Lasso Algorithm Research Articles

Cuproptosis-Related Biomarkers and Characterization of Immune Infiltration in Sepsis.

Sepsis is a worldwide epidemic, with high morbidity and mortality.Cuproptosis is a form of cell death that is associated with a wide range of diseases. This study aimedto explore genes associated withcuproptosisin sepsis, construct predictive modelsandscreen for potential targets. The LASSO algorithm and SVM-RFE model has been analysedthe expression ofcuproptosis-related genes in sepsis and immune infiltration characteristics and identified the marker genes under a diagnostic model. Gene-drug networks, mRNA-miRNA networksandPPI networks were constructed to screen for potential biological targets. The expression of marker genes was validated based on the GSE57065 dataset. Consensus clustering method was used to classify sepsis samples. We found 381 genes associated with the development of sepsis anddiscovered significantly differentially expressedcuproptosis-related genes of 16 cell types in sepsis and immune infiltration with CD8/CD4 T cells being lower.NFE2L2, NLRP3, SLC31A1, DLD, DLAT, PDHB, MTF1, CDKN2A andDLST were identified as marker genes by the LASSO algorithm and the SVM-RFE model. AUC > 0.9 was constructed for PDHB and MTF1 alone respectively. The validation group data for PDHB (P=0.00099) and MTF1 (P=7.2e-14) were statistically significant. Consistent clustering analysis confirmed two subtypes.The C1 subtype may be more relevant to cellular metabolism and the C2 subtype has some relevance to immune molecules.The results of animal experiments showed that the gene expression was consistent with the bioinformatics analysis. Our study systematically explored the relationship between sepsis andcuproptosisand constructed a diagnostic model. And, severalcuproptosis-related genes may interfere with the progression of sepsis through immune cell infiltration.

Development and validation of a radiomics-based model for predicting osteoporosis in patients with lumbar compression fractures

BACKGROUNDOsteoporosis, a metabolic bone disorder, markedly elevates fracture risks, with vertebral compression fractures being predominant. Antiosteoporotic treatments for patients with osteoporotic vertebral compression fractures (OVCF) lessen both the occurrence of subsequent fractures and associated pain. Thus, diagnosing osteoporosis in OVCF patients is vital. PURPOSEThe aim of this study was to develop a predictive radiographic model using T1 sequence MRI images to accurately determine whether patients with lumbar spine compression fractures also have osteoporosis. STUDY DESIGNRetrospective cohort study. PATIENT SAMPLEPatients over 45 years of age diagnosed with a fresh lumbar compression fracture. OUTCOME MEASURESDiagnostic accuracy of the model (area under the ROC curve). METHODSThe study retrospectively collected clinical and imaging data (MRI and DEXA) from hospitalized lumbar compression fracture patients (L1–L4) aged 45 years or older between January 2021 and June 2023. Using the pyradiomics package in Python, features from the lumbar compression fracture vertebral region of interest (ROI) were extracted. Downscaling of the extracted features was performed using the Mann-Whitney U test and the least absolute shrinkage selection operator (LASSO) algorithm. Subsequently, six machine learning models (Naive Bayes, Support Vector Machine [SVM], Decision Tree, Random Forest, Extreme Gradient Boosting [XGBoost], and Light Gradient Boosting Machine [LightGBM]) were employed to train and validate these features in predicting osteoporosis comorbidity in OVCF patients. RESULTSA total of 128 participants, 79 in the osteoporotic group and 49 in the non-osteoporotic group, met the study's inclusion and exclusion criteria. From the T1 sequence MRI images, 1906 imaging features were extracted in both groups. Utilizing the Mann-Whitney U test, 365 radiologic features were selected out of the initial 1,906. Ultimately, the lasso algorithm identified 14 significant radiological features. These features, incorporated into six conventional machine learning algorithms, demonstrated successful prediction of osteoporosis in the validation set. The NaiveBayes model yielded an area under the receiver operating characteristic curve (AUC) of 0.84, sensitivity of 0.87, specificity of 0.70, and accuracy of 0.81. CONCLUSIONSA NaiveBayes machine learning algorithm can predict osteoporosis in OVCF patients using t1-sequence MRI images of lumbar compression fractures. This approach aims to obviate the necessity for further osteoporosis assessments, diminish patient exposure to radiation, and bolster the clinical care of patients with OVCF.

Exploring gene biomarkers and targeted drugs for ferroptosis and cuproptosis in osteosarcoma: A bioinformatic approach.

Osteosarcoma predominantly affects adolescents and young adults and is characterized as a malignant bone tumor. In recent decades, substantial advancements have been achieved in both diagnosing and treating osteosarcoma. Resulting in enhanced survival rates. Despite these advancements, the intricate relationship between ferroptosis and cuproptosis genes in osteosarcoma remains inadequately understood. Leveraging TARGET and GEO datasets, we conducted Cox regression analysis to select prognostic genes from a cohort of 71 candidates. Subsequently, a novel prognostic model was engineered using the LASSO algorithm. Kaplan-Meier analysis demonstrated that patients stratified as low risk had a substantially better prognosis compared with their high-risk counterparts. The model's validity was corroborated by the area under the receiver operating characteristic (ROC) curve. Additionally, we ascertained independent prognostic indicators, including clinical presentation, metastatic status, and risk scores, and crafted a clinical scoring system via nomograms. The tumor immune microenvironment was appraised through ESTIMATE, CIBERSORT, and single-sample gene set enrichment analysis. Gene expression within the model was authenticated through PCR validation. The prognostic model, refined by Cox regression and the LASSO algorithm, comprised two risk genes. Kaplan-Meier curves confirmed a significantly improved prognosis for the low-risk group in contrast to those identified as high-risk. For the training set, the ROC area under the curve (AUC) values stood at 0.636, 0.695, and 0.729 for the 1-, 3-, and 5-year checkpoints, respectively. Although validation set AUCs were 0.738, 0.668, and 0.596, respectively. Immune microenvironmental analysis indicated potential immune deficiencies in high-risk patients. Additionally, sensitivity to three small molecule drugs was investigated in the high-risk cohort, informing potential immunotherapeutic strategies for osteosarcoma. PCR analysis showed increased mRNA levels of the genes FDX1 and SQLE in osteosarcoma tissues. This study elucidates the interaction of ferroptosis and cuproptosis genes in osteosarcoma and paves the way for more targeted immunotherapy.

An MRI radiomics approach to discriminate hemorrhage prone intracranial tumors before stereotactic biopsy.

To explore imaging biomarkers predictive of intratumoral hemorrhage for lesions intended for elective stereotactic biopsy. This study included a retrospective cohort of 143 patients with 175 intracranial lesions intended for stereotactic biopsy. All the lesions were randomly split into a training dataset (n=121) and a test dataset (n=54) at a ratio of 7:3. 34 lesions were defined as "hemorrhage-prone tumors" as hemorrhage occurred between initial diagnostic MRI acquisition and the scheduled biopsy procedure. Radiomics features were extracted from the contrast-enhanced T1WI and T2WI images. Features informative of hemorrhage were then selected by the LASSO algorithm and an SVM model was built with selected features. The SVM model was further simplified by discarding features with low importance calculated using a "permutation importance" method. The model's performance was evaluated with confusion matrix-derived metrics and AUC value on the independent test dataset. Nine radiomics features were selected as hemorrhage related features of intracranial tumors by the LASSO algorithm. The simplified model's sensitivity, specificity, accuracy, and AUC reached 0.909, 0.930, 0.926, and 0.949 (95%CI: 0.865-1.000) on the test dataset in the discrimination of "hemorrhage-prone tumors". The permutation method rated feature "T2_gradient_firstorder_10Percentile" as the most important, the absence of which decreased the model's accuracy by 10.9%. Radiomics features extracted on contrast-enhanced T1WI and T2WI sequences were predictive of future hemorrhage of intracranial tumors with favorable accuracy. This model may assist in the arrangement of biopsy procedures and the selection of target lesions in patients with multiple lesions.

Prognosis and chemotherapy drug sensitivity in liver hepatocellular carcinoma through a disulfidptosis-related lncRNA signature

Disulfidptosis, a new type of regulated cell death associated with the actin cytoskeleton, provides a new therapeutic tool for cancers. The direct relationship between disulfidptosis-related lncRNAs(DRLs) in liver hepatocellular carcinoma(HCC) remains unclear. We acquired transcriptomic data, corresponding clinical data, and tumor mutation data of HCC from the TCGA database. First of all, DRLs were determined through correlation analysis. Then, a prognostic model containing six DRLs was created by adopting univariate Cox regression, LASSO algorithm and multivariate Cox regression analysis. Based on the model, 424 HCC patients were divided into high- and low-risk groups. Next, we structured ROC curves and PCA through combining the model and clinical data. Enrichment analysis and immune infiltration analysis were adopted to further explore the relationship between the model and prognosis. In addition, we explored the relationship between the model and tumor mutation burden (TMB). There were significant differences between high- and low- risk groups, and patients in the high-risk group showed poor prognosis. Enrichment analysis suggested that metabolic progress was obviously different between the two groups. According to the analysis of immune infiltration, there were several differences in immune cells, function, and checkpoints. Patients with high-risk and high TMB demonstrated the least favorable prognosis. The two risk groups both manifested visiblly in chemotherapy drug sensitivity. To sum up, we set up a DRL-based signature and that may provide a predictable value for the prognosis and use of chemotherapy drugs for HCC patients.

Photovoltaic module temperature prediction using various machine learning algorithms: Performance evaluation

This paper presents data-driven models for photovoltaic module temperature prediction and analyzes the relation and effects of ambient conditions to module temperature. A total of 12 different machine learning and regression algorithms are implemented, with a large experimental dataset of 345,600 × 7. Prior to implementing those algorithms, data preprocessing is performed to prepare the datasets and determine the informative attributes for the models. Using PCA with module temperature as target to predict, the selected features for models' inputs were determined to be ambient temperature, solar radiation, wind speed, and relative humidity, and each algorithm is cross-validated and tuned with optimal performance parameters. Results show that while relative humidity is more likely to introduce less information to the model, other aforementioned features are the important parameters to predict the module temperature. While for linear modeling, LASSO algorithm provided the best performance, the ANN model demonstrated the best overall results as it produced the most accurate predictions with lowest errors. A similar performance is attained by the proposed non-linear model, KRR and Gradient Boosting algorithm, with a slight advantage to the KRR model. Furthermore, in comparison to experimental data, the ANN model and the proposed non-linear model provided an R2 values of 0.986 and 0.981, with a MAE of 0.982 and 1.476, and MSE of 2.181and 3.464, respectively. Moreover, the proposed model supplied accurate results when compared to models from literature in an out-of-sample testing, it also proven to be robust and accurate when used to predict the PV power output.

Predicting immunotherapy response in melanoma using a novel tumor immunological phenotype-related gene index.

Melanoma is a highly aggressive and recurrent form of skin cancer, posing challenges in prognosis and therapy prediction. In this study, we developed a novel TIPRGPI consisting of 20 genes using Univariate Cox regression and the LASSO algorithm. The high and low-risk groups based on TIPRGPI exhibited distinct mutation profiles, hallmark pathways, and immune cell infiltration in the tumor microenvironment. Notably, significant differences in tumor immunogenicity and TIDE were observed between the risk groups, suggesting a better response to immune checkpoint blockade therapy in the low-TIPRGPI group. Additionally, molecular docking predicted 10 potential drugs that bind to the core target, PTPRC, of the TIPRGPI signature. Our findings highlight the reliability of TIPRGPI as a prognostic signature and its potential application in risk classification, immunotherapy response prediction, and drug candidate identification for melanoma treatment. The "TIP genes" guided strategy presented in this study may have implications beyond melanoma and could be applied to other cancer types.

A nomogram based on radiomics and clinical information to predict prognosis in percutaneous balloon compression for the treatment of trigeminal neuralgia.

To develop a clinical-radiomics nomogram based on clinical information and radiomics features to predict the prognosis of percutaneous balloon compression (PBC) for the treatment of trigeminal neuralgia (TN). The retrospective study involved clinical data from 149 TN patients undergoing PBC at Zhongnan Hospital, Wuhan University from January 2018 to January 2022. The free open-source software 3D Slicer was used to extract all radiomic features from the intraoperative X-ray balloon region. The relationship between clinical information and TN prognosis was analyzed by univariate logistic analysis and multivariate logistic analysis. Using R software, the optimal radiomics features were selected using the least absolute shrinkage and selection operator (Lasso) algorithm. A prediction model was constructed based on the clinical information and radiomic features, and a nomogram was visualized. The performance of the clinical radiomics nomogram in predicting the prognosis of PBC in TN treatment was evaluated using the area under the receiver operating characteristic curve (AUC) and decision curve analysis (DCA). A total of 149 patients were eventually included. The clinical factors influencing the prognosis of TN in univariate analysis were compression severity score and TN type. The lasso algorithm Max-Relevance and Min-Redundancy(mRMR) was used to select two predictors from 13 morphology-related radiomics features, including elongation and surface-volume ratio. A total of 4 predictors were used to construct a prediction model and nomogram. The AUC was 0.886(95% confidence interval (CI), 0.75 to 0.96), indicating that the model's good predictive ability. DCA demonstrated the nomogram's high clinical applicability. Clinical-radiomics nomogram constructed by combining clinical information and morphology-related radiomics features have good potential in predicting the prognosis of TN for PBC treatment. However, this needs to be further studied and validated in several independent external patient populations.

Precision unveiled: Synergistic genomic landscapes in breast cancer-Integrating single-cell analysis and decoding drug toxicity for elite prognostication and tailored therapeutics.

Globally, breast cancer, with diverse subtypes and prognoses, necessitates tailored therapies for enhanced survival rates. A key focus is glutamine metabolism, governed by select genes. This study explored genes associated with T cells and linked them to glutamine metabolism to construct a prognostic staging index for breast cancer patients for more precise medical treatment. Two frameworks, T-cell related genes (TRG) and glutamine metabolism (GM), stratified breast cancer patients. TRG analysis identified key genes via hdWGCNA and machine learning. T-cell communication and spatial transcriptomics emphasized TRG's clinical value. GM was defined using Cox analyses and the Lasso algorithm. Scores categorized patients as TRG_high+GM_high (HH), TRG_high+GM_low (HL), TRG_low+GM_high (LH), or TRG_low+GM_low (LL). Similarities between HL and LH birthed a "Mixed" class and the TRG_GM classifier. This classifier illuminated gene variations, immune profiles, mutations, and drug responses. Utilizing a composite of two distinct criteria, we devised a typification index termed TRG_GM classifier, which exhibited robust prognostic potential for breast cancer patients. Our analysis elucidated distinct immunological attributes across the classifiers. Moreover, by scrutinizing the genetic variations across groups, we illuminated their unique genetic profiles. Insights into drug sensitivity further underscored avenues for tailored therapeutic interventions. Utilizing TRG and GM, a robust TRG_GM classifier was developed, integrating clinical indicators to create an accurate predictive diagnostic map. Analysis of enrichment disparities, immune responses, and mutation patterns across different subtypes yields crucial subtype-specific characteristics essential for prognostic assessment, clinical decision-making, and personalized therapies. Further exploration is warranted into multiple fusions between metrics to uncover prognostic presentations across various dimensions.

Performance evaluation of cable-stayed bridge expansion joints based on Lasso dimensionality reduction and temperature-displacement-correlation model

Expansion joints play a crucial role in accommodating the longitudinal movement of the main beam, which is mainly caused by temperature variation. This paper establishes an accurate model that relates the temperature field of the main beam to the displacement of the expansion joint, enabling reliable performance prediction and early warning of the expansion joint. Firstly, three commonly used methods for characterizing the temperature field of the main beam are introduced, along with their advantages and disadvantages. Secondly, a novel method is proposed using the Lasso algorithm to calculate critical temperatures. The objective is to select temperature channels data that have significant impact on the longitudinal displacement of the main beam. The selected channels data is then linearly weighted based on feature importance to obtain critical temperature. Based on this, a precise relationship model between the main beam temperature and the expansion joint displacement is derived through regression. For the residual term in the model fitting, an expansion joint performance early warning procedure is developed based on the X-bar control chart. Finally, using one-year long-term monitoring data from a newly constructed cable-stayed bridge as an example, the proposed method demonstrates superior capability in predicting the predefined damage of the expansion joint compared to the other two commonly used methods.

Predicting hypovitaminosis C with LASSO algorithm in adult critically ill patients in surgical intensive care units: a bi-center prospective cohort study

Vitamin C played pleiotropic roles in critical illness and vitamin C insufficiency was predictive of the development of multiple organ failure. Currently, the prevalence of vitamin C insufficiency in Chinese critically ill patients is rarely determined and there are no established bedside tools to predict hypovitaminosis C. To develop a nomogram to identify patients with high risk of hypovitaminosis C, we performed a bi-center prospective cohort study at two ICUs of the first and sixth medical center in PLA General Hospital, Beijing, China from May 6th to July 31st, 2021 We identified 322 eligible patients. 62.4% patients were hypovitaminosis C. 7 features, including source of infection, the level of serum albumin, age, male gender, sepsis, vascular disease, and wasting of vitamin C by the kidney, were selected using LASSO algorithm and therefore included in the nomogram. In the testing set, our model showed moderate discrimination ability with areas under the curve of 0.75 [0.64–0.84]. Variable importance evaluated by SHAP value highlighted two novel important predictors, i.e., abdominal infection and the level of serum albumin. In conclusion, we first reported a high burden of vitamin C insufficiency in Chinese adult patient in the ICU. We also constructed a prediction model to timely identify patients with high risk of hypovitaminosis C, which allows the clinicians to choose appropriate candidates for Vitamin C repletion in clinical practice or clinical trials.

Machine learning-based endoplasmic reticulum-related diagnostic biomarker and immune microenvironment landscape for osteoarthritis.

Osteoarthritis (OA) is the most common degenerative joint disease worldwide. Further improving the current limited understanding of osteoarthritis has positive clinical value. OA samples were collected from GEO database and endoplasmic reticulum related genes (ERRGs) were identified. The WGCNA network was further built to identify the crucial gene module. Based on the expression profiles of characteristic ERRGs, LASSO algorithm was used to select key factors according to the minimum λ value. Random forest (RF) algorithm was used to calculate the importance of ERRGs. Subsequently, overlapping genes based on LASSO and RF algorithms were identified as ERRGs-related diagnostic biomarkers. In addition, OA specimens were also collected and performed qRT-PCR quantitative analysis of selected ERRGs. We identified four ERRGs associated with OA risk assessment through machine learning methods, and verified the abnormal expressions of these screened markers in OA patients through in vitro experiments. The influence of selected markers on OA immune infiltration was also evaluated. Our results provide new evidence for the role of ER stress in the OA progression, as well as new markers and potential intervention targets for OA.

A novel EIF3C-related CD8+ T-cell signature in predicting prognosis and immunotherapy response of nasopharyngeal carcinoma

PurposeAt present, dysfunctional CD8+ T-cells in the nasopharyngeal carcinoma (NPC) tumor immune microenvironment (TIME) have caused unsatisfactory immunotherapeutic effects, such as a low response rate of anti-PD-L1 therapy. Therefore, there is an urgent need to identify reliable markers capable of accurately predicting immunotherapy efficacy.MethodsUtilizing various algorithms for immune-infiltration evaluation, we explored the role of EIF3C in the TIME. We next found the influence of EIF3C expression on NPC based on functional analyses and RNA sequencing. By performing correlation and univariate Cox analyses of CD8+ Tcell markers from scRNA-seq data, we identified four signatures, which were then used in conjunction with the lasso algorithm to determine corresponding coefficients in the resulting EIF3C-related CD8+ T-cell signature (ETS). We subsequently evaluated the prognostic value of ETS using univariate and multivariate Cox regression analyses, Kaplan–Meier curves, and the area under the receiver operating characteristic curve (AUROC).ResultsOur results demonstrate a significant relationship between low expression of EIF3C and high levels of CD8+ T-cell infiltration in the TIME, as well as a correlation between EIF3C expression and progression of NPC. Based on the expression levels of four EIF3C-related CD8+ T-cell marker genes, we constructed the ETS predictive model for NPC prognosis, which demonstrated success in validation. Notably, our model can also serve as an accurate indicator for detecting immunotherapy response.ConclusionOur findings suggest that EIF3C plays a significant role in NPC progression and immune modulation, particularly in CD8+ T-cell infiltration. Furthermore, the ETS model holds promise as both a prognostic predictor for NPC patients and a tool for adjusting individualized immunotherapy strategies.

Identification of cuproptosis-related lncRNAs with the significance in prognosis and immunotherapy of oral squamous cell carcinoma

Cuproptosis, a recently characterized programmed cell death mechanism, has emerged as a potential contributor to tumorigenesis, metastasis, and immune modulation. Long non-coding RNAs (lncRNAs) have demonstrated diverse regulatory roles in cancer and hold promise as biomarkers. However, the involvement and prognostic significance of cuproptosis-related lncRNAs (CRLs) in oral squamous cell carcinoma (OSCC) remain poorly understood. Based on TCGA-OSCC data, we integrated single-sample gene set enrichment analysis (ssGSEA), the LASSO algorithm, and the tumor immune dysfunction and exclusion (TIDE) algorithm. We identified 11 CRLs through differential expression, Spearman correlation, and univariate Cox regression analyses. Two distinct CRL-related subtypes were unveiled, delineating divergent survival patterns, tumor microenvironments (TME), and mutation profiles. A robust CRL-based signature (including AC107027.3, AC008011.2, MYOSLID, AC005785.1, AC019080.5, AC020558.2, AC025265.1, FAM27E3, and LINC02367) prognosticated OSCC outcomes, immunotherapy responses, and anti-tumor strategies. Superior predictive power compared to other lncRNA models was demonstrated. Functional assessments confirmed the influence of FAM27E3, LINC02367, and MYOSLID knockdown on OSCC cell behaviors. Remarkably, the CRLs-based signature maintained stability across OSCC patient subgroups, underscoring its clinical potential for survival prediction. This study elucidates CRLs' roles in TME of OSCC and establishes a potential signature for precision therapy.

Validation and identification of anoikis-related lncRNA signatures for improving prognosis in clear cell renal cell carcinoma.

Clear cell carcinoma (ccRCC) usually has a high metastasis rate and high mortality rate. To enable precise risk stratification, there is a need for novel biomarkers. As one form of apoptosis, anoikis results from the disruption of cell-cell connection or cell-ECM attachment. However, the impact of anoikis-related lncRNAs on ccRCC has not yet received adequate attention. The study utilized univariate Cox regression analysis in order to identify the overall survival (OS) associated anoikis-related lncRNAs (ARLs), followed by the LASSO algorithm for selection. On this basis, a risk model was subsequently established using five anoikis-related lncRNAs. To dig the inner molecular mechanism, KEGG, GO, and GSVA analyses were conducted. Additionally, the immune infiltration landscape was estimated using the ESTIMATE, CIBERSORT, and ssGSEA algorithms. The study constructed a novel risk model based on five ARLs (AC092611.2, AC027601.2, AC103809.1, AL133215.2, and AL162586.1). Patients categorized as low-risk exhibited significantly better OS. Notably, the study observed marked different immune infiltration landscapes and drug sensitivity by risk stratification. Additionally, the study preliminarily explored potential signal pathways associated with risk stratification. The study exhibited the crucial role of ARLs in the carcinogenesis of ccRCC, potentially through differential immune infiltration. Furthermore, the established risk model could serve as a valuable stratification factor for predicting OS prognosis.

Expression of overall survival-EMT-immune cell infiltration genes predict the prognosis of glioma

This study investigates the crucial role of immune- and epithelial-mesenchymal transition (EMT)-associated genes and non-coding RNAs in glioma development and diagnosis, given the challenging 5-year survival rates associated with this prevalent CNS malignant tumor. Clinical and RNA data from glioma patients were meticulously gathered from CGGA databases, and EMT-related genes were sourced from dbEMT2.0, while immune-related genes were obtained from MSigDB. Employing consensus clustering, novel molecular subgroups were identified. Subsequent analyses, including ESTIMATE, TIMER, and MCP counter, provided insights into the tumor microenvironment (TIME) and immune status. Functional studies, embracing GO, KEGG, GSVA, and GSEA analyses, unraveled the underlying mechanisms governing these molecular subgroups. Utilizing the LASSO algorithm and multivariate Cox regression, a prognostic risk model was crafted. The study unveiled two distinct molecular subgroups with significantly disparate survival outcomes. A more favorable prognosis was linked to low immune scores, high tumor purity, and an abundance of immune infiltrating cells with differential expression of non-coding RNAs, including miRNAs. Functional analyses illuminated enrichment of immune- and EMT-associated pathways in differentially expressed genes and non-coding RNAs between these subgroups. GSVA and GSEA analyses hinted at abnormal EMT status potentially contributing to glioma-associated immune disorders. The risk model, centered on OS-EMT-ICI genes, exhibited promise in accurately predicting survival in glioma. Additionally, a nomogram integrating the risk model with clinical characteristics demonstrated notable accuracy in prognostic predictions for glioma patients. In conclusion, OS-EMT-ICI gene and non-coding RNA expression emerges as a valuable indicator intricately linked to immune microenvironment dysregulation, offering a robust tool for precise prognosis prediction in glioma patients within the OBMRC framework.

An exosome-derived lncRNA signature identified by machine learning associated with prognosis and biomarkers for immunotherapy in ovarian cancer.

Ovarian cancer (OC) has the highest mortality rate among gynecological malignancies. Current treatment options are limited and ineffective, prompting the discovery of reliable biomarkers. Exosome lncRNAs, carrying genetic information, are promising new markers. Previous studies only focused on exosome-related genes and employed the Lasso algorithm to construct prediction models, which are not robust. 420 OC patients from the TCGA datasets were divided into training and validation datasets. The GSE102037 dataset was used for external validation. LncRNAs associated with exosome-related genes were selected using Pearson analysis. Univariate COX regression analysis was used to filter prognosis-related lncRNAs. The overlapping lncRNAs were identified as candidate lncRNAs for machine learning. Based on 10 machine learning algorithms and 117 algorithm combinations, the optimal predictor combinations were selected according to the C index. The exosome-related LncRNA Signature (ERLS) model was constructed using multivariate COX regression. Based on the median risk score of the training datasets, the patients were divided into high- and low-risk groups. Kaplan-Meier survival analysis, the time-dependent ROC, immune cell infiltration, immunotherapy response, and immune checkpoints were analyzed. 64 lncRNAs were subjected to a machine-learning process. Based on the stepCox (forward) combined Ridge algorithm, 20 lncRNA were selected to construct the ERLS model. Kaplan-Meier survival analysis showed that the high-risk group had a lower survival rate. The area under the curve (AUC) in predicting OS at 1, 3, and 5 years were 0.758, 0.816, and 0.827 in the entire TCGA cohort. xCell and ssGSEA analysis showed that the low-risk group had higher immune cell infiltration, which may contribute to the activation of cytolytic activity, inflammation promotion, and T-cell co-stimulation pathways. The low-risk group had higher expression levels of PDL1, CTLA4, and higher TMB. The ERLS model can predict response to anti-PD1 and anti-CTLA4 therapy. Patients with low expression of PDL1 or high expression of CTLA4 and low ERLS exhibited significantly better survival prospects, whereas patients with high ERLS and low levels of PDL1 or CTLA4 exhibited the poorest outcomes. Our study constructed an ERLS model that can predict prognostic risk and immunotherapy response, optimizing clinical management for OC patients.

A penalized variable selection ensemble algorithm for high-dimensional group-structured data.

This paper presents a multi-algorithm fusion model (StackingGroup) based on the Stacking ensemble learning framework to address the variable selection problem in high-dimensional group structure data. The proposed algorithm takes into account the differences in data observation and training principles of different algorithms. It leverages the strengths of each model and incorporates Stacking ensemble learning with multiple group structure regularization methods. The main approach involves dividing the data set into K parts on average, using more than 10 algorithms as basic learning models, and selecting the base learner based on low correlation, strong prediction ability, and small model error. Finally, we selected the grSubset + grLasso, grLasso, and grSCAD algorithms as the base learners for the Stacking algorithm. The Lasso algorithm was used as the meta-learner to create a comprehensive algorithm called StackingGroup. This algorithm is designed to handle high-dimensional group structure data. Simulation experiments showed that the proposed method outperformed other R2, RMSE, and MAE prediction methods. Lastly, we applied the proposed algorithm to investigate the risk factors of low birth weight in infants and young children. The final results demonstrate that the proposed method achieves a mean absolute error (MAE) of 0.508 and a root mean square error (RMSE) of 0.668. The obtained values are smaller compared to those obtained from a single model, indicating that the proposed method surpasses other algorithms in terms of prediction accuracy.

A new 4-gene-based prognostic model accurately predicts breast cancer prognosis and immunotherapy response by integrating WGCNA and bioinformatics analysis.

Breast cancer (BRCA) is a common malignancy in women, and its resistance to immunotherapy is a major challenge. Abnormal expression of genes is important in the occurrence and development of BRCA and may also affect the prognosis of patients. Although many BRCA prognosis model scores have been developed, they are only applicable to a limited number of disease subtypes. Our goal is to develop a new prognostic score that is more accurate and applicable to a wider range of BRCA patients. BRCA patient data from The Cancer Genome Atlas database was used to identify breast cancer-related genes (BRGs). Differential expression analysis of BRGs was performed using the 'limma' package in R. Prognostic BRGs were identified using co-expression and univariate Cox analysis. A predictive model of four BRGs was established using Cox regression and the LASSO algorithm. Model performance was evaluated using K-M survival and receiver operating characteristic curve analysis. The predictive ability of the signature in immune microenvironment and immunotherapy was investigated. In vitro experiments validated POLQ function. Our study identified a four-BRG prognostic signature that outperformed conventional clinicopathological characteristics in predicting survival outcomes in BRCA patients. The signature effectively stratified BRCA patients into high- and low-risk groups and showed potential in predicting the response to immunotherapy. Notably, significant differences were observed in immune cell abundance between the two groups. In vitro experiments demonstrated that POLQ knockdown significantly reduced the viability, proliferation, and invasion capacity of MDA-MB-231 or HCC1806 cells. Our 4-BRG signature has the potential as an independent biomarker for predicting prognosis and treatment response in BRCA patients, complementing existing clinicopathological characteristics.

Identification of a novel macrophage-related prognostic signature in colorectal cancer

Colorectal cancer (CRC) is one of the most prevalent and deadliest illnesses all around the world. Growing proofs demonstrate that tumor-associated macrophages (TAMs) are of critical importance in CRC pathogenesis, but their mechanisms remain yet unknown. The current research was designed to recognize underlying biomarkers associated with TAMs in CRC. We screened macrophage-related gene modules through WGCNA, selected hub genes utilizing the LASSO algorithm and COX regression, and established a model. External validation was performed by expression analysis using datasets GSE14333, GSE74602, and GSE87211. After validating the bioinformatics results using real-time quantitative reverse transcription PCR, we identified SPP1, C5AR1, MMP3, TIMP1, ADAM8 as potential biomarkers associated with macrophages in CRC.