Integration of single-cell and bulk RNA sequencing to identify a distinct tumor stem cells and construct a novel prognostic signature for evaluating prognosis and immunotherapy in LUAD

Journal of Translational Medicine, Journal Year: 2025, Volume and Issue: 23(1)

Published: Feb. 22, 2025

Cancer stem cells (CSCs) are crucial for lung adenocarcinoma (LUAD). This study investigates tumor cell gene signatures in LUAD using single-cell RNA sequencing (scRNA-seq) and bulk (RNA-seq), aiming to develop a prognostic marker signature (TSCMS) model. scRNA-seq RNA-seq data were analyzed. CytoTRACE software quantified the stemness score of tumor-derived epithelial clusters. Gene Set Variation Analysis (GSVA) identified potential biological functions different The TSCMS model was constructed Lasso-Cox regression, its value assessed through Kaplan–Meier, Cox receiver-operating characteristic (ROC) curve analyses. Immune infiltration evaluated Cibersortx algorithm, drug response prediction performed pRRophetic package. TAF10 functional investigations involved bioinformatics analysis, qRT-PCR, Western blot, immunohistochemistry, assays proliferation. Seven distinct clusters by CytoTRACE, with cluster 1 (Epi_C1) showing highest potential. included 49 stemness-related genes; high-risk patients exhibited lower immune ESTIMATE scores increased purity. Significant differences landscapes chemotherapy sensitivity observed between risk groups. positively correlated expression-based various tumors, including LUAD. It over-expressed lines clinical tissues, high expression linked poor prognosis. Silencing inhibited proliferation sphere formation. demonstrates model's LUAD, reveals insights into therapeutic response, identifies as target.

Language: Английский

Artificial intelligence-driven translational medicine: a machine learning framework for predicting disease outcomes and optimizing patient-centric care

Journal of Translational Medicine, Journal Year: 2025, Volume and Issue: 23(1)

Published: March 10, 2025

Abstract Background Advancements in artificial intelligence (AI) and machine learning (ML) have revolutionized the medical field transformed translational medicine. These technologies enable more accurate disease trajectory models while enhancing patient-centered care. However, challenges such as heterogeneous datasets, class imbalance, scalability remain barriers to achieving optimal predictive performance. Methods This study proposes a novel AI-based framework that integrates Gradient Boosting Machines (GBM) Deep Neural Networks (DNN) address these challenges. The was evaluated using two distinct datasets: MIMIC-IV, critical care database containing clinical data of critically ill patients, UK Biobank, which comprises genetic, clinical, lifestyle from 500,000 participants. Key performance metrics, including Accuracy, Precision, Recall, F1-Score, AUROC, were used assess against traditional advanced ML models. Results proposed demonstrated superior compared classical Logistic Regression, Random Forest, Support Vector (SVM), Networks. For example, on Biobank dataset, model achieved an AUROC 0.96, significantly outperforming (0.92). also efficient, requiring only 32.4 s for training with low prediction latency, making it suitable real-time applications. Conclusions effectively addresses medicine, offering accuracy efficiency. Its robust across diverse datasets highlights its potential integration into decision support systems, facilitating personalized medicine improving patient outcomes. Future research will focus interpretability broader

Language: Английский

A novel artificial intelligence segmentation model for early diagnosis of bladder tumors

Abdominal Radiology, Journal Year: 2024, Volume and Issue: unknown

Published: Dec. 30, 2024

Language: Английский