PESSA: A web tool for pathway enrichment score-based survival analysis in cancer.

The activation levels of biologically significant gene sets are emerging tumor molecular markers and play an irreplaceable role in the tumor research field; however, web-based tools for prognostic analyses using it as a tumor molecular marker remain scarce. We developed a web-based tool PESSA for survival analysis using gene set activation levels. All data analyses were implemented via R. Activation levels of The Molecular Signatures Database (MSigDB) gene sets were assessed using the single sample gene set enrichment analysis (ssGSEA) method based on data from the Gene Expression Omnibus (GEO), The Cancer Genome Atlas (TCGA), The European Genome-phenome Archive (EGA) and supplementary tables of articles. PESSA was used to perform median and optimal cut-off dichotomous grouping of ssGSEA scores for each dataset, relying on the survival and survminer packages for survival analysis and visualisation. PESSA is an open-access web tool for visualizing the results of tumor prognostic analyses using gene set activation levels. A total of 238 datasets from the GEO, TCGA, EGA, and supplementary tables of articles; covering 51 cancer types and 13 survival outcome types; and 13,434 tumor-related gene sets are obtained from MSigDB for pre-grouping. Users can obtain the results, including Kaplan-Meier analyses based on the median and optimal cut-off values and accompanying visualization plots and the Cox regression analyses of dichotomous and continuous variables, by selecting the gene set markers of interest. PESSA (https://smuonco.shinyapps.io/PESSA/ OR http://robinl-lab.com/PESSA) is a large-scale web-based tumor survival analysis tool covering a large amount of data that creatively uses predefined gene set activation levels as molecular markers of tumors.

Inflammatory corpuscle AIM2 facilitates macrophage foam cell formation by inhibiting cholesterol efflux protein ABCA1.

The inflammatory corpuscle recombinant absents in melanoma 2 (AIM2) and cholesterol efflux protein ATP binding cassette transporter A1(ABCA1) have been reported to play opposing roles in atherosclerosis (AS) plaques. However, the relationship between AIM2 and ABCA1 remains unclear. In this study, we explored the potential connection between AIM2 and ABCA1 in the modulation of AS by bioinformatic analysis combined with in vitro experiments. The GEO database was used to obtain AS transcriptional profiling data; screen differentially expressed genes (DEGs) and construct a weighted gene co-expression network analysis (WGCNA) to obtain AS-related modules. Phorbol myristate acetate (PMA) was used to induce macrophage modelling in THP-1 cells, and ox-LDL was used to induce macrophage foam cell formation. The experiment was divided into Negative Control (NC) group, Model Control (MC) group, AIM2 overexpression + ox-LDL (OE AIM2 + ox-LDL) group, and AIM2 short hairpin RNA + ox-LDL (sh AIM2 + ox-LDL) group. The intracellular cholesterol efflux rate was detected by scintillation counting; high-performance liquid chromatography (HPLC) was used to detect intracellular cholesterol levels; apoptosis levels were detected by TUNEL kit; levels of inflammatory markers (IL-1ß, IL-18, ROS, and GSH) were detected by ELISA kits; and levels of AIM2 and ABCA1 proteins were detected by Western blot. Bioinformatic analysis revealed that the turquoise module correlated most strongly with AS, and AIM2 and ABCA1 were co-expressed in the turquoise module with a trend towards negative correlation. In vitro experiments demonstrated that AIM2 inhibited macrophage cholesterol efflux, resulting in increased intracellular cholesterol levels and foam cell formation. Moreover, AIM2 had a synergistic effect with ox-LDL, exacerbating macrophage oxidative stress and inflammatory response. Silencing AIM2 ameliorated the above conditions. Furthermore, the protein expression levels of AIM2 and ABCA1 were consistent with the bioinformatic analysis, showing a negative correlation. AIM2 inhibits ABCA1 expression, causing abnormal cholesterol metabolism in macrophages and ultimately leading to foam cell formation. Inhibiting AIM2 may reverse this process. Overall, our study suggests that AIM2 is a reliable anti-inflammatory therapeutic target for AS. Inhibiting AIM2 expression may reduce foam cell formation and, consequently, inhibit the progression of AS plaques.

Integrated approach to generate artificial samples with low tumor fraction for somatic variant calling benchmarking.

BACKGROUND: High-throughput sequencing (HTS) has become the gold standard approach for variant analysis in cancer research. However, somatic variants may occur at low fractions due to contamination from normal cells or tumor heterogeneity; this poses a significant challenge for standard HTS analysis pipelines. The problem is exacerbated in scenarios with minimal tumor DNA, such as circulating tumor DNA in plasma. Assessing sensitivity and detection of HTS approaches in such cases is paramount, but time-consuming and expensive: specialized experimental protocols and a sufficient quantity of samples are required for processing and analysis. To overcome these limitations, we propose a new computational approach specifically designed for the generation of artificial datasets suitable for this task, simulating ultra-deep targeted sequencing data with low-fraction variants and demonstrating their effectiveness in benchmarking low-fraction variant calling. RESULTS: Our approach enables the generation of artificial raw reads that mimic real data without relying on pre-existing data by using NEAT, a fine-grained read simulator that generates artificial datasets using models learned from multiple different datasets. Then, it incorporates low-fraction variants to simulate somatic mutations in samples with minimal tumor DNA content. To prove the suitability of the created artificial datasets for low-fraction variant calling benchmarking, we used them as ground truth to evaluate the performance of widely-used variant calling algorithms: they allowed us to define tuned parameter values of major variant callers, considerably improving their detection of very low-fraction variants. CONCLUSIONS: Our findings highlight both the pivotal role of our approach in creating adequate artificial datasets with low tumor fraction, facilitating rapid prototyping and benchmarking of algorithms for such dataset type, as well as the important need of advancing low-fraction variant calling techniques.

A comparison of RNA-Seq data preprocessing pipelines for transcriptomic predictions across independent studies.

BACKGROUND: RNA sequencing combined with machine learning techniques has provided a modern approach to the molecular classification of cancer. Class predictors, reflecting the disease class, can be constructed for known tissue types using the gene expression measurements extracted from cancer patients. One challenge of current cancer predictors is that they often have suboptimal performance estimates when integrating molecular datasets generated from different labs. Often, the quality of the data is variable, procured differently, and contains unwanted noise hampering the ability of a predictive model to extract useful information. Data preprocessing methods can be applied in attempts to reduce these systematic variations and harmonize the datasets before they are used to build a machine learning model for resolving tissue of origins. RESULTS: We aimed to investigate the impact of data preprocessing steps-focusing on normalization, batch effect correction, and data scaling-through trial and comparison. Our goal was to improve the cross-study predictions of tissue of origin for common cancers on large-scale RNA-Seq datasets derived from thousands of patients and over a dozen tumor types. The results showed that the choice of data preprocessing operations affected the performance of the associated classifier models constructed for tissue of origin predictions in cancer. CONCLUSION: By using TCGA as a training set and applying data preprocessing methods, we demonstrated that batch effect correction improved performance measured by weighted F1-score in resolving tissue of origin against an independent GTEx test dataset. On the other hand, the use of data preprocessing operations worsened classification performance when the independent test dataset was aggregated from separate studies in ICGC and GEO. Therefore, based on our findings with these publicly available large-scale RNA-Seq datasets, the application of data preprocessing techniques to a machine learning pipeline is not always appropriate.

CDCA3 is a potential biomarker for glioma malignancy and targeted therapy.

CDCA3, a cell cycle regulator gene that plays a catalytic role in many tumors, was initially identified as a regulator of cell cycle progression, specifically facilitating the transition from the G2 phase to mitosis. However, its role in glioma remains unknown. In this study, bioinformatics analyses (TCGA, CGGA, Rembrandt) shed light on the upregulation and prognostic value of CDCA3 in gliomas. It can also be included in a column chart as a parameter predicting 3- and 5-year survival risk (C index = 0.86). According to Gene Set Enrichment Analysis and gene ontology analysis, the biological processes of CDCA3 are mainly concentrated in the biological activities related to cell cycle such as DNA replication and nuclear division. CDCA3 is closely associated with many classic glioma biomarkers (CDK4, CDK6), and inhibitors of CDK4 and CDK6 have been shown to be effective in tumor therapy. We have demonstrated that high expression of CDCA3 indicates a higher malignancy and poorer prognosis in gliomas.

THOC6 is a novel biomarker of glioma and a target of anti-glioma drugs: An analysis based on bioinformatics and molecular docking.

Glioma is a typical malignant tumor of the nervous system. It is of great significance to identify new biomarkers for accurate diagnosis of glioma. In this context, THOC6 has been studied as a highly diagnostic prognostic biomarker, which contributes to improve the dilemma in diagnosing gliomas. We used online databases and a variety of statistical methods, such as Wilcoxon rank sum test, Dunn test and t test. We analyzed the mutation, location and expression profile of THOC6, revealing the network of THOC6 interaction with disease. Wilcoxon rank sum test showed that THOC6 is highly expressed in gliomas (P < 0.001). Dunn test, Wilcoxon rank sum test and t test showed that THOC6 expression was correlated with multiple clinical features. Logistic regression analysis further confirmed that THOC6 gene expression was a categorical dependent variable related to clinical features of poor prognosis. Kaplan-Meier survival analysis showed that the overall survival (OS) of glioma patients with high expression of THOC6 was poor (P < 0.001). Both univariate (P < 0.001) and multivariate (P = 0.04) Cox analysis confirmed that THOC6 gene expression was an independent risk factor for OS in patients with glioma. ROC curve analysis showed that THOC6 had a high diagnostic value in glioma (AUC = 0.915). Based on this, we constructed a nomogram to predict patient survival. Enrichment analysis showed that THOC6 expression was associated with multiple signal pathways. Immuno-infiltration analysis showed that the expression of THOC6 in glioma was closely related to the infiltration level of multiple immune cells. Molecular docking results showed that THOC6 might be the target of anti-glioma drugs. THOC6 is a novel diagnostic factor and prognostic biomarker of glioma.

Identification of PANoptosis-related subtypes, construction of a prognosis signature, and tumor microenvironment landscape of hepatocellular carcinoma using bioinformatic analysis and experimental verification.

Background: Hepatocellular carcinoma (HCC) is one of the most lethal malignancies worldwide. PANoptosis is a recently unveiled programmed cell death pathway, Nonetheless, the precise implications of PANoptosis within the context of HCC remain incompletely elucidated. Methods: We conducted a comprehensive bioinformatics analysis to evaluate both the expression and mutation patterns of PANoptosis-related genes (PRGs). We categorized HCC into two clusters and identified differentially expressed PANoptosis-related genes (DEPRGs). Next, a PANoptosis risk model was constructed using LASSO and multivariate Cox regression analyses. The relationship between PRGs, risk genes, the risk model, and the immune microenvironment was studies. In addition, drug sensitivity between high- and low-risk groups was examined. The expression profiles of these four risk genes were elucidate by qRT-PCR or immunohistochemical (IHC). Furthermore, the effect of CTSC knock down on HCC cell behavior was verified using in vitro experiments. Results: We constructed a prognostic signature of four DEPRGs (CTSC, CDCA8, G6PD, and CXCL9). Receiver operating characteristic curve analyses underscored the superior prognostic capacity of this signature in assessing the outcomes of HCC patients. Subsequently, patients were stratified based on their risk scores, which revealed that the low-risk group had better prognosis than those in the high-risk group. High-risk group displayed a lower Stromal Score, Immune Score, ESTIMATE score, and higher cancer stem cell content, tumor mutation burden (TMB) values. Furthermore, a correlation was noted between the risk model and the sensitivity to 56 chemotherapeutic agents, as well as immunotherapy efficacy, in patient with. These findings provide valuable guidance for personalized clinical treatment strategies. The qRT-PCR analysis revealed that upregulated expression of CTSC, CDCA8, and G6PD, whereas downregulated expression of CXCL9 in HCC compared with adjacent tumor tissue and normal liver cell lines. The knockdown of CTSC significantly reduced both HCC cell proliferation and migration. Conclusion: Our study underscores the promise of PANoptosis-based molecular clustering and prognostic signatures in predicting patient survival and discerning the intricacies of the tumor microenvironment within the context of HCC. These insights hold the potential to advance our comprehension of the therapeutic contribution of PANoptosis plays in HCC and pave the way for generating more efficacious treatment strategies.

Integration of multi-omics data for survival prediction of lung adenocarcinoma.

BACKGROUND AND OBJECTIVE: The morbidity of lung adenocarcinoma (LUAD) has been increasing year by year and the prognosis is poor. This has prompted researchers to study the survival of LUAD patients to ensure that patients can be cured in time or survive after appropriate treatment. There is still no fully valid model that can be applied to clinical practice. METHODS: We introduced struc2vec-based multi-omics data integration (SBMOI), which could integrate gene expression, somatic mutations and clinical data to construct mutation gene vectors representing LUAD patient features. Based on the patient features, the random survival forest (RSF) model was used to predict the long- and short-term survival of LUAD patients. To further demonstrate the superiority of SBMOI, we simultaneously replaced scale-free gene co-expression network (FCN) with a protein-protein interaction (PPI) network and a significant co-expression network (SCN) to compare accuracy in predicting LUAD patient survival under the same conditions. RESULTS: Our results suggested that compared with SCN and PPI network, the FCN based SBMOI combined with RSF model had better performance in long- and short-term survival prediction tasks for LUAD patients. The AUC of 1-year, 5-year, and 10-year survival in the validation dataset were 0.791, 0.825, and 0.917, respectively. CONCLUSIONS: This study provided a powerful network-based method to multi-omics data integration. SBMOI combined with RSF successfully predicted long- and short-term survival of LUAD patients, especially with high accuracy on long-term survival. Besides, SBMOI algorithm has the potential to combine with other machine learning models to complete clustering or stratificational tasks, and being applied to other diseases.

A multi-omics approach for biomarker discovery in neuroblastoma: a network-based framework.

Neuroblastoma (NB) is one of the leading causes of cancer-associated death in children. MYCN amplification is a prominent genetic marker for NB, and its targeting to halt NB progression is difficult to achieve. Therefore, an in-depth understanding of the molecular interactome of NB is needed to improve treatment outcomes. Analysis of NB multi-omics unravels valuable insight into the interplay between MYCN transcriptional and miRNA post-transcriptional modulation. Moreover, it aids in the identification of various miRNAs that participate in NB development and progression. This study proposes an integrated computational framework with three levels of high-throughput NB data (mRNA-seq, miRNA-seq, and methylation array). Similarity Network Fusion (SNF) and ranked SNF methods were utilized to identify essential genes and miRNAs. The specified genes included both miRNA-target genes and transcription factors (TFs). The interactions between TFs and miRNAs and between miRNAs and their target genes were retrieved where a regulatory network was developed. Finally, an interaction network-based analysis was performed to identify candidate biomarkers. The candidate biomarkers were further analyzed for their potential use in prognosis and diagnosis. The candidate biomarkers included three TFs and seven miRNAs. Four biomarkers have been previously studied and tested in NB, while the remaining identified biomarkers have known roles in other types of cancer. Although the specific molecular role is yet to be addressed, most identified biomarkers possess evidence of involvement in NB tumorigenesis. Analyzing cellular interactome to identify potential biomarkers is a promising approach that can contribute to optimizing efficient therapeutic regimens to target NB vulnerabilities.

Involvement of S100A6/S100A11 in T-Cell Immune Regulatory in HCC Revealed by Single Cell RNA-seq.

Background: Immunotherapy plays a significant role in the treatment of hepatocellular carcinoma (HCC). Members of the S100 protein family (S100s) have been widely implicated in the pathogenesis and progression of tumors. However, the exact mechanism by which S100s contribute to tumor immunity remains unclear. Methods: To explore the role of S100s in HCC immune cells, we collected and comparatively analyzed single-cell RNA sequencing (scRNA-seq) data of HCC and hepatitis B virus-associated HCC. By mapping cell classification and searching for S100s binding targets and downstream targets. Results: S100A6/S100A11 was differentially expressed in tumor T cells and involved in the nuclear factor (NF) κB pathway. Further investigation of the TCGA dataset revealed that patients with low S100A6/S100A11 expression had a better prognosis. Temporal cell trajectory analysis showed that the activation of the NF-κB pathway is at a critical stage and has an important impact on the tumor microenvironment. Conclusion: Our study revealed that S100A6/S100A11 could be involved in regulating the differentiation and cellular activity of T-cell subpopulations in HCC, and its low expression was positively correlated with prognosis. It may provide a new direction for immunotherapy of HCC and a theoretical basis for future clinical applications.

CPADS: a web tool for comprehensive pancancer analysis of drug sensitivity.

Drug therapy is vital in cancer treatment. Accurate analysis of drug sensitivity for specific cancers can guide healthcare professionals in prescribing drugs, leading to improved patient survival and quality of life. However, there is a lack of web-based tools that offer comprehensive visualization and analysis of pancancer drug sensitivity. We gathered cancer drug sensitivity data from publicly available databases (GEO, TCGA and GDSC) and developed a web tool called Comprehensive Pancancer Analysis of Drug Sensitivity (CPADS) using Shiny. CPADS currently includes transcriptomic data from over 29 000 samples, encompassing 44 types of cancer, 288 drugs and more than 9000 gene perturbations. It allows easy execution of various analyses related to cancer drug sensitivity. With its large sample size and diverse drug range, CPADS offers a range of analysis methods, such as differential gene expression, gene correlation, pathway analysis, drug analysis and gene perturbation analysis. Additionally, it provides several visualization approaches. CPADS significantly aids physicians and researchers in exploring primary and secondary drug resistance at both gene and pathway levels. The integration of drug resistance and gene perturbation data also presents novel perspectives for identifying pivotal genes influencing drug resistance. Access CPADS at https://smuonco.shinyapps.io/CPADS/ or https://robinl-lab.com/CPADS.

Selenium May Be Involved in Esophageal Squamous Cancer Prevention by Affecting GPx3 and FABP1 Expression: A Case-Control Study Based on Bioinformatic Analysis.

The role of selenium in the developmental process of esophageal cancer (EC) requires further investigation. To explore the relationship between selenium-related factors and EC through bioinformatic analysis, a case-control study was conducted to verify the results. Utilizing the GEPIA and TCGA databases, we delineated the differential expression of glutathione peroxidase 3 (GPx3) in EC and normal tissues, identified differentially expressed genes (DEGs), and a performed visualization analysis. Additionally, 100 pairs of dietary and plasma samples from esophageal precancerous lesions (EPLs) of esophageal squamous cancer (ESCC) cases and healthy controls from Huai'an district, Jiangsu, were screened. The levels of dietary selenium, plasma selenium, and related enzymes were analyzed using inductively coupled plasma mass spectrometry (ICP-MS) or ELISA kits. The results showed lower GPx3 expression in tumor tissues compared to normal tissues. Further analysis revealed that DEGs were mainly involved in the fat digestion and absorption pathway, and the core protein fatty acid binding protein 1 (FABP1) was significantly upregulated and negatively correlated with GPx3 expression. Our case-control study found that selenium itself was not associated with EPLs risk. However, both the decreased concentration of GPx3 and the increase in FABP1 were positively correlated with the EPLs risk (p for trend = 0.035 and 0.046, respectively). The different expressions of GPx3 and FABP1 reflect the potential of selenium for preventing ESCC at the EPLs stage. GPx3 may affect myocardial infarction through FABP1, which remains to be further studied.

SpatialCells: automated profiling of tumor microenvironments with spatially resolved multiplexed single-cell data.

Cancer is a complex cellular ecosystem where malignant cells coexist and interact with immune, stromal and other cells within the tumor microenvironment (TME). Recent technological advancements in spatially resolved multiplexed imaging at single-cell resolution have led to the generation of large-scale and high-dimensional datasets from biological specimens. This underscores the necessity for automated methodologies that can effectively characterize molecular, cellular and spatial properties of TMEs for various malignancies. This study introduces SpatialCells, an open-source software package designed for region-based exploratory analysis and comprehensive characterization of TMEs using multiplexed single-cell data. The source code and tutorials are available at https://semenovlab.github.io/SpatialCells. SpatialCells efficiently streamlines the automated extraction of features from multiplexed single-cell data and can process samples containing millions of cells. Thus, SpatialCells facilitates subsequent association analyses and machine learning predictions, making it an essential tool in advancing our understanding of tumor growth, invasion and metastasis.

Human pan-cancer analysis of the predictive biomarker for the CDKN3.

BACKGROUND: Cell cycle protein-dependent kinase inhibitor protein 3 (CDKN3), as a member of the protein kinase family, has been demonstrated to exhibit oncogenic properties in several tumors. However, there are no pan-carcinogenic analyses for CDKN3. METHODS: Using bioinformatics tools such as The Cancer Genome Atlas (TCGA) and the UCSC Xena database, a comprehensive pan-cancer analysis of CDKN3 was conducted. The inverstigation encompassed the examination of CDKN3 function actoss 33 different kinds of tumors, as well as the exploration of gene expressions, survival prognosis status, clinical significance, DNA methylation, immune infiltration, and associated signal pathways. RESULTS: CDKN3 was significantly upregulated in most of tumors and correlated with overall survival (OS) of patients. Methylation levels of CDKN3 differed significantly between tumors and normal tissues. In addition, infiltration of CD4 + T cells, cancer-associated fibroblasts, macrophages, and endothelial cells were associated with CDKN3 expression in various tumors. Mechanistically, CDKN3 was associated with P53, PI3K-AKT, cell cycle checkpoints, mitotic spindle checkpoint, and chromosome maintenance. CONCLUSION: Our pan-cancer analysis conducted in the study provides a comprehensive understanding of the involvement of CDKN3 gene in tumorigenesis. The findings suggest that targeting CDKN3 may potentially lead to novel therapeutic strategies for the treatment of tumors.

The Deep Proteomics Approach Identified Extracellular Vesicular Proteins Correlated to Extracellular Matrix in Type One and Two Endometrial Cancer.

Among gynecological cancers, endometrial cancer is the most common in developed countries. Extracellular vesicles (EVs) are cell-derived membrane-surrounded vesicles that contain proteins involved in immune response and apoptosis. A deep proteomic approach can help to identify dysregulated extracellular matrix (ECM) proteins in EVs correlated to key pathways for tumor development. In this study, we used a proteomics approach correlating the two acquisitions-data-dependent acquisition (DDA) and data-independent acquisition (DIA)-on EVs from the conditioned medium of four cell lines identifying 428 ECM proteins. After protein quantification and statistical analysis, we found significant changes in the abundance (p < 0.05) of 67 proteins. Our bioinformatic analysis identified 26 pathways associated with the ECM. Western blotting analysis on 13 patients with type 1 and type 2 EC and 13 endometrial samples confirmed an altered abundance of MMP2. Our proteomics analysis identified the dysregulated ECM proteins involved in cancer growth. Our data can open the path to other studies for understanding the interaction among cancer cells and the rearrangement of the ECM.

Identification of candidate biomarkers for GBM based on WGCNA.

Glioblastoma multiforme (GBM), the most aggressive form of primary brain tumor, poses a considerable challenge in neuro-oncology. Despite advancements in therapeutic approaches, the prognosis for GBM patients remains bleak, primarily attributed to its inherent resistance to conventional treatments and a high recurrence rate. The primary goal of this study was to acquire molecular insights into GBM by constructing a gene co-expression network, aiming to identify and predict key genes and signaling pathways associated with this challenging condition. To investigate differentially expressed genes between various grades of Glioblastoma (GBM), we employed Weighted Gene Co-expression Network Analysis (WGCNA) methodology. Through this approach, we were able to identify modules with specific expression patterns in GBM. Next, genes from these modules were performed Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis using ClusterProfiler package. Our findings revealed a negative correlation between biological processes associated with neuronal development and functioning and GBM. Conversely, the processes related to the cell cycle, glomerular development, and ECM-receptor interaction exhibited a positive correlation with GBM. Subsequently, hub genes, including SYP, TYROBP, and ANXA5, were identified. This study offers a comprehensive overview of the existing research landscape on GBM, underscoring the challenges encountered by clinicians and researchers in devising effective therapeutic strategies.

Prediction of anticancer drug sensitivity using an interpretable model guided by deep learning.

BACKGROUND: The prediction of drug sensitivity plays a crucial role in improving the therapeutic effect of drugs. However, testing the effectiveness of drugs is challenging due to the complex mechanism of drug reactions and the lack of interpretability in most machine learning and deep learning methods. Therefore, it is imperative to establish an interpretable model that receives various cell line and drug feature data to learn drug response mechanisms and achieve stable predictions between available datasets. RESULTS: This study proposes a new and interpretable deep learning model, DrugGene, which integrates gene expression, gene mutation, gene copy number variation of cancer cells, and chemical characteristics of anticancer drugs to predict their sensitivity. This model comprises two different branches of neural networks, where the first involves a hierarchical structure of biological subsystems that uses the biological processes of human cells to form a visual neural network (VNN) and an interpretable deep neural network for human cancer cells. DrugGene receives genotype input from the cell line and detects changes in the subsystem states. We also employ a traditional artificial neural network (ANN) to capture the chemical structural features of drugs. DrugGene generates final drug response predictions by combining VNN and ANN and integrating their outputs into a fully connected layer. The experimental results using drug sensitivity data extracted from the Cancer Drug Sensitivity Genome Database and the Cancer Treatment Response Portal v2 reveal that the proposed model is better than existing prediction methods. Therefore, our model achieves higher accuracy, learns the reaction mechanisms between anticancer drugs and cell lines from various features, and interprets the model's predicted results. CONCLUSIONS: Our method utilizes biological pathways to construct neural networks, which can use genotypes to monitor changes in the state of network subsystems, thereby interpreting the prediction results in the model and achieving satisfactory prediction accuracy. This will help explore new directions in cancer treatment. More available code resources can be downloaded for free from GitHub ( https://github.com/pangweixiong/DrugGene ).

Contrastive learning for enhancing feature extraction in anticancer peptides.

Cancer, recognized as a primary cause of death worldwide, has profound health implications and incurs a substantial social burden. Numerous efforts have been made to develop cancer treatments, among which anticancer peptides (ACPs) are garnering recognition for their potential applications. While ACP screening is time-consuming and costly, in silico prediction tools provide a way to overcome these challenges. Herein, we present a deep learning model designed to screen ACPs using peptide sequences only. A contrastive learning technique was applied to enhance model performance, yielding better results than a model trained solely on binary classification loss. Furthermore, two independent encoders were employed as a replacement for data augmentation, a technique commonly used in contrastive learning. Our model achieved superior performance on five of six benchmark datasets against previous state-of-the-art models. As prediction tools advance, the potential in peptide-based cancer therapeutics increases, promising a brighter future for oncology research and patient care.

Identification of Breast Cancer Subtypes Based on Endoplasmic Reticulum Stress-Related Genes and Analysis of Prognosis and Immune Microenvironment in Breast Cancer Patients.

Introduction: Endoplasmic reticulum stress (ERS) was a response to the accumulation of unfolded proteins and plays a crucial role in the development of tumors, including processes such as tumor cell invasion, metastasis, and immune evasion. However, the specific regulatory mechanisms of ERS in breast cancer (BC) remain unclear. Methods: In this study, we analyzed RNA sequencing data from The Cancer Genome Atlas (TCGA) for breast cancer and identified 8 core genes associated with ERS: ELOVL2, IFNG, MAP2K6, MZB1, PCSK6, PCSK9, IGF2BP1, and POP1. We evaluated their individual expression, independent diagnostic, and prognostic values in breast cancer patients. A multifactorial Cox analysis established a risk prognostic model, validated with an external dataset. Additionally, we conducted a comprehensive assessment of immune infiltration and drug sensitivity for these genes. Results: The results indicate that these eight core genes play a crucial role in regulating the immune microenvironment of breast cancer (BRCA) patients. Meanwhile, an independent diagnostic model based on the expression of these eight genes shows limited independent diagnostic value, and its independent prognostic value is unsatisfactory, with the time ROC AUC values generally below 0.5. According to the results of logistic regression neural networks and risk prognosis models, when these eight genes interact synergistically, they can serve as excellent biomarkers for the diagnosis and prognosis of breast cancer patients. Furthermore, the research findings have been confirmed through qPCR experiments and validation. Conclusion: In conclusion, we explored the mechanisms of ERS in BRCA patients and identified 8 outstanding biomolecular diagnostic markers and prognostic indicators. The research results were double-validated using the GEO database and qPCR.

mRNA markers for survival prediction in glioblastoma multiforme patients: a systematic review with bioinformatic analyses.

BACKGROUND: Glioblastoma multiforme (GBM) is a type of fast-growing brain glioma associated with a very poor prognosis. This study aims to identify key genes whose expression is associated with the overall survival (OS) in patients with GBM. METHODS: A systematic review was performed using PubMed, Scopus, Cochrane, and Web of Science up to Journey 2024. Two researchers independently extracted the data and assessed the study quality according to the New Castle Ottawa scale (NOS). The genes whose expression was found to be associated with survival were identified and considered in a subsequent bioinformatic study. The products of these genes were also analyzed considering protein-protein interaction (PPI) relationship analysis using STRING. Additionally, the most important genes associated with GBM patients' survival were also identified using the Cytoscape 3.9.0 software. For final validation, GEPIA and CGGA (mRNAseq_325 and mRNAseq_693) databases were used to conduct OS analyses. Gene set enrichment analysis was performed with GO Biological Process 2023. RESULTS: From an initial search of 4104 articles, 255 studies were included from 24 countries. Studies described 613 unique genes whose mRNAs were significantly associated with OS in GBM patients, of which 107 were described in 2 or more studies. Based on the NOS, 131 studies were of high quality, while 124 were considered as low-quality studies. According to the PPI network, 31 key target genes were identified. Pathway analysis revealed five hub genes (IL6, NOTCH1, TGFB1, EGFR, and KDR). However, in the validation study, only, the FN1 gene was significant in three cohorts. CONCLUSION: We successfully identified the most important 31 genes whose products may be considered as potential prognosis biomarkers as well as candidate target genes for innovative therapy of GBM tumors.