JoVE - Journal of Visualized Experiments
JoVE Visualize
Contact Us

DriverDetector: An R package providing multiple statistical methods for cancer driver genes detection and tools for downstream analysis

  • 0Faculty of Electronic Information and Electrical Engineering, Dalian University of Technology, Lingshui Street, Dalian, 116024, Liaoning, China.
Heliyon +

|

January 16, 2025

|

January 16, 2025

View abstract on PubMed

Summary

This summary is machine-generated.

Identifying cancer driver genes is challenging. DriverDetector, an R package, uses a voting strategy integrating 11 methods for robust detection, improving consistency and aiding targeted drug development.

Area Of Science

  • Genomics
  • Bioinformatics
  • Cancer Research

Background

  • Cancer driver gene identification is complex due to tumor heterogeneity and gene interactions.
  • Existing algorithms lack consistent results, necessitating improved detection methods.
  • Increasing availability of sequencing data drives the need for advanced computational approaches.

Purpose Of The Study

  • To develop DriverDetector, an R package for reliable cancer driver gene detection and analysis.
  • To integrate multiple statistical methods for robust identification of driver genes.
  • To provide a user-friendly workflow for cancer genomics research.

Main Methods

  • Developed a background mutation rate module using covariate space distance and binomial tests.
  • Integrated 11 driver gene identification methods, including novel Fisher's method variants.
  • Applied a voting strategy combining 10 statistical methods for enhanced prediction consistency.

Main Results

  • Verification on 12 TCGA datasets showed significant variation in gene sets identified by individual methods.
  • The voting strategy demonstrated superior consistency in driver gene prediction.
  • Sample size was found to significantly impact the number of predicted driver genes.

Conclusions

  • DriverDetector offers a robust and user-friendly approach to cancer driver gene detection.
  • The voting strategy enhances prediction reliability and consistency.
  • This tool can facilitate early cancer diagnosis and targeted therapy development.

Keywords: