A versatile distance-based approach for gene expression selection across diverse biological systems

Area of Science:

• Computational Biology
• Transcriptomics
• Immunogenomics

Background:

• Differential gene expression analysis is crucial for understanding immune cell phenotypes.
• Conventional methods using log2 fold-change (log2FC) and False Discovery Rate (FDR) thresholds have limitations in capturing complex transcriptional states.

Purpose of the Study:

• To develop a novel computational method for gene selection from mRNA sequencing (mRNA-seq) data.
• To address the limitations of traditional differential gene expression analysis.
• To enable more refined and biologically coherent gene-marker selection.

Main Methods:

• Developed the Cartesian Distance-Based Gene Expression (CDBGE) selector, a computational method for gene selection from mRNA-seq data.
• The CDBGE selector utilizes multidimensional expression distances instead of univariate statistical cutoffs.
• Applied the CDBGE selector to macrophage polarization datasets and human embryonic stem cell differentiation datasets.

Main Results:

• The CDBGE selector was used to create a gene-based framework for distinguishing macrophage polarization states.
• The method was validated using in vitro human macrophages and tested on stem cell differentiation data.
• Compared to standard pipelines, CDBGE selector identified subtype-specific markers more effectively and revealed dynamic transcriptional transitions.

Conclusions:

• Distance-based gene selection offers an improved strategy for analyzing complex mRNA-seq datasets.
• The CDBGE selector is a robust, scalable, and broadly applicable tool for differential gene expression analysis and phenotype characterization.
• This method enhances the characterization of immune cell phenotypes and transcriptional dynamics.