PMCID
string
Title
string
Sentences
string
PMC12468391
PA2G4 Functions as a Cofactor for MYC Family Oncoproteins in MYC-Driven Malignancies
Our study builds on this emerging evidence and further supports PA2G4 as an oncogenic cofactor in MYC-driven cancers.
PMC12468391
PA2G4 Functions as a Cofactor for MYC Family Oncoproteins in MYC-Driven Malignancies
We show that PA2G4 plays a critical role in stabilizing MYC proteins, reinforcing its function as a MYC family co-regulator.
PMC12468391
PA2G4 Functions as a Cofactor for MYC Family Oncoproteins in MYC-Driven Malignancies
Previous studies have shown that PA2G4 interacts with FBXW7α, sequestering it in the cytoplasm and thereby preventing it from targeting MYC for proteasomal degradation .
PMC12468391
PA2G4 Functions as a Cofactor for MYC Family Oncoproteins in MYC-Driven Malignancies
This interaction effectively prolongs the half-life of MYC proteins, enhancing their oncogenic potential.
PMC12468391
PA2G4 Functions as a Cofactor for MYC Family Oncoproteins in MYC-Driven Malignancies
This interaction leads to increased MYC protein stability and sustained oncogenic signaling.
PMC12468391
PA2G4 Functions as a Cofactor for MYC Family Oncoproteins in MYC-Driven Malignancies
By shielding MYC from degradation, PA2G4 amplifies MYC-driven transcriptional programs and tumorigenic potential.
PMC12468391
PA2G4 Functions as a Cofactor for MYC Family Oncoproteins in MYC-Driven Malignancies
Taken together with prior studies, our findings in neuroblastoma highlight PA2G4 as a conserved and functionally significant MYC regulatory factor which may have a role in many c-MYC-driven malignancies.
PMC12468391
PA2G4 Functions as a Cofactor for MYC Family Oncoproteins in MYC-Driven Malignancies
Targeting the PA2G4–MYC axis represents a promising therapeutic strategy for MYC-driven cancers.
PMC12468391
PA2G4 Functions as a Cofactor for MYC Family Oncoproteins in MYC-Driven Malignancies
One promising therapeutic avenue is the use of PA2G4 inhibition to enhance the efficacy of conventional chemotherapies.
PMC12468391
PA2G4 Functions as a Cofactor for MYC Family Oncoproteins in MYC-Driven Malignancies
In cancers where MYC plays a critical role in tumor growth and chemoresistance, combining PA2G4 inhibition with chemotherapy could provide a more effective strategy for overcoming drug resistance and improving patient outcomes.
PMC12468391
PA2G4 Functions as a Cofactor for MYC Family Oncoproteins in MYC-Driven Malignancies
Additionally, the potential synergy between PA2G4 inhibitors and other targeted therapies, such as those aimed at MYC-interacting proteins like FBXW7 or Aurora A, could further enhance the therapeutic benefit.
PMC12468391
PA2G4 Functions as a Cofactor for MYC Family Oncoproteins in MYC-Driven Malignancies
These combinatorial approaches warrant further investigation as a strategy to exploit MYC-driven vulnerabilities and broaden treatment options for patients with high-risk malignancies.
PMC12468391
PA2G4 Functions as a Cofactor for MYC Family Oncoproteins in MYC-Driven Malignancies
While our findings highlight the therapeutic potential of targeting PA2G4 in MYC-driven cancers, several important limitations remain.
PMC12468391
PA2G4 Functions as a Cofactor for MYC Family Oncoproteins in MYC-Driven Malignancies
Although we demonstrated that PA2G4 is essential for maintaining c-MYC stability and function in vitro, the in vivo efficacy of PA2G4 inhibition in human c-MYC-driven tumors has yet to be fully established.
PMC12468391
PA2G4 Functions as a Cofactor for MYC Family Oncoproteins in MYC-Driven Malignancies
Previous studies have shown that WS6 is effective in MYCN-driven neuroblastoma models; however, its activity in other MYC-driven cancers remains to be determined.
PMC12468391
PA2G4 Functions as a Cofactor for MYC Family Oncoproteins in MYC-Driven Malignancies
Moreover, we observed that WS6-induced cytotoxicity varied across cancer cell lines, suggesting the possibility of off-target effects or cell line-specific differences in dependency on PA2G4.
PMC12468391
PA2G4 Functions as a Cofactor for MYC Family Oncoproteins in MYC-Driven Malignancies
These findings underscore the need to further characterize the broader network of PA2G4 interactions, particularly with proteins involved in MYC regulation, to better understand its context-specific roles in oncogenesis.
PMC12468391
PA2G4 Functions as a Cofactor for MYC Family Oncoproteins in MYC-Driven Malignancies
A major limitation of WS6 as a therapeutic agent is its known toxicity in normal tissues, which significantly restricts its translational potential.
PMC12468391
PA2G4 Functions as a Cofactor for MYC Family Oncoproteins in MYC-Driven Malignancies
To overcome this, our previous work focused on developing and characterizing WS6 analogs with improved selectivity, reduced toxicity, and enhanced drug-like properties .
PMC12468391
PA2G4 Functions as a Cofactor for MYC Family Oncoproteins in MYC-Driven Malignancies
These efforts aim to generate more clinically viable PA2G4 inhibitors for future therapeutic application.
PMC12468391
PA2G4 Functions as a Cofactor for MYC Family Oncoproteins in MYC-Driven Malignancies
To further advance this therapeutic strategy, structural studies of the PA2G4–MYC complex will be crucial.
PMC12468391
PA2G4 Functions as a Cofactor for MYC Family Oncoproteins in MYC-Driven Malignancies
Future studies will focus on elucidating the precise mechanism by which PA2G4 regulates MYC stability, including evaluation of MYC phosphorylation status and its regulation via the proteasome pathway in PA2G4-deficient and -proficient contexts across multiple cancer types.
PMC12468391
PA2G4 Functions as a Cofactor for MYC Family Oncoproteins in MYC-Driven Malignancies
High-resolution structural analyses could reveal key binding sites and conformational changes that occur upon interaction, providing a blueprint for rational drug design.
PMC12468391
PA2G4 Functions as a Cofactor for MYC Family Oncoproteins in MYC-Driven Malignancies
Ultimately, understanding the structural basis of PA2G4′s oncogenic function will be essential for developing potent and selective inhibitors that disrupt its role in cancer while minimizing effects on its physiological functions.
PMC12468391
PA2G4 Functions as a Cofactor for MYC Family Oncoproteins in MYC-Driven Malignancies
In summary, our study identifies PA2G4 as a novel and critical regulator of MYC protein stability, establishing it as a promising therapeutic target in MYC-driven cancers.
PMC12468391
PA2G4 Functions as a Cofactor for MYC Family Oncoproteins in MYC-Driven Malignancies
By demonstrating that disruption of PA2G4 impairs MYC stability and tumor cell viability, we provide a compelling rationale for targeting the PA2G4–MYC axis as an alternative to direct MYC inhibition.
PMC12468391
PA2G4 Functions as a Cofactor for MYC Family Oncoproteins in MYC-Driven Malignancies
These findings lay the groundwork for future development of clinically viable PA2G4 inhibitors and open new avenues for the treatment of high-risk, MYC-driven malignancies.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
Diffuse large B-cell lymphoma (DLBCL) is the most common subtype of non-Hodgkin lymphoma, characterized by significant clinical and molecular heterogeneity, which leads to considerable variability in patient prognosis.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
Programmed cell death (PCD) plays a critical role in the development and progression of various cancers.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
A comprehensive analysis of PCD-related gene expression in DLBCL could enhance risk stratification and inform personalized treatment strategies.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
This study integrated five DLBCL datasets with 18 PCD-related gene expression profiles to identify differentially expressed genes (DEGs) associated with PCD.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
Patients were stratified into two subgroups (C1 and C2) using consensus clustering analysis.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
We further performed immune infiltration analysis, GSVA enrichment analysis, and WGCNA to uncover significant differences in the immune microenvironment and signaling pathways between the subgroups.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
Additionally, 12 machine learning algorithms were employed to construct predictive models for DLBCL, with performance evaluated using AUC and F-score metrics.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
Finally, transcriptome sequencing of the DLBCL cell line VAL and the normal human B lymphocyte cell line IM-9 was conducted to validate potential biomarkers.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
A total of 1074 PCD-related DEGs were identified.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
Unsupervised clustering revealed two distinct molecular subtypes of DLBCL.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
The C2 subgroup exhibited upregulation of pathways involved in DNA repair, cell cycle, and energy metabolism, alongside significant downregulation of immune evasion-related pathways, indicating its classification as a high-risk group.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
Machine learning algorithms and transcriptome sequencing validation identified five potential biomarkers for DLBCL, including CTSB, DPYD, SCARB2, STOM, and GBP1.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
This study identifies two distinct DLBCL subtypes based on PCD-related gene expression, with the C2 subtype characterized as high-risk due to enhanced DNA repair and cell cycle pathways.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
Five key biomarkers (CTSB, DPYD, SCARB2, STOM, GBP1) may improve risk stratification and understanding of DLBCL heterogeneity.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
These findings lay the groundwork for further exploration of DLBCL progression and potential prognostic improvements.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
Diffuse large B-cell lymphoma (DLBCL) is the most common subtype of non-Hodgkin lymphoma (NHL), accounting for approximately 30–40% of adult NHL cases.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
In recent years, the incidence of DLBCL has shown a gradual increase.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
According to statistics, there were 544,000 new cases of DLBCL globally in 2020, representing about 2.8% of all newly diagnosed cancers, with a mortality rate reaching 259,000 deaths.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
The prognosis of DLBCL patients is associated with baseline characteristics such as clinical stage, age, β2-microglobulin levels, and lactate dehydrogenase levels .
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
Moreover, DLBCL exhibits a high degree of phenotypic and genetic heterogeneity.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
The standard treatment regimen for DLBCL is the R-CHOP protocol, consisting of rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone, with a typical cure rate of 50–70%.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
Compared to the R-CHOP regimen as first-line treatment for DLBCL, the pola-R-CHP regimen reduces the relative risk of disease progression, relapse, or death by 27% while maintaining a similar safety profile.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
However, approximately 30–40% of patients experience disease relapse or increased drug resistance within 2 to 5 years after initial treatment, leading to poor prognosis .
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
The success of CAR-T cell therapy as a second-line treatment for relapsed or refractory large B-cell lymphoma has catalyzed its exploration as a second-line therapeutic option, particularly in clinical settings with poor outcomes from standard treatments, including refractory disease or early relapse within 12 months after first-line chemoimmunotherapy.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
Furthermore, conventional salvage immunochemotherapy followed by autologous stem cell transplantation is effective in only 10% of patients with refractory or relapsed disease .
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
Therefore, identifying highly specific biomarkers is crucial for the screening, diagnosis, and prognosis prediction of DLBCL.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
Programmed cell death (PCD) is characterized by distinct morphological features and energy-dependent biochemical mechanisms, and is recognized as a key regulatory factor in various cellular processes, including tissue growth, embryogenesis, cell turnover, and immune responses.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
To date, at least 18 distinct forms of PCD have been reported, including apoptosis, pyroptosis, ferroptosis, and cuproptosis, among others .
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
Each PCD modality operates through unique molecular mechanisms and results in specific functional outcomes, though crosstalk between different PCD pathways is also common.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
Aberrant PCD plays a critical role in a variety of human diseases, including neurodegenerative disorders, ischemic injury, autoimmune diseases, and various types of cancer .
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
Extensive research has demonstrated a close association between PCD and the progression and metastasis of multiple malignancies.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
Consequently, cancer therapies that induce cell death have become a focal point in cancer biology research.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
In recent years, machine learning has emerged as a powerful tool for identifying and prioritizing interactomic hub genes, which play central roles in protein interaction networks and biological pathways .
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
By leveraging network analysis, feature selection, and predictive modeling, machine learning algorithms can effectively uncover key genes that drive disease mechanisms and progression.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
This approach has been particularly valuable in complex diseases such as cancer, where identifying critical molecular players can inform diagnostic and therapeutic strategies.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
This study aims to construct diagnostic models for DLBCL based on PCD-related features using machine learning ensemble methods, with the goal of uncovering the potential roles of PCD-related genes in the development and progression of DLBCL.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
By improving the accuracy of DLBCL diagnosis and prediction, this research ultimately seeks to contribute to personalized and effective therapeutic interventions.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
We obtained five datasets from the Gene Expression Omnibus (GEO, http://www.ncbi.nlm.nih.gov/geo), including GSE132929, GSE43677, GSE156309, GSE190847, and GSE12453 [7–10].
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
The GSE132929 and GSE43677 datasets served as the training set (95 DLBCL and 40 normal samples), GSE156309 and GSE190847 as validation set A (61 DLBCL and 28 normal samples), and GSE12453 as validation set B (11 DLBCL and 15 normal samples).
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
RNA-seq data were corrected for batch effects and log-transformed.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
The 18 forms of programmed cell death (PCD) genes investigated, totaling 1,965, were derived from the literature .
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
After removing duplicates, 1,545 unique PCD-related genes were included (Supplementary File 1, S1).
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
The GSE132929 and GSE43677 datasets were annotated and merged using R, followed by differential expression analysis of PCD-related genes using the "limma" package.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
Differentially expressed genes (DEGs) were identified with the criteria of |log2FC|> 1 and P < 0.05.Based on the differential expression levels of DEGs in DLBCL, unsupervised consensus clustering analysis was performed using the "ConsensusClusterPlus" package, employing the "PAM" algorithm and Euclidean distance as the metric .
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
Subsequently, the clustering classification was assessed through Principal Component Analysis (PCA), and DEG expression profiles across different DLBCL subtypes were compared using the "ggpubr" package .
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
Immune cell infiltration across different subgroups was quantitatively analyzed using the CIBERSORT package, and boxplots were used to visually depict the differences in immune infiltration between the subgroups .
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
Additionally, gene set variation analysis (GSVA) was applied to correct the gene expression data of DLBCL subgroups, followed by KEGG pathway enrichment analysis.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
Pathways were considered significant if they met the criteria of P < 0.05 and |t|> 1 .
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
To better visualize the results, the top 10 upregulated and downregulated pathways were presented as bar plots using the "ggpubr" package.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
We utilized the "WGCNA" package to identify core modules and key genes between DLBCL and healthy control groups, as well as between subgroups C1 and C2 .
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
The top 25% of DEGs with the highest variance were selected, and outlier samples were removed.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
A suitable soft-thresholding power (β) was chosen to construct the gene clustering tree.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
The adjacency matrix was converted into a Topological Overlap Matrix (TOM) to assess gene relationships, and dissimilarity (1-TOM) was calculated.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
The minimum module size was set to 100 genes, and the dynamic tree cut algorithm grouped genes with similar expression patterns into modules.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
We then calculated the correlation between module eigengenes and clinical traits, evaluated the significance of each gene within the modules, and generated scatter plots for the core module genes.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
By intersecting results from different analyses, we identified the characteristic genes of DLBCL.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
We integrated 12 machine learning algorithms to construct predictive models for diffuse large B-cell lymphoma (DLBCL) based on programmed cell death (PCD)-related genes .
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
These algorithms included Support Vector Machine (SVM), Least Absolute Shrinkage and Selection Operator (Lasso), Gradient Boosting Machine (GBM), Ridge Regression, K-Nearest Neighbors (KNN), Logistic Regression (LR), Linear Discriminant Analysis (LDA), Decision Tree (DT), XGBoost, Elastic Net Regression (ENR), Stepwise Regression, and Naive Bayes .
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
All models were implemented using the respective R packages for efficient computation.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
To enhance model performance, we combined these 12 algorithms into 91 possible combinations, each evaluated through cross-validation.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
The F-score, balancing precision and recall, and the Area Under the Curve (AUC), which assesses overall classification ability, were used as key metrics to evaluate model stability and accuracy.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
The DLBCL cell line VAL (purchased from Zhejiang Meisen Cell Technology Co., Ltd.) and the human normal B lymphocyte cell line IM-9 (purchased from Guangzhou Kofan Biotechnology Co., Ltd.) were cultured at 37 ℃ in a 5% CO2 incubator.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
Cells in the logarithmic growth phase were selected for subsequent experiments.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
Total RNA was extracted from VAL and IM-9 cells using the Trizol method, and library preparation was performed by Beijing Qingke Biotechnology Co., Ltd., followed by second-generation high-throughput sequencing using the Illumina NovaSeq platform.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
After data acquisition, paired sample t-tests were conducted on the core genes using Read count values, and genes with significantly different p-values were selected for further analysis.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
The bioinformatics method and statistical method used in this paper were both carried out in R language version 4.3.3.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
Paired sample T-test was performed on the two groups of data conforming to normal distribution, and P < 0.05 indicated that the difference was statistically significant.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
Gene annotation of the GSE132929 and GSE43677 datasets resulted in the identification of 12,447 genes, from which 1545 PCD-related genes were extracted for differential expression analysis.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
A total of 1235 PCD-related DEGs were identified, with 1074 of them meeting the significance criterion of P < 0.05.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
These included 415 apoptosis-related genes, 32 pyroptosis-related genes, 58 ferroptosis-related genes, 249 autophagy-related genes, 63 necroptosis-related genes, 15 cuproptosis-related genes, 3 PARP-1-dependent cell death (parthanatos)-related genes, 11 entotic cell death-related genes, 6 netotic cell death-related genes, 169 lysosome-dependent cell death-related genes, 6 alkaliptosis-related genes, 2 oxeiptosis-related genes, 29 immunogenic cell death-related genes, 257 anoikis-related genes, 45 paraptosis-related genes, 18 NETosis-related genes, 3 methuosis-related genes, and 17 entosis-related genes (Supplementary File 2, S2).
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
We performed consensus clustering analysis based on the expression data of 1074 differentially expressed genes (DEGs) related to programmed cell death (PCD) and determined that the optimal number of clusters was k = 2.
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
At k = 2, the cumulative distribution function (CDF) curve reached its maximum (Fig. 1A), and the tracking plot showed high stability with consistent colors or markers within the same row (Fig. 1B).
PMC12003219
Predictive biomarkers and molecular subtypes in DLBCL: insights from PCD gene expression and machine learning
The heatmap also clearly divided the samples (Fig. 1C).