Robust identification of significant interactions between toxicogenomic biomarkers and their regulatory chemical...

Area of Science:

• Toxicogenomics
• Computational Biology
• Drug Discovery

Background:

• Identifying gene-compound interactions is crucial for toxicogenomics and drug development.
• Existing computational tools lack methods for direct gene-compound interaction identification.
• Biomarker gene and regulatory compound discovery is an active research area.

Purpose of the Study:

• To propose and evaluate novel computational approaches for identifying significant gene-compound interactions.
• To introduce the moving range chart (MRC) and logistic moving range chart (LMRC) methods.
• To enable robust identification of toxicogenomic biomarkers and their regulatory chemical compounds/drugs.

Main Methods:

• Investigated two statistical approaches: moving range chart (MRC) and logistic moving range chart (LMRC).
• Evaluated method performance using simulated toxicogenomic datasets.
• Applied the LMRC approach to real-life toxicogenomic data.

Main Results:

• MRC and LMRC performed similarly on simulated data without outliers.
• LMRC demonstrated superior performance compared to MRC in the presence of outliers.
• The LMRC approach successfully identified significant down-regulated biomarker genes in real-world data that were missed by other methods.

Conclusions:

• The logistic moving range chart (LMRC) is proposed as a robust method for identifying significant gene-compound interactions.
• LMRC offers improved accuracy in detecting toxicogenomic biomarkers and their regulators, particularly in datasets with noise or outliers.
• This approach enhances toxicogenomic data analysis and aids in drug discovery and development.