Machine Learning Identification of TSPAN7 as a Key Target Linking Type 2 Diabetes Mellitus and Colorectal Cancer

Machine Learning Identification of TSPAN7 as a Key Target Linking Type 2 Diabetes Mellitus and Colorectal Cancer | 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 Machine Learning Identification of TSPAN7 as a Key Target Linking Type 2 Diabetes Mellitus and Colorectal Cancer Feng Yu, Shuixia Yang, Yan Dong This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5651334/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 Type 2 Diabetes Mellitus (T2DM) and Colorectal Cancer (CRC) are significant global public health challenges with a notable epidemiological association. This study aims to explore the molecular mechanism behind this epidemiological association. Methods Weighted Gene Co-expression Network Analysis (WGCNA) and differential expression gene (DEG) analysis were conducted to identify shared genes between T2DM and CRC. Machine learning algorithms, including LASSO, Random Forest, and Support Vector Machine (SVM), were employed to identify hub genes. IOBR and clusterProfiler packages were used for immunoinfiltration assessment and enrichment analysis, respectively. Results We identified 27 shared genes between T2DM and CRC, with TSPAN7 emerging as a key hub gene linking the two conditions. TSPAN7 expression was significantly lower in disease groups compared to control groups across multiple cohorts, demonstrating excellent diagnostic accuracy. Enrichment analysis revealed involvement of these genes in various metabolic activities and pathways, including sulfur metabolism, selenium metabolism, renin secretion, pantothenate and CoA biosynthesis, TRP channel regulation, and efferocytosis. Conclusion This study provides new insights into the mechanisms underlying the association between T2DM and CRC by identifying TSPAN7 as a key target. The findings offer theoretical evidence for developing new diagnostic markers and therapeutic strategies for these diseases. Type 2 Diabetes Mellitus Colorectal Cancer TSPAN7 Machine Learning Gene Expression Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Introduction Type 2 diabetes mellitus (T2DM) and colorectal cancer (CRC) are major public health challenges globally, with a significant epidemiological association between them. Large-scale epidemiological studies have consistently shown that individuals with T2DM have a higher risk of developing CRC, with an estimated 1.3-1.5-fold increase in risk ( 1 ). This association may be influenced by factors such as metabolic disorders, chronic inflammation, and insulin resistance in T2DM patients. The global prevalence of T2DM is projected to reach 700 million by 2045( 2 ), underscoring the importance of understanding the link between T2DM and CRC. Traditional research on the association between T2DM and CRC has primarily focused on established biological processes, such as hyperglycemia-induced metabolic disturbances and oxidative stress, as well as insulin resistance leading to hyperinsulinemia ( 3 ). For example, high glucose levels can affect cellular metabolism and oxidative stress, playing a crucial role in the development of CRC ( 4 – 6 ). Insulin resistance, through the activation of the insulin-like growth factor-1 (IGF-1) signaling pathway, promotes tumor cell proliferation and survival ( 7 – 9 ). However, these studies often focus on specific signaling pathways or molecular mechanisms, limiting their ability to fully reveal the complex relationship between T2DM and CRC. TSPAN7, a member of the tetraspanin superfamily, has garnered attention for its role in various cancers. In bladder cancer, TSPAN7 inhibits cell proliferation by suppressing the p-PI3K and p-AKT signaling pathways ( 10 ). In lung cancer, TSPAN7 expression is associated with enhanced cell migration and invasion ( 11 ). These findings suggest that TSPAN7 may play a significant role in different types of cancer, but its role in the association between T2DM and CRC remains unclear. Machine learning (ML) techniques are increasingly being applied in biomedical research, particularly in handling high-dimensional and complex data. ML can analyze large-scale gene expression and clinical data to identify potential novel biomarkers and therapeutic targets. In disease research, ML has been successfully used for early diagnosis, prognosis prediction, and drug development ( 12 ). ML's strength lies in its ability to uncover hidden patterns and relationships from complex data, which is essential for understanding the complex link between T2DM and CRC. This study aims to use machine learning techniques to identify TSPAN7 as a key target linking T2DM and CRC from a systems biology perspective. By integrating multiple biological data sources, such as gene expression datasets and clinical sample information, and applying machine learning algorithms for in-depth analysis, this study seeks to elucidate the mechanisms by which TSPAN7 contributes to the association between T2DM and CRC. This research not only deepens our understanding of the pathogenesis of T2DM and CRC but also provides theoretical evidence for developing new diagnostic markers and therapeutic strategies, offering new hope for improving patient outcomes. Materials and Methods Data Collection and Preprocessing We downloaded transcriptomic data from two T2DM islet samples from the Gene Expression Omnibus (GEO, https://www.ncbi.nlm.nih.gov/gds/ ) database, including GSE76895 and GSE25724. GSE25724 was used as the Discovery cohort, while GSE76895 served as the Validation cohort. Additionally, we downloaded four CRC cohorts' transcriptomic data, including GSE110224, GSE41258, GSE39582, and GSE38832. GSE39582 was used as the Discovery cohort, while the remaining cohorts served as validation sets. Batch effects were processed using the sva package (Figure S1 ). We also utilized the TCGA-COAD cohort as a Validation cohort to assess the prognostic relevance of the key target. Differential Expression Analysis Differential expression analysis was conducted using the limma package, with an adjusted p-value threshold of < 0.05. For the T2DM cohort, we set a |log2(fold change)| threshold of 0.5, while for the CRC cohort, we set a |log2(fold change)| threshold of 1. Weighted Gene Co-expression Network Analysis (WGCNA) We used the WGCNA package version 1.73 to identify modules related to T2DM and CRC from GSE25724 and GSE39582. Hierarchical clustering was performed using the Hclust function. An appropriate soft threshold β (ranging from 1 to 30) was selected to meet scale-free network standards. The soft threshold β value and the gene correlation matrix were calculated using Pearson analysis for all gene pairs, followed by the construction of an adjacency matrix. The topological overlap matrix (TOM) and the corresponding dissimilarity (1-TOM) were classified into the adjacency matrix. A hierarchical clustering dendrogram was then constructed, and genes with similar expression profiles and ME-TOM were grouped into different modules. ME was used to summarize the characteristic vector of each module (CutHeight = 0.35). Finally, modules with high correlation coefficients with clinical features (|Cor| > 0.5 and p < 0.05) were selected for further study. Machine Learning First, we identified shared differentially expressed genes (DEGs) between T2DM and CRC by intersecting DEGs and WGCNA-derived module genes. Three machine learning algorithms—Least Absolute Shrinkage and Selection Operator (LASSO), Random Forest, and Support Vector Machine (SVM)—were used to identify hub genes in T2DM and CRC. LASSO analysis was performed using the glmnet package with 10-fold cross-validation, selecting hub genes with non-zero coefficients. Random Forest analysis was conducted using the randomForestSRC package with 10-fold cross-validation. SVM analysis was performed using the mRFE package with five-fold cross-validation. Immune Infiltration Assessment Immune infiltration was assessed using the IOBR package. The CIBERSORT algorithm was used to calculate immune cell infiltration scores for each cohort. Cohorts were divided into high and low expression groups based on the median expression of hub genes, and differences in immune cell infiltration between groups were compared. The Pearson correlation coefficient between hub gene expression and immune cell infiltration was calculated. Enrichment Analysis Enrichment analysis was performed using the clusterProfiler package. First, gene ontology (GO) annotation and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis were conducted for shared genes between CRC and T2DM. Second, high and low expression groups were created based on the median expression of hub genes, and group-wise differential expression analysis was performed. The results were ranked by log(fold change) and subjected to KEGG gene set enrichment analysis. Pathways significantly enriched in at least three cohorts were visualized. Protein-Protein Interaction (PPI) Analysis PPI networks for shared genes between T2DM and CRC were downloaded from the STRING database, with a minimum required interaction score of 0.15. The results were imported into Cytoscape for visualization and topological analysis. Results Identification of CRC-Related Genes Differential expression analysis of the GSE39582 cohort revealed 680 genes significantly upregulated and 625 genes significantly downregulated in CRC (Fig. 2 A, Supplementary materials: Table S1 ). WGCNA did not exclude any samples, with a β value of 5 (Figure S2 ), resulting in 31 modules (Fig. 2 B). By calculating the correlation between modules and disease, we selected the pink module to screen for CRC-related genes, identifying 352 genes (Fig. 2 C, Supplementary materials: Table S2 ). Ultimately, we obtained 26 significantly upregulated genes (Fig. 2 D) and 215 significantly downregulated genes (Fig. 2 E) related to CRC. Identification of T2DM-Related Genes Differential expression analysis of the GSE25724 cohort revealed 341 genes significantly upregulated and 705 genes significantly downregulated in T2DM (Fig. 3 A, Supplementary materials: Table S3 ). WGCNA did not exclude any samples, with a β value of 5 (Figure S3 ), resulting in 26 modules (Fig. 3 B). By calculating the correlation between modules and disease, we selected six modules (blue, purple, turquoise, salmon, yellow, cyan) to screen for T2DM-related genes, identifying 4298 genes (Fig. 3 C, Supplementary materials: Table S4 ). Ultimately, we obtained 325 significantly upregulated genes (Fig. 3 D) and 682 significantly downregulated genes (Fig. 3 E) related to T2DM. Identification of Shared Genes Between T2DM and CRC By intersecting the significantly upregulated and downregulated genes between T2DM and CRC, we identified 27 shared genes (Fig. 4 A, Supplementary materials: Table S5 ), including 3 shared upregulated genes and 24 shared downregulated genes. PPI analysis showed that 23 genes had 41 interactions, with PRKACB, SUCLG2, and HSD17B11 having the most interactions (9, 7, and 7, respectively) (Fig. 4 B). GO annotation enrichment analysis revealed that these shared genes were associated with molecular functions such as magnesium ion binding, manganese ion binding, and acyl-CoA binding (Fig. 4 C). KEGG pathway enrichment analysis showed that these genes were involved in pathways such as vascular smooth muscle contraction, sulfur metabolism, selenocompound metabolism, renin secretion, primary bile acid biosynthesis, pantothenate and CoA biosynthesis, glycan degradation, inflammatory mediator regulation of TRP channels, and efferocytosis (Fig. 4 D). Identification and Performance Evaluation of Hub Genes in T2DM and CRC We used LASSO, Random Forest, and SVM to identify hub genes in CRC (Figure S4 ) and T2DM (Figure S5 ). In CRC, we identified three hub genes—DYRK2, TSPAN7, and EDN3—by intersecting the results of the three algorithms (Fig. 5 A, Supplementary materials: Table S6 ). In T2DM, we identified five hub genes—TSPAN7, SCP2, PPM1A, GNE, and LGR4 (Fig. 5 B, Supplementary materials: Table S7 ). By intersecting the hub genes from both diseases, we identified TSPAN7 as a key hub gene linking T2DM and CRC (Fig. 5 C). TSPAN7 expression was significantly lower in the disease group compared to the control group in all cohorts except GSE76895 (Figs. 5 D- 5 H), and it demonstrated excellent diagnostic accuracy in all cohorts (Figs. 5 J- 5 K). In the TCGA-COAD cohort, there was no significant difference in disease-free survival (DFS) or overall survival (OS) between high and low TSPAN7 expression groups (Figs. 5 O- 5 P). Similarly, in the GSE38832 cohort, there was no significant difference in disease-specific survival (DSS) or DFS between high and low TSPAN7 expression groups (Figs. 5 Q- 5 R). Association of TSPAN7 with Immune Cell Infiltration To evaluate whether TSPAN7 is associated with immune cell infiltration, we divided all cohorts into high and low expression groups and compared the differences in immune cell infiltration between groups. In the CRC cohorts (Figs. 6 A- 6 B, Figure S6 A-S6B), we observed widespread differences in immune cell infiltration between high and low TSPAN7 expression groups. TSPAN7 was positively correlated with plasma cells, resting mast cells, M2 macrophages, and resting dendritic cells, while it was negatively correlated with NK cells, activated mast cells, M0 or M1 macrophages, and activated dendritic cells. In the T2DM cohort (Figs. 6 C- 6 D, Figure S6 C), TSPAN7 was positively correlated with resting mast cells and negatively correlated with activated dendritic cells. Pathways Associated with TSPAN7 in T2DM and CRC To elucidate the pathways associated with TSPAN7 in T2DM and CRC, we performed GSEA analysis on high and low expression groups in each cohort. In the GSE39582 cohort, pathways such as the cell cycle, DNA replication, ribosome, and proteasome were significantly activated in the high expression group, while pathways such as fatty acid degradation, retinol metabolism, and tyrosine metabolism were significantly inhibited (Fig. 7 A). In the GSE25724 cohort, pathways such as cytokine-cytokine receptor interaction and mineral absorption were significantly activated in the high expression group, while pathways such as renin secretion and fat digestion and absorption were significantly inhibited (Fig. 7 B). Comparing the enrichment results across all cohorts (Fig. 7 C and Figure S7 ), we identified 43 pathways with consistent expression trends, suggesting their association with TSPAN7, including pathways related to cell cycle regulation, metabolism, signal transduction, and secretion. Discussion The epidemiological association between T2DM and CRC has been widely established ( 1 ), but the underlying mechanisms remain poorly understood. In this study, we identified shared genes between T2DM and CRC using WGCNA and DEG analysis. WGCNA helped us identify gene modules highly correlated with T2DM and CRC, while DEG analysis allowed us to find genes with significant expression changes in both diseases. Ultimately, we identified 27 shared genes that showed consistent expression trends in both T2DM and CRC and were validated across multiple internal and external datasets, demonstrating their significant diagnostic and prognostic value. Enrichment analysis revealed that these genes are involved in various metabolic activities and pathways, including sulfur metabolism, selenium metabolism, renin secretion, TRP channel regulation, and efferocytosis. In T2DM, sulfur-containing amino acids are precursors to the antioxidant glutathione ( 13 , 14 ), and hydrogen sulfide can regulate insulin sensitivity ( 15 ). In CRC, gut microbial sulfur metabolites may promote inflammation and tumor development ( 16 ). Selenium metabolism has a non-linear relationship with T2DM ( 17 ), with moderate selenium levels beneficial for antioxidant activity and insulin sensitivity ( 18 ), while excessive selenium exacerbates insulin resistance ( 19 , 20 ). In CRC, selenium compounds exert anti-cancer effects through various selenoproteins, but genetic variations can influence their efficacy ( 21 – 23 ). Renin secretion, through the activation of the renin-angiotensin-aldosterone system, leads to hypertension and insulin resistance in T2DM ( 24 – 26 ), while angiotensin II, a downstream molecule, may promote tumor cell proliferation in CRC ( 27 , 28 ). TRP channels influence insulin secretion, pain perception, and lipid metabolism in T2DM ( 29 – 31 ) and are involved in cell proliferation, migration, and tumor microenvironment regulation in CRC ( 32 , 33 ). Efferocytosis dysfunction in T2DM can lead to chronic inflammation, worsening insulin resistance and affecting wound healing ( 34 – 36 ), while insufficient efferocytosis in CRC promotes inflammation and tumor development and affects treatment response ( 37 ). This study further identified TSPAN7 as a key target linking T2DM and CRC using machine learning techniques. Our results show that TSPAN7 expression is significantly upregulated in both T2DM and CRC patients and is closely associated with disease progression. Previous studies have extensively documented the abnormal expression and biological functions of TSPAN7 in diabetes and cancers. TSPAN7 is a pancreatic autoantigen involved in type 1 diabetes, regulating β-cell Ca2+-dependent exocytosis. In tumors, TSPAN7 plays a dual role: it can exert anti-tumor effects by inhibiting the PTEN/PI3K/AKT pathway in bladder cancer ( 10 ), while it can also promote tumor cell proliferation and metastasis through epithelial-to-mesenchymal transition (EMT) in lung cancer ( 11 ). However, whether TSPAN7 is a key target linking T2DM and CRC and the mechanisms behind this association remain unclear. Gene set enrichment analysis revealed that TSPAN7 is involved in multiple biological processes relevant to the development of T2DM and CRC, such as insulin signaling, cell proliferation, and inflammatory responses, providing clues for further mechanistic exploration. TSPAN7 may exacerbate insulin resistance in T2DM by influencing the insulin signaling pathway. Insulin resistance is a core pathological feature of T2DM, characterized by reduced tissue sensitivity to insulin and decreased glucose uptake and utilization ( 38 ). Hyperglycemia further promotes oxidative stress and the formation of advanced glycation end products (AGEs) ( 39 ), which can induce cellular toxicity, cause DNA damage, and promote mutations, thereby facilitating the development of CRC ( 40 ). Chronic inflammation serves as a bridge in the development of both T2DM and CRC ( 41 ). TSPAN7 may maintain a chronic inflammatory state by regulating the expression of inflammation-related genes ( 42 ). Inflammatory factors such as IL-6 and TNF-α contribute to the formation of insulin resistance and promote CRC cell proliferation and survival by activating transcription factors like NF-κB. ( 43 ) TSPAN7 exhibits pro-oncogenic properties in various cancers by regulating cell cycle-related genes and inhibiting apoptosis ( 10 , 44 ), promoting the malignant transformation of CRC cells. Therefore, the metabolic disorder caused by hyperglycemia and insulin resistance provides a rich nutritional environment for CRC cells, accelerating their proliferation. However, the specific mechanisms of TSPAN7 in these processes need to be further studied. Despite the insights provided by this study, several limitations exist. First, T2DM and CRC are multifactorial, multistep diseases, and interventions targeting a single gene may not fully explain their pathogenesis. Second, the data used in this study primarily come from public databases, which may be subject to batch effects and data biases. Conclusion This study identified 27 shared genes between T2DM and CRC using machine learning techniques, revealing their involvement in various biological processes. TSPAN7 was identified as a key target linking T2DM and CRC, with significant low expression in both diseases. Functional analysis indicated that TSPAN7 is involved in multiple biological processes relevant to the development of T2DM and CRC. This finding provides new insights into the mechanisms underlying the association between T2DM and CRC and offers theoretical evidence for developing new diagnostic markers and therapeutic strategies. Abbreviations T2DM: Type 2 diabetes mellitus CRC: colorectal cancer IGF-1: insulin-like growth factor-1 ML: Machine learning GEO: Gene Expression Omnibus WGCNA: Weighted Gene Co-expression Network Analysis TOM: topological overlap matrix DEG: differentially expressed genes LASSO: Least Absolute Shrinkage and Selection Operator SVM: Support Vector Machine GO: gene ontology KEGG: Kyoto Encyclopedia of Genes and Genomes DFS: disease-free survival OS: overall survival DSS: disease-specific survival EMT: epithelial-to-mesenchymal transition AGE: advanced glycation end products Declarations Ethics approval and consent to participate Not applicable Consent for publication Not applicable Availability of data and materials The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. Competing interests The authors declare that they have no competing interests Funding statement Not applicable Authors' contributions Feng Yu and Yan Dong formulated the idea of the article and supervised the research. Feng Yu, Shuixia Yang, Yan Dong performed the research, analyzed the data and wrote the manuscript. Yan Dong participated in revising the data and improving manuscript writing. All authors reviewed the manuscript, and all authors read and approved the final version of the manuscript. Acknowledgements Not applicable References Lawler T, Walts ZL, Steinwandel M, Lipworth L, Murff HJ, Zheng W, Warren AS. Type 2 Diabetes and Colorectal Cancer Risk. JAMA Netw Open 6 (11): e2343333, 2023. Saeedi P, Petersohn I, Salpea P, Malanda B, Karuranga S, Unwin N, Colagiuri S, Guariguata L, Motala AA, Ogurtsova K, Shaw JE, Bright D, Williams R. Global and regional diabetes prevalence estimates for 2019 and projections for 2030 and 2045: Results from the International Diabetes Federation Diabetes Atlas, 9(th) edition. Diabetes Res Clin Pract 157: 107843, 2019. Vekic J, Zeljkovic A, Stefanovic A, Giglio RV, Ciaccio M, Rizzo M. Diabetes and Colorectal Cancer Risk: A New Look at Molecular Mechanisms and Potential Role of Novel Antidiabetic Agents. Int J Mol Sci 22 (22), 2021. Mao X, Yang S, Zhang Y, Yang H, Yan D, Zhang L. The role of chromatin modulator DPY30 in glucose metabolism of colorectal cancer cells. Transl Cancer Res 13 (8): 4205-4218, 2024. Zhao Y, Ye X, Xiong Z, Ihsan A, Ares I, Martínez M, Lopez-Torres B, Martínez-Larrañaga MR, Anadón A, Wang X, Martínez MA. Cancer Metabolism: The Role of ROS in DNA Damage and Induction of Apoptosis in Cancer Cells. Metabolites 13 (7), 2023. Li S, Fang W, Zheng J, Peng Z, Yu B, Chen C, Zhang Y, Jiang W, Yuan S, Zhang L, Zhang X. Whole-transcriptome defines novel glucose metabolic subtypes in colorectal cancer. J Cell Mol Med 28 (5): e18065, 2024. Zhao M, Chen YL, Yang LH. Advancements in the study of glucose metabolism in relation to tumor progression and treatment. Prog Biophys Mol Biol 192: 11-18, 2024. Chiefari E, Mirabelli M, La Vignera S, Tanyolaç S, Foti DP, Aversa A, Brunetti A. Insulin Resistance and Cancer: In Search for a Causal Link. Int J Mol Sci 22 (20), 2021. Monteiro M, Zhang X, Yee D. Insulin promotes growth in breast cancer cells through the type I IGF receptor in insulin receptor deficient cells. Exp Cell Res 434 (1): 113862, 2024. Yu X, Li S, Pang M, Du Y, Xu T, Bai T, Yang K, Hu J, Zhu S, Wang L, Liu X. TSPAN7 Exerts Anti-Tumor Effects in Bladder Cancer Through the PTEN/PI3K/AKT Pathway. Front Oncol 10: 613869, 2020. Wang X, Lin M, Zhao J, Zhu S, Xu M, Zhou X. TSPAN7 promotes the migration and proliferation of lung cancer cells via epithelial-to-mesenchymal transition. Onco Targets Ther 11: 8815-8822, 2018. Chang CH, Lin CH, Lane HY. Machine Learning and Novel Biomarkers for the Diagnosis of Alzheimer's Disease. Int J Mol Sci 22 (5), 2021. Peeters WM, Gram M, Dias GJ, Vissers M, Hampton MB, Dickerhof N, Bekhit AE, Black MJ, Oxbøll J, Bayer S, Dickens M, Vitzel K, Sheard PW, Danielson KM, Hodges LD, Brønd JC, Bond J, Perry BG, Stoner L, Cornwall J, Rowlands DS. Changes to insulin sensitivity in glucose clearance systems and redox following dietary supplementation with a novel cysteine-rich protein: A pilot randomized controlled trial in humans with type-2 diabetes. Redox Biol 67: 102918, 2023. Ding Y, Wang S, Lu J. Unlocking the Potential: Amino Acids' Role in Predicting and Exploring Therapeutic Avenues for Type 2 Diabetes Mellitus. Metabolites 13 (9), 2023. Wang X, Tian R, Liang C, Jia Y, Zhao L, Xie Q, Huang F, Yuan H. Biomimetic nanoplatform with microbiome modulation and antioxidant functions ameliorating insulin resistance and pancreatic β-cell dysfunction for T2DM management. Biomaterials 313: 122804, 2025. Moon JY, Kye BH, Ko SH, Yoo RN. Sulfur Metabolism of the Gut Microbiome and Colorectal Cancer: The Threat to the Younger Generation. Nutrients 15 (8), 2023. Pyrzynska K, Sentkowska A. Selenium Species in Diabetes Mellitus Type 2. Biol Trace Elem Res 202 (7): 2993-3004, 2024. Ouyang J, Cai Y, Song Y, Gao Z, Bai R, Wang A. Potential Benefits of Selenium Supplementation in Reducing Insulin Resistance in Patients with Cardiometabolic Diseases: A Systematic Review and Meta-Analysis. Nutrients 14 (22), 2022. Zhao J, Zou H, Huo Y, Wei X, Li Y. Emerging roles of selenium on metabolism and type 2 diabetes. Front Nutr 9: 1027629, 2022. Zhao L, Carmean CM, Landeche M, Chellan B, Sargis RM. Selenomethionine modulates insulin secretion in the MIN6-K8 mouse insulinoma cell line. FEBS Lett 595 (24): 3042-3055, 2021. Zhang S, Zhang G, Wang P, Wang L, Fang B, Huang J. Effect of Selenium and Selenoproteins on Radiation Resistance. Nutrients 16 (17), 2024. Garbo S, Di Giacomo S, Łażewska D, Honkisz-Orzechowska E, Di Sotto A, Fioravanti R, Zwergel C, Battistelli C. Selenium-Containing Agents Acting on Cancer-A New Hope? Pharmaceutics 15 (1), 2022. Fedirko V, Jenab M, Méplan C, Jones JS, Zhu W, Schomburg L, Siddiq A, Hybsier S, Overvad K, Tjønneland A, Omichessan H, Perduca V, Boutron-Ruault MC, Kühn T, Katzke V, Aleksandrova K, Trichopoulou A, Karakatsani A, Kotanidou A, Tumino R, Panico S, Masala G, Agnoli C, Naccarati A, Bueno-de-Mesquita B, Vermeulen R, Weiderpass E, Skeie G, Nøst TH, Lujan-Barroso L, Quirós JR, Huerta JM, Rodríguez-Barranco M, Barricarte A, Gylling B, Harlid S, Bradbury KE, Wareham N, Khaw KT, Gunter M, Murphy N, Freisling H, Tsilidis K, Aune D, Riboli E, Hesketh JE, Hughes DJ. Association of Selenoprotein and Selenium Pathway Genotypes with Risk of Colorectal Cancer and Interaction with Selenium Status. Nutrients 11 (4), 2019. Barhoumi T, Todryk S. Role of monocytes/macrophages in renin-angiotensin system-induced hypertension and end organ damage. Front Physiol 14: 1199934, 2023. Delevatti RS, Leonel L, Rodrigues J, Kanitz AC, Alberton CL, Lovatel GA, Siqueira IR, Kruel L. Aerobic Exercise in the Aquatic Environment Suppresses the Plasma Renin Activity in Individuals with Type 2 Diabetes: A Secondary Analysis of a Randomized Clinical Trial. Int J Environ Res Public Health 21 (7), 2024. Caturano A, Galiero R, Vetrano E, Sardu C, Rinaldi L, Russo V, Monda M, Marfella R, Sasso FC. Insulin-Heart Axis: Bridging Physiology to Insulin Resistance. Int J Mol Sci 25 (15), 2024. Kuniyasu H. Multiple roles of angiotensin in colorectal cancer. World J Clin Oncol 3 (12): 150-4, 2012. Van Berlo B, Civati C, Esposito P, De Keulenaer GW, Guns PJDF, Segers VFM. Angiotensin II as a linking factor in cardiovascular disease enhanced cancer growth. European Heart Journal 45 (Supplement_1): ehae666.3205, 2024. Anand S, Rajagopal S. A Comprehensive Review on the Regulatory Action of TRP Channels: A Potential Therapeutic Target for Nociceptive Pain. Neurosci Insights 18: 26331055231220340, 2023. Rosenbaum T, Morales-Lázaro SL, Islas LD. TRP channels: a journey towards a molecular understanding of pain. Nat Rev Neurosci 23 (10): 596-610, 2022. Benzi A, Heine M, Spinelli S, Salis A, Worthmann A, Diercks B, Astigiano C, Pérez MR, Memushaj A, Sturla L, Vellone V, Damonte G, Jaeckstein MY, Koch-Nolte F, Mittrücker HW, Guse AH, De Flora A, Heeren J, Bruzzone S. The TRPM2 ion channel regulates metabolic and thermogenic adaptations in adipose tissue of cold-exposed mice. Front Endocrinol (Lausanne) 14: 1251351, 2024. Jiang S, Lin X, Wu Q, Zheng J, Cui Z, Cai X, Li Y, Zheng C, Sun Y. Transient receptor potential channels' genes forecast cervical cancer outcomes and illuminate its impact on tumor cells. Front Genet 15: 1391842, 2024. Liu Y, Yao X, Zhao W, Xu J, Zhang H, Huang T, Wu C, Yang J, Tang C, Ye Q, Hu W, Wang Q. A comprehensive analysis of TRP-related gene signature, and immune infiltration in patients with colorectal cancer. Discov Oncol 15 (1): 357, 2024. Tajbakhsh A, Gheibi HS, Butler AE, Sahebkar A. Effect of soluble cleavage products of important receptors/ligands on efferocytosis: Their role in inflammatory, autoimmune and cardiovascular disease. Ageing Res Rev 50: 43-57, 2019. Liu X, Liu H, Deng Y. Efferocytosis: An Emerging Therapeutic Strategy for Type 2 Diabetes Mellitus and Diabetes Complications. J Inflamm Res 16: 2801-2815, 2023. Zhao Y, Li M, Mao J, Su Y, Huang X, Xia W, Leng X, Zan T. Immunomodulation of wound healing leading to efferocytosis. Smart Med 3 (1): e20230036, 2024. Ma Z, Sun Y, Yu Y, Xiao W, Xiao Z, Zhong T, Xiang X, Li Z. Extracellular vesicles containing MFGE8 from colorectal cancer facilitate macrophage efferocytosis. Cell Commun Signal 22 (1): 295, 2024. Liu XH, Xu Q, Zhang L, Liu HJ. Association between metabolic score for insulin resistance and regression to normoglycemia from prediabetes in Chinese adults: A retrospective cohort study. PLoS One 19 (8): e0308343, 2024. Kawada A, Yoshitake S, Fujihara R, Ishikawa M. Relationship Between Oxidative Stress in the Rotator Cuff and Transcutaneous Advanced Glycation End-Products Measurement in Diabetic Rats. Cureus 16 (8): e67529, 2024. Azizian-Farsani F, Abedpoor N, Hasan SM, Gure AO, Nasr-Esfahani MH, Ghaedi K. Receptor for Advanced Glycation End Products Acts as a Fuel to Colorectal Cancer Development. Front Oncol 10: 552283, 2020. Yang S, Li Y, Zhang Y, Wang Y. Impact of chronic stress on intestinal mucosal immunity in colorectal cancer progression. Cytokine Growth Factor Rev 80: 24-36, 2024. Shao S, Piao L, Guo L, Wang J, Wang L, Wang J, Tong L, Yuan X, Zhu J, Fang S, Wang Y. Tetraspanin 7 promotes osteosarcoma cell invasion and metastasis by inducing EMT and activating the FAK-Src-Ras-ERK1/2 signaling pathway. Cancer Cell Int 22 (1): 183, 2022. Zhan S, Wang L, Wang W, Li R. Insulin resistance in NSCLC: unraveling the link between development, diagnosis, and treatment. Front Endocrinol (Lausanne) 15: 1328960, 2024. Chen L, Liu H, Li Y, Lin X, Xia S, Wanggou S, Li X. Functional characterization of TSPAN7 as a novel indicator for immunotherapy in glioma. Front Immunol 14: 1105489, 2023. Table Table 1. Summary of T2DM and CRC cohorts used in this study. GEO Accession Platform Samples Disease Group GSE76895 GPL570 islet samples from 36 T2DM and 32 normal T2DM Validation cohort GSE25724 GPL96 islet samples from 6 T2DM and 7 normal T2DM Discovery cohort GSE110224 GPL570 primary adenocarcinomas and matched normal samples from 7 patient CRC Validation cohort GSE41258 GPL96 316 CRC and 74 normal control CRC Validation cohort GSE39582 GPL570 566 CRC and 19 normal control CRC Discovery cohort GSE38832 GPL570 122 CRC CRC Validation cohort TCGA-COAD Illumina 556 Patients and 28 controls CRC Validation cohort Additional Declarations No competing interests reported. Supplementary Files TableS1.xlsx Table S1. The differentially expressed genes in CRC. TableS2.xlsx Table S2. Genes associated with CRC identified by WGCNA. TableS3.xlsx Table S3. The differentially expressed genes in T2DM. TableS4.xlsx Table S4. Genes associated with T2DM identified by WGCNA. TableS5.xlsx Table S5. 27 genes shared between CRC and T2DM. TableS6.xlsx Table S6. Hub genes of CRC identified by machine learning. TableS7.xlsx Table S7. Hub genes of T2DM identified by machine learning. FigureS1.tif Supplementary Materials Figure S1. Batch effect correction in T2DM and CRC cohorts. PCA plots before (A) and after (B) batch effect correction in the T2DM cohort; PCA plots before (C) and after (D) batch effect correction in the CRC cohort. FigureS2.tif Figure S2. WGCNA analysis of the GSE39582 cohort. (A) Sample dendrogram and trait heatmap. (B) Scale independence showing the scale-free topology model fit (Model Fitting R²) against different soft threshold (power) values. (C) Mean connectivity of the network across various soft threshold values. FigureS3.tif Figure S3. WGCNA analysis of the GSE25724 cohort. (A) Sample dendrogram and trait heatmap. (B) Scale independence showing the scale-free topology model fit (Model Fitting R²) against different soft threshold (power) values. (C) Mean connectivity of the network across various soft threshold values. FigureS4.tif Figure S4. Identification of hub genes related to CRC. (A) Confidence intervals at different lambda values in LASSO regression. (B) Trajectory of independent variables in LASSO regression. (C) Cross-validation error distribution in the random forest algorithm. (D) Cross-validation error distribution in the SVM algorithm. FigureS5.tif Figure S5. Identification of hub genes related to T2DM. (A) Confidence intervals at different lambda values in LASSO regression. (B) Trajectory of independent variables in LASSO regression. (C) Cross-validation error distribution in the random forest algorithm. (D) Cross-validation error distribution in the SVM algorithm. FigureS6.tif Figure S6. Correlation between TSPAN7 gene expression and immune cell infiltration in validation cohorts. (A) GSE110224, (B) GSE41258, and (C) GSE76895 cohorts: Comparison of immune cell infiltration in high vs. low TSPAN7 expression groups and Pearson correlation between TSPAN7 expression and immune cell infiltration. FigureS7.tif Figure S7. Gene set enrichment analysis in high vs. low TSPAN7 expression groups in validation cohorts. (A) GSE110224, (B) GSE41258, and (C) GSE76895 cohorts: Gene set analysis related to TSPAN7 expression. 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-5651334","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":391533748,"identity":"6b3475ba-debe-454a-890b-7961c5c3c86e","order_by":0,"name":"Feng Yu","email":"","orcid":"","institution":"No .906 Hospital of the People’s Liberation Army","correspondingAuthor":false,"prefix":"","firstName":"Feng","middleName":"","lastName":"Yu","suffix":""},{"id":391533749,"identity":"f6b17057-dd84-49ea-b806-29ed0f99b6d3","order_by":1,"name":"Shuixia Yang","email":"","orcid":"","institution":"No .906 Hospital of the People’s Liberation Army","correspondingAuthor":false,"prefix":"","firstName":"Shuixia","middleName":"","lastName":"Yang","suffix":""},{"id":391533750,"identity":"16a9b104-20d3-4a21-a2ce-5ce6ef62d1ee","order_by":2,"name":"Yan Dong","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA3ElEQVRIie3PMQrCMBSA4RcqnaJdU4R6hYDgouhVGgQ3T+BgRYiDehen4lgJtEuKs6N0dai4KAiaOjk1GQXzD4HA+0gegM32iwknUucAeyASCM0Iqsgk8BdpaEiSDxFdKiQ1+1crQ9y/Px0WpfJWnvdD8Fbr+sd8gXi7yV222OQxYXIMROa7WkIrgiLMliSPgXEHKJnqifoYYbxzKUrG52aEYJd2MUhQUOiJ2mXZb/IwIJD2FMmwdpfWURxO9+cLjxJRXB98FnirbT1RNcj3DevGq5zSZMpms9n+uDdUpE2n1frfegAAAABJRU5ErkJggg==","orcid":"","institution":"No .906 Hospital of the People’s Liberation Army","correspondingAuthor":true,"prefix":"","firstName":"Yan","middleName":"","lastName":"Dong","suffix":""}],"badges":[],"createdAt":"2024-12-16 07:23:15","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5651334/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5651334/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":71902744,"identity":"e20d4259-c7ad-422d-b5bc-91208cab71f7","added_by":"auto","created_at":"2024-12-19 14:48:41","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":202297,"visible":true,"origin":"","legend":"\u003cp\u003eThe workflow diagram of this study, outlining the methodology and analysis steps.\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-5651334/v1/f338da8c2810364240db47ff.png"},{"id":71902740,"identity":"90248853-79ad-4446-96c2-64742ceb1b0b","added_by":"auto","created_at":"2024-12-19 14:48:41","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":505342,"visible":true,"origin":"","legend":"\u003cp\u003eIdentification of CRC-associated genes. (A) Volcano plot of differentially expressed genes in the GSE39582 cohort. (B) Dendrogram of co-expression network module clustering in CRC, with distinct modules indicated by colors. (C) Module-trait relationships map. (D) Venn diagram of upregulated genes associated with CRC. (E) Venn diagram of downregulated genes associated with CRC.\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-5651334/v1/2a70edc470a3c67777d777f4.png"},{"id":71904558,"identity":"458c49cb-803e-416e-ba63-d1eb1b8d69f9","added_by":"auto","created_at":"2024-12-19 14:56:41","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":483982,"visible":true,"origin":"","legend":"\u003cp\u003eIdentification of T2DM-associated genes. (A) Volcano plot of differentially expressed genes in the GSE25724 cohort. (B) Dendrogram of co-expression network module clustering in T2DM, with distinct modules indicated by colors. (C) Module-trait relationships map. (D) Venn diagram of upregulated genes associated with T2DM. (E) Venn diagram of downregulated genes associated with T2DM.\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-5651334/v1/8de54c9cf2b7bd9f27a41eb5.png"},{"id":71902747,"identity":"91a33e18-24d9-41d5-9fb6-f4f16c4b0666","added_by":"auto","created_at":"2024-12-19 14:48:41","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":477875,"visible":true,"origin":"","legend":"\u003cp\u003eIdentification of shared genes between T2DM and CRC. (A) Venn diagram identifying shared genes between T2DM and CRC. (B) Protein-protein interaction network among the shared genes. (C) GO annotation enrichment analysis and (D) KEGG pathway enrichment analysis of the shared genes.\u003c/p\u003e","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-5651334/v1/dc04bad570fc7b818768f1d3.png"},{"id":71902754,"identity":"9d8c599b-8280-4ab8-935f-98688b321eed","added_by":"auto","created_at":"2024-12-19 14:48:42","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":485915,"visible":true,"origin":"","legend":"\u003cp\u003eIdentification of hub genes in T2DM and CRC. (A) Venn diagram of hub genes in CRC identified by three machine learning algorithms. (B) Venn diagram of hub genes in T2DM identified by three machine learning algorithms. (C) Venn diagram of hub genes common to T2DM and CRC. (D-H) Differential expression analysis of TSPAN7 in disease vs. control groups across GSE76895, GSE25724, GSE110224, GSE41258, and GSE39582 cohorts. (I-M) ROC curve analysis for TSPAN7 gene expression in disease diagnosis across the same cohorts. (N-O) KM survival curve analysis of high vs. low TSPAN7 expression in the TCGA-COAD cohort. (P-Q) KM survival curve analysis of high vs. low TSPAN7 expression in the GSE38832 cohort.\u003c/p\u003e","description":"","filename":"Figure5.png","url":"https://assets-eu.researchsquare.com/files/rs-5651334/v1/3504c9de7e1eb29a49b8b90f.png"},{"id":71902749,"identity":"0a4d6d14-4e7a-40af-8bde-d97bf4914d47","added_by":"auto","created_at":"2024-12-19 14:48:42","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":545828,"visible":true,"origin":"","legend":"\u003cp\u003eCorrelation between TSPAN7 gene expression and immune cell infiltration. (A) Comparison of immune cell infiltration in high vs. low TSPAN7 expression groups in the GSE39582 cohort. (B) Pearson correlation between TSPAN7 expression and immune cell infiltration in the GSE39582 cohort. (C) Comparison of immune cell infiltration in high vs. low TSPAN7 expression groups in the GSE25724 cohort. (D) Pearson correlation between TSPAN7 expression and immune cell infiltration in the GSE25724 cohort.\u003c/p\u003e","description":"","filename":"Figure6.png","url":"https://assets-eu.researchsquare.com/files/rs-5651334/v1/aced7075dab77345d100c12d.png"},{"id":71902766,"identity":"237edc0e-f73e-4d3f-89eb-dbd2a0a71fb8","added_by":"auto","created_at":"2024-12-19 14:48:42","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":733902,"visible":true,"origin":"","legend":"\u003cp\u003ePathway identification related to TSPAN7 in T2DM and CRC. (A) Gene set analysis of TSPAN7 expression in the GSE39582 cohort. (B) Gene set analysis of TSPAN7 expression in the GSE25724 cohort. (C) Pathways associated with TSPAN7 in both T2DM and CRC.\u003c/p\u003e","description":"","filename":"Figure7.png","url":"https://assets-eu.researchsquare.com/files/rs-5651334/v1/aae6d8a1e7b03e155b7bae1b.png"},{"id":72031607,"identity":"132dd592-dad8-4c7e-bcb1-7f68f016c3cf","added_by":"auto","created_at":"2024-12-20 21:01:38","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3798411,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5651334/v1/d00fdefb-4ef2-4d01-a121-9e02c10213fe.pdf"},{"id":71904921,"identity":"6c90f788-df08-4a3e-a846-f35171737fdc","added_by":"auto","created_at":"2024-12-19 15:04:41","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":169701,"visible":true,"origin":"","legend":"\u003cp\u003eTable S1. The differentially expressed genes in CRC.\u003c/p\u003e","description":"","filename":"TableS1.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-5651334/v1/15233ce4f39193c55b649af2.xlsx"},{"id":71902758,"identity":"50578060-f9c8-4628-bf1c-6e6678f6d803","added_by":"auto","created_at":"2024-12-19 14:48:42","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":16885,"visible":true,"origin":"","legend":"\u003cp\u003eTable S2. Genes associated with CRC identified by WGCNA.\u003c/p\u003e","description":"","filename":"TableS2.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-5651334/v1/119a683ef13522658add806a.xlsx"},{"id":71902741,"identity":"2b991bd6-9fae-417e-b804-037b9de61c61","added_by":"auto","created_at":"2024-12-19 14:48:41","extension":"xlsx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":131476,"visible":true,"origin":"","legend":"\u003cp\u003eTable S3. The differentially expressed genes in T2DM.\u003c/p\u003e","description":"","filename":"TableS3.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-5651334/v1/01795fbd0eaf5d6c16a46beb.xlsx"},{"id":71902769,"identity":"374d52ae-35f2-4f0f-895b-9a7994fc14ab","added_by":"auto","created_at":"2024-12-19 14:48:42","extension":"xlsx","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":98304,"visible":true,"origin":"","legend":"\u003cp\u003eTable S4. Genes associated with T2DM identified by WGCNA.\u003c/p\u003e","description":"","filename":"TableS4.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-5651334/v1/8cfcfdce14328021eaae17c8.xlsx"},{"id":71902764,"identity":"27d032ba-846c-45ac-9471-02ad6590e018","added_by":"auto","created_at":"2024-12-19 14:48:42","extension":"xlsx","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":10072,"visible":true,"origin":"","legend":"\u003cp\u003eTable S5. 27 genes shared between CRC and T2DM.\u003c/p\u003e","description":"","filename":"TableS5.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-5651334/v1/b21c5c69e44c6031cdd9b9a1.xlsx"},{"id":71904561,"identity":"17b02a59-de8b-45ad-8d79-31603a945f0d","added_by":"auto","created_at":"2024-12-19 14:56:42","extension":"xlsx","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":10052,"visible":true,"origin":"","legend":"\u003cp\u003eTable S6. Hub genes of CRC identified by machine learning.\u003c/p\u003e","description":"","filename":"TableS6.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-5651334/v1/05f5193544c9a4e4010c379c.xlsx"},{"id":71902776,"identity":"8da1a927-720b-48ce-afc5-fa0aec0a9594","added_by":"auto","created_at":"2024-12-19 14:48:43","extension":"xlsx","order_by":7,"title":"","display":"","copyAsset":false,"role":"supplement","size":10219,"visible":true,"origin":"","legend":"\u003cp\u003eTable S7. Hub genes of T2DM identified by machine learning.\u003c/p\u003e","description":"","filename":"TableS7.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-5651334/v1/9aba0b0ea691967f8cfa378a.xlsx"},{"id":71902773,"identity":"120697ea-2b34-49af-935a-1ba1293b60bb","added_by":"auto","created_at":"2024-12-19 14:48:42","extension":"tif","order_by":8,"title":"","display":"","copyAsset":false,"role":"supplement","size":3989536,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSupplementary Materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFigure S1. Batch effect correction in T2DM and CRC cohorts. PCA plots before (A) and after (B) batch effect correction in the T2DM cohort; PCA plots before (C) and after (D) batch effect correction in the CRC cohort.\u003c/p\u003e","description":"","filename":"FigureS1.tif","url":"https://assets-eu.researchsquare.com/files/rs-5651334/v1/5923dfdf0dbd9b7292ab1bed.tif"},{"id":71904572,"identity":"55f215ca-900a-452c-b450-b80176dc437b","added_by":"auto","created_at":"2024-12-19 14:56:43","extension":"tif","order_by":9,"title":"","display":"","copyAsset":false,"role":"supplement","size":5095332,"visible":true,"origin":"","legend":"\u003cp\u003eFigure S2. WGCNA analysis of the GSE39582 cohort. (A) Sample dendrogram and trait heatmap. (B) Scale independence showing the scale-free topology model fit (Model Fitting R²) against different soft threshold (power) values. (C) Mean connectivity of the network across various soft threshold values.\u003c/p\u003e","description":"","filename":"FigureS2.tif","url":"https://assets-eu.researchsquare.com/files/rs-5651334/v1/34aa695ccabc30b0c3a62408.tif"},{"id":71902772,"identity":"5d72806f-a787-4ab5-ae4c-04898a843dc3","added_by":"auto","created_at":"2024-12-19 14:48:42","extension":"tif","order_by":10,"title":"","display":"","copyAsset":false,"role":"supplement","size":2805276,"visible":true,"origin":"","legend":"\u003cp\u003eFigure S3. WGCNA analysis of the GSE25724 cohort. (A) Sample dendrogram and trait heatmap. (B) Scale independence showing the scale-free topology model fit (Model Fitting R²) against different soft threshold (power) values. (C) Mean connectivity of the network across various soft threshold values.\u003c/p\u003e","description":"","filename":"FigureS3.tif","url":"https://assets-eu.researchsquare.com/files/rs-5651334/v1/1e6e1a29f074d739a0e2e568.tif"},{"id":71902770,"identity":"1ff44361-a021-4175-94ec-05a834f82636","added_by":"auto","created_at":"2024-12-19 14:48:42","extension":"tif","order_by":11,"title":"","display":"","copyAsset":false,"role":"supplement","size":3113428,"visible":true,"origin":"","legend":"\u003cp\u003eFigure S4. Identification of hub genes related to CRC. (A) Confidence intervals at different lambda values in LASSO regression. (B) Trajectory of independent variables in LASSO regression. (C) Cross-validation error distribution in the random forest algorithm. (D) Cross-validation error distribution in the SVM algorithm.\u003c/p\u003e","description":"","filename":"FigureS4.tif","url":"https://assets-eu.researchsquare.com/files/rs-5651334/v1/4bd3d99ae45648c3f30f0771.tif"},{"id":71902779,"identity":"c506ab76-de15-49ee-8b45-80afdacd32ce","added_by":"auto","created_at":"2024-12-19 14:48:43","extension":"tif","order_by":12,"title":"","display":"","copyAsset":false,"role":"supplement","size":3414484,"visible":true,"origin":"","legend":"\u003cp\u003eFigure S5. Identification of hub genes related to T2DM. (A) Confidence intervals at different lambda values in LASSO regression. (B) Trajectory of independent variables in LASSO regression. (C) Cross-validation error distribution in the random forest algorithm. (D) Cross-validation error distribution in the SVM algorithm.\u003c/p\u003e","description":"","filename":"FigureS5.tif","url":"https://assets-eu.researchsquare.com/files/rs-5651334/v1/17606b086f2e803b61d92ac0.tif"},{"id":71902760,"identity":"aaa76d9b-5378-447d-86be-51f142ade0e5","added_by":"auto","created_at":"2024-12-19 14:48:42","extension":"tif","order_by":13,"title":"","display":"","copyAsset":false,"role":"supplement","size":7768164,"visible":true,"origin":"","legend":"\u003cp\u003eFigure S6. Correlation between TSPAN7 gene expression and immune cell infiltration in validation cohorts. (A) GSE110224, (B) GSE41258, and (C) GSE76895 cohorts: Comparison of immune cell infiltration in high vs. low TSPAN7 expression groups and Pearson correlation between TSPAN7 expression and immune cell infiltration.\u003c/p\u003e","description":"","filename":"FigureS6.tif","url":"https://assets-eu.researchsquare.com/files/rs-5651334/v1/942c33699f9c28be723d6e88.tif"},{"id":71904573,"identity":"c07eb899-0760-4a3d-9a8f-de9bc433d596","added_by":"auto","created_at":"2024-12-19 14:56:44","extension":"tif","order_by":14,"title":"","display":"","copyAsset":false,"role":"supplement","size":7356216,"visible":true,"origin":"","legend":"\u003cp\u003eFigure S7. Gene set enrichment analysis in high vs. low TSPAN7 expression groups in validation cohorts. (A) GSE110224, (B) GSE41258, and (C) GSE76895 cohorts: Gene set analysis related to TSPAN7 expression.\u003c/p\u003e","description":"","filename":"FigureS7.tif","url":"https://assets-eu.researchsquare.com/files/rs-5651334/v1/3060658ee95cbf17c06bd664.tif"}],"financialInterests":"No competing interests reported.","formattedTitle":"Machine Learning Identification of TSPAN7 as a Key Target Linking Type 2 Diabetes Mellitus and Colorectal Cancer","fulltext":[{"header":"Introduction","content":"\u003cp\u003eType 2 diabetes mellitus (T2DM) and colorectal cancer (CRC) are major public health challenges globally, with a significant epidemiological association between them. Large-scale epidemiological studies have consistently shown that individuals with T2DM have a higher risk of developing CRC, with an estimated 1.3-1.5-fold increase in risk (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e). This association may be influenced by factors such as metabolic disorders, chronic inflammation, and insulin resistance in T2DM patients. The global prevalence of T2DM is projected to reach 700\u0026nbsp;million by 2045(\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e), underscoring the importance of understanding the link between T2DM and CRC.\u003c/p\u003e \u003cp\u003eTraditional research on the association between T2DM and CRC has primarily focused on established biological processes, such as hyperglycemia-induced metabolic disturbances and oxidative stress, as well as insulin resistance leading to hyperinsulinemia (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e). For example, high glucose levels can affect cellular metabolism and oxidative stress, playing a crucial role in the development of CRC (\u003cspan additionalcitationids=\"CR5\" citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e). Insulin resistance, through the activation of the insulin-like growth factor-1 (IGF-1) signaling pathway, promotes tumor cell proliferation and survival (\u003cspan additionalcitationids=\"CR8\" citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e). However, these studies often focus on specific signaling pathways or molecular mechanisms, limiting their ability to fully reveal the complex relationship between T2DM and CRC.\u003c/p\u003e \u003cp\u003eTSPAN7, a member of the tetraspanin superfamily, has garnered attention for its role in various cancers. In bladder cancer, TSPAN7 inhibits cell proliferation by suppressing the p-PI3K and p-AKT signaling pathways (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e). In lung cancer, TSPAN7 expression is associated with enhanced cell migration and invasion (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e). These findings suggest that TSPAN7 may play a significant role in different types of cancer, but its role in the association between T2DM and CRC remains unclear.\u003c/p\u003e \u003cp\u003eMachine learning (ML) techniques are increasingly being applied in biomedical research, particularly in handling high-dimensional and complex data. ML can analyze large-scale gene expression and clinical data to identify potential novel biomarkers and therapeutic targets. In disease research, ML has been successfully used for early diagnosis, prognosis prediction, and drug development (\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e). ML's strength lies in its ability to uncover hidden patterns and relationships from complex data, which is essential for understanding the complex link between T2DM and CRC.\u003c/p\u003e \u003cp\u003eThis study aims to use machine learning techniques to identify TSPAN7 as a key target linking T2DM and CRC from a systems biology perspective. By integrating multiple biological data sources, such as gene expression datasets and clinical sample information, and applying machine learning algorithms for in-depth analysis, this study seeks to elucidate the mechanisms by which TSPAN7 contributes to the association between T2DM and CRC. This research not only deepens our understanding of the pathogenesis of T2DM and CRC but also provides theoretical evidence for developing new diagnostic markers and therapeutic strategies, offering new hope for improving patient outcomes.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eData Collection and Preprocessing\u003c/h2\u003e \u003cp\u003eWe downloaded transcriptomic data from two T2DM islet samples from the Gene Expression Omnibus (GEO, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.ncbi.nlm.nih.gov/gds/\u003c/span\u003e\u003cspan address=\"https://www.ncbi.nlm.nih.gov/gds/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) database, including GSE76895 and GSE25724. GSE25724 was used as the Discovery cohort, while GSE76895 served as the Validation cohort. Additionally, we downloaded four CRC cohorts' transcriptomic data, including GSE110224, GSE41258, GSE39582, and GSE38832. GSE39582 was used as the Discovery cohort, while the remaining cohorts served as validation sets. Batch effects were processed using the sva package (Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). We also utilized the TCGA-COAD cohort as a Validation cohort to assess the prognostic relevance of the key target.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eDifferential Expression Analysis\u003c/h3\u003e\n\u003cp\u003eDifferential expression analysis was conducted using the limma package, with an adjusted p-value threshold of \u0026lt;\u0026thinsp;0.05. For the T2DM cohort, we set a |log2(fold change)| threshold of 0.5, while for the CRC cohort, we set a |log2(fold change)| threshold of 1.\u003c/p\u003e\n\u003ch3\u003eWeighted Gene Co-expression Network Analysis (WGCNA)\u003c/h3\u003e\n\u003cp\u003eWe used the WGCNA package version 1.73 to identify modules related to T2DM and CRC from GSE25724 and GSE39582. Hierarchical clustering was performed using the Hclust function. An appropriate soft threshold β (ranging from 1 to 30) was selected to meet scale-free network standards. The soft threshold β value and the gene correlation matrix were calculated using Pearson analysis for all gene pairs, followed by the construction of an adjacency matrix. The topological overlap matrix (TOM) and the corresponding dissimilarity (1-TOM) were classified into the adjacency matrix. A hierarchical clustering dendrogram was then constructed, and genes with similar expression profiles and ME-TOM were grouped into different modules. ME was used to summarize the characteristic vector of each module (CutHeight\u0026thinsp;=\u0026thinsp;0.35). Finally, modules with high correlation coefficients with clinical features (|Cor| \u0026gt; 0.5 and p\u0026thinsp;\u0026lt;\u0026thinsp;0.05) were selected for further study.\u003c/p\u003e\n\u003ch3\u003eMachine Learning\u003c/h3\u003e\n\u003cp\u003eFirst, we identified shared differentially expressed genes (DEGs) between T2DM and CRC by intersecting DEGs and WGCNA-derived module genes. Three machine learning algorithms\u0026mdash;Least Absolute Shrinkage and Selection Operator (LASSO), Random Forest, and Support Vector Machine (SVM)\u0026mdash;were used to identify hub genes in T2DM and CRC. LASSO analysis was performed using the glmnet package with 10-fold cross-validation, selecting hub genes with non-zero coefficients. Random Forest analysis was conducted using the randomForestSRC package with 10-fold cross-validation. SVM analysis was performed using the mRFE package with five-fold cross-validation.\u003c/p\u003e\n\u003ch3\u003eImmune Infiltration Assessment\u003c/h3\u003e\n\u003cp\u003eImmune infiltration was assessed using the IOBR package. The CIBERSORT algorithm was used to calculate immune cell infiltration scores for each cohort. Cohorts were divided into high and low expression groups based on the median expression of hub genes, and differences in immune cell infiltration between groups were compared. The Pearson correlation coefficient between hub gene expression and immune cell infiltration was calculated.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eEnrichment Analysis\u003c/h2\u003e \u003cp\u003eEnrichment analysis was performed using the clusterProfiler package. First, gene ontology (GO) annotation and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis were conducted for shared genes between CRC and T2DM. Second, high and low expression groups were created based on the median expression of hub genes, and group-wise differential expression analysis was performed. The results were ranked by log(fold change) and subjected to KEGG gene set enrichment analysis. Pathways significantly enriched in at least three cohorts were visualized.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eProtein-Protein Interaction (PPI) Analysis\u003c/h3\u003e\n\u003cp\u003ePPI networks for shared genes between T2DM and CRC were downloaded from the STRING database, with a minimum required interaction score of 0.15. The results were imported into Cytoscape for visualization and topological analysis.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eIdentification of CRC-Related Genes\u003c/h2\u003e \u003cp\u003eDifferential expression analysis of the GSE39582 cohort revealed 680 genes significantly upregulated and 625 genes significantly downregulated in CRC (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e2\u003c/span\u003eA, Supplementary materials: Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). WGCNA did not exclude any samples, with a β value of 5 (Figure \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e), resulting in 31 modules (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e2\u003c/span\u003eB). By calculating the correlation between modules and disease, we selected the pink module to screen for CRC-related genes, identifying 352 genes (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e2\u003c/span\u003eC, Supplementary materials: Table \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e). Ultimately, we obtained 26 significantly upregulated genes (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e2\u003c/span\u003eD) and 215 significantly downregulated genes (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e2\u003c/span\u003eE) related to CRC.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eIdentification of T2DM-Related Genes\u003c/h2\u003e \u003cp\u003eDifferential expression analysis of the GSE25724 cohort revealed 341 genes significantly upregulated and 705 genes significantly downregulated in T2DM (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e3\u003c/span\u003eA, Supplementary materials: Table \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003e). WGCNA did not exclude any samples, with a β value of 5 (Figure \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003e), resulting in 26 modules (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). By calculating the correlation between modules and disease, we selected six modules (blue, purple, turquoise, salmon, yellow, cyan) to screen for T2DM-related genes, identifying 4298 genes (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e3\u003c/span\u003eC, Supplementary materials: Table \u003cspan refid=\"MOESM4\" class=\"InternalRef\"\u003eS4\u003c/span\u003e). Ultimately, we obtained 325 significantly upregulated genes (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e3\u003c/span\u003eD) and 682 significantly downregulated genes (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e3\u003c/span\u003eE) related to T2DM.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eIdentification of Shared Genes Between T2DM and CRC\u003c/h2\u003e \u003cp\u003eBy intersecting the significantly upregulated and downregulated genes between T2DM and CRC, we identified 27 shared genes (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e4\u003c/span\u003eA, Supplementary materials: Table \u003cspan refid=\"MOESM5\" class=\"InternalRef\"\u003eS5\u003c/span\u003e), including 3 shared upregulated genes and 24 shared downregulated genes. PPI analysis showed that 23 genes had 41 interactions, with PRKACB, SUCLG2, and HSD17B11 having the most interactions (9, 7, and 7, respectively) (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e4\u003c/span\u003eB). GO annotation enrichment analysis revealed that these shared genes were associated with molecular functions such as magnesium ion binding, manganese ion binding, and acyl-CoA binding (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e4\u003c/span\u003eC). KEGG pathway enrichment analysis showed that these genes were involved in pathways such as vascular smooth muscle contraction, sulfur metabolism, selenocompound metabolism, renin secretion, primary bile acid biosynthesis, pantothenate and CoA biosynthesis, glycan degradation, inflammatory mediator regulation of TRP channels, and efferocytosis (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e4\u003c/span\u003eD).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eIdentification and Performance Evaluation of Hub Genes in T2DM and CRC\u003c/h2\u003e \u003cp\u003eWe used LASSO, Random Forest, and SVM to identify hub genes in CRC (Figure \u003cspan refid=\"MOESM4\" class=\"InternalRef\"\u003eS4\u003c/span\u003e) and T2DM (Figure \u003cspan refid=\"MOESM5\" class=\"InternalRef\"\u003eS5\u003c/span\u003e). In CRC, we identified three hub genes\u0026mdash;DYRK2, TSPAN7, and EDN3\u0026mdash;by intersecting the results of the three algorithms (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e5\u003c/span\u003eA, Supplementary materials: Table \u003cspan refid=\"MOESM6\" class=\"InternalRef\"\u003eS6\u003c/span\u003e). In T2DM, we identified five hub genes\u0026mdash;TSPAN7, SCP2, PPM1A, GNE, and LGR4 (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e5\u003c/span\u003eB, Supplementary materials: Table \u003cspan refid=\"MOESM7\" class=\"InternalRef\"\u003eS7\u003c/span\u003e). By intersecting the hub genes from both diseases, we identified TSPAN7 as a key hub gene linking T2DM and CRC (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e5\u003c/span\u003eC). TSPAN7 expression was significantly lower in the disease group compared to the control group in all cohorts except GSE76895 (Figs.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e5\u003c/span\u003eD-\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e5\u003c/span\u003eH), and it demonstrated excellent diagnostic accuracy in all cohorts (Figs.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e5\u003c/span\u003eJ-\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e5\u003c/span\u003eK). In the TCGA-COAD cohort, there was no significant difference in disease-free survival (DFS) or overall survival (OS) between high and low TSPAN7 expression groups (Figs.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e5\u003c/span\u003eO-\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e5\u003c/span\u003eP). Similarly, in the GSE38832 cohort, there was no significant difference in disease-specific survival (DSS) or DFS between high and low TSPAN7 expression groups (Figs.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e5\u003c/span\u003eQ-\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e5\u003c/span\u003eR).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eAssociation of TSPAN7 with Immune Cell Infiltration\u003c/h2\u003e \u003cp\u003eTo evaluate whether TSPAN7 is associated with immune cell infiltration, we divided all cohorts into high and low expression groups and compared the differences in immune cell infiltration between groups. In the CRC cohorts (Figs.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e6\u003c/span\u003eA-\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e6\u003c/span\u003eB, Figure \u003cspan refid=\"MOESM6\" class=\"InternalRef\"\u003eS6\u003c/span\u003eA-S6B), we observed widespread differences in immune cell infiltration between high and low TSPAN7 expression groups. TSPAN7 was positively correlated with plasma cells, resting mast cells, M2 macrophages, and resting dendritic cells, while it was negatively correlated with NK cells, activated mast cells, M0 or M1 macrophages, and activated dendritic cells. In the T2DM cohort (Figs.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e6\u003c/span\u003eC-\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e6\u003c/span\u003eD, Figure \u003cspan refid=\"MOESM6\" class=\"InternalRef\"\u003eS6\u003c/span\u003eC), TSPAN7 was positively correlated with resting mast cells and negatively correlated with activated dendritic cells.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003ePathways Associated with TSPAN7 in T2DM and CRC\u003c/h2\u003e \u003cp\u003eTo elucidate the pathways associated with TSPAN7 in T2DM and CRC, we performed GSEA analysis on high and low expression groups in each cohort. In the GSE39582 cohort, pathways such as the cell cycle, DNA replication, ribosome, and proteasome were significantly activated in the high expression group, while pathways such as fatty acid degradation, retinol metabolism, and tyrosine metabolism were significantly inhibited (Fig.\u0026nbsp;\u003cspan refid=\"Fig13\" class=\"InternalRef\"\u003e7\u003c/span\u003eA). In the GSE25724 cohort, pathways such as cytokine-cytokine receptor interaction and mineral absorption were significantly activated in the high expression group, while pathways such as renin secretion and fat digestion and absorption were significantly inhibited (Fig.\u0026nbsp;\u003cspan refid=\"Fig13\" class=\"InternalRef\"\u003e7\u003c/span\u003eB). Comparing the enrichment results across all cohorts (Fig.\u0026nbsp;\u003cspan refid=\"Fig13\" class=\"InternalRef\"\u003e7\u003c/span\u003eC and Figure \u003cspan refid=\"MOESM7\" class=\"InternalRef\"\u003eS7\u003c/span\u003e), we identified 43 pathways with consistent expression trends, suggesting their association with TSPAN7, including pathways related to cell cycle regulation, metabolism, signal transduction, and secretion.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe epidemiological association between T2DM and CRC has been widely established (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e), but the underlying mechanisms remain poorly understood. In this study, we identified shared genes between T2DM and CRC using WGCNA and DEG analysis. WGCNA helped us identify gene modules highly correlated with T2DM and CRC, while DEG analysis allowed us to find genes with significant expression changes in both diseases. Ultimately, we identified 27 shared genes that showed consistent expression trends in both T2DM and CRC and were validated across multiple internal and external datasets, demonstrating their significant diagnostic and prognostic value.\u003c/p\u003e \u003cp\u003eEnrichment analysis revealed that these genes are involved in various metabolic activities and pathways, including sulfur metabolism, selenium metabolism, renin secretion, TRP channel regulation, and efferocytosis. In T2DM, sulfur-containing amino acids are precursors to the antioxidant glutathione (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e), and hydrogen sulfide can regulate insulin sensitivity (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e). In CRC, gut microbial sulfur metabolites may promote inflammation and tumor development (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e). Selenium metabolism has a non-linear relationship with T2DM (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e), with moderate selenium levels beneficial for antioxidant activity and insulin sensitivity (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e), while excessive selenium exacerbates insulin resistance (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e). In CRC, selenium compounds exert anti-cancer effects through various selenoproteins, but genetic variations can influence their efficacy (\u003cspan additionalcitationids=\"CR22\" citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e). Renin secretion, through the activation of the renin-angiotensin-aldosterone system, leads to hypertension and insulin resistance in T2DM (\u003cspan additionalcitationids=\"CR25\" citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e), while angiotensin II, a downstream molecule, may promote tumor cell proliferation in CRC (\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e). TRP channels influence insulin secretion, pain perception, and lipid metabolism in T2DM (\u003cspan additionalcitationids=\"CR30\" citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e) and are involved in cell proliferation, migration, and tumor microenvironment regulation in CRC (\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e). Efferocytosis dysfunction in T2DM can lead to chronic inflammation, worsening insulin resistance and affecting wound healing (\u003cspan additionalcitationids=\"CR35\" citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e), while insufficient efferocytosis in CRC promotes inflammation and tumor development and affects treatment response (\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThis study further identified TSPAN7 as a key target linking T2DM and CRC using machine learning techniques. Our results show that TSPAN7 expression is significantly upregulated in both T2DM and CRC patients and is closely associated with disease progression. Previous studies have extensively documented the abnormal expression and biological functions of TSPAN7 in diabetes and cancers. TSPAN7 is a pancreatic autoantigen involved in type 1 diabetes, regulating β-cell Ca2+-dependent exocytosis. In tumors, TSPAN7 plays a dual role: it can exert anti-tumor effects by inhibiting the PTEN/PI3K/AKT pathway in bladder cancer (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e), while it can also promote tumor cell proliferation and metastasis through epithelial-to-mesenchymal transition (EMT) in lung cancer (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e). However, whether TSPAN7 is a key target linking T2DM and CRC and the mechanisms behind this association remain unclear. Gene set enrichment analysis revealed that TSPAN7 is involved in multiple biological processes relevant to the development of T2DM and CRC, such as insulin signaling, cell proliferation, and inflammatory responses, providing clues for further mechanistic exploration.\u003c/p\u003e \u003cp\u003eTSPAN7 may exacerbate insulin resistance in T2DM by influencing the insulin signaling pathway. Insulin resistance is a core pathological feature of T2DM, characterized by reduced tissue sensitivity to insulin and decreased glucose uptake and utilization (\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e). Hyperglycemia further promotes oxidative stress and the formation of advanced glycation end products (AGEs) (\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e), which can induce cellular toxicity, cause DNA damage, and promote mutations, thereby facilitating the development of CRC (\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e). Chronic inflammation serves as a bridge in the development of both T2DM and CRC (\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e). TSPAN7 may maintain a chronic inflammatory state by regulating the expression of inflammation-related genes (\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e). Inflammatory factors such as IL-6 and TNF-α contribute to the formation of insulin resistance and promote CRC cell proliferation and survival by activating transcription factors like NF-κB. (\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e) TSPAN7 exhibits pro-oncogenic properties in various cancers by regulating cell cycle-related genes and inhibiting apoptosis (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e), promoting the malignant transformation of CRC cells. Therefore, the metabolic disorder caused by hyperglycemia and insulin resistance provides a rich nutritional environment for CRC cells, accelerating their proliferation. However, the specific mechanisms of TSPAN7 in these processes need to be further studied.\u003c/p\u003e \u003cp\u003eDespite the insights provided by this study, several limitations exist. First, T2DM and CRC are multifactorial, multistep diseases, and interventions targeting a single gene may not fully explain their pathogenesis. Second, the data used in this study primarily come from public databases, which may be subject to batch effects and data biases.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThis study identified 27 shared genes between T2DM and CRC using machine learning techniques, revealing their involvement in various biological processes. TSPAN7 was identified as a key target linking T2DM and CRC, with significant low expression in both diseases. Functional analysis indicated that TSPAN7 is involved in multiple biological processes relevant to the development of T2DM and CRC. This finding provides new insights into the mechanisms underlying the association between T2DM and CRC and offers theoretical evidence for developing new diagnostic markers and therapeutic strategies.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eT2DM: Type 2 diabetes mellitus\u003c/p\u003e\n\u003cp\u003eCRC: colorectal cancer\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIGF-1: insulin-like growth factor-1\u003c/p\u003e\n\u003cp\u003eML: Machine learning\u003c/p\u003e\n\u003cp\u003eGEO: Gene Expression Omnibus\u003c/p\u003e\n\u003cp\u003eWGCNA: Weighted Gene Co-expression Network Analysis\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTOM: topological overlap matrix\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eDEG: differentially expressed genes\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eLASSO: Least Absolute Shrinkage and Selection Operator\u003c/p\u003e\n\u003cp\u003eSVM: Support Vector Machine\u003c/p\u003e\n\u003cp\u003eGO: gene ontology\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eKEGG: Kyoto Encyclopedia of Genes and Genomes\u003c/p\u003e\n\u003cp\u003eDFS: disease-free survival\u003c/p\u003e\n\u003cp\u003eOS: overall survival\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eDSS: disease-specific survival\u003c/p\u003e\n\u003cp\u003eEMT: epithelial-to-mesenchymal transition\u003c/p\u003e\n\u003cp\u003eAGE: advanced glycation end products\u0026nbsp;\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026apos; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFeng Yu and Yan Dong formulated the idea of the article and supervised the research. Feng Yu, Shuixia Yang, Yan Dong performed the research, analyzed the data and wrote the manuscript. Yan Dong participated in revising the data and improving manuscript writing. All authors reviewed the manuscript, and all authors read and approved the final version of the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eLawler T, Walts ZL, Steinwandel M, Lipworth L, Murff HJ, Zheng W, Warren AS. Type 2 Diabetes and Colorectal Cancer Risk. JAMA Netw Open 6 (11): e2343333, 2023.\u003c/li\u003e\n\u003cli\u003eSaeedi P, Petersohn I, Salpea P, Malanda B, Karuranga S, Unwin N, Colagiuri S, Guariguata L, Motala AA, Ogurtsova K, Shaw JE, Bright D, Williams R. Global and regional diabetes prevalence estimates for 2019 and projections for 2030 and 2045: Results from the International Diabetes Federation Diabetes Atlas, 9(th) edition. Diabetes Res Clin Pract 157: 107843, 2019.\u003c/li\u003e\n\u003cli\u003eVekic J, Zeljkovic A, Stefanovic A, Giglio RV, Ciaccio M, Rizzo M. Diabetes and Colorectal Cancer Risk: A New Look at Molecular Mechanisms and Potential Role of Novel Antidiabetic Agents. Int J Mol Sci 22 (22), 2021.\u003c/li\u003e\n\u003cli\u003eMao X, Yang S, Zhang Y, Yang H, Yan D, Zhang L. The role of chromatin modulator DPY30 in glucose metabolism of colorectal cancer cells. Transl Cancer Res 13 (8): 4205-4218, 2024.\u003c/li\u003e\n\u003cli\u003eZhao Y, Ye X, Xiong Z, Ihsan A, Ares I, Mart\u0026iacute;nez M, Lopez-Torres B, Mart\u0026iacute;nez-Larra\u0026ntilde;aga MR, Anad\u0026oacute;n A, Wang X, Mart\u0026iacute;nez MA. Cancer Metabolism: The Role of ROS in DNA Damage and Induction of Apoptosis in Cancer Cells. Metabolites 13 (7), 2023.\u003c/li\u003e\n\u003cli\u003eLi S, Fang W, Zheng J, Peng Z, Yu B, Chen C, Zhang Y, Jiang W, Yuan S, Zhang L, Zhang X. Whole-transcriptome defines novel glucose metabolic subtypes in colorectal cancer. J Cell Mol Med 28 (5): e18065, 2024.\u003c/li\u003e\n\u003cli\u003eZhao M, Chen YL, Yang LH. Advancements in the study of glucose metabolism in relation to tumor progression and treatment. Prog Biophys Mol Biol 192: 11-18, 2024.\u003c/li\u003e\n\u003cli\u003eChiefari E, Mirabelli M, La Vignera S, Tanyola\u0026ccedil; S, Foti DP, Aversa A, Brunetti A. Insulin Resistance and Cancer: In Search for a Causal Link. Int J Mol Sci 22 (20), 2021.\u003c/li\u003e\n\u003cli\u003eMonteiro M, Zhang X, Yee D. Insulin promotes growth in breast cancer cells through the type I IGF receptor in insulin receptor deficient cells. Exp Cell Res 434 (1): 113862, 2024.\u003c/li\u003e\n\u003cli\u003eYu X, Li S, Pang M, Du Y, Xu T, Bai T, Yang K, Hu J, Zhu S, Wang L, Liu X. TSPAN7 Exerts Anti-Tumor Effects in Bladder Cancer Through the PTEN/PI3K/AKT Pathway. Front Oncol 10: 613869, 2020.\u003c/li\u003e\n\u003cli\u003eWang X, Lin M, Zhao J, Zhu S, Xu M, Zhou X. TSPAN7 promotes the migration and proliferation of lung cancer cells via epithelial-to-mesenchymal transition. Onco Targets Ther 11: 8815-8822, 2018.\u003c/li\u003e\n\u003cli\u003eChang CH, Lin CH, Lane HY. Machine Learning and Novel Biomarkers for the Diagnosis of Alzheimer\u0026apos;s Disease. Int J Mol Sci 22 (5), 2021.\u003c/li\u003e\n\u003cli\u003ePeeters WM, Gram M, Dias GJ, Vissers M, Hampton MB, Dickerhof N, Bekhit AE, Black MJ, Oxb\u0026oslash;ll J, Bayer S, Dickens M, Vitzel K, Sheard PW, Danielson KM, Hodges LD, Br\u0026oslash;nd JC, Bond J, Perry BG, Stoner L, Cornwall J, Rowlands DS. Changes to insulin sensitivity in glucose clearance systems and redox following dietary supplementation with a novel cysteine-rich protein: A pilot randomized controlled trial in humans with type-2 diabetes. Redox Biol 67: 102918, 2023.\u003c/li\u003e\n\u003cli\u003eDing Y, Wang S, Lu J. Unlocking the Potential: Amino Acids\u0026apos; Role in Predicting and Exploring Therapeutic Avenues for Type 2 Diabetes Mellitus. Metabolites 13 (9), 2023.\u003c/li\u003e\n\u003cli\u003eWang X, Tian R, Liang C, Jia Y, Zhao L, Xie Q, Huang F, Yuan H. Biomimetic nanoplatform with microbiome modulation and antioxidant functions ameliorating insulin resistance and pancreatic \u0026beta;-cell dysfunction for T2DM management. Biomaterials 313: 122804, 2025.\u003c/li\u003e\n\u003cli\u003eMoon JY, Kye BH, Ko SH, Yoo RN. Sulfur Metabolism of the Gut Microbiome and Colorectal Cancer: The Threat to the Younger Generation. Nutrients 15 (8), 2023.\u003c/li\u003e\n\u003cli\u003ePyrzynska K, Sentkowska A. Selenium Species in Diabetes Mellitus Type 2. Biol Trace Elem Res 202 (7): 2993-3004, 2024.\u003c/li\u003e\n\u003cli\u003eOuyang J, Cai Y, Song Y, Gao Z, Bai R, Wang A. Potential Benefits of Selenium Supplementation in Reducing Insulin Resistance in Patients with Cardiometabolic Diseases: A Systematic Review and Meta-Analysis. Nutrients 14 (22), 2022.\u003c/li\u003e\n\u003cli\u003eZhao J, Zou H, Huo Y, Wei X, Li Y. Emerging roles of selenium on metabolism and type 2 diabetes. Front Nutr 9: 1027629, 2022.\u003c/li\u003e\n\u003cli\u003eZhao L, Carmean CM, Landeche M, Chellan B, Sargis RM. Selenomethionine modulates insulin secretion in the MIN6-K8 mouse insulinoma cell line. FEBS Lett 595 (24): 3042-3055, 2021.\u003c/li\u003e\n\u003cli\u003eZhang S, Zhang G, Wang P, Wang L, Fang B, Huang J. Effect of Selenium and Selenoproteins on Radiation Resistance. Nutrients 16 (17), 2024.\u003c/li\u003e\n\u003cli\u003eGarbo S, Di Giacomo S, Łażewska D, Honkisz-Orzechowska E, Di Sotto A, Fioravanti R, Zwergel C, Battistelli C. Selenium-Containing Agents Acting on Cancer-A New Hope? Pharmaceutics 15 (1), 2022.\u003c/li\u003e\n\u003cli\u003eFedirko V, Jenab M, M\u0026eacute;plan C, Jones JS, Zhu W, Schomburg L, Siddiq A, Hybsier S, Overvad K, Tj\u0026oslash;nneland A, Omichessan H, Perduca V, Boutron-Ruault MC, K\u0026uuml;hn T, Katzke V, Aleksandrova K, Trichopoulou A, Karakatsani A, Kotanidou A, Tumino R, Panico S, Masala G, Agnoli C, Naccarati A, Bueno-de-Mesquita B, Vermeulen R, Weiderpass E, Skeie G, N\u0026oslash;st TH, Lujan-Barroso L, Quir\u0026oacute;s JR, Huerta JM, Rodr\u0026iacute;guez-Barranco M, Barricarte A, Gylling B, Harlid S, Bradbury KE, Wareham N, Khaw KT, Gunter M, Murphy N, Freisling H, Tsilidis K, Aune D, Riboli E, Hesketh JE, Hughes DJ. Association of Selenoprotein and Selenium Pathway Genotypes with Risk of Colorectal Cancer and Interaction with Selenium Status. Nutrients 11 (4), 2019.\u003c/li\u003e\n\u003cli\u003eBarhoumi T, Todryk S. Role of monocytes/macrophages in renin-angiotensin system-induced hypertension and end organ damage. Front Physiol 14: 1199934, 2023.\u003c/li\u003e\n\u003cli\u003eDelevatti RS, Leonel L, Rodrigues J, Kanitz AC, Alberton CL, Lovatel GA, Siqueira IR, Kruel L. Aerobic Exercise in the Aquatic Environment Suppresses the Plasma Renin Activity in Individuals with Type 2 Diabetes: A Secondary Analysis of a Randomized Clinical Trial. Int J Environ Res Public Health 21 (7), 2024.\u003c/li\u003e\n\u003cli\u003eCaturano A, Galiero R, Vetrano E, Sardu C, Rinaldi L, Russo V, Monda M, Marfella R, Sasso FC. Insulin-Heart Axis: Bridging Physiology to Insulin Resistance. Int J Mol Sci 25 (15), 2024.\u003c/li\u003e\n\u003cli\u003eKuniyasu H. Multiple roles of angiotensin in colorectal cancer. World J Clin Oncol 3 (12): 150-4, 2012.\u003c/li\u003e\n\u003cli\u003eVan Berlo B, Civati C, Esposito P, De Keulenaer GW, Guns PJDF, Segers VFM. Angiotensin II as a linking factor in cardiovascular disease enhanced cancer growth. European Heart Journal 45 (Supplement_1): ehae666.3205, 2024.\u003c/li\u003e\n\u003cli\u003eAnand S, Rajagopal S. A Comprehensive Review on the Regulatory Action of TRP Channels: A Potential Therapeutic Target for Nociceptive Pain. Neurosci Insights 18: 26331055231220340, 2023.\u003c/li\u003e\n\u003cli\u003eRosenbaum T, Morales-L\u0026aacute;zaro SL, Islas LD. TRP channels: a journey towards a molecular understanding of pain. Nat Rev Neurosci 23 (10): 596-610, 2022.\u003c/li\u003e\n\u003cli\u003eBenzi A, Heine M, Spinelli S, Salis A, Worthmann A, Diercks B, Astigiano C, P\u0026eacute;rez MR, Memushaj A, Sturla L, Vellone V, Damonte G, Jaeckstein MY, Koch-Nolte F, Mittr\u0026uuml;cker HW, Guse AH, De Flora A, Heeren J, Bruzzone S. The TRPM2 ion channel regulates metabolic and thermogenic adaptations in adipose tissue of cold-exposed mice. Front Endocrinol (Lausanne) 14: 1251351, 2024.\u003c/li\u003e\n\u003cli\u003eJiang S, Lin X, Wu Q, Zheng J, Cui Z, Cai X, Li Y, Zheng C, Sun Y. Transient receptor potential channels\u0026apos; genes forecast cervical cancer outcomes and illuminate its impact on tumor cells. Front Genet 15: 1391842, 2024.\u003c/li\u003e\n\u003cli\u003eLiu Y, Yao X, Zhao W, Xu J, Zhang H, Huang T, Wu C, Yang J, Tang C, Ye Q, Hu W, Wang Q. A comprehensive analysis of TRP-related gene signature, and immune infiltration in patients with colorectal cancer. Discov Oncol 15 (1): 357, 2024.\u003c/li\u003e\n\u003cli\u003eTajbakhsh A, Gheibi HS, Butler AE, Sahebkar A. Effect of soluble cleavage products of important receptors/ligands on efferocytosis: Their role in inflammatory, autoimmune and cardiovascular disease. Ageing Res Rev 50: 43-57, 2019.\u003c/li\u003e\n\u003cli\u003eLiu X, Liu H, Deng Y. Efferocytosis: An Emerging Therapeutic Strategy for Type 2 Diabetes Mellitus and Diabetes Complications. J Inflamm Res 16: 2801-2815, 2023.\u003c/li\u003e\n\u003cli\u003eZhao Y, Li M, Mao J, Su Y, Huang X, Xia W, Leng X, Zan T. Immunomodulation of wound healing leading to efferocytosis. Smart Med 3 (1): e20230036, 2024.\u003c/li\u003e\n\u003cli\u003eMa Z, Sun Y, Yu Y, Xiao W, Xiao Z, Zhong T, Xiang X, Li Z. Extracellular vesicles containing MFGE8 from colorectal cancer facilitate macrophage efferocytosis. Cell Commun Signal 22 (1): 295, 2024.\u003c/li\u003e\n\u003cli\u003eLiu XH, Xu Q, Zhang L, Liu HJ. Association between metabolic score for insulin resistance and regression to normoglycemia from prediabetes in Chinese adults: A retrospective cohort study. PLoS One 19 (8): e0308343, 2024.\u003c/li\u003e\n\u003cli\u003eKawada A, Yoshitake S, Fujihara R, Ishikawa M. Relationship Between Oxidative Stress in the Rotator Cuff and Transcutaneous Advanced Glycation End-Products Measurement in Diabetic Rats. Cureus 16 (8): e67529, 2024.\u003c/li\u003e\n\u003cli\u003eAzizian-Farsani F, Abedpoor N, Hasan SM, Gure AO, Nasr-Esfahani MH, Ghaedi K. Receptor for Advanced Glycation End Products Acts as a Fuel to Colorectal Cancer Development. Front Oncol 10: 552283, 2020.\u003c/li\u003e\n\u003cli\u003eYang S, Li Y, Zhang Y, Wang Y. Impact of chronic stress on intestinal mucosal immunity in colorectal cancer progression. Cytokine Growth Factor Rev 80: 24-36, 2024.\u003c/li\u003e\n\u003cli\u003eShao S, Piao L, Guo L, Wang J, Wang L, Wang J, Tong L, Yuan X, Zhu J, Fang S, Wang Y. Tetraspanin 7 promotes osteosarcoma cell invasion and metastasis by inducing EMT and activating the FAK-Src-Ras-ERK1/2 signaling pathway. Cancer Cell Int 22 (1): 183, 2022.\u003c/li\u003e\n\u003cli\u003eZhan S, Wang L, Wang W, Li R. Insulin resistance in NSCLC: unraveling the link between development, diagnosis, and treatment. Front Endocrinol (Lausanne) 15: 1328960, 2024.\u003c/li\u003e\n\u003cli\u003eChen L, Liu H, Li Y, Lin X, Xia S, Wanggou S, Li X. Functional characterization of TSPAN7 as a novel indicator for immunotherapy in glioma. Front Immunol 14: 1105489, 2023.\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Table","content":"\u003cp\u003eTable 1. Summary of T2DM and CRC cohorts used in this study.\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"100%\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003eGEO Accession\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 14px;\"\u003e\n \u003cp\u003ePlatform\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 31px;\"\u003e\n \u003cp\u003eSamples\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003eDisease\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 22px;\"\u003e\n \u003cp\u003eGroup\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003eGSE76895\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 14px;\"\u003e\n \u003cp\u003eGPL570\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 31px;\"\u003e\n \u003cp\u003eislet samples from 36 T2DM and 32 normal\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003eT2DM\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 22px;\"\u003e\n \u003cp\u003eValidation cohort\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003eGSE25724\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 14px;\"\u003e\n \u003cp\u003eGPL96\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 31px;\"\u003e\n \u003cp\u003eislet samples from 6 T2DM and 7 normal\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003eT2DM\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 22px;\"\u003e\n \u003cp\u003eDiscovery cohort\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003eGSE110224\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 14px;\"\u003e\n \u003cp\u003eGPL570\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 31px;\"\u003e\n \u003cp\u003eprimary adenocarcinomas and matched normal samples from 7 patient\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003eCRC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 22px;\"\u003e\n \u003cp\u003eValidation cohort\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003eGSE41258\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 14px;\"\u003e\n \u003cp\u003eGPL96\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 31px;\"\u003e\n \u003cp\u003e316 CRC and 74 normal control\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003eCRC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 22px;\"\u003e\n \u003cp\u003eValidation cohort\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003eGSE39582\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 14px;\"\u003e\n \u003cp\u003eGPL570\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 31px;\"\u003e\n \u003cp\u003e566 CRC and 19 normal control\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003eCRC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 22px;\"\u003e\n \u003cp\u003eDiscovery cohort\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003eGSE38832\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 14px;\"\u003e\n \u003cp\u003eGPL570\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 31px;\"\u003e\n \u003cp\u003e122 CRC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003eCRC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 22px;\"\u003e\n \u003cp\u003eValidation cohort\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003eTCGA-COAD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 14px;\"\u003e\n \u003cp\u003eIllumina\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 31px;\"\u003e\n \u003cp\u003e556 Patients and 28 controls\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003eCRC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 22px;\"\u003e\n \u003cp\u003eValidation cohort\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\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":"Type 2 Diabetes Mellitus, Colorectal Cancer, TSPAN7, Machine Learning, Gene Expression","lastPublishedDoi":"10.21203/rs.3.rs-5651334/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5651334/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eType 2 Diabetes Mellitus (T2DM) and Colorectal Cancer (CRC) are significant global public health challenges with a notable epidemiological association. This study aims to explore the molecular mechanism behind this epidemiological association.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eWeighted Gene Co-expression Network Analysis (WGCNA) and differential expression gene (DEG) analysis were conducted to identify shared genes between T2DM and CRC. Machine learning algorithms, including LASSO, Random Forest, and Support Vector Machine (SVM), were employed to identify hub genes. IOBR and clusterProfiler packages were used for immunoinfiltration assessment and enrichment analysis, respectively.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eWe identified 27 shared genes between T2DM and CRC, with TSPAN7 emerging as a key hub gene linking the two conditions. TSPAN7 expression was significantly lower in disease groups compared to control groups across multiple cohorts, demonstrating excellent diagnostic accuracy. Enrichment analysis revealed involvement of these genes in various metabolic activities and pathways, including sulfur metabolism, selenium metabolism, renin secretion, pantothenate and CoA biosynthesis, TRP channel regulation, and efferocytosis.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e \u003cp\u003eThis study provides new insights into the mechanisms underlying the association between T2DM and CRC by identifying TSPAN7 as a key target. The findings offer theoretical evidence for developing new diagnostic markers and therapeutic strategies for these diseases.\u003c/p\u003e","manuscriptTitle":"Machine Learning Identification of TSPAN7 as a Key Target Linking Type 2 Diabetes Mellitus and Colorectal Cancer","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-12-19 14:48:36","doi":"10.21203/rs.3.rs-5651334/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":"30df33ca-8023-417e-9ee2-fdc27de9f537","owner":[],"postedDate":"December 19th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-12-20T20:53:23+00:00","versionOfRecord":[],"versionCreatedAt":"2024-12-19 14:48:36","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5651334","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5651334","identity":"rs-5651334","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}