A bioinformatics filtering strategy for identifying radiation response biomarker candidates

Area of Science:

• Bioinformatics
• Genomics
• Radiation Oncology

Background:

• Biomarker discovery is challenged by a large number of candidates versus limited patient data.
• Filtering methodologies are crucial for identifying relevant genes and single nucleotide polymorphisms (SNPs) in radiation response.

Purpose of the Study:

• To apply bioinformatics filtering to isolate key interacting genes and SNPs involved in radiation response.
• To establish a foundation for identifying novel biomarkers using clinical datasets.

Main Methods:

• Literature review of genetic and protein correlates of radiation response across various studies.
• Analysis of public microarray datasets to identify radiation-responsive genes based on mRNA expression.
• Construction of a protein-protein interaction network using a curated database to identify core gene clusters.

Main Results:

• Identified 20 genes common to literature review and microarray data analysis.
• Generated a statistically reliable protein interaction network, identifying a core group of 7 interacting genes.
• Ranked gene importance within the network using a graph-based scoring function.

Conclusions:

• A small, tightly interacting network of 7 genes was identified as crucial for radiation response.
• This network serves as a promising candidate set for developing new radiation response biomarkers.