JoVE - Journal of Visualized Experiments
JoVE Visualize
Contact Us

Identification of potential biomarkers for hepatocellular carcinoma based on machine learning and bioinformatics analysis

  • 0Department of Hepatobiliary Surgery, Northern Jiangsu People's Hospital Affiliated to Yangzhou University, Yangzhou, China.
Discover Oncology +

|

December 18, 2024

|

December 18, 2024

View abstract on PubMed

Summary

This summary is machine-generated.

Hepatocellular carcinoma (HCC) metastasis biomarkers are crucial for patient survival. This study identifies four key genes (SPP1, TYMS, HMMR, MYCN) as prognostic markers for HCC metastasis using machine learning.

Area Of Science

  • Oncology
  • Genomics
  • Bioinformatics

Background

  • Metastasis is the primary driver of mortality in hepatocellular carcinoma (HCC).
  • Effective biomarkers for predicting HCC metastasis are currently underexplored.
  • Identifying reliable prognostic indicators is critical for improving patient outcomes.

Purpose Of The Study

  • To screen and validate candidate genes associated with HCC metastasis.
  • To construct a metastasis-derived prognostic signature (MDPS) using machine learning.
  • To evaluate the potential of identified genes as prognostic biomarkers and therapeutic targets in HCC.

Main Methods

  • Integration of Gene Expression Omnibus (GEO) and The Cancer Genome Atlas (TCGA) datasets.
  • Application of machine learning algorithms to construct a consensus MDPS.
  • Stratification of HCC patients into high-risk and low-risk groups based on risk scores.
  • Comprehensive analysis of survival outcomes, clinical characteristics, and immune cell infiltration.
  • In vitro experiments to validate the role of specific genes in tumor progression.

Main Results

  • A four-gene consensus MDPS comprising SPP1, TYMS, HMMR, and MYCN was successfully constructed.
  • The MDPS effectively stratified HCC patients into distinct prognostic groups.
  • The identified genes demonstrated potential as prognostic biomarkers for HCC metastasis.
  • In vitro studies confirmed that HMMR overexpression promotes HCC progression, including proliferation, migration, and invasion.

Conclusions

  • The developed machine learning-based program provides a robust tool for constructing prognostic signatures in HCC.
  • The four-gene MDPS (SPP1, TYMS, HMMR, MYCN) shows promise as an efficient prognostic biomarker and potential therapeutic target for HCC metastasis.
  • Targeting HMMR may represent a viable strategy to inhibit HCC tumor progression.

Keywords: