SEARCH · PubMed Health

Results for “TCGA”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 55 records · Page 3Linked to original sources

PubMed2026

Spatial transcriptomics of primary and metastatic ALK-rearranged NSCLC reveals site-specific adaptations.

INTRODUCTION: Genetic alterations and the tumor microenvironment (TME) influence treatment response in anaplastic lymphoma kinase-rearranged non-small cell lung cancer (ALK+ NSCLC). This study maps site-specific TME adaptations and exploratory risk-associated signatures in lymph node metastases (LNT) to investigate metastatic evolution. METHOD: We applied spatial transcriptomics to profile tumor (PanCK+) and stromal (PanCK-) compartments in a pilot cohort of 16 cases: primary lung tumors (LT, n = 3), LNT (n = 10), and brain metastases (BT, n = 3), with three site-matched non-tumor controls. LNT-derived prognostic signatures were evaluated using The Cancer Genome Atlas-Lung Adenocarcinoma (TCGA LUAD) cohorts. RESULTS: Distinct, site-specific TME features were observed. LNT stroma was enriched in fibroblasts and macrophages, while tumor segments showed increased neutrophils. BT exhibited a macrophage-associated immunosuppressive TME. Tumor cells evolved divergently: LT retained pulmonary identity and showed trend towards translation-associated programs, LNT cells shifted toward senescence and epigenetic remodeling, and BT cells showed activation of Class A/1 (Rhodopsin-like) receptor, GPCR and drug metabolism pathways. In LNT, exploratory risk-associated differences were observed. Low-risk cases (n = 6) showed adaptive immune signatures, whereas high-risk cases (n = 4) showed enrichment for stromal MET signaling and stress-response pathways. Because treatment exposure differed markedly between the risk groups, these observations should be interpreted as hypothesis-generating. TCGA LUAD analysis suggested the broader biological relevance of immune-associated markers, but reflected general LUAD rather than ALK+ specific biology. Discordant associations for GCLC and TIMP1 underscored the importance of spatial context. CONCLUSION: Site-specific microenvironments may influence tumor adaptation across metastatic niches in ALK+ NSCLC. The exploratory risk-associated findings require validation in larger, uniformly treated cohorts.

Humans↗

PubMed2026

An oxidative stress - and immunotherapy-related six-gene signature defines immune subtypes and predicts prognosis and immunotherapy response in hepatocellular carcinoma.

BACKGROUND: Oxidative stress and the tumor immune microenvironment jointly shape hepatocellular carcinoma (HCC) progression and response to immunotherapy, yet integrated biomarkers linking these processes are lacking. METHODS: Transcriptomic and clinical data from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) datasets were used to identify oxidative stress- and immunotherapyrelated differentially expressed genes (OSIRDEGs). Functional enrichment, weighted gene co-expression network analysis (WGCNA) and LASSO-Cox regression were used to construct a prognostic signature. Consensus clustering, TIDE, CIBERSORT and ssGSEA characterized immune phenotypes. Somatic mutation, copy-number and drug-response data were integrated to assess genomic alterations and drug sensitivity. Expression of model genes was validated by qRT-PCR and western blotting in HCC cell lines. RESULTS: We identified 24 OSIRDEGs enriched in cell-cycle and mitotic pathways. WGCNA intersection yielded 18 module genes, from which a six-gene signature (BUB1B, CDKN2A, CENPE, HMMR, PTTG1, SPP1) was derived. The signature robustly stratified patients into high- and low-risk groups with significantly different progression-free and disease-free survival in both TCGA-LIHC and GSE14520. Based on signature expression, two molecular subtypes were defined, exhibiting distinct survival, immune landscapes and predicted immunotherapy responsiveness. Model genes harbored recurrent alterations and showed significant correlations with anticancer agents. All six genes were upregulated at mRNA and protein levels in metastatic HCC cell lines versus normal hepatocytes. CONCLUSIONS: We systematically explored the landscape of OSIRDEGs in HCC, and proposed a validated six-gene signature that refines prognostic stratification, delineates immunerelevant HCC subtypes and highlights candidate biomarkers for therapeutic selection and mechanistic investigation.

Humans↗

PubMed2026

Decreased expression of Krüppel-like factor 4 is associated with colorectal cancer progression.

Krüppel-like factor 4 (KLF4), a key transcription factor,plays an important role in cell proliferation, differentiation, and apoptosis. Here, we explored the prognostic value of KLF4 and its role in colorectal cancer (CRC) progression. We analyzed transcriptomic data and clinical information related to CRC from The Cancer Genome Atlas (TCGA) and the Gene Expression Omnibus (GEO) database. database. Immunohistochemistry was performed to evaluate KLF4 expression in CRC tissue samples. Additionally, we examined the relationship between clinicopathological factors and patient prognosis using Cox proportional hazards model analysis. Lentiviral transfection was used to create KLF4 knockdown HCT-116 cells. Analysis of the TCGA database and two GEO datasets (GSE21510 and GSE117606) revealed that KLF4 was expressed at low levels in CRC. Furthermore, reduced KLF4 levels correlated with lymph node metastasis, distant metastasis, and advanced TNM staging. ROC curve analysis indicated that KLF4 can effectively differentiate cancerous tissue from normal tissue. Functional enrichment analysis identified KLF4 as significantly linked to the glycoprotein metabolic pathway. Our detection of KLF4 expression in CRC tissue samples confirmed its decreased levels and their association with poorer patient survival. However, KLF4 was not identified as an independent prognostic factor. In vitro, KLF4 knockdown promoted HCT-116 cell migration and invasion and downregulated the mRNA expression of glycoprotein synthesis- and glycosylation-related genes. Conversely, KLF4 re-expression markedly reversed these effects. Our findings suggested that low KLF4 expression served as a predictor factor for disease progression in CRC patients. Furthermore, reduced KLF4 levels enhance the migration and invasion of CRC cells, which may be related to impaired glycoprotein metabolism.

Colorectal cancer↗

PubMed2026

Tumor Loss of the Y Chromosome Defines a Biological Phenotype Associated with Resistance to Radiotherapy Across Cancer Types.

PURPOSE: Sex-linked determinants of radiotherapy response remain poorly understood. We investigated whether tumor loss of the Y chromosome (LOY) is associated with biological and clinical features of radiotherapy resistance across cancer types. MATERIALS AND METHODS: We integrated publicly available cancer cell-line experimental datasets and clinical data to evaluate the impact of LOY on radiotherapy response. The radiosensitivity of 125 cancer cell lines, stratified by Y chromosome status, was analyzed. Gene expression analyses were performed to identify biological pathways associated with LOY. Clinical associations were examined in 537 male patients treated with radiotherapy across multiple tumor types in The Cancer Genome Atlas. RESULTS: LOY was associated with increased post-radiotherapy survival in cancer cell lines (p < 0.001). Transcriptomic analyses demonstrated LOY-associated alterations in DNA damage response, senescence, longevity, and proliferation pathways. In TCGA tumors, LOY was associated with remodeling of the tumor microenvironment, including altered immune and stromal signatures. Clinically, LOY was associated with inferior survival in common-support overlap-weighted analyses adjusted for age, tumor stage, and TCGA-defined tumor type. CONCLUSION: These findings suggest that tumor LOY is associated with a distinct biological profile characterized by features of radioresistance and adverse clinical outcomes following radiotherapy. Further studies are warranted to determine whether LOY represents a clinically relevant sex-linked determinant of radiotherapy response.

loss of Y chromosome↗

PubMed2025

Heme oxygenase 1 (HO-1) is a drug target for reversing cisplatin resistance in non-small cell lung cancer.

INTRODUCTION: Platinum-based drugs, the most widely used chemotherapeutic drugs in clinical oncology, have long faced the problem of drug resistance, which is urgently in need of resolution. Identifying biomarkers of drug resistance may help reduce platinum resistance and improve therapeutic efficacy. OBJECTIVES: This study aims to identify potential biomarkers associated with the development of cisplatin resistance in non-small cell lung cancer (NSCLC) and explore mechanisms to overcome chemoresistance. METHODS: NSCLC cisplatin resistance cell lines were constructed, and transcriptome sequencing was performed. Results were validated using Gene Expression Omnibus (GEO) and The Cancer Genome Atlas (TCGA) databases. Molecular docking, proteomics sequencing, and in vitro and in vivo experiments were conducted to evaluate the role of Heme Oxygenase 1 (HO-1) in cisplatin resistance. RESULTS: NSCLC cisplatin resistance cell lines, GEO and TCGA data identified HMOX1, downstream of Nrf2, as a key drug resistance gene induced by cisplatin. Activation of the Nrf2/HO-1 pathway was found to induce ferroptosis resistance, a critical mechanism of cisplatin resistance. Candidate compounds SB 202190 and Nordihydroguaiaretic acid (NDGA) effectively reactivated ferroptosis by inhibiting HO-1, thereby increasing cisplatin sensitivity. CONCLUSION: The Nrf2/HO-1 pathway is a significant contributor to cisplatin resistance in NSCLC. Targeting HO-1 with SB 202190 and NDGA presents a promising strategy to overcome resistance and improve chemotherapy outcomes.

Cisplatin↗

PubMed2026

A multi-modal survival prediction framework with group-based batch training and structural consistency alignment.

OBJECTIVE: Integrating whole-slide images (WSIs) with transcriptomic profiles is pivotal for enhancing cancer survival prediction. However, the intrinsic gigapixel resolution and variable sequence lengths of WSIs create a fundamental trade-off between training efficiency and the preservation of data heterogeneity in existing frameworks. Furthermore, substantial statistical and structural discrepancies between histological and genomic modalities often impede effective cross-modal alignment and fusion, thereby limiting prognostic accuracy. METHODS: We propose PRISM, an efficient multi-modal learning framework for integrating WSIs with transcriptomic profiles. To reconcile training efficiency with full data heterogeneity, PRISM first stochastically partitions variable-length WSI sequences into a main subset and a complementary residual subset, both of which are packed into fixed-length groups for batch training. The main subset is processed in the main branch, utilizing isolation masking to maintain intra-group sequence independence. Simultaneously, the residual subset is consolidated into "hyperslides" within a residual branch that leverages tailored supervision, effectively capturing inter-slide correlations. Furthermore, PRISM integrates an Informative Token Aggregation (ITA) module to reduce redundancy in WSIs and employs Cross-batch Structural Consistency Alignment (CBSCA) mechanism to enhance inter-modal structural connectivity. Finally, efficient cross-modal feature interaction is achieved through a Low-rank Bilinear Gated Fusion (LBGF) module. Code is available at https://github.com/Alisa2080/PRISM. RESULTS: Compared with existing methods, PRISM achieves the best overall C-index across five TCGA cohorts. On the larger TCGA-BRCA dataset, PRISM requires only 6 hours of training time, substantially reducing computational cost relative to strong multimodal baselines. Furthermore, comprehensive evaluations demonstrate that PRISM achieves the best overall IBS ranking and favorable time-dependent AUC performance at 1, 3, and 5 years, thereby delivering a more favorable trade-off between prognostic performance and computational efficiency. CONCLUSION: PRISM provides a favorable balance between predictive performance, calibration quality, and computational efficiency, highlighting its potential for practical deployment in multimodal survival modeling for computational pathology.

Humans↗

PubMed2026

Targeting pancreatic cancer progression: The formononetin and salvianolic acid B combination suppresses JAK/STAT signaling via MBOAT2 downregulation.

OBJECTIVE: Formononetin and salvianolic acid B (FcS) are the primary bioactive components of the Astragalus mongholicus-Salvia miltiorrhiza herbal pair, a classic combination for treating pancreatic cancer associated with qi deficiency and blood stasis. This study elucidates the therapeutic potential and mechanisms of FcS in the treatment of pancreatic cancer. METHODS: A zebrafish xenograft model was used to screen bioactive combinations derived from A. mongholicus and S. miltiorrhiza, identifying FcS as a candidate with antitumor activity. Its efficacy was evaluated in vivo using the zebrafish model, orthotopic LSL-KrasG12D/+, LSL-Trp53R172H/+ and Pdx-1-Cre (KPC) mice, and subcutaneous xenograft models. Cell viability and proliferation were assessed using cell counting kit-8, 5-ethynyl-2'-deoxyuridine and colony formation assays, and migration and invasion were evaluated by wound healing and transwell assays. Membrane-bound O-acyltransferase 2 (MBOAT2) was identified as a potential target through a molecular docking study and the Cancer Genome Atlas (TCGA) analysis. MBOAT2 knockdown cells were used to explore its roles and the Janus kinase/signal transducer and activator of transcription (JAK/STAT) signaling pathway in FcS-mediated inhibition. RESULTS: In the zebrafish model, FcS strongly inhibited pancreatic tumor growth. FcS reduced tumor volume, the expression of proliferation marker Ki-67, and proliferating cell nuclear antigen in KPC mice. In vitro, FcS inhibited pancreatic cancer cell viability, proliferation, migration and invasion, which was accompanied by downregulation of MBOAT2 expression. TCGA analysis linked high MBOAT2 expression to aggressive phenotypes. MBOAT2 knockdown reduced the survival, proliferation and invasion of BxPC-3 cells. Rescue experiments revealed that MBOAT2 knockdown attenuated the antitumor effects of FcS, possibly through modulation of the JAK/STAT signaling pathway. FcS also inhibited tumor proliferation in xenograft models, and MBOAT2 expression was elevated in tumor tissues from pancreatic cancer patients. CONCLUSION: FcS suppresses pancreatic cancer progression via MBOAT2 downregulation and JAK/STAT pathway inhibition, which highlights MBOAT2 as a potential therapeutic target. Please cite this article as: Xu Y, Xu CS, Jin HB, Gu WG, Shen HZ, Lu L, Chen Y, Xu DC, Zhang XF, Yang JF, Wang Y. Targeting pancreatic cancer progression: The formononetin and salvianolic acid B combination suppresses JAK/STAT signaling via MBOAT2 downregulation. J Integr Med. 2026; 24(5):725-741.

Animals↗

PubMed2025

Comparative Analysis of Somatic and Germline Polymerase Proofreading Deficiencies in Cancer: Molecular and Clinical Implications.

Polymerases ε and δ maintain genome integrity through exonuclease proofreading. Germline and somatic pathogenic variants (PVs) in the exonuclease domain (ED) of POLE and POLD1 impair proofreading, causing hypermutated tumors. Despite shared mutational features that make these tumors highly immunogenic, molecular and clinical distinctions between POLE and POLD1 mutations and between somatic and germline variants remain incompletely understood. We compared the molecular and clinical characteristics of POLE and POLD1 ED PVs (n = 31), assessing their location, pathogenicity, clinical phenotypes, mismatch repair (MMR) status, tumor mutational burden, and signatures. We analyzed 360 proofreading-deficient tumors (source: The Cancer Genome Atlas [TCGA] and Catalogue Of Somatic Mutations In Cancer [COSMIC]) and 70 families (249 individuals) with polymerase proofreading-associated polyposis. All germline and somatic PVs had high AlphaMissense scores (0.87-1) and clustered within or near Exo motifs. Recurrent, nonfounder germline PVs, POLE L424V and POLD1 S478N, showed low/modest REVEL scores. Somatic variants occurred mainly in endometrial cancers (75% of proofreading-deficient TCGA cancers), whereas colorectal cancer predominated in polymerase proofreading-associated polyposis (56% of carriers). Cancer risks and tumor spectra differed between POLE and POLD1 PV carriers. Aggressive hereditary phenotypes were linked to either specific POLE PVs (eg, S297F, V411L, P436R, M444K, A456P, and S461T) or the co-occurrence of germline ED PVs with germline MMR gene PVs. Distinct hypermutator profiles were confirmed for polymerase ε and polymerase δ proofreading deficiencies via unique mutational signatures (Polymerase ε: SBS10a/b, SBS28; Polymerase δ: SBS10c/d). Tumors with combined proofreading and MMR deficiencies had significantly higher tumor mutational burden and a shift in the associated mutational spectra. Unlike POLE, POLD1 ED PVs exhibited haplosufficiency, typically requiring a somatic second hit (eg, loss of heterozygosity) or MMR deficiency to drive hypermutation. In conclusion, differences between POLE and POLD1 and between somatic and germline mutations influence clinical presentation, mutagenic potential, and reliance on cooperating defects in tumorigenesis. These insights advance the understanding of proofreading-deficient cancers, with implications for diagnostics, genetic counseling, and precision oncology.

Humans↗

PubMed2026

FGF19 as a site-specific candidate biomarker in colorectal neuroendocrine carcinomas.

PURPOSE: Gastrointestinal neuroendocrine carcinomas (GI-NECs) are aggressive tumors with marked site-specific heterogeneity, yet molecular markers for colorectal origin are lacking. This study characterized genomic and protein expression profiles to identify origin-specific biomarkers. METHODS: Nineteen GI-NECs (7 esophageal, 6 gastric, 6 colorectal) were analyzed by targeted next-generation sequencing (NGS) of 425 genes and immunohistochemistry (IHC). Genetic variations across primary sites were compared, and associations between FGF19 expression, clinicopathological features, microsatellite (MS) status, and tumor mutational burden (TMB) were assessed. FGF19 transcriptional expression was further examined in The Cancer Genome Atlas (TCGA) colorectal cohort using the UALCAN platform. RESULTS: A total of 163 genomic alterations were identified. FGF19 was the only gene showing site-specific alterations, being exclusively mutated or amplified in colorectal NECs (50%, 95% CI: 11.8-88.2%) with significantly elevated protein expression (83.3%, 95% CI: 35.9-99.6%) compared with other sites. A microsatellite instability-high (MSI-H) subgroup (10.5%, 95% CI: 1.3-33.1%) exhibited markedly higher TMB. TCGA data confirmed upregulated FGF19 in colorectal tumors but showed no survival association, consistent with the prognostic neutrality in our cohort. CONCLUSIONS: FGF19 may act as a site-specific candidate biomarker for colorectal NECs, with 83.3% protein positivity and exclusive site-specific alterations in 50% of cases. Detection of MSI-H suggests that mismatch repair (MMR) testing may be considered in selected patients with suggestive clinical or family histories to inform immunotherapy decisions.

FGF19↗

PubMed2026

CD44 gene rs9666607 polymorphism is associated with papillary thyroid carcinoma and interacts with CREB3L1.

BACKGROUND: The incidence of papillary thyroid carcinoma (PTC) has been rising. CD44 is involved in cell adhesion and migration, but the role of its genetic variation in PTC remains unclear. METHODS: This study aimed to investigate the association of CD44 gene polymorphisms with PTC and to examine the interaction between CD44 and CREB3L1. This study enrolled 354 patients with PTC, and the genotype distribution of the CD44 polymorphism (rs9666607) was analyzed. Key PTC genes were screened using the Gene Expression Omnibus (GEO) database (GSE33630). Gene Ontology (GO) functional enrichment and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses were performed on these key genes. CD44 expression was validated using TCGA database, ELISA, qRTPCR, Western blot, and IHC in patient tissues, PTC mouse model, and human cell lines. The direct interactive molecules were screened through a bioinformatics method. RESULTS: The GSE33630 dataset identified a total of 124 upregulated and 85 downregulated differentially expressed genes. Enrichment analysis revealed 12 key PTC genes, including CD44. TCGA database validation revealed that CD44 was significantly overexpressed in PTC patients. The rs9666607‑A allele was associated with an increased risk of PTC and lymph node metastasis under a dominant model. CD44 mRNA and protein levels were significantly higher in PTC tissues versus adjacent tissue and further elevated in metastatic cases. Bioinformatic analysis predicted CD44 interaction with the transcription factor CREB3L1, and this was confirmed by molecular docking. CREB3L1 expression was synchronously upregulated with CD44 in PTC. CONCLUSION: CD44 polymorphisms, particularly the rs9666607-A allele, are significantly associated with PTC risk and metastasis in the studied population. CD44 is overexpressed in PTC, and its interaction with CREB3L1 suggests a potential novel interaction in PTC pathogenesis.

Hyaluronan Receptors↗

PubMed2026

Lactylation-related immune-metabolic dysregulation defines prognostic and therapeutic stratification in lung adenocarcinoma.

BACKGROUND: Lactylation links lactate metabolism with inflammatory signaling and immune regulation in tumors. However, its cellular distribution and translational value in lung adenocarcinoma (LUAD) remain unclear. METHODS: Single-cell RNA-sequencing datasets GSE189357 and GSE171145 were integrated to characterize lactylation-related activity, intercellular communication, and malignant epithelial cell states in LUAD. Single-cell-derived lactylation-related differentially expressed genes were mapped to TCGA-LUAD and multiple GEO cohorts. Univariate Cox regression and machine learning algorithms were used to construct a lactylation-related prognostic signature (LRPS). The associations of LRPS with prognosis, immunotherapy response, drug sensitivity, genomic alterations, immune infiltration, and inflammation- and metabolism-related pathways were evaluated. KRT7 was further validated using virtual knockout analysis, spatial transcriptomics, and in vitro and in vivo experiments. RESULTS: lactylation-related transcriptional activity showed heterogeneous distribution across LUAD cell populations and was associated with altered cell-cell communication. In malignant epithelial cells, LRTS-high and LRTS-low states exhibited distinct metabolic, inflammatory, and tumor-related pathway activities. LRPS showed stable prognostic performance in TCGA-LUAD and multiple GEO cohorts and remained an independent prognostic factor. Low LRPS was associated with greater potential benefit from immunotherapy, whereas different LRPS groups displayed distinct drug sensitivity, genomic alteration, and immune microenvironment patterns. KRT7 was highly expressed in LUAD and associated with poor prognosis. KRT7 knockdown suppressed LUAD cell proliferation, migration, invasion, colony formation, and tumor growth in vivo. CONCLUSIONS: This study identifies lactylation-related immune-metabolic dysregulation as a clinically relevant feature of LUAD and develops a single-cell-guided LRPS for prognosis and therapeutic stratification. KRT7 emerged as an LRPS-related functional candidate with experimentally supported roles in malignant LUAD phenotypes.

Immunotherapy↗

PubMed2026

Biallelic loss of RB1 in hepatocellular carcinoma as synthetic lethal target for artificial intelligence-guided therapy.

The retinoblastoma (RB1) gene is a critical tumor suppressor that regulates cell cycle progression and genomic stability. Although RB1 alterations have been reported in hepatocellular carcinoma (HCC), the biological and clinical consequences of biallelic RB1 inactivation (RB1-Bi) remain poorly defined. We performed a comprehensive allele-specific genomic analysis of HCC patients from the TCGA-LIHC (n = 355) and in-house AMC (n = 206) cohorts, collectively comprising the AMC-TCGA discovery cohort. In this combined cohort, RB1-Bi was identified in 14.6% of tumors, was enriched in poorly differentiated HCCs and was independently associated with significantly reduced overall survival (adjusted hazard ratio 3.32, 95% CI 1.93-5.72, p < 0.001). Additionally, a deep learning-based histopathology model using hematoxylin and eosin-stained slides (i.e., FR-MIL model) accurately predicted RB1-Bi status (F1 score 84.39% [95% CI, ±0.02]), making it readily identifiable in routine clinical practice. The prevalence and prognostic impact of RB1-Bi, as well as FR-MIL model performance, were consistent across independent validation cohorts, including advanced-stage tumors and external institutions. High-throughput drug screening in isogenic HCC models revealed that RB1-Bi HCC cells were particularly sensitive to inhibitors targeting mitotic regulators (e.g., AURKA, PLK1, KSP) and DNA damage response pathways (e.g., PARP inhibitors). Synthetic lethal interactions between RB1-Bi and these compounds were demonstrated in vitro and in vivo, and combination treatment with mitotic and PARP inhibitors had synergistic effects with acceptable tolerability. We conclude that RB1-Bi represents a clinically actionable biomarker that identifies a high-risk HCC subtype with specific therapeutic vulnerabilities, offering new opportunities for precision medicine.

Humans↗

PubMed2025

Transfer learning with multiomics integration and deep neural networks reveals drug resistance mechanisms in cancer.

Drug resistance remains one of the primary challenges in effective cancer therapy. In this study, we employed a deep neural network (DNN)-based transfer learning (TL) approach to predict drug response and uncover drug resistance mechanisms. We integrated gene expression, somatic mutation, and copy number aberration (CNA) data with drug response profiles using multi-omics integration (MI). We used the Genomics of Drug Sensitivity in Cancer (GDSC) data for training and incorporated drugs with same pathways into the training models. We then evaluated drug response predictions on independent in-vivo PDX Encyclopedia (PDX) and ex-vivo the Cancer Genome Atlas (TCGA) datasets. In addition, we conducted pathway enrichment analyses to elucidate the mechanisms underlying drug resistance for paclitaxel, 5-fluorouracil (5-FU), gemcitabine, and cetuximab. We also applied Fisher's exact test (FET) to assess potential associations between drug resistance and the presence of mutations or CNAs. Our pan-drug models outperformed other methods based on the area under the precision-recall curve (AUCPR). Our pathway enrichment analyses revealed LDHB-mediated pyruvate metabolism and FYN-mediated focal adhesion might have pivotal roles in paclitaxel resistance, while PINK1-mediated mitophagy might be critical in 5-FU resistance. In addition to transcriptional activation, FET suggested that CNAs in LDHB and PINK1 may also be associated with resistance to paclitaxel and 5-FU, respectively. Furthermore, enrichment results for paclitaxel and cetuximab indicated shared resistance mechanisms between the two drugs. Importantly, our findings are consistent with prior experimental studies, providing literature-based validation of our results. Overall, our DNN-based TL approach achieved strong predictive performance across PDX & TCGA datasets and enrichment analyses provided valuable biological insights into drug resistance mechanisms.

Humans↗

PubMed2025

The offonome reveals on and off states of gene expression near the detection limit of RNA-seq.

RNA-seq, widely used for gene expression profiling, provides nucleotide level genome coverage and summary gene expression values. Generally, low-expressed genes are ignored due to their unfavorable signal-to-noise ratio, however, these genes may offer crucial information, such as detecting rare cells in bulk tissues. In this study, we applied an approach that transforms the expression levels of low-expressed genes into a robust dichotomized on/off state by leveraging similarities in transcript coverage shape. Applied to three human cancer cohorts from the Cancer Genome Atlas (TCGA), chosen based on tissue morphology and anatomic site, we identified genes, the "offonome" near the detection limit, consistently or occasionally off across samples. Genes in the offonome spectrum proved useful for supervised and unsupervised applications, including characterizing oncogenic pathways, and identifying rare populations of cells in bulk tissue. Interrogating the offonome is relevant to bulk tumor analyses like TCGA, potentially expediting gene investigation in low-input situations like single cell RNA-seq.

Humans↗

PubMed2025

Identification of a prognostic signature consisting of three macrophage-related genes for glioblastoma based on bulk and single-cell transcriptomes analyses.

BACKGROUND: Tumor-associated macrophages have been implicated in the progression and treatment resistance of glioblastoma (GBM). This study aimed to identify macrophage-related genes associated with prognosis and therapeutic response in GBM. MATERIALS AND METHODS: Bulk RNA-seq data from 533 patients with GBM were downloaded from the Cancer Genome Atlas (TCGA) and Chinese Glioma Genome Atlas (CGGA) databases. Bioinformatic tools were used to detect the co-expression gene modules associated with the infiltration of immune cells, identify a prognostic macrophage-related gene signature, and explore their association with sensitivity to chemotherapeutic drugs and immune checkpoint blockade. Single-cell RNA-seq data and multiplexed immunofluorescence were used to validate ISG20 expression (a member of the identified gene signature) in macrophages. RESULTS: We detected gene modules associated with macrophages and identified a signature consisting of three macrophage-related genes (ISG20, PARP12 and IFIT5) in the discovery set (TCGA-GBM, n = 159), and validated its prognostic value in the validation set (CGGA-GBM, n = 374). This gene signature demonstrated favorable accuracy in predicting prognosis and resistance of immuno- and chemo-therapy. The co-expression of ISG20 and PD-1 in macrophages was verified by single-cell RNA-seq data and multiplex immunofluorescence. CONCLUSIONS: This study presents a macrophage-related gene signature to predict prognosis and therapeutic response in GBM. ISG20, PARP12 and IFIT5 are interferon-stimulated genes, and further investigations may provide new insights into the interplay between macrophages and interferon signaling in GBM.

Humans↗

PubMed2026

X-intNMF: a cross- and intra-omics regularized NMF framework for multi-omics integration.

MOTIVATION: The rapid accumulation of multi-omics data presents a valuable opportunity to advance our understanding of complex diseases and biological systems, driving the development of integrative computational methods. However, the complexity of biological processes, spanning multiple molecular layers and involving intricate regulatory interactions, requires models that can capture both intra- and cross-omics relationships. Most existing integration methods primarily focus on sample-level similarities or intra-omics feature interactions, often neglecting the interactions across different omics layers. This limitation can result in the loss of critical biological information and suboptimal performance. To address this gap, we propose X-intNMF, a network-regularized non-negative matrix factorization (NMF) framework that simultaneously integrates intra- and cross-omics feature interactions into a shared low-dimensional representation (see Fig. 1). By modeling these multi-layered relationships, X-intNMF enhances the representation of biological interactions and improves integration quality and prediction accuracy. RESULTS: For evaluation, we applied X-intNMF to predict breast cancer phenotypes and classify clinical outcomes in lung and ovarian cancers using mRNA expression, microRNA expression, and DNA methylation data from TCGA. The results show that X-intNMF consistently outperforms state-of-the-art methods. Ablation studies confirm that incorporating both cross-omics and intra-omics interactions contributes significantly to the model's improved performance. Additionally, survival analysis on 25 TCGA cancer datasets demonstrates that the integrated multi-omics representation offers strong prognostic value for both overall survival and disease-free status. These findings highlight X-intNMF's ability to effectively model multi-layered molecular interactions while maintaining interpretability, robustness, and scalability within the NMF framework. AVAILABILITY AND IMPLEMENTATION: The source code and datasets supporting this study are publicly available at GitHub (https://github.com/compbiolabucf/X-intNMF) and archived on Zenodo (https://doi.org/10.5281/zenodo.18238385).

Multiomics↗

PubMed2026

GRNContext: an interactive web platform for contextualized gene regulatory networks visualization across human cancers.

SUMMARY: While current Gene Regulatory Network (GRN) databases provide comprehensive reference maps of potential interactions between transcription factors and target genes, they do not specify which regulatory interactions are active within specific biological contexts. This limitation is particularly critical in cancer, where transcriptional programs are inherently tissue-specific. To address this gap, we developed GRNContext, an interactive web platform designed for the visualization, exploration, and comparative analysis of gene regulatory networks contextualized across 33 cancer types from The Cancer Genome Atlas (TCGA). Our approach uses the TFLink human reference GRN as a starting point and integrates TCGA transcriptomic profiles to infer cancer-specific regulatory activity. Regulatory relevance was assessed using complementary machine learning and statistical methods, which were unified into a consensus score to prioritize and filter the most relevant candidate regulators for each target gene. By providing both curated context-specific GRNs and a user-friendly platform, GRNContext constitutes a comprehensive and accessible resource that supports mechanistic investigations, hypothesis generation, and translational research focused on transcriptional regulation in cancer. AVAILABILITY AND IMPLEMENTATION: GRNContext is supported by all major browsers and freely available on the web at https://apps.cienciavida.org/grncontext. It is implemented as a client-server web application featuring a FastAPI backend and a React frontend utilizing Cytoscape.js for interactive network visualization, all containerized via Docker for cross-platform compatibility.

Humans↗

PubMed2026

POU2F3 expression in lung squamous cell carcinoma: transcriptomic and immunohistochemical profiling with prognosis.

BACKGROUND: Lung squamous cell carcinoma (LUSC) lacks well-defined molecular targets. This study investigated the clinical and biological relevance of POU class 2 homeobox 3 (POU2F3), a tuft cell-associated transcription factor, in LUSC. METHODS: RNA sequencing data of patients with LUSC from The Cancer Genome Atlas (TCGA cohort, n = 190) was analysed and compared to a cohort of surgically resected cases analyzed via immunohistochemistry (IHC cohort, n = 137). Prognostic impact was assessed via survival analyses. Transcriptomic features, pathway enrichment, and immune profiles were evaluated via differentially expressed gene analysis, Gene Set Enrichment Analysis, and CIBERSORTx. RESULTS: High POU2F3 expression independently predicted poor overall survival in the TCGA cohort (HR = 2.06, 95% CI: 1.04-4.08, P = 0.039). In contrast, POU2F3 expression was not prognostic in the IHC cohort (P = 0.995). Morphologically, POU2F3-positive tumours were enriched for non-keratinizing and poorly differentiated subtypes. Transcriptomic analysis showed suppression of proliferation and immune-related pathways (FDR < 0.001), with suggestive enrichment of the TGF-β (FDR = 0.143) and p53 (FDR = 0.229) signaling pathways. On immune deconvolution, POU2F3-high tumours showed a nominal increase in activated dendritic cells, which did not withstand multiple testing correction. POU2F3 protein was detected in 12.4% of tumours and was significantly associated with p53 or RB1 abnormalities (single or double) (P = 0.028). CONCLUSIONS: POU2F3 marks a transcriptionally distinct, early-stage subtype of LUSC with keratinization-related features. Its prognostic relevance appears context-dependent and requires prospective validation in uniformly treated cohorts.

Humans↗