Multimodal AI/ML for discovering novel biomarkers and predicting disease using multi-omics profiles of patients with cardiovascular diseases

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Nov. 3, 2024

Cardiovascular diseases (CVDs) are complex, multifactorial conditions that require personalized assessment and treatment. Advancements in multi-omics technologies, namely RNA sequencing whole-genome sequencing, have provided translational researchers with a comprehensive view of the human genome. The efficient synthesis analysis this data through integrated approach characterizes genetic variants alongside expression patterns linked to emerging phenotypes, can reveal novel biomarkers enable segmentation patient populations based on risk factors. In study, we present cutting-edge methodology rooted integration traditional bioinformatics, classical statistics, multimodal machine learning techniques. Our has potential uncover intricate mechanisms underlying CVD, enabling patient-specific response profiling. We sourced transcriptomic single nucleotide polymorphisms (SNPs) from both CVD patients healthy controls. By integrating these datasets clinical demographic information, generated profiles. Utilizing robust feature selection approach, identified signature 27 features SNPs effective predictors CVD. Differential analysis, combined minimum redundancy maximum relevance selection, highlighted explain disease phenotype. This prioritizes biological efficiency learning. employed Combination Annotation Dependent Depletion scores allele frequencies identify pathogenic characteristics patients. Classification models trained demonstrated high-accuracy predictions for best performing was an XGBoost classifier optimized via Bayesian hyperparameter tuning, which able correctly classify all our test dataset. Using SHapley Additive exPlanations, created assessments patients, offering further contextualization setting. Across cohort, RPL36AP37 HBA1 were scored as most important predicting CVDs. A literature review revealed substantial portion diagnostic previously been associated framework propose study is unbiased generalizable other disorders.

Language: Английский

---

Deep visual multi-omics profiling reveals mechanisms that underly cancer cell differentiation and aggressiveness in clear cell renal cell carcinoma

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 4, 2025

Summary Clear cell renal carcinoma (ccRCC) exhibits significant intra-tumoral heterogeneity (ITH) at both morphological and genetic levels, complicating treatment contributing to disease progression 1,2 . Among these, rhabdoid ccRCCs stand out as highly aggressive tumors distinguished by cells with unique features poor clinical outcomes 3 However, the correlation between distinct phenotypes, specific molecular alterations, their influence on tumor behavior remains poorly understood. Understanding link clinically relevant underpinnings will be critical for developing more effective treatments targeting key cellular subsets responsible progression. In this study, we integrated advanced AI-based image analysis single-cell isolation multi-omics profiling dissect of ccRCC cells. Using a novel digital pathology workflow, quantified low-grade, high-grade, morphologies in diagnostic images unprecedented precision. Subsequently, two sets 1,000 morphologically detailed mRNA protein expression analyses, revealed increasing dysregulation associating higher histopathological grades. Rhabdoid (grade 4) demonstrated profiles, including upregulated FOXM1-driven proliferation, disrupted cell-matrix interactions, enhanced immune evasion pathways. Despite high T-cell infiltration areas, identified rhabdoid-specific immunosuppressive network driven cytokines, IFN-beta, integrin signaling, likely exhaustion. These findings provide basis therapeutic strategies these pathways combination immunotherapy improve patients ccRCC. Key Points is characterized well-established but underlying aberrations remained elusive. By integrating single deep profiling, intricacies ccRCC, from targeted collection Our results demonstrate gene correlating grades Aggressive differentiation display they upregulate FOXM1-mediated ECM remodeling responses, suggesting new avenues enhancing ICI efficacy patients.

Language: Английский

---

Conformation-sensitive targeting of CD18 depletes M2-like tumor-associated macrophages resulting in inhibition of solid tumor progression

Journal for ImmunoTherapy of Cancer, Journal Year: 2025, Volume and Issue: 13(4), P. e011422 - e011422

Published: April 1, 2025

Background Tumor-associated macrophages (TAMs) primarily exist in the M2-like phenotype tumor microenvironment (TME). M2-TAMs contribute to progression by establishing an immunosuppressive environment. However, TAM targeting is hindered, mainly owing a lack of specific biomarkers for M2-TAMs. Previously, we demonstrated that novel peptide drug conjugate (TB511) consisting TAM-binding and apoptosis-promoting targets This was achieved through M2-TAM targeting, although target mechanism action remained elusive. Herein, elucidate anticancer efficacy TB511 identifying new proteins preferentially bind clarifying apoptosis-inducing these cells. Methods We investigated binding site using LC-MS/MS analyses, surface plasmon resonance peptide–protein interaction 3D modeling. Activated CD18 expression M2 TAMs assessed Quantibrite PE beads PBMCs. The tested colorectal cancer (CRC) non-small cell lung (NSCLC) mouse model. immunotherapeutic effect spatial transcriptomics human pancreatic ductal adenocarcinoma (PDAC) Results highly expressed tissues significantly higher than other immune showed high affinity among membrane appeared cysteine-rich domain activated form. Moreover, specifically induced apoptosis TAMs, but its ability inhibited blockade or knockout In humanized models solid tumors such as CRC, NSCLC, PDAC, suppressed growth via enhancing presence CD8 + T cells TME. Conclusions Collectively, our findings suggest holds promise protein therapy, shows potential therapeutic agent treatment.

Language: Английский

---

In Silico Approach to Molecular Profiling of the Transition from Ovarian Epithelial Cells to Low-Grade Serous Ovarian Tumors for Targeted Therapeutic Insights

Current Issues in Molecular Biology, Journal Year: 2024, Volume and Issue: 46(3), P. 1777 - 1798

Published: Feb. 26, 2024

This paper aims to elucidate the differentially coexpressed genes, their potential mechanisms, and possible drug targets in low-grade invasive serous ovarian carcinoma (LGSC) terms of biologic continuity normal, borderline, malignant LGSC. We performed a bioinformatics analysis, integrating datasets generated using GPL570 platform from different studies GEO database identify changes this transition, gene expression, targets, relationships with tumor microenvironmental characteristics. In transition epithelial cells FGFR3 “Estrogen Response Late” pathway, ITGB2 “Cell Adhesion Molecule”, CD74 “Regulation Cell Migration”, IGF1 “Xenobiotic Metabolism” pathway were upregulated borderline The ERBB4 “Proteoglycan Cancer”, AR “Pathways Cancer” Early” pathways, addition, SPP1 genes correlated macrophage infiltration LGSC group. research provides valuable framework for development personalized therapeutic approaches context LGSC, aim improving patient outcomes quality life. Furthermore, main goal current study is preliminary designed generate silico inferences, it also important note that subsequent vitro vivo will be necessary confirm results before considering these as fully reliable.

Language: Английский

---

ITGB2 related to immune cell infiltration as a potential therapeutic target of inflammatory bowel disease using bioinformatics and functional research

Journal of Cellular and Molecular Medicine, Journal Year: 2024, Volume and Issue: 28(15)

Published: Aug. 1, 2024

Abstract Inflammatory bowel disease (IBD) is a chronic systemic inflammatory condition regarded as major risk factor for colitis‐associated cancer. However, the underlying mechanisms of IBD remain unclear. First, five GSE data sets available in GEO were used to perform ‘batch correction’ and Robust Rank Aggregation (RRA) identify differentially expressed genes (DEGs). Candidate molecules identified using CytoHubba, their diagnostic effectiveness was predicted. The CIBERSORT algorithm evaluated immune cell infiltration intestinal epithelial tissues patients with controls. Immune control groups determined least absolute shrinkage selection operator Cox regression analysis. Finally, total 51 DEGs screened, nine hub CytoHubba Cytoscape. GSE87466 GSE193677 extra set validate expression genes. CD4‐naïve T cells, gamma–delta M1 macrophages resting dendritic cells (DCs) are main infiltrates IBD. Signal transducer activator transcription 1, CCR5 integrin subunit beta 2 (ITGB2) significantly upregulated mouse model, suppression ITGB2 alleviated inflammation mice. Additionally, IBD‐associated colorectal cancer (CRC). silence suppressed proliferation tumour growth vitro vivo. DCs may provide therapeutic strategy IBD, be potential marker CRC.

Language: Английский

---

Comprehensive analyses for the coagulation and macrophage-related genes to reveal their joint roles in the prognosis and immunotherapy of lung adenocarcinoma patients

Frontiers in Immunology, Journal Year: 2023, Volume and Issue: 14

Published: Oct. 31, 2023

Purpose This study aims to explore novel biomarkers related the coagulation process and tumor-associated macrophage (TAM) infiltration in lung adenocarcinoma (LUAD). Methods The M2-related genes were obtained by Weighted Gene Co-expression Network Analysis (WGCNA) bulk RNA-seq data, while TAM marker identified analyzing scRNA-seq coagulation-associated from MSigDB KEGG databases. Survival analysis was performed for intersectional genes. A risk score model subsequently constructed based on survival-related prognosis prediction validated external datasets. Results In total, 33 macrophage-related (COMAR) obtained, 19 of which selected construction. Finally, 10 survival-associated (APOE, ARRB2, C1QB, F13A1, FCGR2A, FYN, ITGB2, MMP9, OLR1, VSIG4) involved COMAR model. According score, patients equally divided into low- high-risk groups, group significantly worse than that low-risk group. ROC curve indicated had high sensitivity specificity, multiple Moreover, also efficacy predicting clinical outcomes LUAD who received anti-PD-1/PD-L1 immunotherapy. Conclusion this has excellent predictive value immunotherapeutic with LUAD, provides potential treatment prognostic prediction.

Language: Английский

---

Exploring the impact of ITGB2 on glioma progression and treatment: Insights from non‐apoptotic cell death and immunotherapy

Environmental Toxicology, Journal Year: 2024, Volume and Issue: 39(10), P. 4496 - 4511

Published: March 15, 2024

Abstract In the realm of glioma treatment, our groundbreaking research has uncovered pivotal role Integrin Beta 2 (ITGB2) in non‐apoptotic cell death and its profound implications for immunotherapy efficacy. Gliomas, known their aggressive infiltrative nature, demand innovative therapeutic strategies improved patient outcomes. Our study bridges a critical gap by examining interplay between response gliomas. Through comprehensive analysis ten diverse datasets, we developed unique enrichment score identified ITGB2 as significant risk marker. This demonstrates that can predict immune activity, mutation characteristics, drug patients. We reveal not only mediates proliferation migration but also crucially influences responses modulating interaction gliomas macrophages single‐cell sequencing (iTalk ICELLNET). Employing variety molecular cellular methodologies, including vitro models, findings highlight potent marker biology, particularly impacting macrophage polarization. present compelling evidence ITGB2's dual regulating tumor behavior shaping landscape, thereby influencing The underlines potential ITGB2‐targeted enhancing efficacy opens new avenues personalized treatment approaches management. conclusion, this marks stride understanding pathology therapy, positioning key biomarker promising target quest effective treatments.

Language: Английский

---

Single-cell transcriptomics reveals over-activated reactive oxygen species pathway in hepatocytes in the development of hepatocellular carcinoma

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Nov. 30, 2024

Background: Hepatocellular carcinoma (HCC) is a highly heterogeneous tumor and primary cause of cancer-relevant deaths worldwide. The role reactive oxygen species (ROS) in HCC development less studied. Methods: Seurat package CellMarker database were employed for single-cell RNA sequencing (scRNA-seq) analysis based on the GSE189175 dataset from Gene Expression Omnibus (GEO). DAVID MsigDB utilized pathway analysis. SCENIC was performed to map transcription factors (TFs) regulatory network. CellChat used cellular communication Results: Six major cell subpopulations identified, among which hepatocytes accounted highest proportion both cancer adjacent tissues. enrichment scores 50 hallmark gene sets showed that ROS abnormally activated positively correlated with energy metabolism-related pathways (glucose metabolism, lipid amino acid etc.). Then, divided into four subgroups. Noticeably, GPX4+ activity related worse prognosis HCC. Mechanism revealed JUND involved positive regulation hepatocytes. It found interdependent ligand-receptor interaction between liver cells immune facilitated malignant Conclusion: over-activated having closely interacted T M2 macrophage cells. Molecular subtypes risk score signature its potential target are encouraged be developed improving precision treatment

Language: Английский

---

The development and progression of low-grade serous ovarian carcinoma

Academia oncology., Journal Year: 2024, Volume and Issue: 1(1)

Published: Sept. 13, 2024

Low-grade serous ovarian carcinoma (LGSOC) is found more frequently than high-grade in younger women. This rare subtype represents about 5–10% of all cancers and not as sensitive to chemotherapy cancer. New alternative treatments are emerging from recent clinical trials, additional trials confirming the benefit these ongoing. However, a lack deep understanding development progression LGSOC major bottleneck novel therapeutic strategies. review summarizes our current LGSOC, including findings genomic proteomic studies. Continuing investigate origins its potential precursors, will enable us develop strategies inhibit this devastating disease.

Language: Английский

---

Prognostic value and potential molecular mechanism of ITGB superfamily members in hepatocellular carcinoma

Medicine, Journal Year: 2023, Volume and Issue: 102(33), P. e34765 - e34765

Published: Aug. 18, 2023

We analyzed the prognostic value and potential molecular mechanisms of members integrin β (ITGB)superfamily in hepatocellular carcinoma (HCC) using data from The Cancer Genome Atlas (TCGA), cBioPortal, Gene Expression Profiling Interactive Analysis (GEPIA), Human Protein (HPA) HPA, Search Tool for Retrieval Interacting Genes/Proteins, GeneMANIA, Ontology (GO), Kyoto Encyclopedia Genes Genomes (KEGG), TIMER set enrichment analysis (GSEA) databases. ITGB4/5 mRNA was upregulated HCC tissues contrast to normal liver tissues, whereas ITGB2/3/8 levels were lower former. ITGB4 most frequently mutated ITGB gene HCC. Receiver operating characteristic curve (ROC) showed that expression ITGB2/3/4/5/7/8 had significant diagnostic distinguishing healthy ITGB8 highest efficacy. ITGB1/3/6/8 also tissues. verified by immunohistochemistry (IHC). Furthermore, ITGB6 ITGB7 strongly associated with overall survival (OS) patients. superfamily exhibited homology interactions protein structure. In addition, together negatively related infiltration multiple immune cell populations. GSEA results enriched migration recurrence, significantly HIPPO, TOLL JAK-STAT signaling pathways. conclusion, genes are possible be biomarkers

Language: Английский

---