Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Identification of Prognostic Markers and Potential Therapeutic Targets in Gastric Adenocarcinoma by Machine Learning Based on mRNAsi Index

Identification of Prognostic Markers and Potential Therapeutic Targets in Gastric Adenocarcinoma... Hindawi Journal of Oncology Volume 2022, Article ID 8926127, 14 pages https://doi.org/10.1155/2022/8926127 Research Article Identification of Prognostic Markers and Potential Therapeutic Targets in Gastric Adenocarcinoma by Machine Learning Based on mRNAsi Index 1 2 2 Si Hong Guo, Li Ma, and Jie Chen Personal Health Management, Hong Kong Baptist University, Hong Kong 999077, China Department of Gynaecologic Oncology, Harbin Medical University Cancer Hospital, Harbin 150000, China Correspondence should be addressed to Jie Chen; cj2365255@hrbmu.edu.cn Received 15 August 2022; Revised 11 September 2022; Accepted 14 September 2022; Published 30 September 2022 Academic Editor: Zhongjie Shi Copyright © 2022 Si Hong Guo et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Background. Cancer stem cells (CSCs), characterized by self-renewal and therapeutic resistance, play important roles in stomach adenocarcinoma (STAD). However, the molecular mechanism of STAD stem cells is still unclear. In this study, our purpose is to explore the expression of stem cell-related genes in STAD. Methods. The stemness index based on mRNA expression (mRNAsi) was used to analyze STAD cases in The Cancer Genome Atlas (TCGA). Firstly, mRNAsi was used and analyzed by differential expression, survival analysis, clinical stage, and gender in STAD. Then, weighted gene coexpression network analysis (WGCNA) was used to discover the fascinating modules and key genes. Enrichment analysis was carried out to annotate the functions and pathways of key genes. The gene expression comprehensive database (GEO) in STAD was used to verify the expression levels of key genes in all cancers. Protein-protein interaction networks is used to determine the relationships between key genes. Results. The mRNAsi was obviously upregulated in tumor cases. With the increase of tumor stage and T stage, the mRNAsi score decreased, and the overall survival rate of high score group patients was better. According to the degree of association with mRNAsi, different modules and key genes were screened out. A total of 6,740 differential genes were found, of which 1,147 genes were downregulated and 5,593 genes were upregulated. 19 key genes (BUB1, BUB1B, KIF14, NCAPH, RACGAP, KIF15, CENPF, TPX2, RAD54L, KIF18B, KIF4A, TTK, SGO2, PLK4, ARHGAP11A, XRCC2, Clorf112, NCAPG, and ORC6) were screened due to significant upregulation in STAD. And they had been proven that enriched from the cell cycle Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway, relating to cell proliferation Gene Ontology (GO) terms, as well. Among them, 9 genes have been extensively associated to OS, and 3 genes had been associated to receive chemotherapy resistance. PPI protein network suggests that there is a sturdy correlation between these key genes. Conclusion. A total of 19 key genes were found to play an essential position in retaining the traits of STAD stem cells. These genes can be used to evaluate the prognosis of STAD patients or become specific therapeutic targets. 1. Introduction unlimited proliferation [3–5]. CSC theory points out that tumor proliferation, therapeutic resistance, and recurrence The incidence rate and mortality of stomach cancer are additionally pushed by way of a small range of tumor stem decreased significantly in five years, but it still ranked third cells hidden in most cancers. It explains these clinical observa- among common malignant tumors and the second leading tions, such as tumor recurrence, tumor dormancy, and metas- cause of cancer-related death [1]. Ninety percent of all tasis after successful surgical resection, chemotherapy, and tumors of the stomach are malignancies, and stomach ade- radiotherapy [6]. CSCs have been found in several human nocarcinoma (STAD) accounts for 95% of all cases of malig- malignancies, such as leukemia [7], breast cancer [8], colorec- nancies [2]. tal cancer [9], and brain cancer [10]. In addition, strong pre- In current years, the characteristic of most cancers stem clinical data and clinical evidence have been added as telephone has been mentioned such as self-renewal and supports of the existence of gastric CSCs [11]. Therefore, 2 Journal of Oncology 1.00 0.75 0.50 0.8 p = 3.761e–09 0.25 p < 0.001 0.00 0.6 0 12345 6789 10 Time (years) 0.4 mRNAsi High Low 0.2 High 197125 41 20 13 741111 Low 197105 23 11 7422220 0.0 0 123456789 10 Normal Tumor Time (years) (a) (b) mRNAsi (p = 0.001) mRNAsi (p = 0.034) 0.7 0.7 0.6 0.6 0.5 0.5 0.4 0.4 0.3 0.3 0.2 0.2 0.1 0.1 Stage I Stage II Stage III Stage IV G1 G2 G3 Stage Grade (c) (d) mRNAsi (p = 0.024) Volcano 0.7 15 0.6 0.5 0.4 0.3 0.2 0.1 T1 T2 T3 T4 –10 –5 0 5 10 logFC (e) (f) Figure 1: The correlation of mRNAsi profiles with STAD. (a) Scatter plot illustrating the difference of mRNAsi index expression between normal tissues and tumors. (b) Kaplan–Meier survival curve of correlation between mRNAsi score and OS of STAD patients. Detect the correlation between mRNAsi score and the Grade (c), Stage (d), and T degree (e) by the Kruskal-Wallis test. (f) Volcano map of DEGs between STAD tissues and normal tissues. Downregulated genes are indicated in green, and upregulated genes are indicated in red. STAD: stomach adenocarcinoma; DEGs: differentially expressed genes. mRNAsi mRNAsi mRNAsi mRNAsi mRNAsi –log10 (fdr) Survival probability Journal of Oncology 3 Sample clustering to detect outliers 8e + 04 6e + 04 4e + 04 2e + 04 0e + 00 (a) Scale independence Mean connectivity 6 19 7 161718 20 4 8 10 13 9 1112 0.8 0.6 0.4 0.2 56 7 1 0 89 1011121314151617181920 510 15 20 510 15 20 Soft threshold (Power) Soft threshold (Power) (b) Module-trait relationships MEmagenta MEpurple MEtan MEblue 0.5 MEpink MEcyan MEgreenyellow MEmidnightblue MElightcyan MEblack MEbrown Cluster dendrogram MEred –0.5 1.0 MEgrey60 0.9 MEsalmon 0.8 MEturquoise 0.7 MEgrey –1 0.6 0.5 0.4 0.3 Dynamic tree cut Merged dynamic (c) (d) Figure 2: Continued. Scale free topology model fit, signed R Height Height Mean connectivity mRNAsi EREG–mRNAsi 4 Journal of Oncology Module membership vs. gene significance Module membership vs. gene significance cor = 0.88, p<1e–200 cor = 0.74, p = 4, 1e–150 0.5 0.4 0.4 0.3 0.3 0.2 0.2 0.1 0.1 0.0 0.0 0.0 0.2 0.4 0.6 0.8 1.0 0.2 0.4 0.6 0.8 Module membership in blue module Module membership in brown module (e) (f) Module membership vs. gene significance cor = 0.081, p = 0.23 0.25 0.20 0.15 0.10 0.05 0.00 0.0 0.2 0.4 0.6 0.8 Module membership in pink module (g) Figure 2: Construction of weighted gene coexpression network for STAD stemness related datasets. (a) Identify and remove outlier samples through average linkage hierarchical clustering. Samples exceeding the red line were considered deviations in gene expression. (b) Network topology analysis of different soft threshold powers. The left figure shows the influence of soft threshold power on the scale-free topological fitting index. The right figure shows the influence of soft threshold power on average connectivity. (c) Clustering dendrograms was done via mean linkage hierarchical. (d) Module-trait relationships. Each column represents a clinical phenotype, and each row denotes an ME. The correlation coefficient and P value are contained in each cell. (e–g) Scatterplots of GS for weight vs. MM to pick out the key genes from the blue, brown, and pink modules. STAD: stomach adenocarcinoma; ME: module eigengene; GS: gene significance; MM: module membership. CSC research is able to provide a new paradigm for managing Tathiane et al. used publicly available molecular profiles patients with STAD. from TCGA to obtain two independent stemness indices A growing number of studies have shown cancer stem- by using original one-class logistic regression machine- ness is associated with being transcriptomic, genomic, epi- learning algorithm (OCLR) to complete the integration genomic, and proteomic [12]. Within the last decade, The of transcriptome, methylome, and transcription factor Cancer Genome Atlas (TCGA) has elucidated the primary [15]. One was mDNAsi which reflects epigenetic features; tumor landscapes by generating comprehensive multiomics the other was mRNAsi which reflects gene expression. characteristics, along with pathophysiological feature and Malta et al. identified the relationship between the two clinical information annotations [13]. Machine learning stem cell indices and new carcinogenesis pathways, has been increasing applied in various areas of society somatic cell changes, microRNAs (miRNAs), and tran- and has become a useful strategy in biotechnology [14]. scription regulatory networks. These characteristics are Gene significance for EREG-mRNAsi Gene significance for EREG-mRNAsi Gene significance for EREG-mRNAsi BP CC MF Journal of Oncology 5 p. adjust Sister chromatid segregation Mitotic sister chromatid segregation Nuclear chromosome segregation Mitotic nuclear division Nuclear division Organelle fission Chromosome segregation 0.005 Meiotic cell cycle Meiotic chromosome segregation Chromosome separation Spindle Chromosome, centromeric region Condensed chromosome Condensed chromosome, centromeric region 0.010 Kinetochore Condensed chromosome outer kinetochore Kinesin complex Chromosomal region Microtubule Condensed chromosome kinetochore 0.015 Microtubule motor activity Microtubule binding Motor activity Tubulin binding ATPase activity ATP-dependent microtubule motor activity, plus-end-directed ATP-dependent microtubule motor activity 0.020 Protein serine/threonine kinase activity Nucleoside-triphosphatase regulator activity ATPase activity, coupled 0.0 2.5 5.0 7.5 10.0 12.5 (a) p. adjust 0.0004 Cell cycle 0.0008 0.0012 Homologous recombination 0.0016 0 1 2 3 4 (b) Figure 3: Functional analysis of brown module. (a) GO enrichment analysis. (b) KEGG enrichment analysis. GO: Gene Ontology; KEGG: Kyoto Encyclopedia of Genes and Genomes; BP: biological process; CC: cellular component; MF: molecular function. 6 Journal of Oncology related to cancer stem cells in specific molecular subtypes 2.3. 2.3 Differential Expressed Gene (DEG) Analysis. We used of TCGA tumors, which may be the factors controlling the R package “limma” for differential expression analysis cancer stem cells. Importantly, higher stem cell index [20]. We used the cut-off values, which were fold change > value is related to the active biological processes and 1 and adj:P <0:05, to screen for DEGs between normal greater tumor dedifferentiation in tumor stem cells, as health and stomach adenocarcinoma samples. The volcano reflected in histopathological grade. Metastatic tumor cells plot and the box-plots showing differences in key genes pre- show more dedifferentiation in phenotype, which may sented in this study were drawn by the R package “pheat- contribute to their invasiveness. The stemness indices map” and “ggpubr,” respectively. had positive correlation with tumor dedifferentiation and 2.4. WGCNA. WGCNA was performed using the WGCNA biological active in CSCs [16]. The mRNAsi and mDNAsi R package [17], which were “matrixStats,”“Hmisc,”“fore- scores in TCGA samples had been calculated by applying ach,”“doParallel,”“fastcluster,” dynamicTreeCut,”“sur- the stemness indices. vival,” and “WGCNA.” Before the building of coexpression Weighted gene coexpression network analysis network, the rectangular Euclidean relative distance of every (WGCNA), a method commonly used to explore biological take a look at pattern was once calculated by means of prac- networks, paired relationships between genes and pheno- tical adjacency method, and the integration connectivity of types. WGCNA transforms gene expression data into coex- the total pattern community calculated via distance was once pression module, providing insights into signaling standardized via practical scaling method. Due to some networks, and mine the pathway-related modules [17]. It is exceptional genes with no tremendous trade in expression widely applied in many physiological and pathological pro- between samples which are surprisingly correlated in cesses, including cancer, genetic therapy, and clinical data WGCNA as a whole, it appears that the genes with the analysis, which can be useful for identifying biomarkers of most biased expression have been used in the subsequent disease or target points for therapy [18, 19]. WGCNA analysis. The gene with the highest DEG vari- In this study, our purpose is to identify key genes associ- ance of 25% was selected. Clear ordinary value pattern ated with STAD stemness in TCGA based on mRNAsi information with connectivity is much less than -2.5. scores. The purpose of this study was to provide an interest- Function pickSoftThreshold was used to calculate scale- ing bioinformatics method for identifying stem cell-related free topology becoming indicesR corresponding to one- genes and revealing the role of some CSC-related genes in STAD. of-a-kind smooth thresholding powersβ. Theβvalue was used as lengthy asR reaching 0.8. After that, the gene expression matrix was converted into an adjacency matrix 2. Materials and Methods and a Topological Overlap Matrix (TOM), and then the corresponding dissimilarity of TOM (dissTOM) was calcu- 2.1. Software and R Packages. We used R Studio version lated. For module detection, hierarchical clustering was 1.2.5042 (URL: https://rstudio.com/) with R version 3.6.2 used to produce a hierarchical clustering tree (dendro- (URL: https://www.r-project.org/) in this study. The pro- gram) of genes by using characteristic “hclust” based gramming software Perl version 64-bit (URL: https://www totally on dissTOM. The Dynamic Tree Cut approach .perl.org/) was used for data processing. All R packages were was carried out for department reduction to generate downloaded from Bioconductor (URL: https://www modules. During this, a quite massive minimal module .bioconductor.org/). measurement of minClusterSize = 30 to department split- ting had been chosen to avoid producing too many small 2.2. Database and mRNAsi Index. The RNA-sequencing or massive modules. To consider the magnitude of every (RNA-seq) of STAD and all pathological and clinical infor- module, gene significance (GS) was once calculated to mation were downloaded from TCGA database (URL: measure the correlation coefficient between genes and pat- https://portal.gdc.cancer.gov/). These data were updated on tern traits. The module eigengene (ME) is described as the 5 October 2019. The results of RNA-seq were including first foremost thing of a given element and can be 375 cancer samples and 32 normal samples, structured for regarded as a consultant of the gene expression profile of a matrix file. We used Ensemble data to exchange the gene the module integration. It was calculated by using pur- names expressed by Ensembl IDs which are specifically con- poseful module genes. If their MES correlation coefficient verted into a gene symbol matrix. Moreover, to explore the is higher than 0.75, the modules will be merged, with mode of action of CSC-related genes in chemotherapy capacity that they have considerable comparable gene resistance, we download the microarray (GSE14210) expression levels. Here, we can pick out mRNAsi and epi- results from the Gene Expression from the Gene Expres- genetically regulated mRNAsi as scientific phenotypes. sion Omnibus (GEO) (URL: https://www.ncbi.nlm.nih After selecting the components of interest, let us cal- .gov/geo/). We referred to the mRNAsi index data for all culate the GS and module membership of each key gene types of tissues in the supporting documents to Malta (MM, the significance between the module’s own gene et al.’s article and specifically screened the mRNAsi index and gene expression profile), and set their threshold of patients with stomach adenocarcinoma for incorpora- values. The thresholds for screening key genes in the tion into TCGA data for stomach adenocarcinoma, with module were defined as cor.gene MM > 0:8 and cor.gene the unmatched cases deleted. GS > 0:5. Journal of Oncology 7 ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ Type (a) BUB1 (209642_at) KIF14 (236641_at) 1.0 1.0 HR = 0.83 (0.7–0.98) HR = 0.69 (0.55–0.86) Logrank P = 0.029 Logrank P = 0.00076 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0.0 0.0 0 50 100 150 0 50 100 150 Time (months) Time (months) Number at risk Number at risk Low 439 129 34 1 Low 316 111 30 1 High 436 169 14 0 High 315 154 18 0 Expression Expression Low Low High High (b) (c) NCAPH (212949_at) 1.0 HR = 0.76 (0.64–0.9) Logrank P = 0.0015 RAD54L (204558_at) 0.8 1.0 HR = 1.21 (1.02–1.44) Logrank P = 0.026 0.6 0.8 0.4 0.6 0.2 0.4 0.0 0.2 0 50 100 150 0.0 Time (months) Number at risk 0 50 100 150 Low 441 124 29 1 Time (months) High 434 174 19 0 Number at risk Low 438 160 33 0 Expression Low High 437 138 15 1 High Expression Low High (d) (e) Figure 4: Continued. Probability Probability Gene expression ARHGAP11A BUB1 BUB1B Clorf112 CENPF KIF14 KIF15 KIF18B KIF4A Probability Probability NCAPG NCAPH ORC6 PLK4 RACGAP1 RAD54L SGO2 TPX2 TTK XRCC2 8 Journal of Oncology PLK4 (204886_at) ARHGAP11A (204492_at) 1.0 1.0 HR = 0.81 (0.68–0.96) HR = 0.8 (0.67–0.95) Logrank P = 0.013 Logrank P = 0.0097 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0.0 0.0 0 50 100 150 0 50 100 150 Time (months) Time (months) Number at risk Number at risk Low 438 123 30 1 Low 439 129 31 1 High 437 175 18 0 High 436 169 17 0 Expression Expression Low Low High High (f) (g) XRCC2 (207598_x_at) CAP-G (218662_s_at) 1.0 1.0 HR = 1.87 (1.57–2.22) HR = 0.78 (0.66–0.93) Logrank P = 6.3e–13 Logrank P = 0.005 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0.0 0.0 0 50 100 150 0 50 100 150 Time (months) Time (months) Number at risk Number at risk Low 439 191 27 0 Low 439 123 35 1 High 436 107 21 1 High 436 175 13 0 Expression Expression Low Low High High (h) (i) ORC6L (219105_x_at) 1.0 HR = 1.19 (1.01–1.41) Logrank P = 0.042 0.8 0.6 0.4 GSE14210 0.2 0.0 0 50 100 150 Time (months) 308 3 16 Number at risk Low 439 162 33 1 High 436 136 15 0 Expression Low High Key genes (j) (k) Figure 4: Verification of the influence of key genes on diseases. (a) The mRNA expression level of key genes between tumor and normal tissues in TCGA STAD dataset. The Kaplan–Meier plotter database was used to assess the correlation between the expression of BUB1 (b), KIF14 (c), NCAPH (d), RAD54L (e), PLK4 (f), ARHGAP11A (g), XRCC2 (h), NCAPG (i, also called CAP-G), ORC6 (j, also called ORC6L), and the OS of STAD patients. Kaplan–Meier survival plots (K–M plots) were generated using the on-line tool, Kaplan–Meier plotter. (k) Venn diagram of the relationship between 19 key genes and acquired chemoresistance by GSE14210. 2.5. Overall Survival Curve. Finally, to determine the prog- overall survival deviation between patients with low and nostic significance and value of mRNAsi scores, we can draw high mRNAsi index. In this part, R package “survival” and the Kaplan–Meier diagram of mRNAsi index to explore the “surviminer” were used, and a log-rank test is used to test Probability Probability Probability Probability Probability Journal of Oncology 9 the relationship between them. In key gene validation analy- which 1147 were downregulated and 5593 were upregulated sis, Kaplan–Meier survival curves of key genes were drawn (Figure 1(f)). with the online tool Kaplan–Meier plotter [21] 3.2. WGCNA: Identifying the Most Significant Modules and (URL:http://www.kmplot.com/analysis/index.php?p= service). Genes. With WGCNA, we built a gene coexpression network to become aware of biologically significant gene modules. It 2.6. Functional Annotation Gene Ontology (GO) and Kyoto can help us to understand the genes associated with STAD Encyclopedia of Genes and Genomes (KEGG) Analyses. The stemness. We put 6740 DEGs with the highest variance of 25% into the same module through cluster analysis. Before GO functional annotations and KEGG pathway enrichment analysis shown in this study were obtained from the data that, the outlier samples should be removed (Figure 2(a)). analysis conducted by the R package “cluster Profiler” to According to the lowest value of scale-free network, the β investigate the biological functions of key genes. The thresh- value is determined. What the pickSoftThreshold function old values were as follows: P <0:01 and FDR < 0:05. does is to find the appropriate power. The selection of the power value is determined by β value. Calculate the correla- 2.7. Gene Coexpression Analysis and Construction of Protein- tion intensity (weighted correlation value) of expression Protein Interaction (PPI) Network. In order to further study levels among all genes to obtain the adjacency matrix. As a the stability of these special relationships at the transcrip- result, we choose β =4 (scale-free R =0:9) as the soft tional level, we calculated the coexpression relationships threshold (Figure 2(b)). We find 16 modules for subsequent among key genes within the module depending on the gene evaluation (Figure 2(c)). expression level. The R “corrlot” package is mainly used to Taking MS as the total gene expression level of the cor- calculate the Pearson correlation degree between genes. responding module, the correlation between MS and clinical The STAD data set was selected from TCGA for analysis phenotype was calculated. This is useful for us to discover and research, and the routine data were analyzed by the the relationship between these modules and the dryness Pearson correlation test. Results with a correlation index of the sample. By calculating the Pearson correlation coefficient > 0:3 and P value < 0.01 were considered statisti- coefficient, a threshold value can be obtained. If the correla- cally significant. tion coefficient is greater than 0.8 or so, it can be used as the Accurately retrieve PPI network from STRING version basis of strong correlation between the two genes. The con- 11.0 (URL: https://string-db.org/) [22]. And display the bar sequences confirmed that the blue and brown modules were graph of the nodes in the network with top-level network extensively correlated with mRNasi, and the correlation was connectivity. The minimum required interaction score was once close to 0.8. However, the correlation coefficient of the set to a medium confidence of 0.4, and now, the hidden brown module is 0.77, which is higher. In addition, the pink branch nodes in the network are disconnected. The number module was fantastically negatively correlated with mRNasi of adjacent nodes of each gene in the PPI network was calcu- (Figure 2(d)). Therefore, the brown module was chosen lated, and then, the genes were sorted by the bar graph com- through us as the most fascinating module for subsequent bined with the number of adjacent nodes. analysis. The threshold for screening key genes in the mRNAsi group was described as cor:MM > 0:8 and cor:GS > 0:5.We 3. Results pick 19 key genes (BUB1, BUB1B, KIF14, NCAPH, RAC- 3.1. Clinical Characteristics of mRNAsi and DEGs in STAD. GAP1, KIF15, CENPF, TPX2, RAD54L, KIF18B, KIF4A, The mRNAsi is an index of CSCs that can quantitatively TTK, SGO2, PLK4, ARHGAP11A, XRCC2, Clorf112, describe the similarity between tumor cells and stem cells. NCAPG, and ORC6), as shown in Figures 2(e)–2(g). And We observe large distinction in mRNAsi between tumor we exhibit the distinct expressions of key genes between and ordinary tissues (Figure 1(a)). In the survival analysis, most cancers and ordinary samples in TCGA; all these genes we divided gastric cancer patient into higher mRNAsi score in brown module are upregulated in tumor cases (Figure 2(f group and lower mRNAsi group by using mRNAsi median )). value. Obviously, patients with higher mRNAsi scores have greater overall survival in contrast with sufferers with lower 3.3. Enrichment Analysis of Brown Module. We use GO and mRNAsi scores (Figure 1(b)), the five-year survival rate of KEGG analysis to elucidate the function similarities of mod- higher scores group is 47.9% with CI (0.344, 0.668), and ule brown genes. The results show that nuclear division, the lower scores group is 21.2% with CI (0.107, 0.421). Sur- spindle, and microtubule binding are the most great enrich- prisingly, the mRNAsi scores tend to decline with the grade ments in cellular component (CC), biological process (BP), increasing with the exception of G1 (only 8 samples). Also, and molecular function (MF) groups (Figure 3(a)). KEGG the mRNAsi score shows an overall decreasing trend in stage pathway enrichment analysis suggested cell cycle and and T (Figures 1(c), 1(d), and 1(e)). The Kruskal Wallis test homologous recombination pathways are significant path- was once used to determine the value of variations between ways (Figure 3(b)). All of them are related to cancer stem groups. cells. We download mRNA-seq data and did difference analy- sis to compare STAD and normal since the mRNAsi differ- 3.4. Data Validation. Firstly, the STAD dataset of TCGA ence between tumor and normal. We find 6740 DEGs in showed significant differences in the expression of all key 10 Journal of Oncology (a) XRCC2 1 ORC6 C1orf112 8 SGOL2 RAD54L 14 PLK4 KIF18b 14 KIF14 ARHGAP11A 14 KIF4A TTK 16 RACGAP1 NCAPH 16 KIF15 CENPF 16 BUB1B TPX2 17 NCAPG BUB1 17 0 5 10 15 20 (b) Figure 5: Continued. Journal of Oncology 11 ORC6 C1orf112 0.56 0.8 TPX2 0.59 0.73 XRCC2 0.66 0.62 0.68 0.6 TTK 0.61 0.72 0.79 0.71 KIF14 0.59 0.76 0.73 0.67 0.73 0.4 SGO2 0.58 0.69 0.73 0.67 0.79 0.79 CENPF 0.53 0.73 0.73 0.63 0.7 0.88 0.76 0.2 BUB1 0.63 0.74 0.78 0.7 0.77 0.78 0.82 0.75 RACGAP1 0.61 0.66 0.73 0.63 0.72 0.74 0.73 0.72 0.77 0 KIF15 0.56 0.69 0.67 0.68 0.71 0.75 0.74 0.78 0.73 0.72 –0.2 BUB1B 0.65 0.69 0.75 0.7 0.75 0.74 0.78 0.74 0.84 0.74 0.76 KIF18B 0.53 0.62 0.73 0.7 0.67 0.7 0.69 0.73 0.72 0.74 0.73 0.73 –0.4 KIF4A 0.52 0.65 0.74 0.62 0.7 0.72 0.71 0.74 0.78 0.75 0.71 0.75 0.74 NCAPH 0.57 0.69 0.75 0.66 0.71 0.69 0.72 0.69 0.87 0.77 0.72 0.77 0.76 0.79 –0.6 ARHGAP11A 0.58 0.6 0.64 0.61 0.67 0.76 0.72 0.75 0.79 0.7 0.75 0.82 0.68 0.72 0.73 PLK4 0.6 0.62 0.64 0.68 0.71 0.7 0.71 0.7 0.79 0.7 0.75 0.81 0.72 0.72 0.75 0.77 –0.8 NCAPG 0.57 0.6 0.63 0.61 0.7 0.7 0.72 0.7 0.76 0.72 0.75 0.78 0.67 0.73 0.75 0.75 0.8 RAD54L 0.54 0.64 0.68 0.66 0.66 0.64 0.65 0.64 0.76 0.72 0.74 0.75 0.76 0.75 0.81 0.67 0.75 0.72 –1 (c) Figure 5: PPI interactive network. (a) String diagram composed of 19 key genes as nodes. (b) The bar-plot lists the connections of key genes in the brown module by the counts of connections. (c) Correlation analysis between key genes. The higher phase of the graph indicates the degree of correlation. The darker the color, the greater the correlation. The lower part shows the corresponding correlation value. PPI: protein-protein interaction. genes between normal and tumor cases (Figure 4(a)). In all 3.5. Protein-Protein Interactions (PPI) among Genes of patients with STAD, the Kaplan–Meier curve and log rank Brown Module. The application of the on-line device test analysis showed that 7 genes in the brown module were STRING (URL: http://string-db.org/) to protein-protein significantly associated with OS (P <0:05, FDR < 0:05) interaction networks for each module will assist us to (Figures 4(b)–4(j)). explore the interplay between key genes extra deeply. There It is well known that CSCs have chemoresistance, and were 19 nodes and 129 edges in the shaped PPI network, and resistance is related to cancer-associated fibroblasts in the the PPI enrichment (P value < 0.01) (Figure 5(a)). In addi- extracellular matrix [23]. The mapping of GSE14210 is tion, the significant nodes shown in the bar-plot can identify based on Venn diagram. 19 key gene maps selected from the genes most closely related to other members of the mod- the brown module were scored by GS and MM scores. ule (Figure 5(b)). Finally, SGO2, TTK, and CENPF were associated with The correlation between the key genes of this module the acquired chemoresistance to cisplatin and fluorouracil was strong (Figure 5(c)), and the correlation was statistically combination chemotherapy in gastric cancer (Figure 4(k)). significant (P <0:01). The correlation between CENPF and ORC6 C1orf112 TPX2 XRCC2 TTK KIF14 SGO2 CENPF BUB1 RACGAP1 KIF15 BUB1B KIF18B KIF4A NCAPH ARHGAP11A PLK4 NCAPG RAD54L 12 Journal of Oncology ORC6 (0.53), KIF18B, and ORC6 (0.53) was the lowest, tion characteristics. The pathway enrichment advised that whereas the correlation between CENPF and KIF14 (0.88) the four key genes in the cycle pathway time period have was the highest. been most possibly a useful gene set that impacts tumor stemness via regulating the cell cycle. The gene units that keep the traits of stem cells in a range 4. Discussion of cancers may additionally have similarities. Since the for- The morbidity and mortality of gastric most cancers stay mation of a range of organ tissues takes place from pluripo- excessive all over the worldwide. In current years, CSCs have tent stem cells, their CSCs are dedifferentiated with stem been mentioned to make vital contributions to tumor pro- mobile phone characteristics. This reverse improvement gression, recurrence, and therapeutic resistance [24, 25]. has made a range of CSCs possessing some traits of pluripo- Therefore, therapy concentrating on STAD stem cells is tent stem cells. Moreover, their stage of overexpression was essential. In addition, choosing out the emergence of these once associated to the stage of stemness, and their persisted druggable genetic ameliorations in pancancer cases, and expand might also promote modifications in tumor develop- whether or not there are modifications in the expression of ment and posttherapy progression. More than half of the key the equal mRNAsi-related genes, is additionally a query genes have been mentioned in STAD, and some have been proven to be related with the traits of CSCs. BUB1 is related priceless of dialogue in the future work. In this study, we tried to discover key genes associated to with the most cancer stem cell attainable in breast cancer STAD stem cells in TCGA database. We used WGCNA [29]. An issue highlights a study that links the presentation based totally on mRNAsi scores, as calculated by Salomonis of kinetochores within mitosis to an essential requirement et al. [26]. The tumor case had a greater mRNAsi rating than for BUB1 threonine kinase B (BUB 1B), broadening our understanding of the cell-cycle machinery in CSCs [30]. Kine- the regular case. The mRNAsi scores reduced with the sick- ness grade, stage, and T stage, though the mRNAsi rating of sin family member 15 (KIF15) promotes the CSC phenotype G1 was once small which may also be associated to inade- and malignancy by means of PHGDH-mediated ROS imbal- quate pattern size. The excessive mRNAsi team confirmed ance in hepatocellular carcinoma [31]. TTK gene was overex- a decrease survival chance than the low team in the first 5 pressed in the CSC-like cell populace remoted from human years, which used to be constant with the negative conse- esophageal carcinoma phone strains as properly as in the quence related with stemness features. human more than one myeloma stem cells sorted through We developed coexpression modules through WGCNA aldehyde dehydrogenase 1 (ALDH1) [32, 33]. and pick out brown module as the best correlations with Survival curves have been generated to validate the prog- mRNAsi. Key genes have been screened from the blue module nostic fee of the key genes in brown module in STAD. In the primarily based on the GS and MM scores. The expression K-M plots, 7 genes had been substantially related with prog- degrees of key genes are appreciably upregulated in tumor noses (P <0:01, FDR < 0:05). High expression of BUB1, samples. There have been robust coexpression relationships TPX2, and X-ray repair cross complementing 2 (XRCC2) at the transcriptional degree in brown module. There was had been noticeably related with negative prognoses. The additionally a robust PPI community among proteins of this expression of NCAPH, NCAPG, RACGAP1, and SGO2 module. The key genes intently associated to pluripotent stem has been positively correlated with affected person progno- cells have been confirmed to be overexpressed in most tumors. sis. As known, CSCs can withstand clinical remedy and Moreover, all organ tissues are developed from pluripotent make contributions to tumor relapse. The key genes had stem cells, suggesting that key genes might also play a position been validated in GSE14210, and SGO2, TTK, and CENPF in keeping stem cellphone residences in a range of cancers. have been related with the obtained chemoresistance to cis- The consequences led us to reassess the relationship between platin and fluorouracil mixture chemotherapy in gastric can- CSC traits and STAD progression. cer. Several studies had proven that CSCs have one or Undifferentiated major tumors are more probably to rea- greater abnormalities in signaling pathways that modify son most cancer cells to unfold to far-off organs, mainly to self-renewal. The Wnt/β-catenin, Notch, and Hedgehog sickness development and negative prognosis. Moreover, pathways have been mentioned fully [34]. Wnt/β-catenin CSCs are typically resistant to handy remedies [27]. The KIF14, TPX2, KIF18B, and PLK4 in the Wnt/β-catenin acquisition of progenitor cell-like and stem cell-like traits pathway [35–38], TTK and XRCC2 in the Hedgehog path- and loss of the differentiated phenotype are manifestations way [37, 38], and RACGAP1 and TTK in the Notch pathway of most cancer development [28], regular with the expand [39, 40] may additionally be necessary for the tumorigenicity in STAD stemness as the tumor progression. In our study, of CSCs. These genes are vital therapeutic aimed at inhibit- ing the self-renewal, proliferation, and tumor development we observed that sufferers with greater corrected mRNAsi rankings had decreased ordinary survival rates, which used of CSCs. to be regular with the negative prognosis related with CSC characteristics. Disease stage 1 and T1 stage STAD had 5. Conclusions pretty greater CSC characteristics, indicating the stem tele- phone residences start to upward thrust from initiation of In summary, 19 key genes have been determined to play nec- the cancer. essary roles in STAD stem phone maintenance. The valida- Functional annotations of the brown module had been tions confirmed that these genes ought to be beneficial for chiefly associated to the stem cell self-renewal and prolifera- outlining the prognosis of STAD patients. These genes may Journal of Oncology 13 also be therapeutic pursuits for inhibiting STAD stemness Sciences of the United States of America, vol. 100, no. 7, pp. 3983–3988, 2003. characteristics. However, our conclusions are primarily based on the retrospective information, and similarly [9] C. A. O'Brien, A. Pollett, S. Gallinger, and J. E. Dick, “A human colon cancer cell capable of initiating tumour growth in organic and scientific investigation of these genes should immunodeficient mice,” Nature, vol. 445, no. 7123, pp. 106– lead to novel insights into the manageable associations of 110, 2007. CSCs with a STAD prognosis. [10] S. K. Singh, C. Hawkins, I. D. Clarke et al., “Identification of human brain tumour initiating cells,” Nature, vol. 432, Data Availability no. 7015, pp. 396–401, 2004. [11] D. Brungs, M. Aghmesheh, K. L. Vine, T. M. Becker, M. G. The datasets generated and/or analyzed during this research Carolan, and M. Ranson, “Gastric cancer stem cells: evidence, period can be obtained from the corresponding author. potential markers, and clinical implications,” Journal of Gas- troenterology, vol. 51, no. 4, pp. 313–326, 2016. Conflicts of Interest [12] I. Ben-Porath, M. W. Thomson, V. J. Carey et al., “An embry- onic stem cell-like gene expression signature in poorly differ- The authors declare that the study was conducted in the entiated aggressive human tumors,” Nature Genetics, vol. 40, absence of any commercial or financial relationships that no. 5, pp. 499–507, 2008. may be interpreted as potential conflicts of interest. [13] K. Tomczak, P. Czerwińska, and M. Wiznerowicz, “The Cancer Genome Atlas (TCGA): an immeasurable source of Authors’ Contributions knowledge,” Współczesna Onkologia,vol.19, no.1A, pp. A68– A77, 2015. JC and SG designed the study and drafted the manuscript. [14] Y. Sun, M. Peng, Y. Zhou, Y. Huang, and S. Mao, “Application SG and LM jointly collect, analyze data, and revise the man- of machine learning in wireless networks: key techniques and uscript. All authors have read and approved the final manu- open issues,” IEEE Communications Surveys Tutorials, script. SiHong Guo and Li Ma contributed equally to this vol. 21, no. 4, pp. 3072–3108, 2019. work. [15] A. Sokolov, E. O. Paull, and J. M. Stuart, “One-class detection of cell STATES in tumor subtypes,” Pacific Symposium on Bio- computing, vol. 21, pp. 405–416, 2016. Acknowledgments [16] T. M. Malta, A. Sokolov, A. J. Gentles et al., “Machine This work was supported by the Harbin Medical University learning identifies stemness features associated with onco- Cancer Hospital Key Project of Haiyan Fund. (No. genic dedifferentiation,” Cell, vol. 173, no. 2, pp. 338– JJZD2019-06). 354.e15, 2018. [17] P. Langfelder and S. Horvath, “WGCNA: an R package for weighted correlation network analysis,” BMC Bioinformatics, References vol. 9, no. 1, p. 559, 2008. [1] R. Siegel, J. Ma, Z. Zou, and A. Jemal, “Cancer statistics, 2014,” [18] D. Perumal, V. V. Leshchenko, P. Y. Kuo et al., “Weighted CA: a Cancer Journal for Clinicians, vol. 64, no. 1, pp. 9–29, gene co-expression network analysis (WGCNA) identifies 2014. highly proliferative myeloma subgroup responsive to CDK4/ ARK5 inhibition,” Blood, vol. 124, no. 21, pp. 3445–3445, [2] E. Yakirevich and M. B. Resnick, “Pathology of gastric cancer and its precursor lesions,” Gastroenterology Clinics of North America, vol. 42, no. 2, pp. 261–284, 2013. [19] S. Pan, Y. Zhan, X. Chen, B. Wu, and B. Liu, “Identification of biomarkers for controlling cancer stem cell characteristics [3] Y. Horikoshi, Y. Yan, M. Terashvili et al., “Fatty acid-treated in bladder cancer by network analysis of transcriptome data induced pluripotent stem cell-derived human cardiomyocytes stemness indices,” Frontiers in Oncology, vol. 9, p. 613, exhibit adult cardiomyocyte-like energy metabolism pheno- types,” Cell, vol. 8, no. 9, article 1095, 2019. [4] M. Paolillo, R. Colombo, M. Serra et al., “Stem-like cancer cells [20] M. E. Ritchie, B. Phipson, D. Wu et al., “limma powers differ- ential expression analyses for RNA-sequencing and microar- in a dynamic 3D culture system: a model to study metastatic cell adhesion and anti-cancer drugs,” Cell, vol. 8, no. 11, article ray studies,” Nucleic Acids Research, vol. 43, no. 7, article e47, 1434, 2019. [5] L. MacDonagh, S. G. Gray, E. Breen et al., “Lung cancer stem [21] A. M. Szász, A. Lánczky, Á. Nagy et al., “Cross-validation of cells: the root of resistance,” Cancer Letters, vol. 372, no. 2, survival associated biomarkers in gastric cancer using tran- pp. 147–156, 2016. scriptomic data of 1,065 patients,” Oncotarget, vol. 7, no. 31, pp. 49322–49333, 2016. [6] E. Batlle and H. Clevers, “Cancer stem cells revisited,” Nature Medicine, vol. 23, no. 10, pp. 1124–1134, 2017. [22] D. Szklarczyk, A. L. Gable, D. Lyon et al., “STRING v11: [7] F. M. Uckun, H. Sather, G. Reaman et al., “Leukemic cell protein-protein association networks with increased coverage, supporting functional discovery in genome-wide experimental growth in SCID mice as a predictor of relapse in high-risk B- lineage acute lymphoblastic leukemia,” Blood, vol. 85, no. 4, datasets,” Nucleic Acids Research, vol. 47, no. D1, pp. D607– D613, 2019. pp. 873–878, 1995. [8] M. Al-Hajj, M. S. Wicha, A. Benito-Hernandez, S. J. Morrison, [23] J. Huelsken and D. Hanahan, “A subset of cancer-associated and M. F. Clarke, “Prospective identification of tumorigenic fibroblasts determines therapy resistance,” Cell, vol. 172, breast cancer cells,” Proceedings of the National Academy of no. 4, pp. 643-644, 2018. 14 Journal of Oncology [24] Y. C. Chae and J. H. Kim, “Cancer stem cell metabolism: target [40] J. Sun, L. Smith, A. Armento, and W. M. Deng, “Regulation of for cancer therapy,” BMB Reports, vol. 51, no. 7, pp. 319–326, the endocycle/gene amplification switch by Notch and ecdy- sone signaling,” The Journal of Cell Biology, vol. 182, no. 5, pp. 885–896, 2008. [25] F. Ricci, L. Brunelli, R. Affatato et al., “Overcoming platinum- acquired resistance in ovarian cancer patient-derived xeno- grafts,” Medical Oncology, vol. 11, article 1758835919839543, 2019Published 2019 Jun 24. [26] N. Salomonis, P. J. Dexheimer, L. Omberg et al., “Integrated genomic analysis of diverse induced pluripotent stem cells from the progenitor cell biology consortium,” Stem Cell Reports, vol. 7, no. 1, pp. 110–125, 2016. [27] T. Shibue and R. A. Weinberg, “EMT, CSCs, and drug resis- tance: the mechanistic link and clinical implications,” Nature Reviews. Clinical Oncology, vol. 14, no. 10, pp. 611–629, 2017. [28] D. Friedmann-Morvinski and I. M. Verma, “Dedifferentiation and reprogramming: origins of cancer stem cells,” EMBO Reports, vol. 15, no. 3, pp. 244–253, 2014. [29] J. Y. Han, Y. K. Han, G. Y. Park, S. D. Kim, and C. G. Lee, “Bub1 is required for maintaining cancer stem cells in breast cancer cell lines,” Scientific Reports, vol. 5, article 15993, 2015Published 2015 Nov 2. [30] M. Venere, T. E. Miller, and J. N. Rich, “Mitotic control of can- cer stem cells,” Cancer Discovery, vol. 3, no. 2, pp. 141–144, [31] Q. Li, J. Qiu, H. Yang et al., “Kinesin family member 15 pro- motes cancer stem cell phenotype and malignancy via reactive oxygen species imbalance in hepatocellular carcinoma,” Can- cer Letters, vol. 482, pp. 112–125, 2020. [32] D. Huang, Q. Gao, L. Guo et al., “Isolation and identification of cancer stem-like cells in esophageal carcinoma cell lines,” Stem Cells and Development, vol. 18, no. 3, pp. 465–474, 2009. [33] W. Zhou, Y. Yang, Z. Gu et al., “ALDH1 activity identifies tumor-initiating cells and links to chromosomal instability sig- natures in multiple myeloma,” Leukemia, vol. 28, no. 5, pp. 1155–1158, 2014. [34] N. Takebe, L. Miele, P. J. Harris et al., “Targeting Notch, Hedgehog, and Wnt pathways in cancer stem cells: clinical update,” Nature Reviews. Clinical Oncology, vol. 12, no. 8, pp. 445–464, 2015. [35] T. Yang, X. N. Li, L. Li et al., “Sox17 inhibits hepatocellular carcinoma progression by downregulation of KIF14 expres- sion,” Tumour Biology, vol. 35, no. 11, pp. 11199–11207, [36] L. Ding, S. Zhang, S. Chen, L. Zheng, and L. Xiao, “Effect and mechanism of lentivirus-mediated silencing of TPX2 gene on proliferation and apoptosis of human hepatoma cells,” Journal of Cellular Biochemistry, vol. 120, 2018. [37] Y. Wu, A. Wang, B. Zhu et al., “KIF18B promotes tumor pro- gression through activating the Wnt/β-catenin pathway in cer- vical cancer,” Oncotargets and Therapy, vol. 11, pp. 1707– 1720, 2018, Published 2018 Mar 28. [38] Z. Liao, H. Zhang, P. Fan et al., “High PLK4 expression pro- motes tumor progression and induces epithelial- mesenchymal transition by regulating the Wnt/β-catenin sig- naling pathway in colorectal cancer,” International Journal of Oncology, vol. 54, no. 2, pp. 479–490, 2019. [39] B. D. Wilson, L. J. Ricks-Santi, T. E. Mason et al., “Admixture mapping links RACGAP1 regulation to prostate cancer in African Americans,” Cancer Genomics & Proteomics, vol. 15, no. 3, pp. 185–191, 2018. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Oncology Hindawi Publishing Corporation

Identification of Prognostic Markers and Potential Therapeutic Targets in Gastric Adenocarcinoma by Machine Learning Based on mRNAsi Index

Journal of Oncology , Volume 2022 – Sep 30, 2022

Identification of Prognostic Markers and Potential Therapeutic Targets in Gastric Adenocarcinoma by Machine Learning Based on mRNAsi Index

Abstract

<i>Background</i>. Cancer stem cells (CSCs), characterized by self-renewal and therapeutic resistance, play important roles in stomach adenocarcinoma (STAD). However, the molecular mechanism of STAD stem cells is still unclear. In this study, our purpose is to explore the expression of stem cell-related genes in STAD. <i>Methods</i>. The stemness index based on mRNA expression (mRNAsi) was used to analyze STAD cases in The Cancer Genome Atlas (TCGA). Firstly, mRNAsi was used and analyzed by differential expression, survival analysis, clinical stage, and gender in STAD. Then, weighted gene coexpression network analysis (WGCNA) was used to discover the fascinating modules and key genes. Enrichment analysis was carried out to annotate the functions and pathways of key genes. The gene expression comprehensive database (GEO) in STAD was used to verify the expression levels of key genes in all cancers. Protein-protein interaction networks is used to determine the relationships between key genes. <i>Results</i>. The mRNAsi was obviously upregulated in tumor cases. With the increase of tumor stage and T stage, the mRNAsi score decreased, and the overall survival rate of high score group patients was better. According to the degree of association with mRNAsi, different modules and key genes were screened out. A total of 6,740 differential genes were found, of which 1,147 genes were downregulated and 5,593 genes were upregulated. 19 key genes (BUB1, BUB1B, KIF14, NCAPH, RACGAP, KIF15, CENPF, TPX2, RAD54L, KIF18B, KIF4A, TTK, SGO2, PLK4, ARHGAP11A, XRCC2, Clorf112, NCAPG, and ORC6) were screened due to significant upregulation in STAD. And they had been proven that enriched from the cell cycle Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway, relating to cell proliferation Gene Ontology (GO) terms, as well. Among them, 9 genes have been extensively associated to OS, and 3 genes had been associated to receive chemotherapy resistance. PPI protein network suggests that there is a sturdy correlation between these key genes. <i>Conclusion</i>. A total of 19 key genes were found to play an essential position in retaining the traits of STAD stem cells. These genes can be used to evaluate the prognosis of STAD patients or become specific therapeutic targets.

Loading next page...
 
/lp/hindawi-publishing-corporation/identification-of-prognostic-markers-and-potential-therapeutic-targets-vZSSy0RKlJ

References (41)

Publisher
Hindawi Publishing Corporation
ISSN
1687-8450
eISSN
1687-8469
DOI
10.1155/2022/8926127
Publisher site
See Article on Publisher Site

Abstract

Hindawi Journal of Oncology Volume 2022, Article ID 8926127, 14 pages https://doi.org/10.1155/2022/8926127 Research Article Identification of Prognostic Markers and Potential Therapeutic Targets in Gastric Adenocarcinoma by Machine Learning Based on mRNAsi Index 1 2 2 Si Hong Guo, Li Ma, and Jie Chen Personal Health Management, Hong Kong Baptist University, Hong Kong 999077, China Department of Gynaecologic Oncology, Harbin Medical University Cancer Hospital, Harbin 150000, China Correspondence should be addressed to Jie Chen; cj2365255@hrbmu.edu.cn Received 15 August 2022; Revised 11 September 2022; Accepted 14 September 2022; Published 30 September 2022 Academic Editor: Zhongjie Shi Copyright © 2022 Si Hong Guo et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Background. Cancer stem cells (CSCs), characterized by self-renewal and therapeutic resistance, play important roles in stomach adenocarcinoma (STAD). However, the molecular mechanism of STAD stem cells is still unclear. In this study, our purpose is to explore the expression of stem cell-related genes in STAD. Methods. The stemness index based on mRNA expression (mRNAsi) was used to analyze STAD cases in The Cancer Genome Atlas (TCGA). Firstly, mRNAsi was used and analyzed by differential expression, survival analysis, clinical stage, and gender in STAD. Then, weighted gene coexpression network analysis (WGCNA) was used to discover the fascinating modules and key genes. Enrichment analysis was carried out to annotate the functions and pathways of key genes. The gene expression comprehensive database (GEO) in STAD was used to verify the expression levels of key genes in all cancers. Protein-protein interaction networks is used to determine the relationships between key genes. Results. The mRNAsi was obviously upregulated in tumor cases. With the increase of tumor stage and T stage, the mRNAsi score decreased, and the overall survival rate of high score group patients was better. According to the degree of association with mRNAsi, different modules and key genes were screened out. A total of 6,740 differential genes were found, of which 1,147 genes were downregulated and 5,593 genes were upregulated. 19 key genes (BUB1, BUB1B, KIF14, NCAPH, RACGAP, KIF15, CENPF, TPX2, RAD54L, KIF18B, KIF4A, TTK, SGO2, PLK4, ARHGAP11A, XRCC2, Clorf112, NCAPG, and ORC6) were screened due to significant upregulation in STAD. And they had been proven that enriched from the cell cycle Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway, relating to cell proliferation Gene Ontology (GO) terms, as well. Among them, 9 genes have been extensively associated to OS, and 3 genes had been associated to receive chemotherapy resistance. PPI protein network suggests that there is a sturdy correlation between these key genes. Conclusion. A total of 19 key genes were found to play an essential position in retaining the traits of STAD stem cells. These genes can be used to evaluate the prognosis of STAD patients or become specific therapeutic targets. 1. Introduction unlimited proliferation [3–5]. CSC theory points out that tumor proliferation, therapeutic resistance, and recurrence The incidence rate and mortality of stomach cancer are additionally pushed by way of a small range of tumor stem decreased significantly in five years, but it still ranked third cells hidden in most cancers. It explains these clinical observa- among common malignant tumors and the second leading tions, such as tumor recurrence, tumor dormancy, and metas- cause of cancer-related death [1]. Ninety percent of all tasis after successful surgical resection, chemotherapy, and tumors of the stomach are malignancies, and stomach ade- radiotherapy [6]. CSCs have been found in several human nocarcinoma (STAD) accounts for 95% of all cases of malig- malignancies, such as leukemia [7], breast cancer [8], colorec- nancies [2]. tal cancer [9], and brain cancer [10]. In addition, strong pre- In current years, the characteristic of most cancers stem clinical data and clinical evidence have been added as telephone has been mentioned such as self-renewal and supports of the existence of gastric CSCs [11]. Therefore, 2 Journal of Oncology 1.00 0.75 0.50 0.8 p = 3.761e–09 0.25 p < 0.001 0.00 0.6 0 12345 6789 10 Time (years) 0.4 mRNAsi High Low 0.2 High 197125 41 20 13 741111 Low 197105 23 11 7422220 0.0 0 123456789 10 Normal Tumor Time (years) (a) (b) mRNAsi (p = 0.001) mRNAsi (p = 0.034) 0.7 0.7 0.6 0.6 0.5 0.5 0.4 0.4 0.3 0.3 0.2 0.2 0.1 0.1 Stage I Stage II Stage III Stage IV G1 G2 G3 Stage Grade (c) (d) mRNAsi (p = 0.024) Volcano 0.7 15 0.6 0.5 0.4 0.3 0.2 0.1 T1 T2 T3 T4 –10 –5 0 5 10 logFC (e) (f) Figure 1: The correlation of mRNAsi profiles with STAD. (a) Scatter plot illustrating the difference of mRNAsi index expression between normal tissues and tumors. (b) Kaplan–Meier survival curve of correlation between mRNAsi score and OS of STAD patients. Detect the correlation between mRNAsi score and the Grade (c), Stage (d), and T degree (e) by the Kruskal-Wallis test. (f) Volcano map of DEGs between STAD tissues and normal tissues. Downregulated genes are indicated in green, and upregulated genes are indicated in red. STAD: stomach adenocarcinoma; DEGs: differentially expressed genes. mRNAsi mRNAsi mRNAsi mRNAsi mRNAsi –log10 (fdr) Survival probability Journal of Oncology 3 Sample clustering to detect outliers 8e + 04 6e + 04 4e + 04 2e + 04 0e + 00 (a) Scale independence Mean connectivity 6 19 7 161718 20 4 8 10 13 9 1112 0.8 0.6 0.4 0.2 56 7 1 0 89 1011121314151617181920 510 15 20 510 15 20 Soft threshold (Power) Soft threshold (Power) (b) Module-trait relationships MEmagenta MEpurple MEtan MEblue 0.5 MEpink MEcyan MEgreenyellow MEmidnightblue MElightcyan MEblack MEbrown Cluster dendrogram MEred –0.5 1.0 MEgrey60 0.9 MEsalmon 0.8 MEturquoise 0.7 MEgrey –1 0.6 0.5 0.4 0.3 Dynamic tree cut Merged dynamic (c) (d) Figure 2: Continued. Scale free topology model fit, signed R Height Height Mean connectivity mRNAsi EREG–mRNAsi 4 Journal of Oncology Module membership vs. gene significance Module membership vs. gene significance cor = 0.88, p<1e–200 cor = 0.74, p = 4, 1e–150 0.5 0.4 0.4 0.3 0.3 0.2 0.2 0.1 0.1 0.0 0.0 0.0 0.2 0.4 0.6 0.8 1.0 0.2 0.4 0.6 0.8 Module membership in blue module Module membership in brown module (e) (f) Module membership vs. gene significance cor = 0.081, p = 0.23 0.25 0.20 0.15 0.10 0.05 0.00 0.0 0.2 0.4 0.6 0.8 Module membership in pink module (g) Figure 2: Construction of weighted gene coexpression network for STAD stemness related datasets. (a) Identify and remove outlier samples through average linkage hierarchical clustering. Samples exceeding the red line were considered deviations in gene expression. (b) Network topology analysis of different soft threshold powers. The left figure shows the influence of soft threshold power on the scale-free topological fitting index. The right figure shows the influence of soft threshold power on average connectivity. (c) Clustering dendrograms was done via mean linkage hierarchical. (d) Module-trait relationships. Each column represents a clinical phenotype, and each row denotes an ME. The correlation coefficient and P value are contained in each cell. (e–g) Scatterplots of GS for weight vs. MM to pick out the key genes from the blue, brown, and pink modules. STAD: stomach adenocarcinoma; ME: module eigengene; GS: gene significance; MM: module membership. CSC research is able to provide a new paradigm for managing Tathiane et al. used publicly available molecular profiles patients with STAD. from TCGA to obtain two independent stemness indices A growing number of studies have shown cancer stem- by using original one-class logistic regression machine- ness is associated with being transcriptomic, genomic, epi- learning algorithm (OCLR) to complete the integration genomic, and proteomic [12]. Within the last decade, The of transcriptome, methylome, and transcription factor Cancer Genome Atlas (TCGA) has elucidated the primary [15]. One was mDNAsi which reflects epigenetic features; tumor landscapes by generating comprehensive multiomics the other was mRNAsi which reflects gene expression. characteristics, along with pathophysiological feature and Malta et al. identified the relationship between the two clinical information annotations [13]. Machine learning stem cell indices and new carcinogenesis pathways, has been increasing applied in various areas of society somatic cell changes, microRNAs (miRNAs), and tran- and has become a useful strategy in biotechnology [14]. scription regulatory networks. These characteristics are Gene significance for EREG-mRNAsi Gene significance for EREG-mRNAsi Gene significance for EREG-mRNAsi BP CC MF Journal of Oncology 5 p. adjust Sister chromatid segregation Mitotic sister chromatid segregation Nuclear chromosome segregation Mitotic nuclear division Nuclear division Organelle fission Chromosome segregation 0.005 Meiotic cell cycle Meiotic chromosome segregation Chromosome separation Spindle Chromosome, centromeric region Condensed chromosome Condensed chromosome, centromeric region 0.010 Kinetochore Condensed chromosome outer kinetochore Kinesin complex Chromosomal region Microtubule Condensed chromosome kinetochore 0.015 Microtubule motor activity Microtubule binding Motor activity Tubulin binding ATPase activity ATP-dependent microtubule motor activity, plus-end-directed ATP-dependent microtubule motor activity 0.020 Protein serine/threonine kinase activity Nucleoside-triphosphatase regulator activity ATPase activity, coupled 0.0 2.5 5.0 7.5 10.0 12.5 (a) p. adjust 0.0004 Cell cycle 0.0008 0.0012 Homologous recombination 0.0016 0 1 2 3 4 (b) Figure 3: Functional analysis of brown module. (a) GO enrichment analysis. (b) KEGG enrichment analysis. GO: Gene Ontology; KEGG: Kyoto Encyclopedia of Genes and Genomes; BP: biological process; CC: cellular component; MF: molecular function. 6 Journal of Oncology related to cancer stem cells in specific molecular subtypes 2.3. 2.3 Differential Expressed Gene (DEG) Analysis. We used of TCGA tumors, which may be the factors controlling the R package “limma” for differential expression analysis cancer stem cells. Importantly, higher stem cell index [20]. We used the cut-off values, which were fold change > value is related to the active biological processes and 1 and adj:P <0:05, to screen for DEGs between normal greater tumor dedifferentiation in tumor stem cells, as health and stomach adenocarcinoma samples. The volcano reflected in histopathological grade. Metastatic tumor cells plot and the box-plots showing differences in key genes pre- show more dedifferentiation in phenotype, which may sented in this study were drawn by the R package “pheat- contribute to their invasiveness. The stemness indices map” and “ggpubr,” respectively. had positive correlation with tumor dedifferentiation and 2.4. WGCNA. WGCNA was performed using the WGCNA biological active in CSCs [16]. The mRNAsi and mDNAsi R package [17], which were “matrixStats,”“Hmisc,”“fore- scores in TCGA samples had been calculated by applying ach,”“doParallel,”“fastcluster,” dynamicTreeCut,”“sur- the stemness indices. vival,” and “WGCNA.” Before the building of coexpression Weighted gene coexpression network analysis network, the rectangular Euclidean relative distance of every (WGCNA), a method commonly used to explore biological take a look at pattern was once calculated by means of prac- networks, paired relationships between genes and pheno- tical adjacency method, and the integration connectivity of types. WGCNA transforms gene expression data into coex- the total pattern community calculated via distance was once pression module, providing insights into signaling standardized via practical scaling method. Due to some networks, and mine the pathway-related modules [17]. It is exceptional genes with no tremendous trade in expression widely applied in many physiological and pathological pro- between samples which are surprisingly correlated in cesses, including cancer, genetic therapy, and clinical data WGCNA as a whole, it appears that the genes with the analysis, which can be useful for identifying biomarkers of most biased expression have been used in the subsequent disease or target points for therapy [18, 19]. WGCNA analysis. The gene with the highest DEG vari- In this study, our purpose is to identify key genes associ- ance of 25% was selected. Clear ordinary value pattern ated with STAD stemness in TCGA based on mRNAsi information with connectivity is much less than -2.5. scores. The purpose of this study was to provide an interest- Function pickSoftThreshold was used to calculate scale- ing bioinformatics method for identifying stem cell-related free topology becoming indicesR corresponding to one- genes and revealing the role of some CSC-related genes in STAD. of-a-kind smooth thresholding powersβ. Theβvalue was used as lengthy asR reaching 0.8. After that, the gene expression matrix was converted into an adjacency matrix 2. Materials and Methods and a Topological Overlap Matrix (TOM), and then the corresponding dissimilarity of TOM (dissTOM) was calcu- 2.1. Software and R Packages. We used R Studio version lated. For module detection, hierarchical clustering was 1.2.5042 (URL: https://rstudio.com/) with R version 3.6.2 used to produce a hierarchical clustering tree (dendro- (URL: https://www.r-project.org/) in this study. The pro- gram) of genes by using characteristic “hclust” based gramming software Perl version 64-bit (URL: https://www totally on dissTOM. The Dynamic Tree Cut approach .perl.org/) was used for data processing. All R packages were was carried out for department reduction to generate downloaded from Bioconductor (URL: https://www modules. During this, a quite massive minimal module .bioconductor.org/). measurement of minClusterSize = 30 to department split- ting had been chosen to avoid producing too many small 2.2. Database and mRNAsi Index. The RNA-sequencing or massive modules. To consider the magnitude of every (RNA-seq) of STAD and all pathological and clinical infor- module, gene significance (GS) was once calculated to mation were downloaded from TCGA database (URL: measure the correlation coefficient between genes and pat- https://portal.gdc.cancer.gov/). These data were updated on tern traits. The module eigengene (ME) is described as the 5 October 2019. The results of RNA-seq were including first foremost thing of a given element and can be 375 cancer samples and 32 normal samples, structured for regarded as a consultant of the gene expression profile of a matrix file. We used Ensemble data to exchange the gene the module integration. It was calculated by using pur- names expressed by Ensembl IDs which are specifically con- poseful module genes. If their MES correlation coefficient verted into a gene symbol matrix. Moreover, to explore the is higher than 0.75, the modules will be merged, with mode of action of CSC-related genes in chemotherapy capacity that they have considerable comparable gene resistance, we download the microarray (GSE14210) expression levels. Here, we can pick out mRNAsi and epi- results from the Gene Expression from the Gene Expres- genetically regulated mRNAsi as scientific phenotypes. sion Omnibus (GEO) (URL: https://www.ncbi.nlm.nih After selecting the components of interest, let us cal- .gov/geo/). We referred to the mRNAsi index data for all culate the GS and module membership of each key gene types of tissues in the supporting documents to Malta (MM, the significance between the module’s own gene et al.’s article and specifically screened the mRNAsi index and gene expression profile), and set their threshold of patients with stomach adenocarcinoma for incorpora- values. The thresholds for screening key genes in the tion into TCGA data for stomach adenocarcinoma, with module were defined as cor.gene MM > 0:8 and cor.gene the unmatched cases deleted. GS > 0:5. Journal of Oncology 7 ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎ Type (a) BUB1 (209642_at) KIF14 (236641_at) 1.0 1.0 HR = 0.83 (0.7–0.98) HR = 0.69 (0.55–0.86) Logrank P = 0.029 Logrank P = 0.00076 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0.0 0.0 0 50 100 150 0 50 100 150 Time (months) Time (months) Number at risk Number at risk Low 439 129 34 1 Low 316 111 30 1 High 436 169 14 0 High 315 154 18 0 Expression Expression Low Low High High (b) (c) NCAPH (212949_at) 1.0 HR = 0.76 (0.64–0.9) Logrank P = 0.0015 RAD54L (204558_at) 0.8 1.0 HR = 1.21 (1.02–1.44) Logrank P = 0.026 0.6 0.8 0.4 0.6 0.2 0.4 0.0 0.2 0 50 100 150 0.0 Time (months) Number at risk 0 50 100 150 Low 441 124 29 1 Time (months) High 434 174 19 0 Number at risk Low 438 160 33 0 Expression Low High 437 138 15 1 High Expression Low High (d) (e) Figure 4: Continued. Probability Probability Gene expression ARHGAP11A BUB1 BUB1B Clorf112 CENPF KIF14 KIF15 KIF18B KIF4A Probability Probability NCAPG NCAPH ORC6 PLK4 RACGAP1 RAD54L SGO2 TPX2 TTK XRCC2 8 Journal of Oncology PLK4 (204886_at) ARHGAP11A (204492_at) 1.0 1.0 HR = 0.81 (0.68–0.96) HR = 0.8 (0.67–0.95) Logrank P = 0.013 Logrank P = 0.0097 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0.0 0.0 0 50 100 150 0 50 100 150 Time (months) Time (months) Number at risk Number at risk Low 438 123 30 1 Low 439 129 31 1 High 437 175 18 0 High 436 169 17 0 Expression Expression Low Low High High (f) (g) XRCC2 (207598_x_at) CAP-G (218662_s_at) 1.0 1.0 HR = 1.87 (1.57–2.22) HR = 0.78 (0.66–0.93) Logrank P = 6.3e–13 Logrank P = 0.005 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0.0 0.0 0 50 100 150 0 50 100 150 Time (months) Time (months) Number at risk Number at risk Low 439 191 27 0 Low 439 123 35 1 High 436 107 21 1 High 436 175 13 0 Expression Expression Low Low High High (h) (i) ORC6L (219105_x_at) 1.0 HR = 1.19 (1.01–1.41) Logrank P = 0.042 0.8 0.6 0.4 GSE14210 0.2 0.0 0 50 100 150 Time (months) 308 3 16 Number at risk Low 439 162 33 1 High 436 136 15 0 Expression Low High Key genes (j) (k) Figure 4: Verification of the influence of key genes on diseases. (a) The mRNA expression level of key genes between tumor and normal tissues in TCGA STAD dataset. The Kaplan–Meier plotter database was used to assess the correlation between the expression of BUB1 (b), KIF14 (c), NCAPH (d), RAD54L (e), PLK4 (f), ARHGAP11A (g), XRCC2 (h), NCAPG (i, also called CAP-G), ORC6 (j, also called ORC6L), and the OS of STAD patients. Kaplan–Meier survival plots (K–M plots) were generated using the on-line tool, Kaplan–Meier plotter. (k) Venn diagram of the relationship between 19 key genes and acquired chemoresistance by GSE14210. 2.5. Overall Survival Curve. Finally, to determine the prog- overall survival deviation between patients with low and nostic significance and value of mRNAsi scores, we can draw high mRNAsi index. In this part, R package “survival” and the Kaplan–Meier diagram of mRNAsi index to explore the “surviminer” were used, and a log-rank test is used to test Probability Probability Probability Probability Probability Journal of Oncology 9 the relationship between them. In key gene validation analy- which 1147 were downregulated and 5593 were upregulated sis, Kaplan–Meier survival curves of key genes were drawn (Figure 1(f)). with the online tool Kaplan–Meier plotter [21] 3.2. WGCNA: Identifying the Most Significant Modules and (URL:http://www.kmplot.com/analysis/index.php?p= service). Genes. With WGCNA, we built a gene coexpression network to become aware of biologically significant gene modules. It 2.6. Functional Annotation Gene Ontology (GO) and Kyoto can help us to understand the genes associated with STAD Encyclopedia of Genes and Genomes (KEGG) Analyses. The stemness. We put 6740 DEGs with the highest variance of 25% into the same module through cluster analysis. Before GO functional annotations and KEGG pathway enrichment analysis shown in this study were obtained from the data that, the outlier samples should be removed (Figure 2(a)). analysis conducted by the R package “cluster Profiler” to According to the lowest value of scale-free network, the β investigate the biological functions of key genes. The thresh- value is determined. What the pickSoftThreshold function old values were as follows: P <0:01 and FDR < 0:05. does is to find the appropriate power. The selection of the power value is determined by β value. Calculate the correla- 2.7. Gene Coexpression Analysis and Construction of Protein- tion intensity (weighted correlation value) of expression Protein Interaction (PPI) Network. In order to further study levels among all genes to obtain the adjacency matrix. As a the stability of these special relationships at the transcrip- result, we choose β =4 (scale-free R =0:9) as the soft tional level, we calculated the coexpression relationships threshold (Figure 2(b)). We find 16 modules for subsequent among key genes within the module depending on the gene evaluation (Figure 2(c)). expression level. The R “corrlot” package is mainly used to Taking MS as the total gene expression level of the cor- calculate the Pearson correlation degree between genes. responding module, the correlation between MS and clinical The STAD data set was selected from TCGA for analysis phenotype was calculated. This is useful for us to discover and research, and the routine data were analyzed by the the relationship between these modules and the dryness Pearson correlation test. Results with a correlation index of the sample. By calculating the Pearson correlation coefficient > 0:3 and P value < 0.01 were considered statisti- coefficient, a threshold value can be obtained. If the correla- cally significant. tion coefficient is greater than 0.8 or so, it can be used as the Accurately retrieve PPI network from STRING version basis of strong correlation between the two genes. The con- 11.0 (URL: https://string-db.org/) [22]. And display the bar sequences confirmed that the blue and brown modules were graph of the nodes in the network with top-level network extensively correlated with mRNasi, and the correlation was connectivity. The minimum required interaction score was once close to 0.8. However, the correlation coefficient of the set to a medium confidence of 0.4, and now, the hidden brown module is 0.77, which is higher. In addition, the pink branch nodes in the network are disconnected. The number module was fantastically negatively correlated with mRNasi of adjacent nodes of each gene in the PPI network was calcu- (Figure 2(d)). Therefore, the brown module was chosen lated, and then, the genes were sorted by the bar graph com- through us as the most fascinating module for subsequent bined with the number of adjacent nodes. analysis. The threshold for screening key genes in the mRNAsi group was described as cor:MM > 0:8 and cor:GS > 0:5.We 3. Results pick 19 key genes (BUB1, BUB1B, KIF14, NCAPH, RAC- 3.1. Clinical Characteristics of mRNAsi and DEGs in STAD. GAP1, KIF15, CENPF, TPX2, RAD54L, KIF18B, KIF4A, The mRNAsi is an index of CSCs that can quantitatively TTK, SGO2, PLK4, ARHGAP11A, XRCC2, Clorf112, describe the similarity between tumor cells and stem cells. NCAPG, and ORC6), as shown in Figures 2(e)–2(g). And We observe large distinction in mRNAsi between tumor we exhibit the distinct expressions of key genes between and ordinary tissues (Figure 1(a)). In the survival analysis, most cancers and ordinary samples in TCGA; all these genes we divided gastric cancer patient into higher mRNAsi score in brown module are upregulated in tumor cases (Figure 2(f group and lower mRNAsi group by using mRNAsi median )). value. Obviously, patients with higher mRNAsi scores have greater overall survival in contrast with sufferers with lower 3.3. Enrichment Analysis of Brown Module. We use GO and mRNAsi scores (Figure 1(b)), the five-year survival rate of KEGG analysis to elucidate the function similarities of mod- higher scores group is 47.9% with CI (0.344, 0.668), and ule brown genes. The results show that nuclear division, the lower scores group is 21.2% with CI (0.107, 0.421). Sur- spindle, and microtubule binding are the most great enrich- prisingly, the mRNAsi scores tend to decline with the grade ments in cellular component (CC), biological process (BP), increasing with the exception of G1 (only 8 samples). Also, and molecular function (MF) groups (Figure 3(a)). KEGG the mRNAsi score shows an overall decreasing trend in stage pathway enrichment analysis suggested cell cycle and and T (Figures 1(c), 1(d), and 1(e)). The Kruskal Wallis test homologous recombination pathways are significant path- was once used to determine the value of variations between ways (Figure 3(b)). All of them are related to cancer stem groups. cells. We download mRNA-seq data and did difference analy- sis to compare STAD and normal since the mRNAsi differ- 3.4. Data Validation. Firstly, the STAD dataset of TCGA ence between tumor and normal. We find 6740 DEGs in showed significant differences in the expression of all key 10 Journal of Oncology (a) XRCC2 1 ORC6 C1orf112 8 SGOL2 RAD54L 14 PLK4 KIF18b 14 KIF14 ARHGAP11A 14 KIF4A TTK 16 RACGAP1 NCAPH 16 KIF15 CENPF 16 BUB1B TPX2 17 NCAPG BUB1 17 0 5 10 15 20 (b) Figure 5: Continued. Journal of Oncology 11 ORC6 C1orf112 0.56 0.8 TPX2 0.59 0.73 XRCC2 0.66 0.62 0.68 0.6 TTK 0.61 0.72 0.79 0.71 KIF14 0.59 0.76 0.73 0.67 0.73 0.4 SGO2 0.58 0.69 0.73 0.67 0.79 0.79 CENPF 0.53 0.73 0.73 0.63 0.7 0.88 0.76 0.2 BUB1 0.63 0.74 0.78 0.7 0.77 0.78 0.82 0.75 RACGAP1 0.61 0.66 0.73 0.63 0.72 0.74 0.73 0.72 0.77 0 KIF15 0.56 0.69 0.67 0.68 0.71 0.75 0.74 0.78 0.73 0.72 –0.2 BUB1B 0.65 0.69 0.75 0.7 0.75 0.74 0.78 0.74 0.84 0.74 0.76 KIF18B 0.53 0.62 0.73 0.7 0.67 0.7 0.69 0.73 0.72 0.74 0.73 0.73 –0.4 KIF4A 0.52 0.65 0.74 0.62 0.7 0.72 0.71 0.74 0.78 0.75 0.71 0.75 0.74 NCAPH 0.57 0.69 0.75 0.66 0.71 0.69 0.72 0.69 0.87 0.77 0.72 0.77 0.76 0.79 –0.6 ARHGAP11A 0.58 0.6 0.64 0.61 0.67 0.76 0.72 0.75 0.79 0.7 0.75 0.82 0.68 0.72 0.73 PLK4 0.6 0.62 0.64 0.68 0.71 0.7 0.71 0.7 0.79 0.7 0.75 0.81 0.72 0.72 0.75 0.77 –0.8 NCAPG 0.57 0.6 0.63 0.61 0.7 0.7 0.72 0.7 0.76 0.72 0.75 0.78 0.67 0.73 0.75 0.75 0.8 RAD54L 0.54 0.64 0.68 0.66 0.66 0.64 0.65 0.64 0.76 0.72 0.74 0.75 0.76 0.75 0.81 0.67 0.75 0.72 –1 (c) Figure 5: PPI interactive network. (a) String diagram composed of 19 key genes as nodes. (b) The bar-plot lists the connections of key genes in the brown module by the counts of connections. (c) Correlation analysis between key genes. The higher phase of the graph indicates the degree of correlation. The darker the color, the greater the correlation. The lower part shows the corresponding correlation value. PPI: protein-protein interaction. genes between normal and tumor cases (Figure 4(a)). In all 3.5. Protein-Protein Interactions (PPI) among Genes of patients with STAD, the Kaplan–Meier curve and log rank Brown Module. The application of the on-line device test analysis showed that 7 genes in the brown module were STRING (URL: http://string-db.org/) to protein-protein significantly associated with OS (P <0:05, FDR < 0:05) interaction networks for each module will assist us to (Figures 4(b)–4(j)). explore the interplay between key genes extra deeply. There It is well known that CSCs have chemoresistance, and were 19 nodes and 129 edges in the shaped PPI network, and resistance is related to cancer-associated fibroblasts in the the PPI enrichment (P value < 0.01) (Figure 5(a)). In addi- extracellular matrix [23]. The mapping of GSE14210 is tion, the significant nodes shown in the bar-plot can identify based on Venn diagram. 19 key gene maps selected from the genes most closely related to other members of the mod- the brown module were scored by GS and MM scores. ule (Figure 5(b)). Finally, SGO2, TTK, and CENPF were associated with The correlation between the key genes of this module the acquired chemoresistance to cisplatin and fluorouracil was strong (Figure 5(c)), and the correlation was statistically combination chemotherapy in gastric cancer (Figure 4(k)). significant (P <0:01). The correlation between CENPF and ORC6 C1orf112 TPX2 XRCC2 TTK KIF14 SGO2 CENPF BUB1 RACGAP1 KIF15 BUB1B KIF18B KIF4A NCAPH ARHGAP11A PLK4 NCAPG RAD54L 12 Journal of Oncology ORC6 (0.53), KIF18B, and ORC6 (0.53) was the lowest, tion characteristics. The pathway enrichment advised that whereas the correlation between CENPF and KIF14 (0.88) the four key genes in the cycle pathway time period have was the highest. been most possibly a useful gene set that impacts tumor stemness via regulating the cell cycle. The gene units that keep the traits of stem cells in a range 4. Discussion of cancers may additionally have similarities. Since the for- The morbidity and mortality of gastric most cancers stay mation of a range of organ tissues takes place from pluripo- excessive all over the worldwide. In current years, CSCs have tent stem cells, their CSCs are dedifferentiated with stem been mentioned to make vital contributions to tumor pro- mobile phone characteristics. This reverse improvement gression, recurrence, and therapeutic resistance [24, 25]. has made a range of CSCs possessing some traits of pluripo- Therefore, therapy concentrating on STAD stem cells is tent stem cells. Moreover, their stage of overexpression was essential. In addition, choosing out the emergence of these once associated to the stage of stemness, and their persisted druggable genetic ameliorations in pancancer cases, and expand might also promote modifications in tumor develop- whether or not there are modifications in the expression of ment and posttherapy progression. More than half of the key the equal mRNAsi-related genes, is additionally a query genes have been mentioned in STAD, and some have been proven to be related with the traits of CSCs. BUB1 is related priceless of dialogue in the future work. In this study, we tried to discover key genes associated to with the most cancer stem cell attainable in breast cancer STAD stem cells in TCGA database. We used WGCNA [29]. An issue highlights a study that links the presentation based totally on mRNAsi scores, as calculated by Salomonis of kinetochores within mitosis to an essential requirement et al. [26]. The tumor case had a greater mRNAsi rating than for BUB1 threonine kinase B (BUB 1B), broadening our understanding of the cell-cycle machinery in CSCs [30]. Kine- the regular case. The mRNAsi scores reduced with the sick- ness grade, stage, and T stage, though the mRNAsi rating of sin family member 15 (KIF15) promotes the CSC phenotype G1 was once small which may also be associated to inade- and malignancy by means of PHGDH-mediated ROS imbal- quate pattern size. The excessive mRNAsi team confirmed ance in hepatocellular carcinoma [31]. TTK gene was overex- a decrease survival chance than the low team in the first 5 pressed in the CSC-like cell populace remoted from human years, which used to be constant with the negative conse- esophageal carcinoma phone strains as properly as in the quence related with stemness features. human more than one myeloma stem cells sorted through We developed coexpression modules through WGCNA aldehyde dehydrogenase 1 (ALDH1) [32, 33]. and pick out brown module as the best correlations with Survival curves have been generated to validate the prog- mRNAsi. Key genes have been screened from the blue module nostic fee of the key genes in brown module in STAD. In the primarily based on the GS and MM scores. The expression K-M plots, 7 genes had been substantially related with prog- degrees of key genes are appreciably upregulated in tumor noses (P <0:01, FDR < 0:05). High expression of BUB1, samples. There have been robust coexpression relationships TPX2, and X-ray repair cross complementing 2 (XRCC2) at the transcriptional degree in brown module. There was had been noticeably related with negative prognoses. The additionally a robust PPI community among proteins of this expression of NCAPH, NCAPG, RACGAP1, and SGO2 module. The key genes intently associated to pluripotent stem has been positively correlated with affected person progno- cells have been confirmed to be overexpressed in most tumors. sis. As known, CSCs can withstand clinical remedy and Moreover, all organ tissues are developed from pluripotent make contributions to tumor relapse. The key genes had stem cells, suggesting that key genes might also play a position been validated in GSE14210, and SGO2, TTK, and CENPF in keeping stem cellphone residences in a range of cancers. have been related with the obtained chemoresistance to cis- The consequences led us to reassess the relationship between platin and fluorouracil mixture chemotherapy in gastric can- CSC traits and STAD progression. cer. Several studies had proven that CSCs have one or Undifferentiated major tumors are more probably to rea- greater abnormalities in signaling pathways that modify son most cancer cells to unfold to far-off organs, mainly to self-renewal. The Wnt/β-catenin, Notch, and Hedgehog sickness development and negative prognosis. Moreover, pathways have been mentioned fully [34]. Wnt/β-catenin CSCs are typically resistant to handy remedies [27]. The KIF14, TPX2, KIF18B, and PLK4 in the Wnt/β-catenin acquisition of progenitor cell-like and stem cell-like traits pathway [35–38], TTK and XRCC2 in the Hedgehog path- and loss of the differentiated phenotype are manifestations way [37, 38], and RACGAP1 and TTK in the Notch pathway of most cancer development [28], regular with the expand [39, 40] may additionally be necessary for the tumorigenicity in STAD stemness as the tumor progression. In our study, of CSCs. These genes are vital therapeutic aimed at inhibit- ing the self-renewal, proliferation, and tumor development we observed that sufferers with greater corrected mRNAsi rankings had decreased ordinary survival rates, which used of CSCs. to be regular with the negative prognosis related with CSC characteristics. Disease stage 1 and T1 stage STAD had 5. Conclusions pretty greater CSC characteristics, indicating the stem tele- phone residences start to upward thrust from initiation of In summary, 19 key genes have been determined to play nec- the cancer. essary roles in STAD stem phone maintenance. The valida- Functional annotations of the brown module had been tions confirmed that these genes ought to be beneficial for chiefly associated to the stem cell self-renewal and prolifera- outlining the prognosis of STAD patients. These genes may Journal of Oncology 13 also be therapeutic pursuits for inhibiting STAD stemness Sciences of the United States of America, vol. 100, no. 7, pp. 3983–3988, 2003. characteristics. However, our conclusions are primarily based on the retrospective information, and similarly [9] C. A. O'Brien, A. Pollett, S. Gallinger, and J. E. Dick, “A human colon cancer cell capable of initiating tumour growth in organic and scientific investigation of these genes should immunodeficient mice,” Nature, vol. 445, no. 7123, pp. 106– lead to novel insights into the manageable associations of 110, 2007. CSCs with a STAD prognosis. [10] S. K. Singh, C. Hawkins, I. D. Clarke et al., “Identification of human brain tumour initiating cells,” Nature, vol. 432, Data Availability no. 7015, pp. 396–401, 2004. [11] D. Brungs, M. Aghmesheh, K. L. Vine, T. M. Becker, M. G. The datasets generated and/or analyzed during this research Carolan, and M. Ranson, “Gastric cancer stem cells: evidence, period can be obtained from the corresponding author. potential markers, and clinical implications,” Journal of Gas- troenterology, vol. 51, no. 4, pp. 313–326, 2016. Conflicts of Interest [12] I. Ben-Porath, M. W. Thomson, V. J. Carey et al., “An embry- onic stem cell-like gene expression signature in poorly differ- The authors declare that the study was conducted in the entiated aggressive human tumors,” Nature Genetics, vol. 40, absence of any commercial or financial relationships that no. 5, pp. 499–507, 2008. may be interpreted as potential conflicts of interest. [13] K. Tomczak, P. Czerwińska, and M. Wiznerowicz, “The Cancer Genome Atlas (TCGA): an immeasurable source of Authors’ Contributions knowledge,” Współczesna Onkologia,vol.19, no.1A, pp. A68– A77, 2015. JC and SG designed the study and drafted the manuscript. [14] Y. Sun, M. Peng, Y. Zhou, Y. Huang, and S. Mao, “Application SG and LM jointly collect, analyze data, and revise the man- of machine learning in wireless networks: key techniques and uscript. All authors have read and approved the final manu- open issues,” IEEE Communications Surveys Tutorials, script. SiHong Guo and Li Ma contributed equally to this vol. 21, no. 4, pp. 3072–3108, 2019. work. [15] A. Sokolov, E. O. Paull, and J. M. Stuart, “One-class detection of cell STATES in tumor subtypes,” Pacific Symposium on Bio- computing, vol. 21, pp. 405–416, 2016. Acknowledgments [16] T. M. Malta, A. Sokolov, A. J. Gentles et al., “Machine This work was supported by the Harbin Medical University learning identifies stemness features associated with onco- Cancer Hospital Key Project of Haiyan Fund. (No. genic dedifferentiation,” Cell, vol. 173, no. 2, pp. 338– JJZD2019-06). 354.e15, 2018. [17] P. Langfelder and S. Horvath, “WGCNA: an R package for weighted correlation network analysis,” BMC Bioinformatics, References vol. 9, no. 1, p. 559, 2008. [1] R. Siegel, J. Ma, Z. Zou, and A. Jemal, “Cancer statistics, 2014,” [18] D. Perumal, V. V. Leshchenko, P. Y. Kuo et al., “Weighted CA: a Cancer Journal for Clinicians, vol. 64, no. 1, pp. 9–29, gene co-expression network analysis (WGCNA) identifies 2014. highly proliferative myeloma subgroup responsive to CDK4/ ARK5 inhibition,” Blood, vol. 124, no. 21, pp. 3445–3445, [2] E. Yakirevich and M. B. Resnick, “Pathology of gastric cancer and its precursor lesions,” Gastroenterology Clinics of North America, vol. 42, no. 2, pp. 261–284, 2013. [19] S. Pan, Y. Zhan, X. Chen, B. Wu, and B. Liu, “Identification of biomarkers for controlling cancer stem cell characteristics [3] Y. Horikoshi, Y. Yan, M. Terashvili et al., “Fatty acid-treated in bladder cancer by network analysis of transcriptome data induced pluripotent stem cell-derived human cardiomyocytes stemness indices,” Frontiers in Oncology, vol. 9, p. 613, exhibit adult cardiomyocyte-like energy metabolism pheno- types,” Cell, vol. 8, no. 9, article 1095, 2019. [4] M. Paolillo, R. Colombo, M. Serra et al., “Stem-like cancer cells [20] M. E. Ritchie, B. Phipson, D. Wu et al., “limma powers differ- ential expression analyses for RNA-sequencing and microar- in a dynamic 3D culture system: a model to study metastatic cell adhesion and anti-cancer drugs,” Cell, vol. 8, no. 11, article ray studies,” Nucleic Acids Research, vol. 43, no. 7, article e47, 1434, 2019. [5] L. MacDonagh, S. G. Gray, E. Breen et al., “Lung cancer stem [21] A. M. Szász, A. Lánczky, Á. Nagy et al., “Cross-validation of cells: the root of resistance,” Cancer Letters, vol. 372, no. 2, survival associated biomarkers in gastric cancer using tran- pp. 147–156, 2016. scriptomic data of 1,065 patients,” Oncotarget, vol. 7, no. 31, pp. 49322–49333, 2016. [6] E. Batlle and H. Clevers, “Cancer stem cells revisited,” Nature Medicine, vol. 23, no. 10, pp. 1124–1134, 2017. [22] D. Szklarczyk, A. L. Gable, D. Lyon et al., “STRING v11: [7] F. M. Uckun, H. Sather, G. Reaman et al., “Leukemic cell protein-protein association networks with increased coverage, supporting functional discovery in genome-wide experimental growth in SCID mice as a predictor of relapse in high-risk B- lineage acute lymphoblastic leukemia,” Blood, vol. 85, no. 4, datasets,” Nucleic Acids Research, vol. 47, no. D1, pp. D607– D613, 2019. pp. 873–878, 1995. [8] M. Al-Hajj, M. S. Wicha, A. Benito-Hernandez, S. J. Morrison, [23] J. Huelsken and D. Hanahan, “A subset of cancer-associated and M. F. Clarke, “Prospective identification of tumorigenic fibroblasts determines therapy resistance,” Cell, vol. 172, breast cancer cells,” Proceedings of the National Academy of no. 4, pp. 643-644, 2018. 14 Journal of Oncology [24] Y. C. Chae and J. H. Kim, “Cancer stem cell metabolism: target [40] J. Sun, L. Smith, A. Armento, and W. M. Deng, “Regulation of for cancer therapy,” BMB Reports, vol. 51, no. 7, pp. 319–326, the endocycle/gene amplification switch by Notch and ecdy- sone signaling,” The Journal of Cell Biology, vol. 182, no. 5, pp. 885–896, 2008. [25] F. Ricci, L. Brunelli, R. Affatato et al., “Overcoming platinum- acquired resistance in ovarian cancer patient-derived xeno- grafts,” Medical Oncology, vol. 11, article 1758835919839543, 2019Published 2019 Jun 24. [26] N. Salomonis, P. J. Dexheimer, L. Omberg et al., “Integrated genomic analysis of diverse induced pluripotent stem cells from the progenitor cell biology consortium,” Stem Cell Reports, vol. 7, no. 1, pp. 110–125, 2016. [27] T. Shibue and R. A. Weinberg, “EMT, CSCs, and drug resis- tance: the mechanistic link and clinical implications,” Nature Reviews. Clinical Oncology, vol. 14, no. 10, pp. 611–629, 2017. [28] D. Friedmann-Morvinski and I. M. Verma, “Dedifferentiation and reprogramming: origins of cancer stem cells,” EMBO Reports, vol. 15, no. 3, pp. 244–253, 2014. [29] J. Y. Han, Y. K. Han, G. Y. Park, S. D. Kim, and C. G. Lee, “Bub1 is required for maintaining cancer stem cells in breast cancer cell lines,” Scientific Reports, vol. 5, article 15993, 2015Published 2015 Nov 2. [30] M. Venere, T. E. Miller, and J. N. Rich, “Mitotic control of can- cer stem cells,” Cancer Discovery, vol. 3, no. 2, pp. 141–144, [31] Q. Li, J. Qiu, H. Yang et al., “Kinesin family member 15 pro- motes cancer stem cell phenotype and malignancy via reactive oxygen species imbalance in hepatocellular carcinoma,” Can- cer Letters, vol. 482, pp. 112–125, 2020. [32] D. Huang, Q. Gao, L. Guo et al., “Isolation and identification of cancer stem-like cells in esophageal carcinoma cell lines,” Stem Cells and Development, vol. 18, no. 3, pp. 465–474, 2009. [33] W. Zhou, Y. Yang, Z. Gu et al., “ALDH1 activity identifies tumor-initiating cells and links to chromosomal instability sig- natures in multiple myeloma,” Leukemia, vol. 28, no. 5, pp. 1155–1158, 2014. [34] N. Takebe, L. Miele, P. J. Harris et al., “Targeting Notch, Hedgehog, and Wnt pathways in cancer stem cells: clinical update,” Nature Reviews. Clinical Oncology, vol. 12, no. 8, pp. 445–464, 2015. [35] T. Yang, X. N. Li, L. Li et al., “Sox17 inhibits hepatocellular carcinoma progression by downregulation of KIF14 expres- sion,” Tumour Biology, vol. 35, no. 11, pp. 11199–11207, [36] L. Ding, S. Zhang, S. Chen, L. Zheng, and L. Xiao, “Effect and mechanism of lentivirus-mediated silencing of TPX2 gene on proliferation and apoptosis of human hepatoma cells,” Journal of Cellular Biochemistry, vol. 120, 2018. [37] Y. Wu, A. Wang, B. Zhu et al., “KIF18B promotes tumor pro- gression through activating the Wnt/β-catenin pathway in cer- vical cancer,” Oncotargets and Therapy, vol. 11, pp. 1707– 1720, 2018, Published 2018 Mar 28. [38] Z. Liao, H. Zhang, P. Fan et al., “High PLK4 expression pro- motes tumor progression and induces epithelial- mesenchymal transition by regulating the Wnt/β-catenin sig- naling pathway in colorectal cancer,” International Journal of Oncology, vol. 54, no. 2, pp. 479–490, 2019. [39] B. D. Wilson, L. J. Ricks-Santi, T. E. Mason et al., “Admixture mapping links RACGAP1 regulation to prostate cancer in African Americans,” Cancer Genomics & Proteomics, vol. 15, no. 3, pp. 185–191, 2018.

Journal

Journal of OncologyHindawi Publishing Corporation

Published: Sep 30, 2022

There are no references for this article.