DRdriver: identifying drug resistance driver genes using individual-specific gene regulatory network

Area of Science:

• Oncology
• Genomics
• Bioinformatics

Background:

• Drug resistance is a major challenge in cancer therapy.
• Genetic mutations and tumor heterogeneity contribute significantly to drug resistance.
• Personalized approaches are needed to identify individual-specific drivers of drug resistance.

Purpose of the Study:

• To develop and validate a computational approach, DRdriver, for identifying personalized drug resistance driver genes.
• To explore the molecular mechanisms underlying inter-patient heterogeneity in drug resistance.
• To identify novel therapeutic targets for overcoming drug resistance.

Main Methods:

• DRdriver approach utilizes individual-specific networks and genetic algorithms.
• Identified differential mutations in resistant cancer patients.
• Constructed patient-specific networks including mutated genes and their targets.
• Employed genetic algorithms to pinpoint driver genes regulating gene expression.

Main Results:

• Identified 1202 drug resistance driver genes across 8 cancer types and 10 drugs.
• Validated that identified driver genes are frequently mutated and linked to cancer development and drug resistance.
• Discovered distinct drug resistance subtypes in brain lower grade glioma based on driver gene mutational signatures and pathways.
• Characterized subtypes by variations in epithelial-mesenchyme transition, DNA damage repair, and tumor mutation burden.

Conclusions:

• The DRdriver method provides a robust framework for identifying personalized drug resistance driver genes.
• This approach facilitates a deeper understanding of the molecular basis and heterogeneity of cancer drug resistance.
• Findings pave the way for developing targeted therapies to overcome drug resistance in individual cancer patients.