Multiple Microarray Datasets Research Articles

Unraveling the Molecular Complexity of Adenoid Cystic Carcinoma (ACC): A Comprehensive Exploration of Hub Genes, Protein-Protein Interaction (PPI) Networks, microRNA (miRNA) Involvement, and Drug-Gene Interactions (DGIs).

Background Adenoid cystic carcinoma (ACC) poses clinical challenges with its unique histology and potential for perineural invasion, recurrence, and distant metastases. Recent genomic advancements have unveiled key genetic alterations in ACC, offering insights into its pathogenesis. Aim This study aims to unravel the intricate molecular landscape of ACC through a comprehensive analysis of gene expression patterns. By integrating data from multiple microarray datasets, the study explores differentially expressed genes (DEGs), their functional enrichment, protein-protein interactions (PPI), hub genes, microRNA (miRNA) involvement, transcription factors, and potential drug-gene interactions. Methods Three microarray datasets (GSE88804, GSE153002, and GSE36820) related to ACC were selected from theGene Expression Omnibus (GEO) repository. DEGs were identified using GEO2R and further analyzed for commonalities and differences. Functional enrichment analysis, including Gene Set Enrichment Analysis (GSEA), provided insights into biological processes, cellular components, molecular functions, and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways associated with ACC. PPI networks and hub genes were identified using the Search Tool for the Retrieval of Interacting Genes/Proteins (STRING) (STRING Consortium, Lausanne,Switzerland) database and Cytoscape (Cytoscape Consortium, California, United States). The study also explored miRNAs, transcription factors, and potential drug-gene interactions. Results The integrated analysis revealed 339 common upregulated and 643 downregulated DEGs in ACC. Functional and pathway enrichment analyses unveiled the involvement of these genes in critical cellular processes, signaling cascades, and pathways. The PPI network, comprising 904 nodes and 4139 edges, highlighted the complexity of interactions. Hub genes, including KIF11, BUB1, and DLGAP5, were identified, shedding light on their pivotal roles in cell cycle regulation. The study also identified miRNAs (e.g., hsa-mir-7-5p and hsa-mir-138-5p) and transcription factors (e.g., E2F1 and TP53) associated with ACC. Drug-gene interactions have identified potential therapeutic options, including amsacrine and rucaparib. Conclusions The ACCgene expression highlights a nuanced molecular landscape, identifying pivotal hub genes such as KIF11 and CDK1 as potential therapeutic targets for ACC, given their roles in cell cycle progression. The dysregulation of microRNAs and transcription factors adds complexity to ACC's molecular profile. Exploration of drug-gene interactions reveals promising therapeutic strategies, involving FDA-approved drugs such as amsacrine and rucaparib, providing avenues for personalized interventions.

Novel HGDBO: A Hybrid Genetic and Dung Beetle Optimization Algorithm for Microarray Gene Selection and Efficient Cancer Classification

Introduction: Ovarian cancer ranks as the seventh most frequently diagnosed cancer and stands as the eighth leading cause of cancer-related mortality among women globally. Early detection significantly improves survival rates and outcomes, highlighting the need for enhanced screening methods and increased awareness to facilitate early diagnosis and treatment. Microarray gene data, characterized by its high dimensionality, includes the expression levels of thousands of genes across numerous samples, posing both opportunities and challenges in the analysis of gene functions and disease mechanisms. Method: This paper presents a novel hybrid gene feature selection method called HGDBO, which combines the Dung Beetle Optimization (DBO) algorithm with the Genetic Algorithm (GA) to increase the effectiveness of microarray data analysis. The proposed HGDBO method utilizes the exploratory capabilities of DBO and the exploitative strengths of GA to identify the most relevant genes for disease classification. Experimental results on multiple microarray datasets demonstrate that the hybrid approach offers superior classification performance, stability, and computational efficiency compared to traditional and state-of-the-art methods. To classify ovarian cancer, Naïve-Bayes (NB) and Random-Forest (RF) classification algorithms were employed. Results and Discussion: The proposed Random Forest model outperforms the Naive Bayes model across all metrics, achieving better accuracy (0.96 vs. 0.91), precision (0.95 vs. 0.91), recall (0.97 vs. 0.90), F-1 score (0.95 vs. 0.91), and specificity (0.97 vs. 0.86). Conclusion: These results underscore the effectiveness of the HGDBO method and the Random Forest classifier in enhancing the analysis and classification of ovarian cancer using microarray gene data.

Integrating multiple microarray datasets to explore the significance of ferroptosis regulators in the diagnosis and subtype classification of osteoarthritis.

Osteoarthritis (OA) is a chronic joint disease that reduces quality of life for patients. Ferroptosis plays a significant role in OA. However, its underlying mechanism remains unclear. In this study, we integrated 7 OA synovial datasets from the GEO database to screen for significant ferroptosis-related genes. The top 5 ferroptosis regulators were used to construct nomogram models to predict OA prevalence. Consensus clustering was applied to classify OA patients into different ferroptosis patterns based on significant ferroptosis-related genes. Subsequently, an immune cell infiltration study was performed to investigate the relationship between the significant ferroptosis regulators and immune cells. As a result, we screened 11 ferroptosis-related genes in OA patients. Five candidate ferroptosis regulators (SLC7A11, ALOX5, SLC1A5, GOT1, and GSS) were used to predict OA risk. The nomogram model based on these 5 genes is important for assessing the occurrence of OA. Consensus clustering analysis showed that OA patients could be classified into 2 ferroptosis patterns (Clusters A and B). Immune cell infiltration levels were higher in Cluster B than in Cluster A. Two subtypes, gene Clusters A and B, were classified according to the expression of ferroptosis-related DEGs among the ferroptosis patterns. Cluster A and gene Cluster A had higher ferroptosis scores than Cluster B or gene Cluster B, whereas the expression levels of the proinflammatory cytokines interleukin (IL)-1β, tumor necrosis factor, IL-6, IL-18, and IL-10 were higher in Cluster B or gene Cluster B than those in Cluster A or gene Cluster A. Different subtypes of ferroptosis play critical roles in OA. Furthermore, immunotherapy strategies for OA treatment may be guided by our study on ferroptosis patterns.

PMP-SVM: A Hybrid Approach for effective Cancer Diagnosis using Feature Selection and Optimization

Cancer disease is becoming a prominent factor in increasing the death ration over the world due to the late diagnosis. Machine Learning (ML) is playing a vital role in providing computer aided diagnosis models for early diagnosis of cancer. For the diagnosis process the microarray data has its own place. Microarray data contain the genetic information of a patient with a large number of dimensions such as genes with a small sample such as patient details. If the microarray is directly taken without reducing the dimension as the input to any ML model for classification, then Small Sample Size is the resulting issue. So, size of the microarray data needs to be reduces by using either of dimensionality reduction technique or the feature selection technique to increase the model’s performance. In this work, proposed a hybrid model using Principal Component Analysis (PCA), Maximum Relevance Minimum Redundancy (MRMR), Particle Swarm Optimization (PSO) and Support Vector Machine (SVM) for cancer diagnosis. PCA and MRMR is used for feature selection and PSO is applied to get the optimized feature set. Finally, SVM is applied as the classification model. The proposed model is evaluated against multiple cancer microarray datasets to measure the performance in terms of accuracy, precision, recall, and F1 score. Result shows that proposed model performs better than existing state of art model.

A multi-platform normalization method for meta-analysis of gene expression data.

Transcriptomic profiling is a mainstay of translational cancer research and is often used to identify cancer subtypes, stratify responders vs. non-responders patients, predict survival, and identify potential targets for therapeutic intervention. Analysis of gene expression data gathered by RNA sequencing (RNA-seq) and microarray is generally the first step in identifying and characterizing cancer-associated molecular determinants. The methodological advancements and reduced costs associated with transcriptomic profiling have increased the number of publicly available gene expression profiles for cancer subtypes. Data integration from multiple datasets is routinely done to increase the number of samples, improve statistical power, and provide better insight into the heterogeneity of the biological determinant. However, utilizing raw data from multiple platforms, species, and sources introduces systematic variations due to noise, batch effects, and biases. As such, the integrated data is mathematically adjusted through normalization, which allows direct comparison of expression measures among studies while minimizing technical and systemic variations. This study applied meta-analysis to multiple independent Affymetrix microarray and Illumina RNA-seq datasets available through the Gene Expression Omnibus (GEO) and The Cancer Gene Atlas (TCGA). We have previously identified a tripartite motif containing 37 (TRIM37), a breast cancer oncogene, that drives tumorigenesis and metastasis in triple-negative breast cancer. In this article, we adapted and assessed the validity of Stouffer's z-score normalization method to interrogate TRIM37 expression across different cancer types using multiple large-scale datasets.

NUF2 is correlated with a poor prognosis and immune infiltration in clear cell renal cell carcinoma

BackgroundClear cell renal cell carcinoma (ccRCC) is one of the most common malignancies. Recently, immunotherapy has been considered a promising treatment for metastatic ccRCC. NUF2 is a crucial component of the Ndc80 complex. NUF2 can stabilize microtubule attachment and is closely related to cell apoptosis and proliferation. This research is dedicated to investigating the role of NUF2 in ccRCC and the possible mechanisms.MethodsFirst, analysis of NUF2 mRNA expression levels in ccRCC and normal tissues by The Cancer Genome Atlas (TCGA) database and further verified by analysis of independent multiple microarray data sets in the Gene Expression Omnibus (GEO) database. Moreover, we evaluated and identified correlations between NUF2 expression, clinicopathologic variable, and overall survival (OS) in ccRCC by various methods. We investigated the relationship between NUF2 and tumor immune infiltration and the expression of corresponding immune cell markers via the Gene Expression Profiling Interactive Analysis (GEPIA) and Tumor Immune Estimation Resource (TIMER) databases. Then, we performed functional enrichment analysis of NUF2 co-expressed genes using R software and protein-protein interactions (PPIs) using the search tool used to retrieve interacting genes/proteins (STRING) databases.ResultsWe discovered that NUF2 mRNA expression was upregulated in ccRCC tissues and was associated with sex, grade, pathological stage, lymph node metastasis, and worse prognosis. In addition, NUF2 was positively linked to tumor immune cells in ccRCC. Moreover, NUF2 was closely related to genetic markers of different immune cells. Finally, functional enrichment and protein–protein interaction (PPI) analysis suggested that NUF2 and its closely related genes may be involved in the regulation of the cell cycle and mitosis. Our results suggested that NUF2 is correlated with a poor prognosis and immune infiltration in ccRCC.

AURKA, TOP2A and MELK are the key genes identified by WGCNA for the pathogenesis of lung adenocarcinoma.

The comprehensive analysis of single or multiple microarray datasets is currently available in Gene Expression Omnibus (GEO) databases, with several studies having identified genes strongly associated with the development of lung adenocarcinoma (LUAD). However, the mechanisms of LUAD development remain largely unknown and has not yet been systematically studied; thus, further studies are required in this field. In the present study, weighted gene co-expression network analysis (WGCNA) was used for the evaluation of key genes with potential high risk of LUAD, and to provide more reliable evidence concerning its pathogenesis. The GSE140797 dataset from the high-throughput GEO database was downloaded and was first analyzed using the Limma package in the R language in order to determine the differentially expressed genes. The dataset was then analyzed using the WGCNA package to analyze the co-expressed genes, and the modular genes with the highest correlation with the clinical phenotype were identified. Subsequently, the pathogenic genes shared in common between the result of the two analyses were imported into the STRING database for protein-protein interaction network analysis. The hub genes were screened out using Cytoscape, and then The Cancer Genome Atlas analysis, receiver operating characteristic analysis and survival analysis were subsequently performed. Finally, the key genes were evaluated using reverse transcription-quantitative PCR and western blot analysis. Bioinformatics analysis of the GSE140797 dataset revealed eight key genes: AURKA, BUB1, CCNB1, CDK1, MELK, NUSAP1, TOP2A and PBK. Finally, the AURKA, TOP2A and MELK genes were evaluated in samples from patients with lung cancer using WGCNA and RT-qPCR, western blot analysis experiments, providing basis for further research on the mechanisms of LUAD development and targeted therapy.

Deciphering novel common gene signatures for rheumatoid arthritis and systemic lupus erythematosus by integrative analysis of transcriptomic profiles.

Rheumatoid Arthritis (RA) and Systemic Lupus Erythematosus (SLE) are the two highly prevalent debilitating and sometimes life-threatening systemic inflammatory autoimmune diseases. The etiology and pathogenesis of RA and SLE are interconnected in several ways, with limited knowledge about the underlying molecular mechanisms. With the motivation to better understand shared biological mechanisms and determine novel therapeutic targets, we explored common molecular disease signatures by performing a meta-analysis of publicly available microarray gene expression datasets of RA and SLE. We performed an integrated, multi-cohort analysis of 1088 transcriptomic profiles from 14 independent studies to identify common gene signatures. We identified sixty-two genes common among RA and SLE, out of which fifty-nine genes (21 upregulated and 38 downregulated) had similar expression profiles in the diseases. However, antagonistic expression profiles were observed for ACVR2A, FAM135A, and MAPRE1 genes. Thirty genes common between RA and SLE were proposed as robust gene signatures, with persistent expression in all the studies and cell types. These gene signatures were found to be involved in innate as well as adaptive immune responses, bone development and growth. In conclusion, our analysis of multicohort and multiple microarray datasets would provide the basis for understanding the common mechanisms of pathogenesis and exploring these gene signatures for their diagnostic and therapeutic potential.

Machine Learning Model Based on Insulin Resistance Metagenes Underpins Genetic Basis of Type 2 Diabetes.

Insulin resistance (IR) is considered the precursor and the key pathophysiological mechanism of type 2 diabetes (T2D) and metabolic syndrome (MetS). However, the pathways that IR shares with T2D are not clearly understood. Meta-analysis of multiple DNA microarray datasets could provide a robust set of metagenes identified across multiple studies. These metagenes would likely include a subset of genes (key metagenes) shared by both IR and T2D, and possibly responsible for the transition between them. In this study, we attempted to find these key metagenes using a feature selection method, LASSO, and then used the expression profiles of these genes to train five machine learning models: LASSO, SVM, XGBoost, Random Forest, and ANN. Among them, ANN performed well, with an area under the curve (AUC) > 95%. It also demonstrated fairly good performance in differentiating diabetics from normal glucose tolerant (NGT) persons in the test dataset, with 73% accuracy across 64 human adipose tissue samples. Furthermore, these core metagenes were also enriched in diabetes-associated terms and were found in previous genome-wide association studies of T2D and its associated glycemic traits HOMA-IR and HOMA-B. Therefore, this metagenome deserves further investigation with regard to the cardinal molecular pathological defects/pathways underlying both IR and T2D.

Potential Biomarkers for Parkinson Disease from Functional Enrichment and Bioinformatic Analysis of Global Gene Expression Patterns of Blood and Substantia Nigra Tissues.

The Parkinson disease (PD) is the second most common neurodegenerative disorder affecting the central nervous system and motor functions. The biological complexity of PD is yet to reveal potential targets for intervention or to slow the disease severity. Therefore, this study aimed to compare the fidelity of blood to substantia nigra (SN) tissue gene expression from PD patients to provide a systematic approach to predict role of the key genes of PD pathobiology. Differentially expressed genes (DEGs) from multiple microarray data sets of PD blood and SN tissue from GEO database are identified. Using the theoretical network approach and variety of bioinformatic tools, we prioritized the key genes from DEGs. A total of 540 and 1024 DEGs were identified in blood and SN tissue samples, respectively. Functional pathways closely related to PD such as ERK1 and ERK2 cascades, mitogen-activated protein kinase (MAPK) signaling, Wnt, nuclear factor-κB (NF-κB), and PI3K-Akt signaling were observed by enrichment analysis. Expression patterns of 13 DEGs were similar in both blood and SN tissues. Comprehensive network topological analysis and gene regulatory networks identified additional 10 DEGs functionally connected with molecular mechanisms of PD through the mammalian target of rapamycin (mTOR), autophagy, and AMP-activated protein kinase (AMPK) signaling pathways. Potential drug molecules were identified by chemical-protein network and drug prediction analysis. These potential candidates can be further validated in vitro/in vivo to be used as biomarkers and/or novel drug targets for the PD pathology and/or to arrest or delay the neurodegeneration over the years, respectively.

The iron-modulating hormone hepcidin is upregulated and associated with poor survival outcomes in renal clear cell carcinoma.

Background: Reliable biomarkers are rare for renal cell carcinoma (RCC) treatment selection. We aimed to discover novel biomarkers for precision medicine. The iron-regulating hormone hepcidin (HAMP) was reportedly increased in RCC patient sera and tissues. However, its potential implication as a prognostic biomarker remains exclusive. Methods: Multiple RNA-seq and cDNA microarray datasets were utilized to analyze gene expression profiles. Hepcidin protein expression was assessed using an ELISA assay in cell culture models. Comparisons of gene expression profiles and patient survival outcomes were conducted using the R package bioinformatics software. Results: Five (HAMP, HBS, ISCA2, STEAP2, and STEAP3) out of 71 iron-modulating genes exhibited consistent changes along with tumor stage, lymph node invasion, distal metastasis, tumor cell grade, progression-free interval, overall survival, and disease-specific survival. Of which HAMP upregulation exerted as a superior factor (AUC = 0.911) over the other four genes in distinguishing ccRCC tissue from normal renal tissue. HAMP upregulation was tightly associated with its promoter hypomethylation and immune checkpoint factors (PDCD1, LAG3, TIGIT, and CTLA4). Interleukin-34 (IL34) treatment strongly enhanced hepcidin expression in renal cancer Caki-1 cells. Patients with higher levels of HAMP expression experienced worse survival outcomes. Conclusion: These data suggest that HAMP upregulation is a potent prognostic factor of poor survival outcomes and a novel immunotherapeutic biomarker for ccRCC patients.

Elevated CXCL14 in Induced Sputum Was Associated with Eosinophilic Inflammation and Airway Obstruction in Patients with Asthma

Introduction: CXCL14 involved in inflammatory processes was upregulated in the asthma expression profile datasets in our pilot study. However, the expression of CXCL14 in induced sputum and its potential clinical role in asthma were poorly reported. Objective: We sought to detect CXCL14 expression in airway epithelium and induced sputum cells of asthma and explore its potential clinical implications. Methods: The expression of CXCL14 in asthma was analyzed using R software based on multiple microarray datasets, including GSE43696, GSE63142, GSE67940, and GSE76262. Subsequent verification of the CXCL14 expression pattern in induced sputum and bronchial epithelium cells was performed by qRT-PCR and ELISA. Besides, the correlations between CXCL14 and eosinophilic inflammation indicators (FeNO, EOS#, and IgE), Th2 signature genes (SERPINB2, POSTN, and CLCA1), inflammatory cytokines (IL-4, IL-5, IL-10, IL-13, IL-25, IL-33, TSLP, IL-8, IL-17A, IFN-γ, and IL-2), and airway obstruction indicators (pulmonary function and mucin secretion) were further explored. Results: The expression of CXCL14 in epithelium and sputum cells was upregulated in asthma and positively correlated with clinical eosinophilic indicators. The protein levels of CXCL14 were positively associated with Th2 signature genes (SERPINB2, POSTN, and CLCA1) and Th2 cytokines (IL-4, IL-5, IL-10, IL-13, IL-25, IL-33, and TSLP). Increased expression of CXCL14 was also observed in BEAS-2B cells stimulated by the cytokine IL-4. Furthermore, the expression of CXCL14 was positively correlated with MUC5AC secretion and negatively associated with pulmonary function. Conclusions: Upregulated CXCL14 in asthma was positively correlated with inflammatory indicators and negatively correlated with pulmonary function, which indicated that upregulated CXCL14 might act as a pathogenic gene through involvement in Th2 inflammation in asthma.

Bioinformatic analysis identifies potential key genes in the pathogenesis of age-related macular degeneration.

Age-related macular degeneration (AMD) is the leading cause of irreversible blindness in older individuals. More studies focused on screening the genes, which may be correlated with the development of AMD. With advances in various technologies like multiple microarray datasets, researchers could identify differentially expressed genes (DEGs) more accurately. Exploring abnormal gene expression in disease status can help to understand pathophysiological changes in complex diseases. This study aims to identify the key genes and upstream regulators in AMD and reveal factors, especially genetic association, and the prognosis of the development of this disease. Data from expression profile GSE125564 and profile GSE29801 were obtained from the Gene Expression Omnibus (GEO) database. We analyzed DEGs using R software (version 3.6.3). Functional enrichment and PPI network analysis were performed using the R package and online database STRING (version 11.0). We compared AMD with normal and found 68 up-regulated genes (URGs) and 25 down-regulated genes (DRGs). We also compared wet AMD with dry AMD and found 41 DRGs in dry AMD. Further work including PPI network analysis, GO classification, and KEGG analysis was done to find connections with AMD. The URGs were mainly enriched in the biological process such as DNA replication, nucleoplasm, extracellular exosome, and cadherin binding. Besides, DRGs were mainly enriched in these functions such as an integral component of membrane and formation of the blood-aqueous barrier (BAB). This study implied that core genes might involve in the process of AMD. Our findings may contribute to revealing the pathogenesis, developing new biomarkers, and raising strategies of treatment for AMD.

AT-rich interaction domain 5A regulates the transcription of interleukin-6 gene in prostate cancer cells.

Interleukin-6 (IL-6) is a pleiotropic cytokine that confers androgen-independence and aggressiveness in prostate cancer (PCa); however, the molecular mechanisms regulating IL-6 expression remain unclear. The expression of ARID5A, an AT-rich interaction domain (ARID) DNA-binding motif-containing transcription factor is positively correlated with IL-6 expression in human PCa. We, therefore, hypothesized that ARID5A could regulate IL-6 expression in PCa. The relationship between ARID5A and IL-6 in PCa patients was analyzed using statistical analyses of multiple clinical microarray data sets. To investigate whether ARID5A regulates IL-6 expression, CRISPR-driven ARID5A knockout clones were established in DU145 and PC-3 cells. Analysis of three microarray data sets showed a positive correlation between ARID5A and IL-6 expression. The expression of IL-6 in ARID5A knockout clones was significantly reduced compared with control clones in both PCa cell lines. Knockout of ARID5A did not result in any loss of IL-6 mRNA stability. Instead, we observed a significant decrease in the occupancy of both active RNA Polymerase II and the active histone mark, H3K4me3 at the IL-6 transcriptional start site in ARID5A knockout PCa cells, suggesting a role for transcriptional regulation. Our study demonstrated that loss of ARID5A downregulates the expression of IL-6 at the transcriptional level.

PAQR6 Upregulation Is Associated with AR Signaling and Unfavorite Prognosis in Prostate Cancers.

Progesterone-induced rapid non-genomic signaling events have been confirmed through several membrane progesterone receptors (mPR). Some mPRs were reported to correlate with cancer progression and patient prognosis. In this study, we conducted a comprehensive analysis of all progesterone receptor (PGR)-related genes in prostate cancer tissues and examined the correlations of their expression levels with disease progression and patient survival outcomes. We utilized multiple RNA-seq and cDNA microarray datasets to analyze gene expression profiles and performed logistics aggression and Kaplan-Meier survival analysis after stratifying patients based on tumor stages and Gleason scores. We also used NCBI GEO datasets to examine gene expression patterns in individual cell types of the prostate gland and to determine the androgen-induced alteration of gene expression. Spearman coefficient analysis was conducted to access the correlation of target gene expression with treatment responses and disease progression status. The classic PGR was mainly expressed in stromal cells and progestin and adipoQ receptor (PAQR) genes were the predominant genes in prostate epithelial cells. Progesterone receptor membrane component-1 (PGRMC1) was significantly higher than PGRMC2 in all prostate cell types. In prostate cancer tissues, PAQR6 expression was significantly upregulated, while all other genes were largely downregulated compared to normal prostate tissues. Although both PAQR6 upregulation and PAQR5 downregulation were significantly correlated with tumor pathological stages, only PAQR6 upregulation was associated with Gleason score, free-prostate-specific antigen (fPSA)/total-PSA (tPSA) ratio, and patient overall survival outcomes. In addition, PAQR6 upregulation and PGR/PGRMC1 downregulation were significantly associated with a quick relapse. Conversely, in neuroendocrinal prostate cancer (NEPC) tissues, PAQR6 expression was significantly lower, but PAQR7/8 expression was higher than castration-resistant prostate cancer (CRPC) tissues. PAQR8 expression was positively correlated with androgen receptor (AR) score and AR-V7 expression levels but inversely correlated with NEPC score in metastatic CRPC tumors. This study provides detailed expression profiles of membrane progesterone receptor genes in primary cancer, CRPC, and NEPC tissues. PAQR6 upregulation in primary cancer tissues is a novel prognostic biomarker for disease progression, overall, and progression-free survival in prostate cancers. PAQR8 expression in CRPC tissues is a biomarker for AR activation.

Validation of Gene Expression-Based Predictive Biomarkers for Response to Neoadjuvant Chemoradiotherapy in Locally Advanced Rectal Cancer.

Simple SummaryThere is a clinical need for predictive biomarkers that can identify patients with rectal cancer who do not respond to preoperative neoadjuvant chemoradiotherapy. In this study, we assembled multiple independent microarray datasets of biopsy specimens obtained from patients with rectal cancer before neoadjuvant treatment, including 237 non-responders and 152 responders. These datasets were utilized as the discovery cohorts or the validation cohorts, to develop and validate gene expression signatures predictive of treatment response. Using an in silico meta-analysis approach, here we tested not only our 4-gene signature built in this study but also nine different single-gene and multi-gene predictive signatures that were previously reported in the literature. Nevertheless, in the validation cohorts, none of the tested signatures were consistently differentially expressed between tumor specimens from non-responders and responders, and the meta-analyses revealed that those signatures had limited predictive values in clinical practice.Background: Neoadjuvant chemoradiotherapy (nCRT) followed by surgery is widely used for patients with locally advanced rectal cancer. However, response to nCRT varies substantially among patients, highlighting the need for predictive biomarkers that can distinguish non-responsive from responsive patients before nCRT. This study aimed to build novel multi-gene assays for predicting nCRT response, and to validate our signature and previously-reported signatures in multiple independent cohorts. Methods: Three microarray datasets of pre-therapeutic biopsies containing a total of 61 non-responders and 53 responders were used as the discovery cohorts to screen for genes that were consistently associated with nCRT response. The predictive values of signatures were tested in a meta-analysis using six independent datasets as the validation cohorts, consisted of a total of 176 non-responders and 99 responders. Results: We identified four genes, including BRCA1, GPR110, TNIK, and WDR4 in the discovery cohorts. Although our 4-gene signature and nine published signatures were evaluated, they were unable to predict nCRT response in the validation cohorts. Conclusions: Although this is one of the largest studies addressing the validity of gene expression-based classifiers using pre-treatment biopsies from patients with rectal cancer, our findings do not support their clinically meaningful values to be predictive of nCRT response.

Abstract 2309: Predicting molecular networks mediating colorectal cancer neoplastic progression by integrative transcriptome-wide meta-analysis

Abstract Introduction: Colorectal cancer (CRC) is the third most common cancer in the United States. While survival rates have increased in part due to enhanced detection and targeted therapeutics, mortality remains high, highlighting a need to further investigate novel targetable pathways along the neoplastic axis. To explore potential molecular networks involved in pre- and post-malignant processes, we implemented a bioinformatics-based pipeline to meta-analyze multiple microarray datasets for predicting differentially active/inhibited pathways. Methods: We constructed a curated, batch-corrected Meta-dataset comprising twelve microarray datasets including GSE4183, GSE8671, GSE9348, GSE15960, GSE20916, GSE21510, GSE22598, GSE23194, GSE23878, GSE32323, GSE33113, and GSE37364, harboring normal (n = 221), adenoma (n = 137), and CRC (n = 452) samples. The Meta-dataset expression set was validated against the TCGA-COAD dataset with functional adenoma and CRC signatures confirmed by Ingenuity Pathway Analysis (IPA) and Gene Set Enrichment Analysis (GSEA). Differentially expressed genes (DEGs) were identified for Adenoma:Normal (A v N), CRC:Adenoma (C v A), and CRC:Normal (C v N) comparison groups and submitted to IPA. To identify relevant gene clusters between DEGs, a heatmap was constructed and arranged by Euclidean distance-weighted unsupervised hierarchal clustering. The same was done regarding IPA-predicted mediators with analysis being restricted to upstream regulators with activation Z-scores > I 2 I in at least two of the three comparison groups. Results: Using ComBat, inter-study batch effects were removed from the Meta-dataset whilst preserving normal, adenoma, and CRC signatures. We found a high degree of correlation between the Meta- and the TCGA-COAD dataset (rho = 0.89, p = 2.2e-16) with functional validation of both adenoma (FDR < 0.0001) and CRC (FDR < 0.0001) signatures using IPA and GSEA. Following, three-way analysis of A v N, C v A, and C v N DEGs revealed four distinct clusters associated with adenoma formation and malignant transformation. Clusters were then cross-reference against established networks regulated by KRAS, EGFR, ERBB2, and TP53 in IPA. Predicted networks associated with adenoma formation were enriched in epithelial-mesenchymal transition pathways (AREG, SNAI1, FOXM1, TCF4) while those associated with malignant transformation were inflammatory based (IL17, TNF, CD24, TLR4). Moreover, early and late downregulated networks were primarily controlled by tumor suppressors (CDKN1A, CST5, TRPS1 and NR1H family members). Conclusion: We identified both known and novel pathways involved in the neoplastic progression of CRC through transcriptome-wide pooled analysis. These pathways may be attractive candidates for future investigation of tumorigenic mechanisms, prognostication, and therapeutics. Citation Format: Michael W. Rohr, Deborah Altomare. Predicting molecular networks mediating colorectal cancer neoplastic progression by integrative transcriptome-wide meta-analysis [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 2309.

Coupling of Co-expression Network Analysis and Machine Learning Validation Unearthed Potential Key Genes Involved in Rheumatoid Arthritis.

Rheumatoid arthritis (RA) is an incurable disease that afflicts 0.5–1.0% of the global population though it is less threatening at its early stage. Therefore, improved diagnostic efficiency and prognostic outcome are critical for confronting RA. Although machine learning is considered a promising technique in clinical research, its potential in verifying the biological significance of gene was not fully exploited. The performance of a machine learning model depends greatly on the features used for model training; therefore, the effectiveness of prediction might reflect the quality of input features. In the present study, we used weighted gene co-expression network analysis (WGCNA) in conjunction with differentially expressed gene (DEG) analysis to select the key genes that were highly associated with RA phenotypes based on multiple microarray datasets of RA blood samples, after which they were used as features in machine learning model validation. A total of six machine learning models were used to validate the biological significance of the key genes based on gene expression, among which five models achieved good performances [area under curve (AUC) >0.85], suggesting that our currently identified key genes are biologically significant and highly representative of genes involved in RA. Combined with other biological interpretations including Gene Ontology (GO) analysis, protein–protein interaction (PPI) network analysis, as well as inference of immune cell composition, our current study might shed a light on the in-depth study of RA diagnosis and prognosis.

Independent component analysis recovers consistent regulatory signals from disparate datasets.

The availability of bacterial transcriptomes has dramatically increased in recent years. This data deluge could result in detailed inference of underlying regulatory networks, but the diversity of experimental platforms and protocols introduces critical biases that could hinder scalable analysis of existing data. Here, we show that the underlying structure of the E. coli transcriptome, as determined by Independent Component Analysis (ICA), is conserved across multiple independent datasets, including both RNA-seq and microarray datasets. We subsequently combined five transcriptomics datasets into a large compendium containing over 800 expression profiles and discovered that its underlying ICA-based structure was still comparable to that of the individual datasets. With this understanding, we expanded our analysis to over 3,000 E. coli expression profiles and predicted three high-impact regulons that respond to oxidative stress, anaerobiosis, and antibiotic treatment. ICA thus enables deep analysis of disparate data to uncover new insights that were not visible in the individual datasets.

CCR2 improves homing and engraftment of adipose-derived stem cells in dystrophic mice

BackgroundDystrophinopathy, a common neuromuscular disorder caused by the absence of dystrophin, currently lacks effective treatments. Systemic transplantation of adipose-derived stem cells (ADSCs) is a promising treatment approach, but its low efficacy remains a challenge. Chemokine system-mediated stem cell homing plays a critical role in systemic transplantation. Here, we investigated whether overexpression of a specific chemokine receptor could improve muscle homing and therapeutic effects of ADSC systemic transplantation in dystrophic mice.MethodsWe analysed multiple microarray datasets from the Gene Expression Omnibus to identify a candidate chemokine receptor and then evaluated the protein expression of target ligands in different tissues and organs of dystrophic mice. The candidate chemokine receptor was overexpressed using the lentiviral system in mouse ADSCs, which were used for systemic transplantation into the dystrophic mice, followed by evaluation of motor function, stem cell muscle homing, dystrophin expression, and muscle pathology.ResultsChemokine-profile analysis identified C–C chemokine receptor (CCR)2 as the potential target for improving ADSC homing. We found that the levels of its ligands C–C chemokine ligand (CCL)2 and CCL7 were higher in muscles than in other tissues and organs of dystrophic mice. Additionally, CCR2 overexpression improved ADSC migration ability and maintained their multilineage-differentiation potentials. Compared with control ADSCs, transplantation of those overexpressing CCR2 displayed better muscle homing and further improved motor function, dystrophin expression, and muscle pathology in dystrophic mice.ConclusionsThese results demonstrated that CCR2 improved ADSC muscle homing and therapeutic effects following systemic transplantation in dystrophic mice.