| Gastric cancer is the third most fatal cancer in China and the fourth in the world,and it becomes a common health problem.Currently,pathological diagnosis is still the "golden standard" for gastric cancer diagnosis.There are two mainly classification methods used widely in clinical pathological diagnosis:one is the classification of diffuse gastric cancer(DGC)and intestinal gastric cancer(IGC)proposed by Lauren,and the other is the histological classification proposed by WHO.For intestinal gastric cancer in Lauren classification,P.Correa proposed a staged evolution process of intestinal gastric cancer,which is chronic inflammation,intestinal metaplasia,dysplasia,and gastric cancer.Although,this plays a key role in the early diagnosis and treatment of intestinal gastric cancer,it is only classified in detail of morphology of intestinal gastric cancer.For diffuse gastric adenocarcinoma that is difficult to detect and has a poor prognosis,there is still no good precancerous staging theory to support.In tumor tissues,tumor cells themselves can undergo changes at multiple levels,such as gene mutations,DNA copy number variations,differences in RNA transcription,differences in protein expression,and so on.It is currently believed that the occurrence of tumors is due to the combined action of multiple factors.Different tumors have different forms of changes.Regarding the multiple differences of tumor cells in DGC and IGC,we still lack in-depth understanding and need to be studied in depth.The tumor microenvironment has an important influence on the occurrence and development of tumors.Under normal circumstances,the gastric mucosa is composed of epithelial tissue,interstitial tissue and mucosal muscle.Among them,the interstitial tissue is mainly composed of fibroblasts and smooth muscle cells,and also contains immuno-inflammatory cells,such as lymphocytes,plasma cells,macrophages,eosinophils and so on.These immune cells are recruited to the gastric mucosa propria under the induction of cytokines and chemokines,and play an important role in the process of tumor immunity.An in-depth understanding of the changes in the immune microenvironment of the different gastric mucosa of DGC and IGC can reveal its important role in the occurrence and development of the two types of gastric cancer,which can provide important theoretical references for the early warning,prognosis and immunotherapy of DGC and IGC.With the advancement of technology,single-cell sequencing is increasingly playing an important role in tumor research.Pseudotime analysis can sort the cell expression in the single-cell sequencing results by dimensionality reduction to find similar cells in the process of cell occurrence and development,and further classify these similar cells,which can track the trajectory of cell differentiation and the law of development.At present,single-cell sequencing studies on the occurrence of gastric cancer have been reported,but studies focusing on the origin of DGC and IGC have not been reported.In summary,this study intends to use a variety of bioinformatics analysis methods to conduct a multi-omics and multi-dimensional comparative analysis of DGC and IGC from the characteristics of tumor cells,tumor microenvironment and tumor occurrence and development,in order to systematically clarify DGC,the molecular phenotypic characteristics of multiple omics differences from IGC provide a theoretical basis for the precise diagnosis and treatment of different histological types of gastric cancer.Research purposes1.Through multi-omics comparison of gastric cancer cases in the TCGA database,understand the differences in genes,transcription,and proteins between diffuse gastric cancer and intestinal gastric cancer.2.Through the correlation analysis of the degree of immune cell infiltration and the expression degree of chemokines and cytokines in gastric cancer cases in the TCGA database,understand the difference between diffuse gastric cancer and intestinal gastric cancer at the level of the immune microenvironment.3.Through the analysis of single-cell sequencing results in public databases,we can understand the different origins of diffuse gastric cancer and intestinal gastric cancer,so as to analyze the development and evolution of gastric cancer.Research method1.Source of casesDownload the clinical data of stomach cancer(STAD)in the TCGA(The Cancer Genome Atlas)database,and extract case data with clear Lauren classification diagnosis from the clinical data for analysis.Through screening,65 cases of DGC and 187 cases of IGC were included in the mutation data.In the DNA copy number variation data,64 cases were DGC and 189 cases were IGC.In the DNA methylation data,60 cases were DGC and 162 cases were IGC.In the mRNA data,66 cases were DGC and 181 cases were IGC.Long non-coding RNA(lncRNA)data is extracted from mRNA data,and the number of cases is the same as the number of cases in the mRNA data.In the microRNA(miRNA)data,66 cases were DGC and 181 cases were IGC.In the protein data,60 cases were DGC and 159 cases were IGC.The single-cell sequencing data uses the Cell Ranger output data of Sathe et al.(https://dna-discovery.stanford.edu)and Zhang et al.(GSE134520)for in-depth mining and analysis.Cases include:non-atrophic gastritis(NAG)3 cases,chronic atrophic gastritis(CAG)There were 6 cases of intestinal metaplasia(IM),1 case of early gastric cancer(EGC),4 cases of advanced gastric cancer(AGC),including 2 cases of DGC and 2 cases of IGC.2.Comparison of multi-omics differences between diffuse gastric cancer and intestinal gastric cancera)Analysis of mutation differences between diffuse gastric cancer and intestinal gastric cancerUse the maftools package to compare and visualize the mutation data of diffuse gastric cancer and intestinal gastric cancer.Identify cancer driver genes based on the oncodriveCLUST algorithm.Comparing the mutation characteristics of the Catalogue of Somatic Mutations in Cancer(COSMIC)v2 to obtain the mutation characteristics of intestinal gastric cancer and diffuse gastric cancer.b)DNA copy number difference analysisGenomic Analysis of Important Aberrations(GAIA)method is used to analyze the test results,and the tumor tissue DNA copy number variation information is obtained by comparing normal genes.c)Differential analysis of DNA methylationUse the Illumina Human Methylation 450K chip test results to compare the methylation results of intestinal gastric cancer and diffuse gastric cancer.d)ncRNA expression difference analysisUse TCGA Ensembl database(v75)to extract lncRNA information and perform downstream analysis.Use the TCGAbiolinks package to perform differential analysis on the raw count(RC)of IncRNA.Due to the small amount of miRNA data,using the TCGA data results,directly use the edgeR package for difference analysis.e)Transcriptomics difference analysisThe RC data of mRNA downloaded by TCGA was standardized using the TCGAbiolinks package,and then the edgeR package was used for differential analysis.e)Differential analysis of protein expressionThe limma method was used for differential analysis of the protein expression data in the TCGA database.f)Functional enrichment analysisUse clusterProfiler package for GO,KEGG,GSEA function enrichment.g)Multi-omics combination and attribution analysisFor the differential genes between diffuse gastric cancer and intestinal gastric cancer,use GDCRNATools to build a ceRNA network and use cytospace to visualize it.For genes in protein expression network analysis,use the starbase database to compare their corresponding miRNAs,then find the lncRNAs corresponding to the miRNAs,and finally compare the results of the spearman correlation analysis to confirm whether the ceRNA network is established.According to the path of the protein and the result of comparing the protein expression data using limma,combined with the differential expression of mRNA,a protein-mRNA expression correlation network is established.Use MOFA(Multi-Omics Factor Analysis)for multi-omics factor analysis.Analyze the factors to confirm related genes and related functions.3.DGC and IGC immune microenvironment difference analysisa)Differential analysis of immune cell infiltrationUsing the metagenes of 28 immune cells as a gene set,single-sample gene set enrichment analysis(ssGSEA)was performed on the mRNA genes of samples of different gastric cancer types,and the degree of infiltration of 28 immune cells in diffuse gastric cancer and intestinal gastric cancer was obtained.b)Analysis of differential expression of interleukins,chemokines and receptorsExtracted 35 kinds of interleukins,56 kinds of chemokines and chemokine receptors,and then compared the expression differences of various cytokines in diffuse gastric cancer and intestinal gastric cancer by using Mann-Whitney U testc)Joint analysis of immune cells,chemokines and cytokinesThe least square regression method was used to analyze the cross-linking of immune cell infiltration,chemokines and interleukins.First,select all chemokines as independent variables and one type of immune infiltrating cells as dependent variables,then use all immune infiltrating cells as independent variables and interleukins as dependent variables for analysis to establish a chemokine-immune cell-interleukin interaction model.4.Histogenesis analysis based on single cell sequencing resultsa)Homogenization and cluster analysis of single-cell sequencing resultsThe standard method in Seurat was used for homogenization,and then the cells were clustered,and finally 28 kinds of cells were obtained.b)Quasi-time sequence analysisUse Monocle3 to cluster the cells after Seurat homogenization,and then sort the cells in each cluster to analyze the pseudo-chronological path.We use diffusion map to analyze two subgroups including diffuse gastric cancer cells and intestinal gastric cancer cells.c)Analysis of the activity of the greening factorSCENIC was used to analyze the transcription factor activity of each cell.d)Analysis of cell-cell interaction in different gastric mucosal statesUse Cellchat to analyze the interaction between cells in different states.According to the label of the ligand receptor in the database and the expression of each receptor ligand in the cell,the cells in each state are divided into:Sedner,Receiver,Mediatior,and Influence four categories.At the same time,the importance of these four is evaluated.Research resultPart One:Comparison of multi-omics differences between diffuse gastric cancer and intestinal gastric cancer1.DGC and IGC genomics difference analysis1.1 Analysis of mutation differences1.1.1 Differences in mutation types and characteristicsThe most common point mutations in both DGC and IGC gastric cancers are C-T transition mutations;the fragment mutation types of the two are also basically the same,that is,deletion(Del)is the most common,followed by insertion(Ins).The Cancer Somatic Mutation Catalog(COSMIC)summarizes and categorizes many tumor mutation genes,and divides tumor mutation types into 30 types.By comparing with the mutation characteristics of COSMIC,we found that DGC has 4 mutation characteristics,namely COSMIC Signture 1(cosine approximation:0.953),COSMIC Signture 17(cosine approximation:0.976),and COSMIC Signture 6(Cosine Approximation:0.973),COSMIC Signture 3(cosine approximation:0.783);3 types of sudden changes in IGC are:COSMIC feature 10(cosine approximation:0.897),COSMIC Signture 17(cosine approximation:0.791),COSMIC Signture 6(cosine approximation:0.953).1.1.2 Differences in mutant genesComparing the mutation data of DGC and IGC,the top 10 differentially mutated genes are:CDH1,FBN1,RELN,DNAH3,NEB,PCLO,FAT3,USH2A,MYO16,COL22A1.The top 10 genes with the highest mutation rates of DGC and IGC are not exactly the same.DGC is:TTN,CDH1,CSMD3,TP53,MUC16,SPTA1,FAT4,ARID1A,CSMD1,PIK3CA,and IGC is:TTN,MUC16,TP53,LRP1B,FAT4,SYNE1,PCLO,HMCN1,FLG,ARID1A.The two genes that are the same are:TTN,TP53,MUC16,FAT4,ARID1A.1.1.3 Differences in mutation linkageMany genes in cancer show strong co-occurrence or exclusivity in mutation patterns.The linkage of mutations in DGC and IGC is mainly based on the co-occurrence relationship,and the number of linkages of mutant genes in DGC is less than that of IGC.There are a few mutations that are exclusive:in IGC,TP53 gene is exclusive with RELN,LRP2,PIK3CA,USH2A,KMT2D,OBSCN,ARID1A and MUC16;in DGC,CDH1 gene is exclusive with DNAH5,LRP1B and CSMD3.1.1.4 Differences in driver gene mutationsThe cancer driver genes are identified based on the oncodriveCLUST algorithm.The possible driver genes in DGC are SPECC1 and PLOD3,and the possible driver genes in IGC are MYBL1,ZHF330,DAZAP1,DXX39B,SERPIN1 and TVP23A.1.1.5 Differences in gene mutations of pathway nodesThe TCGA working group has identified 10 pathways related to cancer,and this article has enriched each pathway.Among them,the mutation rate of TGF-beta pathway node genes in IGC is 100%(7/7),and the mutation rate in DGC is also 100%(7/7);the mutation rate in DGC is less than that in IGC.1.1.6 Enrichment analysis of mutant genesThe GO and KEGG databases were used for functional enrichment of the DGC and IGC mutation differential genes,and it was found that the KEGG database enrichment did not enrich the relevant functions.The first three functions of using GO enrichment biological process(BP)are:the regulation of chemical synaptic transmission,the regulation of synaptic signal transduction,the complex combination of ?-catenin/TCF;the first three functions of GO enrichment molecular function(MF)The functions are:cortical cytoskeleton,cytoplasmic region,voltage-gated calcium channel complex;the first three functions of GO-enriched cell components(CC)are:calcium ion transmembrane transport protein activity,calcium channel activity,voltage-gated calcium Channel activity.1.2 Difference analysis of DNA copy number variation1.2.1 Differential genes with DNA copy number variationThere are 145 genes with DNA copy number variation in DGC,of which 124 are inserted and 21 are missing;there are 7992 genes with DNA copy number variation in IGC,of which 3908 are inserted and 4084 are missing.The top ten DNA copy number mutation genes in DGC are:RN7SL5P(Del),C12orf 77(Amp),LRMP(Amp),CENPUP2(Amp),CASC1(Amp)ETFRF1(Amp),KRAS(Amp),RNU4-67P(Amp),LMNTD1(Amp),TUBB4BP1(Amp).There are many DNA copy number mutation genes in IGC,and there are more than 10 genes with the same minimum FDR value,which are not listed in this article.1.2.2 DNA copy number differential gene enrichment analysisPerform functional enrichment for DNA copy number variant genes.In DGC,the first three pathways of KEGG enrichment are:melanoma,gastric cancer,and actin cytoskeleton regulation.In IGC,the first three positions of GO-enriched BP are:defense against other organisms,bacterial defense,and cell killing;the first three positions of GO-enriched MF are:serine endopeptidase inhibitor activity,taste receptor activity,and bitter taste Body activity;the top three enriched by KEGG are:the interaction of cytokines and their receptors,Kaposi's sarcoma-associated herpes virus infection and hepatitis B.2.DGC and IGC epigenomics difference analysis2.1 DNA methylation difference analysisThe differences in DNA methylation status of DGC and IGC were compared,and there was a total of 599 methylation sites with differences,and these sites contained 411 genes.Among them,there are 462 hypermethylation sites and 328 hypermethylation genes in DGC;137 hypermethylation sites and 83 hypermethylation genes in IGC(see Table 1 in the main text for details).2.2 DNA methylation differential gene function enrichment analysisThe DGC and IGC methylation differential genes were functionally enriched,and it was found that the KEGG database enrichment did not enrich the relevant functions.The first three functions of using GO to enrich BP are:gland development,regulation of cell-matrix adhesion,and positive regulation of protein localization in the membrane;the first three functions of GO enriching MF are:actin cytoskeleton,voltage gating Potassium channel complex,potassium channel complex;the first three functions of GO enrichment CC are:RNA polymerase ? proximal promoter sequence-specific DNA binding,proximal promoter sequence specific DNA binding,DNA binding transcription activator Activity,RNA polymerase ? specificity.2.3 Analysis of differences in non-coding RNA(ncRNA)expressionThe difference of miRNA expression between DGC and IGC was compared,and it was found that 57 genes with absolute logFC value greater than 2,4 were highly expressed in DGC,and 53 were highly expressed in IGC.The differences in the expression of lncRNA between DGC and IGC were compared,and it was found that a total of 24 genes with an absolute value of logFC greater than 2,8 were highly expressed in DGC,and 16 were highly expressed in IGC(see Tables 2 and 3 in the main text for details).3.DGC and IGC transcriptomics difference analysis3.1 Analysis of mRNA expression differencesThe mRNA expression difference between DGC and IGC was compared,and it was found that a total of 415 differential genes with an absolute value of logFC greater than 2 were found,including 233 highly expressed genes in DGC and 182 highly expressed genes in IGC(see Table 4 in the text for details).3.2 Analysis of mRNA expression differential gene enrichmentTo enrich these differentially expressed mRNAs,the first three functions of GO-enriched BP are:the growth process of the muscle system,the construction of extracellular structure,and muscle contraction;the first three functions of GO-enriched MF are:receptor regulation activity,receptor coordination Body activity and sulfide binding;the first three positions of GO-enriched CC are:extracellular matrix,collagen-containing extracellular matrix,and endoplasmic reticulum cavity;the first three positions of KEGG enrichment are:phosphatidylinositol 3-kinase signaling pathway,pancreas Secretion,protein digestion and absorption.4.DGC and IGC proteomics difference analysisThe difference in protein expression between DGC and IGC was compared.A total of 87 protein expressions were different.46 were highly expressed in DGC and 41 were highly expressed in IGC(see Table 5 in the main text for details).5.Joint analysis and attribution analysis of multiple omics parameters5.1 Combined analysis of DNA methylation and mRNA expressionThe combined analysis of DNA methylation differences and mRNA expression differences between DGC and IGC showed that 8 methylation sites were linked to mRNA expression.Among them,the gene with high methylation and low expression in DGC was IEIS1,and the gene with high methylation in IGC was high methylation.The low-expressed genes are SGK2,APOH,TMED6,SERPENA10,HNF4A.5.2 ceRNA network analysisComparing the differential lncRNA,miRNA,and mRNA found in this paper with the known binding lncRNA,miRNA,and mRNA in the database,it was found that IncRNA PVT1 can be the same as hsa-mir-106a,hsa-mir-106b,hsa-mir-17.hsa-mir-20a,hsa-mir-20b,hsa-mir-519d,and hsa-mir-93 interact,and these miRNAs can interact with POLQ,PARD6B,HMGA2,and CDC25A,suggesting that the above lncRNA-miRNA-There is a ceRNA regulatory network between mRNAs.5.3 Multidimensional regulatory network analysis of differentially expressed proteins in DGC and IGC related pathwaysComparing the protein expression results of DGC and IGC,it is found that in DGC,DNA damage response signal pathway,epithelial-mesenchymal transition signal pathway,Hormone A/B signal pathway,tyrosine kinase signal pathway,nodules the protein expression in the sexual sclerosis complex-rapamycin target protein signaling pathway are more active;while in IGC,the protein expression in the cell cycle signaling pathway is more active.Through multi-dimensional regulatory network analysis,it is found that the DNA damage response pathway is active in DGC because BRCA2 and Chkl are highly expressed in DGC,while BRCA2 is regulated by hsa-mir-369,hsa-mir-488,XIST,NEAT1,HCG18,and CHEK1 in DGC.It is regulated by hsa-mir-195,hsa-mir-497,PVT1,SNHG1,SNHG12,and HCG18.The epithelial-mesenchymal transition pathway is active in DGC because Collagen VI and E-Cadherin are highly expressed in DGC.In DGC,CDH1 is mainly affected by mutations,COL6A1 is affected by MEG3,and COL6A2 is affected by HSPA7,KCNQ10T1,FBXL21,PCDHB18,TUG1,PCDHB9,RPLP0P2,ANKRD26P1,ZFP92,ZDHHC8P1,CIDECP,ZSCAN12P1,CASC2,H19,C4orf10,GATS.1,UBE2Q2P1,OR2A9P are affected,COL6A3 is affected by ACTN3,HSPA7,PCDHB18,CASCAN12P1,PCDHBRP,PCDHB18,TUG1,PCDHBRP,LPC2P702P,H19,GATS.1,FAM35B2,OR2A9P.The hormone A pathway is active in DGC because ERalpha and PR are highly expressed in DGC,while ESR1 is affected by XIST,KCNQ10T1,and MEG3.The hormone B pathway is active in DGC because AR and Bcl-2 are highly expressed in DGC,while BCL2 in DGC is affected by KCNQ10T1,POLR2J4,NEAT1,HCG18,and MIAT.The activation of tyrosine kinase pathway is active in DGC because EGFRpY1 173 and HER3pY1289 are highly expressed in DGC,and EGRF is highly expressed in DGC,but no related miRNA and lncRNA have been found.The nodular sclerosis complex-target protein pathway of rapamycin is active in DGC because 4E-BP1pS65 and RictorpT1135 are highly expressed in DGC.In DGC and IGC,EIF4BP1 is affected by GAS5,and RPS6 is affected by hsa-let-7d,hsa-mir-1224,hsa-mir-125a,hsa-mir-212,hsa-mir-3187,hsa-mir-652,hsa-mir-744,SNHG1,SNHG12 affect.The cell cycle pathway is active in IGC because CDK1,CyclinB1,CyclinE1,CyclinE2,and PCNA are highly expressed in IGC.In IGC,CDK1 is affected by SNHG3,DLEU1,and TUG1.CCNE1 is affected by SNHG1 and HCG18,and CCNE2 is affected by hsa-mir-30a.5.4 Attribution analysis of multiple omics parametersAttribution analysis was performed on the multiple omics data parameters of 63 cases of DGC and 175 cases of IGC obtained from all enrollment analysis,including 10725 mutant genes,3086 DNA copy number variant genes,373333 methylation sites,and 799 miRNA genes There are 14,202 mRNA genes and 218 differentially expressed proteins.Through analysis,the correlation degree between each parameter and Lauren classification(DGC/IGC)is obtained.Among them,the highest correlation degree is DNA methylation,and the lowest correlation degree is gene mutation.Part 2 Comparison of the differences in immune microenvironment between diffuse gastric cancer and intestinal gastric cancer1.Differences in immune infiltrating cellsCompare the similarities and differences of the degree of immune cell infiltration between DGC and IGC.The results show that the immune cells with a high degree of infiltration in IGC and significant statistical differences include:Activated CD8 T cell,CD56dim natural killer cell,Central memory CD8 T cell,Effector memory CD8 T cell,Immature dendritic cell,Macrophage,Myeloid-derived suppressor cells(MDSC),Natural killer T cell,Regulatory T cell,Type 2 T helper cell.The immune cells that have a high degree of infiltration in DGC and have significant statistical differences include:activated B cells,activated dendritic cells,?? T cells,Immature B cell,Memory B cell,Natural killer cell,Plasmacytoid dendritic cell,and Type 17 T helper cell.2.Differences in chemokine ligands and receptor genesComparing the expression differences of DGC and IGC chemokines and receptors,it was found that a total of 28 chemokines and receptors showed statistical differences.Among them,the chemokines and receptors highly expressed in DGC are:CCL2,CCL3,CCL18,CCL20,CCL21,CCL25,CCL26,CXCL2,CXCL4,CXCL7,CXCL11,CXCL13,XCL2,CCR3,CCR4,CCR5,CCR7,CCR8,CXCR3;chemokines and receptors highly expressed in IGC are:CCL1,CCL8,CCL13,CXCL5,CXCL10,CX3CL1,XCL1,CCR1,CXCR2.3.Differences in interleukin gene expressionComparing the expression differences between DGC and IGC cytokines,it was found that there were 20 interleukins with expression differences.Among them,the interleukins highly expressed in DGC were:IL12B,IL13,IL17A,IL17F,IL22;the interleukins highly expressed in IGC were:IL1B,IL4,IL5,IL7,IL16,IL17B,IL17D,IL18,IL21,IL24,IL26,IL27,IL32,IL33,IL34.4.Joint analysis of immune infiltrating cells,chemokines and cytokinesIn IGC,the positively correlated chemokines and immune infiltrating cells are:CCL1,CCL5,CCL4,CXCL16 and CD56dim natural killer cells;CX3CL1,CCL8 and Eosinophil;CXCL2 and Mast cell;CCL21 and active B cells;CCL3 and Effector memeory CD4 T cells;CX3CL1 and natural killer T cells.Immune infiltrating cells and cytokines with higher positive correlation are:type 17 helper T cells and IL23A,IL17A,IL13;natural killer T cells and IL11,IL15;myeloid dendritic cells and IL13,IL17A;regulatory T cells and IL11,IL15;?? T cells and IL6.In DGC,the positively correlated chemokines and immune infiltrating cells are:CCL4,CCL11 and natural killer cells;CCL14,CXCL9 and myeloid dendritic cells;CCL11,CCL3,CCL5 and eosinophils;CXCL10,CCL20,CXCL1 and CD56dim natural killer cells.Immune infiltrating cells and cytokines with high positive correlation are:eosinophils,regulatory T cells,natural killer cells,naive B cells and TXLNA;natural killer T cells and IL10,IL26;natural killer cells and IL17F;regulation T cells and IL10,IL15;?? T cells and IL32.Part ? The histogenesis of diffuse gastric cancer and intestinal gastric cancer based on single-cell sequencing1.Single-cell sequencing maps of gastric mucosal epithelial cells in different gastric mucosal statesAfter analyzing the results of single-cell sequencing,a total of 29 types of cells were isolated using cell characteristic genes,including 15 types of epithelial cells and 14 types of mesenchymal cells(see Table 13 for characteristic genes).When using OLFM4 as a stem cell characteristic gene to mark cells,we found that some cells highly express OLFM4,but this part of the cell is far away from other stem cells in the UMAP diagram,and it is considered that this part of the cell is a subtype of stem cell,so the stem cells are named Stem cell 2(S2),and the remaining stem cells are named Stem cell 1(S1).After sorting the cells,it was found that from NAG to EGC and AGC,the cells are mainly Pit cell,Basal gland mucous cell(BGM),stem cell 1,and the proportion of stem cells gradually Increased,stem cell 1 accounted for the highest proportion in EGC.In DGC and IGC,the proportion of stem cells decreases,and the proportion of immune-related cells increases.2.Pseudotime analysis of gastric mucosal epithelial cellsUse Monocle3 to perform cluster analysis on all cells,and then perform pseudotime analysis.We found that there are two clusters containing DGC cells(PartA)and IGC cells(PartB).According to this,although it can be seen that there are differentiation trajectories,it is not easy to confirm the order of the states of the cells,so we further use the diffusion map for analysis.The results showed that PartA mainly contains S2 stem cells,goblet cells,intestinal endocrine cells,D cells,G cells and DGC cells.It was found that S2,most of the goblet cells,and DGC cells had cell aggregations close to each other,while intestinal endocrine cells and D cells had cell aggregations that were close together,and G cell aggregates had cell aggregations.It is far away from other cells.PartB mainly contains principal cells,basal gland mucous cells,S1 stem cells,progenitor cells,gastric pit epithelial cells,enteroty cells and IGC cells.Among them,the main cell and BGM cell aggregation is closer,the S1 and progenitor cells are closer together,and the gastric pit epithelial cells,IGC cells,and absorptive cells are the locations where cell aggregation occurs.The distance is closer.Through analysis,it is found that S2 has similar differentiation with goblet cells and DGC cells,and S1 has similar differentiation with absorber cells and IGC cells.3.Analysis of transcription factor activity of different gastric mucosal epithelial cellsThe SCENIC method was used to identify the transcription factor activity of various cells,and the transcription factor with predicted activity greater than 0.7 was selected as the transcription factor that functions in the cell.We analyzed 20 active transcription factors in S1,S2,absorber cells,goblet cells,DGC cells,and IGC cells,including 10 transcription factors(GEBPD,ELF4,KLF5,MAFG,MAZ,PRDM1,SMARCA4,SREBF2,SRF,YY1,ZNF143)are active in different gastric mucosal epithelial cells.Among the cell-specific active transcription factors,we are concerned that the active transcription factor GATA6 is only present in S2,goblet cells and DGC;CEBPG is present in S1,absorber cells,goblet cells,and IGC cells;KLF4 is present in S1,Absorber cells,goblet cells,DGC cells,IGC cells;TBX3 exists in S1,absorber cells;active transcription factors ELF3,HNF4A,MXD1 only exist in absorber cells;SUZ12 exists in IGC cells;RB1 exists in S1,S2,goblet cells,DGC cells,IGC cells;CD59 exists in S1,S2,absorber cells,goblet cells;CDX1 exists in absorber cells,goblet cells,DGC.4.Analysis of ligand-receptor interaction between cells in different gastric mucosal statesFurthermore,we use Cellchat to analyze the interaction between cells in different gastric mucosal states,and evaluate the function of each cell during the interaction.In DGC,DGC cells are used as Sender or Receiver and the ligand receptor pathways with an importance degree greater than 0.7 are:insulin-like growth factor(IGF)signaling pathway,macrophage migration inhibitory factor(MIF)signaling Pathway,midkine(MK)signaling pathway,platelet-derived factor(PDGF)signaling pathway,multi-effect growth factor(PTN)signaling pathway.In IGC,IGC cells are used as donors or receptors and the ligand receptor pathways whose importance is greater than 0.7 are:cluster of differentiation antigen CD40 signaling pathway,Fas receptor ligand(FASLG)signaling pathway,IGF signaling pathway,MIF signaling pathway,MK signaling pathway,PDFG signaling pathway,PTN signaling pathway,tumor necrosis factor-related apoptosis-inducing ligand(TRAIL)signaling pathway.Comparing the important ligand receptor pathways in DGC and IGC,we found that the IGF signaling pathway,MIF signaling pathway,MK signaling pathway,PDGF signaling pathway,and PTN signaling pathway also play important roles in DGC and IGC.The difference is that in DGC,the main role of MIF signaling pathway is MIF-(CD74+CXCR4),while in IGC,the main role is MIF-(CD74+CD44).In the MK signal pathway,the contribution of MDK-SDC4 in DGC is greater than that in IGC.In the PDFG signaling pathway,the contribution of PDGFA-PDGFRA in DGC is greater than that in IGC,while the contribution of PDGFA-PDGFRB in IGC is greater than that in DGC.In the PTN signaling pathway,the contribution of PTN-SDC4 in DGC is greater than that in IGC5.Pathway analysis of gastric mucosal epithelial cells in different gastric mucosal statesUsing the KEGG database,we analyze different gastric mucosa such as non-atrophic gastritis(NAG),chronic atrophic gastritis(CAG),intestinal metaplasia(IM),early gastric cancer(EGC),and advanced gastric cancer(AGC)including DGC and IGC.The gastric mucosal epithelial cell pathway was enriched and analyzed in the state.In the state of NAG and CAG,no results that meet the requirements were presented.In the IM state,S1 is enriched in phosphatidylinositol 3-kinase-protein serine threonine kinase signaling pathway,cancer proteoglycan pathway,mitogen-activated protein kinase signaling pathway,human papillomavirus infection,Rap1 signaling pathway,Focal adhesion pathway,actin cell bone regulation pathway,extracellular matrix receptor interaction pathway,breast cancer pathway,hematopoietic cell lineage pathway.Absorbed cells are enriched in Leishmaniasis pathway,tuberculosis pathway,Rapl signaling pathway,actin cell bone regulation pathway,pertussis virus pathway,complement and coagulation cascade pathway,hematopoietic cell lineage pathway,Th17 cell differentiation pathway,phagosome pathway,JAK-STAT signal pathway.It is worth mentioning that DGC cells can be seen in the IM state(see Table 14 in the main text for details),and the functional enrichment results show that DGC cells are enriched in human cytomegalovirus infection,viral proteins and cytokines and receptor pathways,toxoplasmosis path.In the EGC state of early gastric cancer,stem cells are enriched in cytokine-cytokine receptor interaction pathway,MAPK signaling pathway,PI3K-Akt signaling pathway,tumor proteoglycan,TGF-? signaling pathway,fluid shear stress and atherosclerosis Sclerosis,hematopoietic cell lineage,signaling pathways that regulate stem cell pluripotency,adhesion junctions,complement and coagulation cascade signaling pathways.In advanced gastric cancer,stem cell 1 is enriched in PI3K-Akt signaling pathway,MAPK signaling pathway,focal adhesion pathway,tumor proteoglycan,EGFR tyrosine kinase inhibitor resistance pathw... |
PDF Full Text Request