Identifying Potential miRNA Biomarkers for Gastric Cancer Diagnosis Using Machine Learning Variable Selection Approach

Identification of biomarkers related to Escherichia coli infection for the diagnosis of gastrointestinal tumors applying machine learning methods

Background

Escherichia coli (E. coli) is a part of normal gastrointestinal microbiota but it could also cause human gastrointestinal diseases. Understanding the mechanism of E. coli in the progression of gastrointestinal tumors can provide novel prevention and treatment strategies for gastrointestinal tumors.

Methods

The E. coli infection score was calculated by single sample GSEA (ssGSEA). Weighted correlation network analysis (WGCNA) and differentially expressed genes (DEGs) analysis were used to identify genes related to E. coli infection in gastrointestinal tumors. Hub genes were selected by machine learning methods to establish a diagnostic model. The diagnostic performance of the model was evaluated by the area under the receiver operating characteristic (ROC) curve (AUC) and validated in three external datasets. After determining the biomarkers, immune infiltration analysis and GSEA were further performed. The mRNA expressions of the biomarkers in stomach adenocarcinoma (STAD) cells and the invasion and migration of the tumor cells were detected by conducting in vitro experiments.

Results

The E. coli infection score was lower in tumor samples than in normal samples. Eight hub genes were selected from a total of 28 genes associated with E. coli-related dysbiosis in gastrointestinal tumors to establish an accurate diagnostic model. The AUC values of PRKCB and IL16 were all greater than 0.7 in three external datasets and the mRNA expression pattern was consistent with TCGA cohort, therefore PRKCB and IL16 were selected as the diagnostic biomarkers. PRKCB and IL16 exhibited significant positive correlations with most immune cells, and inflammation-related pathways were activated in the high expression groups of PRKCB and IL16. Moreover, IL16 was high-expressed but PRKCB was low-expressed in STAD cells, and silencing IL16 suppressed the invasion and migration of STAD cells.

Conclusions

Overall, we identified and validated 8 robust genes related to E. coli applying bioinformatics and machine learning algorithms, providing theoretical foundations for the relationship between E. coli-related dysbiosis and gastrointestinal tumors.

A deep learning method to integrate extracelluar miRNA with mRNA for cancer studies

MOTIVATION

Extracellular miRNAs (exmiRs) and intracellular mRNAs both can serve as promising biomarkers and therapeutic targets for various diseases. However, exmiR expression data is often noisy, and obtaining intracellular mRNA expression data usually involves intrusive procedures. To gain valuable insights into disease mechanisms, it is thus essential to improve the quality of exmiR expression data and develop noninvasive methods for assessing intracellular mRNA expression.

RESULTS

We developed CrossPred, a deep-learning multi-encoder model for the cross-prediction of exmiRs and mRNAs. Utilizing contrastive learning, we created a shared embedding space to integrate exmiRs and mRNAs. This shared embedding was then used to predict intracellular mRNA expression from noisy exmiR data and to predict exmiR expression from intracellular mRNA data. We evaluated CrossPred on three types of cancers and assessed its effectiveness in predicting the expression levels of exmiRs and mRNAs. CrossPred outperformed the baseline encoder-decoder model, exmiR or mRNA-based models, and variational autoencoder models. Moreover, the integration of exmiR and mRNA data uncovered important exmiRs and mRNAs associated with cancer. Our study offers new insights into the bidirectional relationship between mRNAs and exmiRs.

AVAILABILITY AND IMPLEMENTATION

The datasets and tool are available at https://doi.org/10.5281/zenodo.13891508.

Immunological Strategies in Gastric Cancer: How Toll-like Receptors 2, -3, -4, and -9 on Monocytes and Dendritic Cells Depend on Patient Factors?

(1) Introduction: Toll-like receptors (TLRs) are key in immune response by recognizing pathogen-associated molecular patterns (PAMPs) and damage-associated molecular patterns (DAMPs). In gastric cancer (GC), TLR2, TLR3, TLR4, and TLR9 are crucial for modulating immune response and tumor progression. (2) Objective: This study aimed to assess the percentage of dendritic cells and monocytes expressing TLR2, TLR3, TLR4, and TLR9, along with the concentration of their soluble forms in the serum of GC patients compared to healthy volunteers. Factors such as disease stage, tumor type, age, and gender were also analyzed. (3) Materials and Methods: Blood samples from newly diagnosed GC patients and healthy controls were immunophenotyped using flow cytometry to assess TLR expression on dendritic cell subpopulations and monocytes. Serum-soluble TLRs were measured by ELISA. Statistical analysis considered clinical variables such as tumor type, stage, age, and gender. (4) Results: TLR expression was significantly higher in GC patients, except for TLR3 on classical monocytes. Soluble forms of all TLRs were elevated in GC patients, with significant differences based on disease stage but not tumor type, except for serum TLR2, TLR4, and TLR9. (5) Conclusions: Elevated TLR expression and soluble TLR levels in GC patients suggest a role in tumor pathogenesis and progression, offering potential biomarkers and therapeutic targets.

Gastric Cancer in the Era of Epigenetics

Gastric cancer (GC) remains a significant contributor to cancer-related mortality. Novel high-throughput techniques have enlightened the epigenetic mechanisms governing gene-expression regulation. Epigenetic characteristics contribute to molecular taxonomy and give rise to cancer-specific epigenetic patterns. Helicobacter pylori (Hp) infection has an impact on aberrant DNA methylation either through its pathogenic CagA protein or by inducing chronic inflammation. The hypomethylation of specific repetitive elements generates an epigenetic field effect early in tumorigenesis. Epstein-Barr virus (EBV) infection triggers DNA methylation by dysregulating DNA methyltransferases (DNMT) enzyme activity, while persistent Hp-EBV co-infection leads to aggressive tumor behavior. Distinct histone modifications are also responsible for oncogene upregulation and tumor-suppressor gene silencing in gastric carcinomas. While histone methylation and acetylation processes have been extensively studied, other less prevalent alterations contribute to the development and migration of gastric cancer via a complex network of interactions. Enzymes, such as Nicotinamide N-methyltransferase (NNMT), which is involved in tumor's metabolic reprogramming, interact with methyltransferases and modify gene expression. Non-coding RNA molecules, including long non-coding RNAs, circular RNAs, and miRNAs serve as epigenetic regulators contributing to GC development, metastasis, poor outcomes and therapy resistance. Serum RNA molecules hold the potential to serve as non-invasive biomarkers for diagnostic, prognostic or therapeutic applications. Gastric fluids represent a valuable source to identify potential biomarkers with diagnostic use in terms of liquid biopsy. Ongoing clinical trials are currently evaluating the efficacy of next-generation epigenetic drugs, displaying promising outcomes. Various approaches including multiple miRNA inhibitors or targeted nanoparticles carrying epigenetic drugs are being designed to enhance existing treatment efficacy and overcome treatment resistance.

CircGLIS3 promotes gastric cancer progression by regulating the miR-1343-3p/PGK1 pathway and inhibiting vimentin phosphorylation

BACKGROUND

Circular RNAs (circRNAs) have been proved to play crucial roles in the development of various cancers. However, the molecular mechanism of circGLIS3 involved in gastric cancer (GC) tumorigenesis has not been elucidated.

METHODS

The higher expression level of circGLIS3 was identified in GC through RNA sequencing and subsequent tissue verification using Quantitative real-time PCR (qRT-PCR). A series of functional experiments in vitro and in vivo were performed to evaluated the effects of circGLIS3 on tumor growth and metastasis in GC. The interaction and regulation of circGLIS3/miR-1343-3p/PGK1 axis was confirmed by RNA pulldown, western blot, and rescue experiments. RIP and western blot were performed to demonstrate the role of circGLIS3 in regulating phosphorylation of VIMENTIN. We then used qRT-PCR and co culture system to trace circGLIS3 transmission via exosomal communication and identify the effect of exosomal circGLIS3 on gastric cancer and macrophages. Finally, RIP experiments were used to determine that EIF4A3 regulates circGLIS3 expression.

RESULTS

CircGLIS3(hsa_circ_0002874) was significantly upregulated in GC tissues and high circGLIS3 expression was associated with advanced TNM stage and lymph node metastasis in GC patients. We discovered that overexpression of circGLIS3 promoted GC cell proliferation, migration, invasion in vitro and in vivo, while suppression of circGLIS3 exhibited the opposite effect. Mechanistically, circGLIS3 could sponge miR-1343-3p and up-regulate the expression of PGK1 to promote GC tumorigenesis. We also found that circGLIS3 reduced the phosphorylation of VIMENTIN at ser 83 site by binding with VIMENTIN. Moreover, it was proven that exosomal circGLIS3 could promote gastric cancer metastasis and the M2 type polarization of macrophages. In the final step, the mechanism of EIF4A3 regulating the generation of circGLIS3 was determined.

CONCLUSION

Our findings demonstrate that circGLIS3 promotes GC progression through sponging miR-1343-3p and regulating VIMENTIN phosphorylation. CircGLIS3 is a potential therapeutic target for GC patients.

A Mini-review on Helicobacter pylori with Gastric Cancer and Available Treatments

Helicobacter pylori (H. pylori) is the most thoroughly researched etiological component for stomach inflammation and malignancies. Even though there are conventional recommendations and treatment regimens for eradicating H. pylori, failure rates continue to climb. Antibiotic resistance contributes significantly to misdiagnoses, false positive results, and clinical failures, all of which raise the chance of infection recurrence. This review aims to explore the molecular mechanisms underlying drug resistance in H. pylori and discuss novel approaches for detecting genotypic resistance. Modulation of drug uptake/ efflux, biofilm, and coccoid development. Newer genome sequencing approaches capable of detecting H. pylori genotypic resistance are presented. Prolonged infection in the stomach causes major problems such as gastric cancer. The review discusses how H. pylori causes stomach cancer, recent biomarkers such as miRNAs, molecular pathways in the development of gastric cancer, and diagnostic methods and clinical trials for the disease. Efforts have been made to summarize the recent advancements made toward early diagnosis and novel therapeutic approaches for H. pylori-induced gastric cancer.

Novel biomarkers for early detection of gastric cancer

Gastric cancer (GC) remains a leading cause of cancer-related death worldwide. Less than half of GC cases are diagnosed at an advanced stage due to its lack of early symptoms. GC is a heterogeneous disease associated with a number of genetic and somatic mutations. Early detection and effective monitoring of tumor progression are essential for reducing GC disease burden and mortality. The current widespread use of semi-invasive endoscopic methods and radiologic approaches has increased the number of treatable cancers: However, these approaches are invasive, costly, and time-consuming. Thus, novel molecular noninvasive tests that detect GC alterations seem to be more sensitive and specific compared to the current methods. Recent technological advances have enabled the detection of blood-based biomarkers that could be used as diagnostic indicators and for monitoring postsurgical minimal residual disease. These biomarkers include circulating DNA, RNA, extracellular vesicles, and proteins, and their clinical applications are currently being investigated. The identification of ideal diagnostic markers for GC that have high sensitivity and specificity would improve survival rates and contribute to the advancement of precision medicine. This review provides an overview of current topics regarding the novel, recently developed diagnostic markers for GC.

Machine learning algorithms reveal potential miRNAs biomarkers in gastric cancer

Gastric cancer is the high mortality rate cancers globally, and the current survival rate is 30% even with the use of combination therapies. Recently, mounting evidence indicates the potential role of miRNAs in the diagnosis and assessing the prognosis of cancers. In the state-of-art research in cancer, machine-learning (ML) has gained increasing attention to find clinically useful biomarkers. The present study aimed to identify potential diagnostic and prognostic miRNAs in GC with the application of ML. Using the TCGA database and ML algorithms such as Support Vector Machine (SVM), Random Forest, k-NN, etc., a panel of 29 was obtained. Among the ML algorithms, SVM was chosen (AUC:88.5%, Accuracy:93% in GC). To find common molecular mechanisms of the miRNAs, their common gene targets were predicted using online databases such as miRWalk, miRDB, and Targetscan. Functional and enrichment analyzes were performed using Gene Ontology (GO) and Kyoto Database of Genes and Genomes (KEGG), as well as identification of protein-protein interactions (PPI) using the STRING database. Pathway analysis of the target genes revealed the involvement of several cancer-related pathways including miRNA mediated inhibition of translation, regulation of gene expression by genetic imprinting, and the Wnt signaling pathway. Survival and ROC curve analysis showed that the expression levels of hsa-miR-21, hsa-miR-133a, hsa-miR-146b, and hsa-miR-29c were associated with higher mortality and potentially earlier detection of GC patients. A panel of dysregulated miRNAs that may serve as reliable biomarkers for gastric cancer were identified using machine learning, which represents a powerful tool in biomarker identification.

Overview of MicroRNAs as Diagnostic and Prognostic Biomarkers for High-Incidence Cancers in 2021

MicroRNAs (miRNAs) are small non-coding RNAs (ncRNAs) about 22 nucleotides in size, which play an important role in gene regulation and are involved in almost all major cellular physiological processes. In recent years, the abnormal expression of miRNAs has been shown to be associated with human diseases including cancer. In the past ten years, the link between miRNAs and various cancers has been extensively studied, and the abnormal expression of miRNAs has been reported in various malignant tumors, such as lung cancer, gastric cancer, colorectal cancer, liver cancer, breast cancer, and prostate cancer. Due to the high malignancy grade of these cancers, it is more necessary to develop the related diagnostic and prognostic methods. According to the study of miRNAs, many potential cancer biomarkers have been proposed for the diagnosis and prognosis of diseases, especially cancer, thus providing a new theoretical basis and perspective for cancer screening. The use of miRNAs as biomarkers for diagnosis or prognosis of cancer has the advantages of being less invasive to patients, with better accuracy and lower price. In view of the important clinical significance of miRNAs in human cancer research, this article reviewed the research status of miRNAs in the above-mentioned cancers in 2021, especially in terms of diagnosis and prognosis, and provided some new perspectives and theoretical basis for the diagnosis and treatment of cancers.

Screening of Serum miRNAs as Diagnostic Biomarkers for Lung Cancer Using the Minimal-Redundancy-Maximal-Relevance Algorithm and Random Forest Classifier Based on a Public Database

BACKGROUND

Lung cancer is one of the deadliest cancers, early diagnosis of which can efficiently enhance patient's survival. We aimed to screening out the serum miRNAs as diagnostic biomarkers for patients with lung cancer.

METHODS

A total of 416 remarkably differentially expressed miRNAs were acquired using the limma package, and next feature ranking was derived by the minimal-redundancy-maximal-relevance method. An incremental feature selection algorithm of a random forest (RF) classifier was utilized to choose the top 5 miRNA combination with the optimum predictive performance. The performance of the RF classifier of top 5 miRNAs was analyzed using the receiver operator characteristic (ROC) curve. Afterward, the classification effect of the 5-miRNA combination was validated through principal component analysis and hierarchical clustering analysis. Analysis of top 5 miRNA expressions between lung cancer patients and normal people was performed based on GSE137140 dataset, and their expression was validated by qPCR. The hierarchical clustering analysis was used to analyze the similarity of 5 miRNAs expression profiles. ROC analysis was undertaken on each miRNA.

RESULTS

We acquired top 5 miRNAs finally, with the Matthews correlation coefficient value as 0.988 and the area under the curve (AUC) value as 0.996. The 5 feature miRNAs were capable of distinguishing most cancer patients and normal people. Furthermore, except for the lowly expressed miR-6875-5p in lung cancer tissue, the other 4 miRNAs all expressed highly in cancer patients. Performance analysis revealed that their AUC values were 0.92, 0.96, 0.94, 0.95, and 0.93, respectively.

CONCLUSION

By and large, the 5 feature miRNAs screened here were anticipated to be effective biomarkers for lung cancer.