Identification of Shared and Unique Key Biomarkers of Alcohol Liver Cirrhosis and Non-Alcoholic Steatohepatitis Through Machine Learning Network-Based Algorithms
Bibliographic record
Abstract
Abstract Introduction Liver fibrosis can progress to cirrhosis, liver failure, or hepatocellular carcinoma, which often requires transplantation and burdens healthcare systems around the world. Advances in single-cell RNA sequencing and machine learning have enhanced the understanding of immune responses in many liver diseases particularly alcohol liver cirrhosis (ALC) and non-alcoholic steatohepatitis (NASH). This study aims to identify key biomarkers involved in these conditions and assess their potential as non-invasive diagnostic tools. Methods Two gene expression profiles GSE136103 and GSE115469 were used to conduct differential gene expression (DEG) analysis. Using the results from DEG analysis, we then applied two machine learning network-based algorithms, master regulator analysis (MRA) and weighted key driver analysis (wKDA), to identify potential biomarker genes for NASH and ALC. Results A total of 1,435 and 5,074 DEGs were identified for ALC and NASH compared to healthy controls, including 1,077 shared DEGs between the two diseases. The MRA showed HLA-DPA1, HLA-DRB1, IFI44L, ISG15, and CD74 as the potential master regulators of ALC and HLA-DPB1, HLA-DQB1, HLA-DRB5, PFN1, and TMSB4X as the potential master regulators of NASH. In addition, wKDA analysis indicated CD300A, FCGR2A, RGS1, HLA-DMB, and C1QA as the key drivers of ALC and INPP5D, NCKAP1L, RAC2, PTPRC, and TYROBP as key drivers of NASH. Conclusion This study presented a comprehensive framework for analyzing single-cell RNA-seq data, demonstrating the potential of combining advanced network-based machine-learning techniques with conventional DEG analysis to uncover actionable prognostic markers for ALC and NASH with potential use as target biomarkers in drug development.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".