UPR gene marker at pan cancer level as well as construction method and application of UPR gene marker

Abstract

According to the UPR gene marker and the construction method and application thereof, multiple bioinformatics tools and analysis methods are integrated, it is ensured that the construction and verification process of the UPR gene marker is scientific and rigorous, the UPR gene marker has universality and practicability in diagnosis and prognosis evaluation of different cancers, and the UPR gene marker is suitable for being used in the field of cancer diagnosis and prognosis evaluation. The method can be used for predicting response of a patient to different treatment schemes, so that personalized treatment strategies are optimized, the treatment effect is improved, and side effects are reduced. Besides, the UPR gene marker is used for risk stratification, and patients can be accurately classified into different risk groups, so that more positive treatment intervention is adopted for high-risk patients, the overall treatment effect and the patient survival rate are improved, and wide clinical application prospects and important scientific research values are achieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of bioinformatics, and relates to a pan-cancer level UPR gene marker and a construction method and application thereof. Background Art

[0002] Unfolded protein response (UPR) plays an important role in the occurrence and progression of many cancers. UPR is a mechanism for cells to cope with endoplasmic reticulum stress. It maintains cell homeostasis by regulating the folding ability of the endoplasmic reticulum, enhancing protein degradation pathways, and inhibiting new protein synthesis. More and more studies have shown that key molecules of UPR are abnormally expressed in various types of cancer and play a role in promoting cancer cell survival, proliferation, and metastasis. For example, UPR-related molecules such as GRP78 have been found to be highly expressed in many tumors and are closely related to the malignancy and prognosis of cancer. In addition, the activation status of the three main signaling pathways of UPR - IRE1α, PERK and ATF6 in cancer cells also shows obvious changes. These changes help cancer cells adapt to the adverse tumor microenvironment and play an important role in cancer progression.

[0003] However, there is currently a lack of a systematic, pan-cancer UPR gene marker to comprehensively evaluate the universality and heterogeneity of the activation of the UPR pathway in different tumors and in different cell types of the same tumor. Existing studies have mostly focused on the analysis of a single cancer or a specific UPR pathway, and lack comprehensive gene markers to evaluate the overall impact of UPR in various tumors. Summary of the invention

[0004] The technical problem solved by the present invention is to develop a pan-cancer UPR gene marker and a construction method and application thereof. The constructed pan-cancer UPR gene marker can be used to systematically evaluate the activation state of the UPR pathway in various cancers and its relationship with tumor progression, and has broad clinical application prospects and important scientific research value.

[0005] To achieve the above object, the technical solution adopted by the present invention is:

[0006] The present invention provides pan-cancer UPR gene markers, including: HSPA5, HSP90B1, CALR, P4HB, PDIA3, SLC3A2, CANX, HERPUD1, PDIA4, MANF, PDIA6, GARS1, HYOU1, TARS1, WARS1, MTHFD2, DDIT3, TMED10, ASNS, ARF4, HDLBP, SSR2, SEC61A1, ANXA5, SHMT2, NUCB2, DNAJB11, TMED9, SARS1, TRIB3, TRAM1, SDF2L1, RPN1, STT3A, CALU, SERP1, SSR1, FK BP2, NFE2L1, MYDGF, LMAN2, CDK2AP2, AARS1, CRELD2, KDELR2, XPOT, SELENOK, DERL2, SELENOS, COPA, RAB1A, OS9, SARAF, COPG1, LMAN1, GFPT1, SEC31A, SLC39A7, LONP1, TFG, DNAJC10, TMED7, SEC63, NANS, SEL1L, CALCOCO2, COPB1, CEBPB, PCK2, PLPP5, GTPBP2, ATP2B1, SLC39A14, CARS1, TAX1BP1, CCPG1, A RHGEF2, RAB6A, RRBP1, TNFRSF10B, SYVN1, CLIC4, PGM3, ZFAND2A, HSPA13, ERLEC1, USO1, SERINC3, ARMCX3, DNAJC3, DNAJB9, TMEM214, SEC24D, AZI2, PJ A2, TMEM41B, SLC7A1, TMEM39A, NFE2L2, TMEM263, YIPF5, WIPI1, GMPPA, ATG13, SERINC1, RHOQ, ZFAND3, PRRC1, CEBPG, CREB3, UFM1, FNDC3B, SLC33A1, SE SN2, ARFGAP3, SLC35E1, TTC17, ALG2, IBTK, GPT2, MBNL2, RAB5A, EDEM1, YIPF2, TBL2, OSBP, SUCO, RNF41, FKBP14, GOLGA4, GMPPB, RIOK3, TSPYL2, DNAJC 1. GOLGA5, CNPY3, FYN, STX5, CRELD1, AKAP17A, TERF2IP, CTH, TMF1, NIBAN1, GOLGB1, SLC6A9, GABARAPL1, ERP44, MON1B, PNPLA8, PDIA5, LYSMD3, KDM4A,ADAM17, PDE4DIP, SLC1A4, EIF2AK3, GOLPH3L, CCDC174. ,

[0007] The present invention also provides a method for constructing a pan-cancer UPR gene marker, comprising:

[0008] 1) Review the published literature related to UPR and collect relevant information of the following datasets: GSE84636, GSE66556, GSE196634, GSE169584, GSE153557, GSE148802, GSE146840, GSE145108, GSE139391, GSE131048, GSE129757, GSE108212, GSE200205, GSE217305, GSE226549, GSE227811, as the basis for constructing UPR gene markers;

[0009] 2) The weighted gene co-expression network analysis (WGCNA) algorithm was combined to identify gene modules related to the UPR activation state, and genes with higher gene significance (GS) and module membership (MM) values ​​were selected for constructing UPR gene markers.

[0010] Preferably, in step 1), the batch effect is removed by using the "ComBat" function of the R package "sva", so that the influence of the source is reduced and the difference in biological state is revealed.

[0011] Preferably, in step 2), during the weighted gene co-expression network analysis, the soft threshold β is selected as 12; the correlation between each module and the cell state and UPR activity is calculated, and the UPR activity is calculated based on a published gene set.

[0012] Preferably, in step 2), genes with GS>0.4 and MM>0.7 are selected for constructing UPR gene markers.

[0013] Preferably, the above method also includes using an existing UPR-related chip sequencing data set for verification to ensure the accuracy and reliability of the data; using the GSEA algorithm to perform robustness verification on the constructed UPR gene markers; using the ssGSEA algorithm to calculate the UPR score of each sample, evaluate its activity level under different conditions, and further analyze the correlation between the UPR score and the HSPA5 expression level to verify the practicality and reliability of the UPR score under different conditions.

[0014] More preferably, the sequencing data of samples treated with TG, TM and DMSO are collected as the validation set of the pan-cancer level UPR gene markers; the validation set IDs are: GSE68718, GSE29929, GSE13548, GSE16157, GSE56843, GSE21979, GSE19519.

[0015] The present invention further provides the use of the above-mentioned pan-cancer level UPR gene marker in the preparation of pan-cancer diagnosis, treatment and prognosis evaluation products.

[0016] Preferably, the product systematically evaluates the activation status of the UPR pathway in various cancers and its relationship with tumor progression through the pan-cancer level UPR gene markers.

[0017] More preferably, the application includes: collecting samples and performing transcriptome or single-cell sequencing to obtain a gene expression matrix, using the pan-cancer UPR gene markers, and using the ssGSEA algorithm to perform scoring, wherein a higher score indicates a stronger activity of the UPR pathway.

[0018] The beneficial effects of the present invention are:

[0019] 1. Applicability and reliability for multiple cancer types

[0020] The UPR gene markers of the present invention are applicable to a variety of cancer types, including solid tumors and blood tumors, through systematic analysis. This wide applicability makes it universal and practical in the diagnosis and prognosis evaluation of different cancers. In addition, the present invention integrates a variety of bioinformatics tools and analysis methods to ensure that the construction and verification process of the UPR gene markers is scientific and rigorous, providing a reliable method for the UPR activity evaluation of multiple cancers.

[0021] 2. Mechanism research support

[0022] The present invention comprehensively evaluates the activity and function of the UPR pathway at the pan-cancer level through systematic analysis. This comprehensive assessment enables us to have a deeper understanding of the universality and specific role of UPR in different cancer types. Through the integration and analysis of a large amount of data, the present invention clarifies the association between the UPR pathway and tumor progression. The identification of key UPR-related genes and molecular pathways provides a reliable basis for further research on the specific mechanism of UPR in cancer.

[0023] 3. Personalized treatment guidance

[0024] The UPR gene markers of the present invention can be used to predict the patient's response to different treatment options (such as immunotherapy), thereby optimizing personalized treatment strategies, improving treatment effects and reducing side effects. In addition, risk stratification using UPR gene markers can accurately classify patients into different risk groups, thereby taking more active treatment interventions for high-risk patients, improving overall treatment effects and patient survival rates.

[0025] 4. Accuracy of prognostic prediction

[0026] Based on UPR gene markers and combined with machine learning algorithm optimization, the patient's prognosis can be accurately predicted. The efficient prognostic model helps clinicians better stratify and manage patients and develop more effective treatment plans.

[0027] In summary, the pan-cancer UPR gene markers of the present invention and their construction methods and applications not only significantly improve the accuracy of cancer diagnosis and prognosis assessment, but also provide a new guiding tool for personalized treatment, and have broad clinical application prospects and important scientific research value. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 : Development of UPR gene markers. (A) PCA plot of samples from different sources and biological states before removing batch effects. (B) PCA plot of samples from different sources and biological states after removing batch effects. (C) Correlation analysis between module characteristic genes and phenotypes. (D) Correlation between MM and GS in the blue module.

[0029] Figure 2 : Validation of the robustness of the UPR gene marker. (A) Average NES value of the UPR-related gene set. (B) GSEA plots of specific gene sets in the two datasets.

[0030] Figure 3 : Validation of the robustness of the UPR gene markers. (A) and (B) ssGSEA analysis of samples in different biological states. (C) and (D) Correlation analysis between HSPA5 expression and UPR score.

[0031] Figure 4 : Heterogeneity of UPR activity in pan-cancer. UPR profiles of 28 normal and 33 tumor tissues.

[0032] Figure 5: Clinical translational significance of UPR gene markers. (A) Overall strategy for selecting three core genes in the UPR signature to predict survival in BRCA patients. (B) and (D) Kaplan-Meier survival curves of patients with high and low UPR predictors in the TCGA BRCA cohort. (C) Violin plots showing UPR predictors in patients with different clinical stages in the TCGA BRCA cohort. (E)-(H) Kaplan-Meier survival curves of patients with high and low UPR predictors in four published BRCA cohorts. DETAILED DESCRIPTION

[0033] In order to explain the present invention more clearly, the present invention is further described in detail below in conjunction with embodiments and with reference to the accompanying drawings. It should be understood by those skilled in the art that the content described below is illustrative rather than restrictive, and should not be used to limit the scope of protection of the present invention.

[0034] The main technical routes for the construction and application of the pan-cancer UPR gene markers of the present invention are as follows:

[0035] Step 1: Construction of UPR gene markers

[0036] Through literature review and data set collection, we first screened the gene data related to the activation state of UPR. We used the Weighted Gene Co-expression Network Analysis (WGCNA) algorithm to identify and select the most relevant gene modules, and the genes with higher gene significance (GS) and module membership (MM) values ​​in the screening modules were used to construct UPR gene markers.

[0037] Step 2: Validation and evaluation of markers

[0038] The existing UPR-related chip sequencing dataset was used for validation to ensure the accuracy and reliability of the data. The gene set enrichment analysis (GSEA) algorithm was applied to verify the robustness of the constructed UPR gene markers. The single sample gene set enrichment analysis (ssGSEA) algorithm was used to calculate the UPR score of each sample and evaluate its activity level under different conditions. The correlation between the UPR score and the expression of HSPA5 was further analyzed to verify the practicality and reliability of the score.

[0039] Step 3: Application of markers

[0040] Collect samples and perform transcriptome or single-cell sequencing to obtain gene expression matrix. Use the constructed gene markers and use the ssGSEA algorithm to score. The higher the score, the stronger the activity of the UPR pathway.

[0041] Example

[0042] 1. Construction of UPR gene markers

[0043] 1. Literature review and dataset collection

[0044] The publicly published literature and data sets related to UPR were consulted as the basis for constructing UPR gene markers.

[0045] We consulted relevant literature and collected relevant information of the following datasets: GSE84636, GSE66556, GSE196634, GSE169584, GSE153557, GSE148802, GSE146840, GSE145108, GSE139391, GSE131048, GSE129757, GSE108212, GSE200205, GSE217305, GSE226549, and GSE227811.

[0046] These datasets included 19 cancer cell lines exposed to typical UPR inducers such as trocarotene (TG), tunicamycin (TM), hypoxia, and glucose deprivation (GD). Principal component analysis (PCA) revealed significant batch effects in data from different sources, masking biological differences ( Figure 1 A). It is noteworthy that after removing the batch effect using the “ComBat” function of the R package “sva”, the influence of the source was mitigated and the differences in biological status emerged ( Figure 1 B).

[0047] 2. Development of UPR gene markers

[0048] Combined with the WGCNA algorithm, gene modules related to the UPR activation state were identified, and genes with higher gene significance (GS) and module membership (MM) values ​​were selected to construct UPR gene markers.

[0049] During the weighted gene co-expression network analysis, a soft threshold β of 12 was selected, which provided an appropriate power value for the construction of the co-expression network. In addition, we calculated the correlation of each module with the cell state and UPR activity, which was calculated based on published gene sets. The blue module has the highest correlation with UPR activation ( Figure 1C). In the blue module, the correlation coefficient between gene significance (GS) and module membership (MM) is 0.56, indicating that the construction quality of this module is relatively high ( Figure 1 D). Based on this, we selected 159 genes highly associated with UPR activation as UPR markers, characterized by GS>0.4 and MM>0.7 ( Figure 1 D).

[0050] The UPR gene markers screened in this way include 159 genes: HSPA5, HSP90B1, CALR, P4HB, PDIA3, SLC3A2, CANX, HERPUD1, PDIA4, MANF, PDIA6, GARS1, HYOU1, TARS1, WARS1, MTHFD2, DDIT3, TMED10, ASNS, ARF4, HDLBP, SSR2, SEC61A1, ANXA5, SHMT2, NUCB2, DNAJB11, TMED9, SARS1, TRIB3, TRAM1, SDF2L1, RPN1, STT3A, CALU, SERP1, SSR1, FK BP2, NFE2L1, MYDGF, LMAN2, CDK2AP2, AARS1, CRELD2, KDELR2, XPOT, SELENOK, DERL2, SELENOS, COPA, RAB1A, OS9, SARAF, COPG1, LMAN1, GFPT1, SEC31A, SLC39A7, LONP1, TFG, DNAJC10, TMED7, SEC63, NANS, SEL1L, CALCOCO2, COPB1, CEBPB, PCK2, PLPP5, GTPBP2, ATP2B1, SLC39A14, CARS1, TAX1BP1, CCPG1, A RHGEF2, RAB6A, RRBP1, TNFRSF10B, SYVN1, CLIC4, PGM3, ZFAND2A, HSPA13, ERLEC1, USO1, SERINC3, ARMCX3, DNAJC3, DNAJB9, TMEM214, SEC24D, AZI2, PJ A2, TMEM41B, SLC7A1, TMEM39A, NFE2L2, TMEM263, YIPF5, WIPI1, GMPPA, ATG13, SERINC1, RHOQ, ZFAND3, PRRC1, CEBPG, CREB3, UFM1, FNDC3B, SLC33A1, SE SN2, ARFGAP3, SLC35E1, TTC17, ALG2, IBTK, GPT2, MBNL2, RAB5A, EDEM1, YIPF2, TBL2, OSBP, SUCO, RNF41, FKBP14, GOLGA4, GMPPB, RIOK3, TSPYL2, DNAJC 1. GOLGA5, CNPY3, FYN, STX5, CRELD1, AKAP17A, TERF2IP, CTH, TMF1, NIBAN1, GOLGB1, SLC6A9, GABARAPL1, ERP44, MON1B, PNPLA8, PDIA5, LYSMD3, KDM4A,ADAM17, PDE4DIP, SLC1A4, EIF2AK3, GOLPH3L, CCDC174. ,

[0051] 2. Validation and evaluation of gene markers

[0052] 1. Collect validation dataset

[0053] Existing UPR-related chip sequencing data sets were collected to ensure the quality and accuracy of the data. Given that TG and TM are typical UPR inducers, we consulted the literature and collected sequencing data of samples treated with TG, TM, and DMSO as the validation set for this gene marker.

[0054] The validation set IDs are: GSE68718, GSE29929, GSE13548, GSE16157, GSE56843, GSE21979, GSE19519.

[0055] 2. Validation of the robustness of UPR gene markers

[0056] To validate the robustness of the UPR gene signature, we used several approaches. First, we evaluated the validity of the UPR gene signature using the gene set enrichment analysis (GSEA) algorithm ( Figure 2 A, 2B). Second, single sample gene set enrichment analysis (ssGSEA) was used to calculate the UPR score of each sample. A higher score indicates a higher degree of UPR activation ( Figure 3 A, 3B). Finally, since HSPA5 is an important molecular chaperone protein in ER stress and is closely related to the intensity of ER stress, we also analyzed the correlation between the UPR score and the expression of heat shock protein A5 (HSPA5) to further verify the reliability and practicality of the UPR score under different conditions ( Figure 3 C, 3D).

[0057] The results showed that by performing GSEA analysis on the results of differential expression analysis between the UPR activation group and the inactivated group, a higher normalized enrichment score (NES) was obtained, indicating that the UPR was significantly activated. The UPR score calculated by the ssGSEA algorithm showed that the score of the UPR activation group was significantly higher than that of the inactivated group. In addition, the UPR score was significantly positively correlated with the expression of HSPA5. These three methods jointly verified the validity and robustness of the UPR gene markers.

[0058] 3. Application in Pan-cancer

[0059] In pan-cancer treatment and research, this marker can be used to systematically evaluate the role of the UPR pathway in various cancers. The UPR score is calculated using the ssGSEA algorithm, and a higher score indicates stronger UPR activity.

[0060] Based on this gene marker, we systematically portrayed the full picture of the UPR and revealed the intra- and inter-tumor heterogeneity of UPR activity ( Figure 4 ). In addition, by using three machine learning algorithms, we proposed a promising prognostic biomarker ( Figure 5 A). In both the training and validation cohorts, breast cancer patients with high UPR scores had shorter survival and faster disease progression ( Figure 5 These results provide an efficient method for assessing UPR activity in cancer and highlight the clinical relevance and translational significance of the UPR in cancer.

[0061] Obviously, the above embodiments of the present invention are merely examples to more clearly illustrate the present invention, and are not limitations on the implementation methods of the present invention. For ordinary technicians in the relevant field, other different forms of changes or modifications can be made on the basis of the above description. It is impossible to list all the implementation methods here. All obvious changes or modifications derived from the technical solution of the present invention are still within the protection scope of the present invention.

Claims

1. Pan-cancer UPR gene markers, including: <h2 style=";text-align:left;direction:ltr">HSPA5, HSP90B1, CALR, P4HB, PDIA3, SLC3A2, CANX, HERPUD1, PDIA4, MANF, PDIA6, GARS1, HYOU1, TARS1, WARS1, MTHFD2, DDIT3, TMED10, ASNS, ARF4, HDL BP, SSR2, SEC61A1, ANXA5, SHMT2, NUCB2, DNAJB11, TMED9, SARS1, TRIB3, TRAM1, SDF2L1, RPN1, STT3A, CALU, SERP1, SSR1, FKBP2, NFE2L1, MYDGF, LMAN2 CDK2AP2, AARS1, CRELD2, KDELR2, XPOT, SELENOK, DERL2, SELENOS, COPA, RAB1A, OS9, SARAF, COPG1, LMAN1, GFPT1, SEC31A, SLC39A7, LONP1, TFG, DNAJ C10、TMED7、SEC63、NANS、SEL1L、CALCOCO2、COPB1、CEBPB、PCK2、PLPP5、GTP BP2、ATP2B1、SLC39A14、CARS1、TAX1BP1、CCPG1、ARHGEF2、RAB6A、RRBP1、TNF RSF10B、SYVN1、CLIC4、PGM3、ZFAND2A、HSPA13、ERLEC1、USO1、SERINC3、ARM CX3、DNAJC3、DNAJB9、TMEM214、SEC24D、AZI2、PJA2、TMEM41B、SLC7A1、TMEM 39A、NFE2L2、TMEM263、YIPF5、WIPI1、GMPPA、ATG13、SERINC1、RHOQ、ZFAND3 、PRRC1、CEBPG、CREB3、UFM1、FNDC3B、SLC33A1、SESN2、ARFGAP3、SLC35E1、TT C17、ALG2、IBTK、GPT2、MBNL2、RAB5A、EDEM1、YIPF2、TBL2、OSBP、SUCO、RNF4 1、FKBP14、GOLGA4、GMPPB、RIOK3、TSPYL2、DNAJC1、GOLGA5、CNPY3、FYN、STX 5、CRELD1、AKAP17A、TERF2IP、CTH、TMF1、NIBAN1、GOLGB1、SLC6A9、GABARAP L1、ERP44、MON1B、PNPLA8、PDIA5、LYSMD3、KDM4A、ADAM17、PDE4DIP、SLC1A4、EIF2AK3、GOLPH3L、CCDC174。、 2. Method for constructing pan-cancer UPR gene markers, including: 1) Review the published literature related to UPR and collect relevant information of the following datasets: GSE84636, GSE66556, GSE196634, GSE169584, GSE153557, GSE148802, GSE146840, GSE145108, GSE139391, GSE131048, GSE129757, GSE108212, GSE200205, GSE217305, GSE226549, GSE227811, as the basis for constructing UPR gene markers; 2) Combined with the weighted gene co-expression network analysis (WGCNA) algorithm, the gene modules related to the UPR activation state were identified, and genes with higher gene significance GS and module membership MM values ​​were selected to construct UPR gene markers.

3. The construction method according to claim 2, characterized in that: In step 1), the batch effect was removed by using the "ComBat" function of the R package "sva", so that the influence of the source was reduced and the differences in biological status were revealed.

4. The construction method according to claim 2, characterized in that: In step 2), during the weighted gene co-expression network analysis, a soft threshold β of 12 was selected; the correlation between each module and the cell state and UPR activity was calculated, and the UPR activity was calculated based on the published gene set.

5. The construction method according to claim 2, characterized in that: In step 2), genes with GS>0.4 and MM>0.7 were selected for constructing UPR gene markers.

6. The construction method according to claim 2, characterized in that: It also includes the use of existing UPR-related chip sequencing data sets for verification to ensure the accuracy and reliability of the data; the use of the GSEA algorithm to perform robustness verification on the constructed UPR gene markers; the use of the ssGSEA algorithm to calculate the UPR score of each sample, evaluate its activity level under different conditions, and further analyze the correlation between the UPR score and the expression level of HSPA5 to verify the practicality and reliability of the UPR score under different conditions.

7. The construction method according to claim 6, characterized in that: The sequencing data of samples treated with TG, TM and DMSO were collected as the validation set of the pan-cancer UPR gene markers; the validation set ID is: GSE68718, GSE29929, GSE13548, GSE16157, GSE56843, GSE21979, GSE19519.

8. Use of the pan-cancer UPR gene marker described in claim 1 or the pan-cancer UPR gene marker obtained by the construction method of any one of claims 2 to 7 in the preparation of pan-cancer diagnosis, treatment and prognosis assessment products.

9. The use according to claim 8, characterized in that: The product evaluates the activation status of the UPR pathway in various cancers and its relationship with tumor progression through the pan-cancer level UPR gene marker.

10. The use according to claim 8, characterized in that: The application includes: collecting samples and performing transcriptome or single-cell sequencing to obtain a gene expression matrix, using the pan-cancer level UPR gene markers, and using the ssGSEA algorithm to perform scoring, wherein a higher score indicates a stronger activity of the UPR pathway.

Images