Identification of immune-related hub genes in hepatocellular carcinoma through integrated non-negative matrix factorization and electrochemical sensor: Tumor immune microenvironment

The treatment options for hepatocellular carcinoma (HCC) are limited and the prognosis is poor. The tumor immune microenvironment plays a key role in the progression and treatment response of HCC. Identifying immune-related central genes may provide new biomarkers and therapeutic targets for HCC. As an emerging medical testing technology, electrochemical sensors can achieve highly sensitive and specific detection of biomolecules, and have great application potential in the early diagnosis and treatment monitoring of diseases. The study identified immune-related central genes in HCC by combining the analysis methods of non-negative matrix factor (NMF) clustering and electrochemical sensors, and explored their potential applications in the classification of HCC immune subtypes and therapeutic targeting. NMF cluster analysis revealed two immune subtypes (C1 and C2) with significant differences, and differential expression analysis indicated significant transcriptome differences between the subtypes. Immune infiltration analysis indicated that the two subtypes had significant differences in the infiltration patterns of ten types of immune cells. By integrating three machine learning methods, five robust immune-related central genes were identified. These genes demonstrated consistent discriminatory ability in all three algorithms, with the area under the ROC curve exceeding 0.8, and can be used as potential biomarkers for the classification of HCC immune subtypes and therapeutic targeting. In terms of the application of electrochemical sensors, through theoretical analysis and literature review, the potential application of electrochemical sensors in the diagnosis and treatment monitoring of HCC is emphasized, especially in the detection of low-concentration biomarkers and real-time monitoring of cell physiological states, which have significant advantages. These findings provide computational evidence for the role of these genes in the immunobiology of HCC, although experimental verification is needed. The application prospects of electrochemical sensors in medical laboratory tests are broad, especially in the early diagnosis of HCC and the monitoring of the immune microenvironment. Figure 1. NMF clustering analysis identifies two distinct HCC molecular subtypes. (A) NMF rank survey evaluating factorization ranks 2-6 across seven quality metrics: cophenetic correlation, dispersion, explained variance (evar), residuals, residual sum of squares (rss), silhouette width, and sparseness. The cophenetic coefficient peaks at rank 2, supporting selection of k=2. (B) Consensus matrix heatmap for k=2 showing sample co-clustering frequencies across 100 NMF runs. Dark red blocks along the diagonal indicate high within-subtype consensus (values approaching 1.0), while blue off-diagonal regions show low between-subtype consensus (values near 0), confirming two stable subtypes. The basis scale bar indicates consensus values from -0.49 (blue) to 1 (red), with silhouette values showing sample cluster membership strength. (C) NMF coefficient matrix heatmap displaying the two latent factors (matrix_$ values, rows) across all 24 samples (columns). Hierarchical clustering reveals two major sample groups corresponding to subtypes C1 and C2. Blue indicates low coefficient values (-1) and red indicates high values (3). (D) NMF basis matrix heatmap showing the top 50 genes (rows) and their weights for the two NMF factors (columns). Genes display preferential association with either factor, with blue representing negative weights (-2) and red representing positive weights (4), defining subtype-specific gene expression signatures. • The tumor immune microenvironment plays a critical role in HCC progression and therapeutic response. • Identification of immune-related hub genes may provide novel biomarkers and therapeutic targets. • The findings provide computational evidence for the role of these genes in HCC immune biology.