSPG4 varied, and NCAM1 increased then decreased (Fig.\u0026nbsp;11i-j). Transcription factors and binding motifs in key cell subtypes were identified (Fig.\u0026nbsp;11k). Furthermore, causal relationships between transcription factors and key genes were inferred and demonstrated by means of directed networks (Fig.\u0026nbsp;12).\u003c/p\u003e \u003cp\u003eBLCA typically originates from epithelial cells. To further investigate malignancy, epithelial cells were categorized into high, intermediate and low malignancy groups based on copy number variation scores of cell subclusters (Fig.\u0026nbsp;13a-d). Analysis of differentiation trajectories showed that AHNAK expression initially decreased and then increased over time, whereas CSPG4 remained stable and NCAM1 decreased to a point and then remained constant (\u003cb\u003eFigure S13e-f\u003c/b\u003e). Notably, AHNAK gene expression varied across cell subpo- pulations with different levels of malignancy, whereas CSPG4 and NCAM1 showed no changes only in the low and moderate malignancy groups (\u003cb\u003eFigure S13g\u003c/b\u003e).\u003c/p\u003e \u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cp\u003eBLCA, recognized as one of the most prevalent malignancies of the urinary tract, predominantly affects individuals over the age of 55 and shows a higher incidence in men\u003csup\u003e[\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e29\u003c/span\u003e]\u003c/sup\u003e. As our understanding of the pathogenesis of BLCA has expanded, numerous studies have provided a more refined molecular typing based on genes and protein expression and established corresponding prognostic models\u003csup\u003e[\u003cspan additionalcitationids=\"CR31\" citationid=\"CR28\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e32\u003c/span\u003e]\u003c/sup\u003e. However, tumorigenesis remains a multifaceted process influenced by various biological programs, so we seek to provide more comprehensive and accurate molecular subtyping for BLCA patients according to the combination of WES and intratumoral microbiome data on this basis. In the present study, three key genes, 90 key metabolites, and two key microbiota were identified by DEGs, SMGs, and DEMs, and the gene-metabolite-microbe regulatory network was constructed accordingly. Meanwhile, the BLCA samples were categorized into 2 subgroups depending on the key genes, which differed significantly in immunity, gene mutation, drug sensitivity, and clinicopathological features.\u003c/p\u003e \u003cp\u003eThe 3 key genes we selected including AHNAK, CSPG4, and NCAM1, were significantly under-expressed in BLCA patients, with their expression levels negatively correlating with overall survival, but their roles in BLCA have rarely been reported in previous studies\u003csup\u003e[\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e33\u003c/span\u003e, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e34\u003c/span\u003e]\u003c/sup\u003e. Ankyrin Repeat Domain-Containing Protein 1(AHNAK), a cytoskeletal protein, plays a crucial role in calcium homeostasis, muscle formation, and various biological processes like cell proliferation and signaling\u003csup\u003e[\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e35\u003c/span\u003e]\u003c/sup\u003e. Also, our KEGG enrichment indicated that the association between AHNAK and cell adhesion and the cytoskeleton in BLCA. In BLCA, AHNAK mainly behaves as an inhibitory oncogene, with significantly down-regulated expression in tumor patients, and the level of AHNAK in urine can be used as a biomarker to distinguish between uroepithelial carcinoma and normal uroepithelial cells\u003csup\u003e[\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e36\u003c/span\u003e]\u003c/sup\u003e, which is in keeping with our results. Previous studies concluded that AHNAK overexpression contributed to DNA damage repair (DDR) and cisplatin resistance, with NAT10 facilitating this process by recruiting LIG4 and XRCC4[34]. Chondroitin sulfate proteoglycan 4(CSPG4), a transmembrane protein, is aberrantly expressed in a variety of tumors and has been involved in tumor invasion, lymphovascular infiltration\u003csup\u003e[\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e37\u003c/span\u003e]\u003c/sup\u003e, and underexpressed in BLCA, impacting EMT, the immune microenvironment, and energy metabolism, and correlating negatively with patient prognosis\u003csup\u003e[\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e38\u003c/span\u003e]\u003c/sup\u003e. We also obtained similar conclusions, hinting at a possible engagement in cell migration and invasive processes in BLCA and a possible role in regulating key metabolites and microbes. Neural Cell Adhesion Molecule 1(NCAM1), belongs to the immunoglobulin superfamily of adhesion molecules and correlates with neurogenesis, neuronal synapse growth, proliferation, and cell migration, as described in several tumor entities such as gliomas, ovarian carcinomas, and small-cell lung carcinomas\u003csup\u003e[\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e39\u003c/span\u003e]\u003c/sup\u003e. It has also been characterized as a potential prognostic and targeted therapeutic cell-cycle related marker in BLCA\u003csup\u003e[\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e33\u003c/span\u003e]\u003c/sup\u003e. Here, we uncovered a significant correlation between NCAM1 and amino acid metabolism and immune infiltration and hypothesized that NCAM1 may modulate these processes to influence patient prognosis.\u003c/p\u003e \u003cp\u003eWe synthesized key genes, metabolites, and microbes to construct an interaction network, revealing that key genes were all negatively correlated with metabolites but did not appear to have a strong correlation with microbes. This suggested they may drive BLCA progression via metabolic pathways, while microbes were likely not gene-regulated. Consistent with previous studies\u003csup\u003e[\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]\u003c/sup\u003e, we elaborated key metabolites enriched in amino acid and nucleotide metabolic pathways, like purine, arginine, glycerophospholipid, and pyrimidine metabolism based on KEGG results. High extracellular adenosine deaminase activity in bladder cancer cell lines, as shown by Hesse et al.\u003csup\u003e[\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e40\u003c/span\u003e]\u003c/sup\u003e, interfered with antitumor immune responses by altering purine metabolism and impacting adenosine levels. Arginine metabolism also exercises anti-tumor immunity mainly through influencing components of the tumor microenvironment such as macrophages and T cells\u003csup\u003e[\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e41\u003c/span\u003e]\u003c/sup\u003e. The pyrimidine metabolism- associated enzyme UPP1 is a critical enzyme for the maintenance of uridine homeostasis, promoted BLCA cell proliferation and gemcitabine resistance via the activation of the AKT signaling pathway\u003csup\u003e[\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e42\u003c/span\u003e]\u003c/sup\u003e. Meanwhile, the functional enrichment analysis of differential microbes demonstrated connections to arginine and pyrimidine metabolism pathways. The key microbes, Sphingomonas koreensis, and Rhodospirillaceae, are environmental bacteria with potential roles in health and disease. Sphingomonas koreensis was first reported as a meningitis patient pathogen in 2015\u003csup\u003e[\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e43\u003c/span\u003e]\u003c/sup\u003e. Reduced levels of Rhodospirillaceae correlate with high pancreatic cancer metastatic potential, though its exact mechanisms in cancer progression remain unclear, potentially involving host immune system interactions and bacterial metabolite production\u003csup\u003e[\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e44\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eWe categorized BLCA patients into two clusters with respect to the 3 key genes and analyzed the functional differences between them. This analysis indicated that the differential pathways enriched in the subtypes primarily involved the extracellular matrix (ECM), cytoskeleton, adhesion mechanisms, and several inflammatory and tumor-related biological pathways, aligning with the functional enrichment results of the key genes. The ECM, a critical component of the tumor microenvironment, has an altered collagen composition linked to epithelial-mesenchymal transition (EMT), facilitating cell adhesion, migration, and BLCA progression\u003csup\u003e[\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e28\u003c/span\u003e]\u003c/sup\u003e. Additionally, our findings highlighted a positive correlation between BLCA subtypes and a broad range of microbes. Earlier efforts have documented microbial interactions with the ECM in the tumor microenvironment, where specific bacterial strains disrupt tight junctions within ECM components, enabling tissue colonization and inflammation, thus promoting ECM remodeling and generating reactive oxygen species that can lead to DNA damage and cancer recurrence\u003csup\u003e[\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e45\u003c/span\u003e]\u003c/sup\u003e. Consequently, the BLCA subtypes identified may influence tumor progression via microbial interactions with the ECM. Our GSEA outcomes revealed clustering around common tumor-related pathways such as EMT, interferon gamma response, TNFA signaling via NFkB, IL6 JAK STAT3 signaling, and the p53 pathway, all of which are crucial in BLCA growth, progression, DNA repair, and response to immunotherapy\u003csup\u003e[\u003cspan additionalcitationids=\"CR47\" citationid=\"CR44\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e48\u003c/span\u003e]\u003c/sup\u003e. Accordingly, we examined immune infiltration and mutation profiles across subtypes, observing notable differences in immune microenvironments and oncogenic mutation patterns. Cluster 2 displayed higher immune infiltration and checkpoint expression levels while exhibiting reduced responsiveness to chemotherapeutic agents like FGFR and EGFR inhibitors. Tumor cell growth and development are largely determined by their interactions with the surrounding microenvironment; a high tumor mutational burden (TMB) leads to the production of aberrant proteins recognizable by immune cells, eliciting an effective anti-tumor immune response\u003csup\u003e[\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e49\u003c/span\u003e]\u003c/sup\u003e. Concurrently, immune cell infiltration modulates this response, correlating directly with tumor progression and therapeutic outcomes\u003csup\u003e[\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e50\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eIn single cell analysis, we detected that key genes were overexpressed in CAF, speculating that CAF was the key cell type for the occurrence of BLCA. CAF could moderate tumor proliferation and drug resistance by promoting EMT, ECM remodeling, angiogenesis, and inhibiting anti-tumor immunity\u003csup\u003e[\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e51\u003c/span\u003e]\u003c/sup\u003e. This is in some congruence with our pathway enrichment results. In particular, iCAF was associated with cell adhesion and intercellular tight junction regulation, potentially enhancing tumor progression and metastasis. Chen et al. also noticed interactions between iCAFs and tumor cells, with secreted cytokines inducing an inflammatory tumor microenvironment with pro-tumorigenic properties, which could reduce the sensitivity of BLCA patients to chemotherapeutic agents and strengthen the tumor invasive and metastatic properties\u003csup\u003e[\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e52\u003c/span\u003e]\u003c/sup\u003e. Secondly, epithelial cells, as the predominant cell type in BLCA tissues, can exhibit a malignant phenotype to a certain extent\u003csup\u003e[\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e53\u003c/span\u003e]\u003c/sup\u003e. A subset of epithelial cells expressing N-calmodulin 2 (CDH12) showed specific invasive traits, such as chemoresistance and poor prognosis\u003csup\u003e[\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e54\u003c/span\u003e]\u003c/sup\u003e. Therefore, we also graded the degree of malignancy of the epithelial cells, and the expression level of AHNAK showed a negative correlation, which may be related to the property that AHNAK itself is a cytoskeletal protein.\u003c/p\u003e \u003cp\u003eTo summarize, we constructed a network of interactions among genes, metabolites, and microbes by combinatorial analysis of multi-omics data, including transcriptomics, whole-exome, metabolomics, and intratumoral microbiomics, which breaks the limitation of single-omics and contributes to a more profound understanding of the molecular mechanism of BLCA. The subtype analysis of BLCA based on key genes illustrated the discrepancies in biological pathways, immune microenvironment, genomic variants, immunotherapy, and drug sensitivity across subtypes. Nevertheless, there are some limitations that restrict our study. One is that the functions and regulatory mechanisms of key genes were mainly grounded in functional enrichment, which requires more experiments for further validation. The other is that the efficacy of the molecular subtypes we identified in real-world BLCA patients needs to be confirmed by more samples.\u003c/p\u003e"},{"header":"5. Conclusion","content":"\u003cp\u003eThis study successfully employed an integrated multi-omics approach to identify significant correlations between three pivotal genes\u0026mdash;AHNAK, CSPG4, and NCAM1\u0026mdash;and key metabolites and microorganisms in bladder cancer. The findings offer novel insights into the molecular mechanisms of BLCA and provide a comprehensive reference for developing targeted therapeutic strategies. These biomarkers and their interplay could enhance diagnostic accuracy, personalize treatments, and deepen our understanding of bladder cancer, potentially improving patient outcomes.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe would like to thank the researchers and staff of the above software and databases.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eH.W. and M.D. : Supervision, Project administration, Funding acquisition. Z.T. : Writing original draft, Investigation, Funding acquisition. X.C. : Bioinformatics analysis. Y.H. and S.F. : Interpretation of the data. H.L. and C.G. : Experimental design and execution. D.L. and C.Y. : Literature review. J.W. : Manuscript revision. All authors have read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research is supported by the National Natural Science Foundation of China (grant No. 82060464, grant No. 82260609, grant No.82360603). Yunnan Fundamental Research Projects (grant No. 202001AY070001-163, grant No. 202201AU070220, grant No. 202201AY070001-113, grant No. 202401AU070010). Yunnan Provincial Department of Education Project (grant No. 2024J0225).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDerived data supporting the methodology of this study are available from the corresponding author upon reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDeclaration of competing interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eHuman Ethics and Consent to Participate declarations\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eHuman Ethics Approval All procedures involving human participants were conducted in accordance with the ethical standards of the Ethical Committee of The Second Affiliated Hospital of Kunming Medical University and with the 1964 Helsinki Declaration and its later amendments. The study was approved by the Ethical Committee of The Second Affiliated Hospital of Kunming Medical University (FEY-BG-39-2.0). Consent to Participate Informed consent was obtained from all individual participants included in the study.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eDyrskj\u0026oslash;t L, Hansel DE, Efstathiou JA, Knowles MA, Galsky MD, Teoh J, Theodorescu D. Bladder cancer. Nat Rev Dis Primers. 2023 Oct 26;9(1):58. \u003c/li\u003e\n\u003cli\u003eGroeneveld CS, Sanchez-Quiles V, Dufour F, Shi M, Dingli F, Nicolle R, Chapeaublanc E, Poullet P, Jeffery D, Krucker C, Maill\u0026eacute; P, Vacherot F, Vordos D, Benhamou S, Lebret T, Micheau O, Zinovyev A, Loew D, Allory Y, de Reyni\u0026egrave;s A, Bernard-Pierrot I, Radvanyi F. Proteogenomic Characterization of Bladder Cancer Reveals Sensitivity to Apoptosis Induced by Tumor Necrosis Factor-related Apoptosis-inducing Ligand in FGFR3-mutated Tumors. Eur Urol. 2024 May;85(5):483-494. \u003c/li\u003e\n\u003cli\u003eLi Q, Damish AW, Frazier Z, Liu D, Reznichenko E, Kamburov A, Bell A, Zhao H, Jordan EJ, Gao SP, Ma J, Abbosh PH, Bellmunt J, Plimack ER, Lazaro JB, Solit DB, Bajorin D, Rosenberg JE, D\u0026apos;Andrea AD, Riaz N, Van Allen EM, Iyer G, Mouw KW. ERCC2 Helicase Domain Mutations Confer Nucleotide Excision Repair Deficiency and Drive Cisplatin Sensitivity in Muscle-Invasive Bladder Cancer. Clin Cancer Res. 2019 Feb 1;25(3):977-988.\u003c/li\u003e\n\u003cli\u003eWu G, Weng W, Xia P, Yan S, Zhong C, Xie L, Xie Y, Fan G. Wnt signalling pathway in bladder cancer. Cell Signal. 2021 Mar;79:109886. \u003c/li\u003e\n\u003cli\u003eKomura K, Hirosuna K, Tokushige S, Tsujino T, Nishimura K, Ishida M, Hayashi T, Ura A, Ohno T, Yamazaki S, Nakamori K, Kinoshita S, Maenosono R, Ajiro M, Yoshikawa Y, Takai T, Tsutsumi T, Taniguchi K, Tanaka T, Takahara K, Konuma T, Inamoto T, Hirose Y, Ono F, Shiraishi Y, Yoshimi A, Azuma H. The Impact of FGFR3 Alterations on the Tumor Microenvironment and the Efficacy of Immune Checkpoint Inhibitors in Bladder Cancer. Mol Cancer. 2023 Nov 18;22(1):185. \u003c/li\u003e\n\u003cli\u003eVantaku V, Donepudi SR, Piyarathna DWB, Amara CS, Ambati CR, Tang W, Putluri V, Chandrashekar DS, Varambally S, Terris MK, Davies K, Ambs S, Bollag R, Apolo AB, Sreekumar A, Putluri N. Large-scale profiling of serum metabolites in African American and European American patients with bladder cancer reveals metabolic pathways associated with patient survival. Cancer. 2019 Mar 15;125(6):921-932. \u003c/li\u003e\n\u003cli\u003eMeeks JJ, Al-Ahmadie H, Faltas BM, Taylor JA 3rd, Flaig TW, DeGraff DJ, Christensen E, Woolbright BL, McConkey DJ, Dyrskj\u0026oslash;t L. Genomic heterogeneity in bladder cancer: challenges and possible solutions to improve outcomes. Nat Rev Urol. 2020 May;17(5):259-270. \u003c/li\u003e\n\u003cli\u003eTan Z, Chen X, Zuo J, Fu S, Wang H, Wang J. Comprehensive analysis of scRNA-Seq and bulk RNA-Seq reveals dynamic changes in the tumor immune microenvironment of bladder cancer and establishes a prognostic model. J Transl Med. 2023 Mar 27;21(1):223. \u003c/li\u003e\n\u003cli\u003eLiang Y, Tan Y, Guan B, Guo B, Xia M, Li J, Shi Y, Yu Z, Zhang Q, Liu D, Yang X, Hao J, Gong Y, Shakeel M, Zhou L, Ci W, Li X. Single-cell atlases link macrophages and CD8\u003csup\u003e+ \u003c/sup\u003eT-cell subpopulations to disease progression and immunotherapy response in urothelial carcinoma. Theranostics. 2022 Nov 14;12(18):7745-7759. \u003c/li\u003e\n\u003cli\u003eLi J, Kong Z, Qi Y, Wang W, Su Q, Huang W, Zhang Z, Li S, Du E. Single-cell and bulk RNA-sequence identified fibroblasts signature and CD8\u003csup\u003e+\u003c/sup\u003e T-cell - fibroblast subtype predicting prognosis and immune therapeutic response of bladder cancer, based on machine learning: bioinformatics multi-omics study. Int J Surg. 2024 Aug 1;110(8):4911-4931. \u003c/li\u003e\n\u003cli\u003eLawson ARJ, Abascal F, Coorens THH, Hooks Y, O\u0026apos;Neill L, Latimer C, Raine K, Sanders MA, Warren AY, Mahbubani KTA, Bareham B, Butler TM, Harvey LMR, Cagan A, Menzies A, Moore L, Colquhoun AJ, Turner W, Thomas B, Gnanapragasam V, Williams N, Rassl DM, V\u0026ouml;hringer H, Zumalave S, Nangalia J, Tub\u0026iacute;o JMC, Gerstung M, Saeb-Parsy K, Stratton MR, Campbell PJ, Mitchell TJ, Martincorena I. Extensive heterogeneity in somatic mutation and selection in the human bladder. Science. 2020 Oct 2;370(6512):75-82. \u003c/li\u003e\n\u003cli\u003eLawson ARJ, Abascal F, Coorens THH, Hooks Y, O\u0026apos;Neill L, Latimer C, Raine K, Sanders MA, Warren AY, Mahbubani KTA, Bareham B, Butler TM, Harvey LMR, Cagan A, Menzies A, Moore L, Colquhoun AJ, Turner W, Thomas B, Gnanapragasam V, Williams N, Rassl DM, V\u0026ouml;hringer H, Zumalave S, Nangalia J, Tub\u0026iacute;o JMC, Gerstung M, Saeb-Parsy K, Stratton MR, Campbell PJ, Mitchell TJ, Martincorena I. Extensive heterogeneity in somatic mutation and selection in the human bladder. Science. 2020 Oct 2;370(6512):75-82. \u003c/li\u003e\n\u003cli\u003eMarkowski MC, Boorjian SA, Burton JP, Hahn NM, Ingersoll MA, Maleki Vareki S, Pal SK, Sfanos KS. The Microbiome and Genitourinary Cancer: A Collaborative Review. Eur Urol. 2019 Apr;75(4):637-646. \u003c/li\u003e\n\u003cli\u003eSun JX, Xia QD, Zhong XY, Liu Z, Wang SG. The bladder microbiome of NMIBC and MIBC patients revealed by 2bRAD-M. Front Cell Infect Microbiol. 2023 Jun 7;13:1182322. \u003c/li\u003e\n\u003cli\u003eWu P, Zhang G, Zhao J, Chen J, Chen Y, Huang W, Zhong J, Zeng J. Profiling the Urinary Microbiota in Male Patients With Bladder Cancer in China. Front Cell Infect Microbiol. 2018 May 31;8:167. Erratum in: Front Cell Infect Microbiol. 2018 Dec 11;8:429. \u003c/li\u003e\n\u003cli\u003eLiu J, Wang J, Wu M, Zhang W, Meng L, Wang J, Lv Z, Xia H, Zhang Y, Wang J. Comprehensive Analysis of N6-methyladenosine Modification Patterns Associated With Multiomic Characteristics of Bladder Cancer. Front Med (Lausanne). 2021 Dec 23;8:757432. \u003c/li\u003e\n\u003cli\u003eTh\u0026eacute;venot EA, Roux A, Xu Y, Ezan E, Junot C. Analysis of the Human Adult Urinary Metabolome Variations with Age, Body Mass Index, and Gender by Implementing a Comprehensive Workflow for Univariate and OPLS Statistical Analyses. J Proteome Res. 2015 Aug 7;14(8):3322-35. \u003c/li\u003e\n\u003cli\u003eLove MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15(12):550. \u003c/li\u003e\n\u003cli\u003eWu T, Hu E, Xu S, Chen M, Guo P, Dai Z, Feng T, Zhou L, Tang W, Zhan L, Fu X, Liu S, Bo X, Yu G. clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. Innovation (Camb). 2021 Jul 1;2(3):100141. \u003c/li\u003e\n\u003cli\u003eMayakonda A, Lin DC, Assenov Y, Plass C, Koeffler HP. Maftools: efficient and comprehensive analysis of somatic variants in cancer. Genome Res. 2018 Nov;28(11):1747-1756.\u003c/li\u003e\n\u003cli\u003eLiu TT, Li R, Huo C, Li JP, Yao J, Ji XL, Qu YQ. Identification of CDK2-Related Immune Forecast Model and ceRNA in Lung Adenocarcinoma, a Pan-Cancer Analysis. Front Cell Dev Biol. 2021 Jul 30;9:682002. \u003c/li\u003e\n\u003cli\u003eChen D, Cao Y, Tang H, Zang L, Yao N, Zhu Y, Jiang Y, Zhai S, Liu Y, Shi M, Zhao S, Wang W, Wen C, Peng C, Chen H, Deng X, Jiang L, Shen B. Comprehensive machine learning-generated classifier identifies pro-metastatic characteristics and predicts individual treatment in pancreatic cancer: A multicenter cohort study based on super-enhancer profiling. Theranostics. 2023 May 27;13(10):3290-3309. \u003c/li\u003e\n\u003cli\u003eHu FF, Liu CJ, Liu LL, Zhang Q, Guo AY. Expression profile of immune checkpoint genes and their roles in predicting immunotherapy response. Brief Bioinform. 2021 May 20;22(3):bbaa176. \u003c/li\u003e\n\u003cli\u003eChhabra Y, Weeraratna AT. Fibroblasts in cancer: Unity in heterogeneity. Cell. 2023 Apr 13;186(8):1580-1609. \u003c/li\u003e\n\u003cli\u003eElyada E, Bolisetty M, Laise P, Flynn WF, Courtois ET, Burkhart RA, Teinor JA, Belleau P, Biffi G, Lucito MS, Sivajothi S, Armstrong TD, Engle DD, Yu KH, Hao Y, Wolfgang CL, Park Y, Preall J, Jaffee EM, Califano A, Robson P, Tuveson DA. Cross-Species Single-Cell Analysis of Pancreatic Ductal Adenocarcinoma Reveals Antigen-Presenting Cancer-Associated Fibroblasts. Cancer Discov. 2019 Aug;9(8):1102-1123.\u003c/li\u003e\n\u003cli\u003eMa Z, Li X, Mao Y, Wei C, Huang Z, Li G, Yin J, Liang X, Liu Z. Interferon-dependent SLC14A1\u003csup\u003e+\u003c/sup\u003e cancer-associated fibroblasts promote cancer stemness via WNT5A in bladder cancer. Cancer Cell. 2022 Dec 12;40(12):1550-1565.e7. \u003c/li\u003e\n\u003cli\u003eAltman DG, McShane LM, Sauerbrei W, Taube SE. Reporting recommendations for tumor marker prognostic studies (REMARK): explanation and elaboration. BMC Med. 2012 May 29;10:51. \u003c/li\u003e\n\u003cli\u003eWang J, Zhu J, Hu J, Wang Z, Wang X, Pan J, Chu Y, Li Z, Jiang W, Liang C, Hou J, Guo J, Dang Y, Jiang S. A novel in vitro prognostic model of bladder cancer based on urine-derived living tumor cells. Genes Dis. 2022 Nov 26;10(6):2586-2596. \u003c/li\u003e\n\u003cli\u003eSung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, Bray F. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin. 2021 May;71(3):209-249. \u003c/li\u003e\n\u003cli\u003eHuang X, Du G, Yang Y, Su P, Chen S, Cai C, Huang T, Zeng Y, Tao Y, Tian D, Zhang N. Advancing bladder cancer management: development of a prognostic model and personalized therapy. Front Immunol. 2024 Jul 22;15:1430792. \u003c/li\u003e\n\u003cli\u003eSafder I, Valentine H, Uzzo N, Sfakianos J, Uzzo R, Gupta S, Brown J, Ranti D, Plimack E, Haber G, Weight C, Kutikov A, Abbosh P, Bukavina L. Identification and Validation of Prognostic Model for Tumor Microenvironment-Associated Genes in Bladder Cancer Based on Single-Cell RNA Sequencing Data Sets. JCO Precis Oncol. 2024 Aug;8:e2300661. \u003c/li\u003e\n\u003cli\u003eFeng ZH, Liang YP, Cen JJ, Yao HH, Lin HS, Li JY, Liang H, Wang Z, Deng Q, Cao JZ, Huang Y, Wei JH, Luo JH, Chen W, Chen ZH. m6A-immune-related lncRNA prognostic signature for predicting immune landscape and prognosis of bladder cancer. J Transl Med. 2022 Oct 29;20(1):492. \u003c/li\u003e\n\u003cli\u003eZeng X, Su H, Liu Z, Wang Y, Lu Z, Cheng S. Integrative analysis of the bladder cancer from a cell cycle NCAM1 perspective at both single cell and bulk resolution. Environ Toxicol. 2024 Apr 6. \u003c/li\u003e\n\u003cli\u003eXie R, Cheng L, Huang M, Huang L, Chen Z, Zhang Q, Li H, Lu J, Wang H, Zhou Q, Huang J, Chen X, Lin T. NAT10 Drives Cisplatin Chemoresistance by Enhancing ac4C-Associated DNA Repair in Bladder Cancer. Cancer Res. 2023 May 15;83(10):1666-1683. \u003c/li\u003e\n\u003cli\u003eLee JM, Lim TY, Oh SB, Lee SJ, Bae YS, Jung HS. Ahnak is required to balance calcium ion homeostasis and smooth muscle development in the urinary system. Cell Biosci. 2023 Jun 12;13(1):108. \u003c/li\u003e\n\u003cli\u003eLee H, Kim K, Woo J, Park J, Kim H, Lee KE, Kim H, Kim Y, Moon KC, Kim JY, Park IA, Shim BB, Moon JH, Han D, Ryu HS. Quantitative Proteomic Analysis Identifies AHNAK (Neuroblast Differentiation-associated Protein AHNAK) as a Novel Candidate Biomarker for Bladder Urothelial Carcinoma Diagnosis by Liquid-based Cytology. Mol Cell Proteomics. 2018 Sep;17(9):1788-1802. \u003c/li\u003e\n\u003cli\u003eEgan CE, Stefanova D, Ahmed A, Raja VJ, Thiesmeyer JW, Chen KJ, Greenberg JA, Zhang T, He B, Finnerty BM, Zarnegar R, Jin MM, Scognamiglio T, Dephoure N, Fahey T 3rd, Min IM. CSPG4 Is a Potential Therapeutic Target in Anaplastic Thyroid Cancer. Thyroid. 2021 Oct;31(10):1481-1493. \u003c/li\u003e\n\u003cli\u003eQiu Y, Ye W, Wang C, Zang J. Prognostic significance and immunoinfiltration analysis of genes associated with epithelial-mesenchymal transition and energy metabolism in bladder urothelial carcinoma. Aging (Albany NY). 2023 Nov 24;15(22):13312-13328. \u003c/li\u003e\n\u003cli\u003eSasca D, Szybinski J, Sch\u0026uuml;ler A, Shah V, Heidelberger J, Haehnel PS, Dolnik A, Kriege O, Fehr EM, Gebhardt WH, Reid G, Scholl C, Theobald M, Bullinger L, Beli P, Kindler T. NCAM1 (CD56) promotes leukemogenesis and confers drug resistance in AML. Blood. 2019 May 23;133(21):2305-2319. \u003c/li\u003e\n\u003cli\u003eHesse J, Steckel B, Dieterich P, Aydin S, Deussen A, Schrader J. Intercellular crosstalk shapes purinergic metabolism and signaling in cancer cells. Cell Rep. 2024 Jan 23;43(1):113643. \u003c/li\u003e\n\u003cli\u003eNowak M, Klink M. The Role of Tumor-Associated Macrophages in the Progression and Chemoresistance of Ovarian Cancer. Cells. 2020 May 22;9(5):1299. \u003c/li\u003e\n\u003cli\u003eDu W, Tu S, Zhang W, Zhang Y, Liu W, Xiong K, Zhou F, Li N, Zhang R, Yu J, Li M, Xiang W, Qian K, Wang G, Xiao Y, Wang X, Ju L. UPP1 enhances bladder cancer progression and gemcitabine resistance through AKT. Int J Biol Sci. 2024 Jan 27;20(4):1389-1409.\u003c/li\u003e\n\u003cli\u003eMarbjerg LH, Gaini S, Justesen US. First report of Sphingomonas koreensis as a human pathogen in a patient with meningitis. J Clin Microbiol. 2015 Mar;53(3):1028-30. \u003c/li\u003e\n\u003cli\u003eVillani A, Fontana A, Panebianco C, Ferro C, Copetti M, Pavlovic R, Drago D, Fiorentini C, Terracciano F, Bazzocchi F, Canistro G, Pisati F, Maiello E, Latiano TP, Perri F, Pazienza V. A powerful machine learning approach to identify interactions of differentially abundant gut microbial subsets in patients with metastatic and non-metastatic pancreatic cancer. Gut Microbes. 2024 Jan-Dec;16(1):2375483. \u003c/li\u003e\n\u003cli\u003eAlfano M, Canducci F, Nebuloni M, Clementi M, Montorsi F, Salonia A. The interplay of extracellular matrix and microbiome in urothelial bladder cancer. Nat Rev Urol. 2016 Feb;13(2):77-90. \u003c/li\u003e\n\u003cli\u003eBaker SC, Mason AS, Slip RG, Eriksson P, Sj\u0026ouml;dahl G, Trejdosiewicz LK, Southgate J. The Urothelial Transcriptomic Response to Interferon Gamma: Implications for Bladder Cancer Prognosis and Immunotherapy. Cancers (Basel). 2022 Oct 27;14(21):5295. \u003c/li\u003e\n\u003cli\u003eLi L, Jiang P, Hu W, Zou F, Li M, Rao T, Ruan Y, Yu W, Ning J, Cheng F. AURKB promotes bladder cancer progression by deregulating the p53 DNA damage response pathway via MAD2L2. J Transl Med. 2024 Mar 21;22(1):295. \u003c/li\u003e\n\u003cli\u003eAmara CS, Kami Reddy KR, Yuntao Y, Chan YS, Piyarathna DWB, Dobrolecki LE, Shih DJH, Shi Z, Xu J, Huang S, Ellis MJ, Apolo AB, Ballester LY, Gao J, Hansel DE, Lotan Y, Hodges HC, Lerner SP, Creighton CJ, Sreekumar A, Zheng WJ, Msaouel P, Kavuri SM, Putluri N. The IL6/JAK/STAT3 signaling axis is a therapeutic vulnerability in SMARCB1- deficient bladder cancer. Nat Commun. 2024 Feb 14;15(1):1373. \u003c/li\u003e\n\u003cli\u003evan Dorp J, van der Heijden MS. The bladder cancer immune micro-environment in the context of response to immune checkpoint inhibition. Front Immunol. 2023 Aug 31;14:1235884. \u003c/li\u003e\n\u003cli\u003eLow V, Li Z, Blenis J. Metabolite activation of tumorigenic signaling pathways in the tumor microenvironment. Sci Signal. 2022 Nov 8;15(759):eabj4220. \u003c/li\u003e\n\u003cli\u003eCaramelo B, Zagorac S, Corral S, Marqu\u0026eacute;s M, Real FX. Cancer-associated Fibroblasts in Bladder Cancer: Origin, Biology, and Therapeutic Opportunities. Eur Urol Oncol. 2023 Aug;6(4):366-375. \u003c/li\u003e\n\u003cli\u003eChen H, Yang W, Xue X, Li Y, Jin Z, Ji Z. Integrated Analysis Revealed an Inflammatory Cancer-Associated Fibroblast-Based Subtypes with Promising Implications in Predicting the Prognosis and Immunotherapeutic Response of Bladder Cancer Patients. Int J Mol Sci. 2022 Dec 15;23(24):15970. \u003c/li\u003e\n\u003cli\u003eWang J, Zuo Z, Yu Z, Chen Z, Tran LJ, Zhang J, Ao J, Ye F, Sun Z. Collaborating single-cell and bulk RNA sequencing for comprehensive characterization of the intratumor heterogeneity and prognostic model development for bladder cancer. Aging (Albany NY). 2023 Nov 6;15(21):12104-12119. \u003c/li\u003e\n\u003cli\u003eGouin KH 3rd, Ing N, Plummer JT, Rosser CJ, Ben Cheikh B, Oh C, Chen SS, Chan KS, Furuya H, Tourtellotte WG, Knott SRV, Theodorescu D. An N-Cadherin 2 expressing epithelial cell subpopulation predicts response to surgery, chemotherapy and immunotherapy in bladder cancer. Nat Commun. 2021 Aug 12;12(1):4906. \u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"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":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Bladder cancer, Multi-omics, Key genes, Key metabolites, Key microorganisms, Single-cell analysis","lastPublishedDoi":"10.21203/rs.3.rs-5898970/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5898970/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground: \u003c/strong\u003eBladder cancer (BLCA) is a common malignancy with significant impact on patient health. The aim of this study was to explore the potential mechanisms of BLCA through a combination of multi-omics and single-cell analyses.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods: \u003c/strong\u003eIn this study, samples from BLCA and paracancerous tissues were collected for transcriptome, whole-exome sequencing, metabolome and intratumoural microbiome sequencing. These data were then co-analyzed with publicly available datasets to identify and analyze key genes, metabolites and microbiomes as well as their regulatory mechanisms in the pathogenesis of BLCA. Different BLCA clusters were then identified on the basis of key genes. Differences among the clusters were then investigated in terms of biological pathways, immunological microenvironment, genetic alterations, immunotherapy and drug susceptibility. The prognostic value of the key genes was then analyzed using publicly available data, and their molecular regulatory mechanisms were further investigated. Finally, the expression patterns of the key genes were observed at the single cell level and key cells were identified.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults:\u003c/strong\u003e In this paper, three key genes (AHNAK, CSPG4, and NCAM1), 90 key metabolites and two key microorganisms (Sphingomonas koreensis and Rhodospirillaceae) were identified in a multi-omics analysis. Of these, key genes and key metabolites were negatively correlated. The BLCA samples from transcriptome sequencing were then divided into cluster 1 and cluster 2 based on key genes. Single-cell analysis identified nine cell types, with fibroblasts exhibiting the highest expression of key genes, thus establishing fibroblasts as the key cell in this study. Notably, AHNAK expression was higher in fibroblast subtypes.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion: \u003c/strong\u003eThe combined multi-omics analysis revealed a significant correlation between three key genes (AHNAK, CSPG4, and NCAM1) and multiple key metabolites and key microorganisms, which offering a new reference and theoretical support for the treatment and research of BLCA.\u003c/p\u003e","manuscriptTitle":"Integrated multi-omics analysis reveals key genetic, metabolic, and microbial drivers in bladder cancer insights into molecular subtyping and therapeutic approaches: A tumor marker prognostic study","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-01-28 18:52:35","doi":"10.21203/rs.3.rs-5898970/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"0fa49840-17ac-4a54-b0b8-e85dcf0f03a5","owner":[],"postedDate":"January 28th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-03-09T09:23:29+00:00","versionOfRecord":[],"versionCreatedAt":"2025-01-28 18:52:35","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5898970","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5898970","identity":"rs-5898970","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}