Design of optimal labeling patterns for optical genome mapping via information theory

Area of Science:

• Genomics
• Bioinformatics
• Information Theory

Background:

• Optical genome mapping (OGM) analyzes linearized DNA fragments with fluorescent labels for genomic insights.
• Current OGM labeling strategies rely on available biochemical methods, not necessarily optimized for specific applications.
• Applications include structural variation detection, epigenomic profiling, and microbial identification.

Purpose of the Study:

• To develop an information theory-based model for designing optimal labeling patterns in OGM.
• To enhance the accuracy and yield of OGM for diverse genomic applications.

Main Methods:

• Developed an information theory model to predict optimal DNA labeling patterns for OGM.
• Validated the model using experimental OGM on human DNA.
• Performed simulations on bacterial DNA to assess model performance.

Main Results:

• The model enables the design of labeling patterns tailored to specific applications and genomes.
• Predicted up to 10-fold improvement in OGM accuracy through optimal pattern selection.
• Demonstrated model applicability on both human and bacterial DNA.

Conclusions:

• The developed model provides a framework for optimizing OGM biochemical labeling methods.
• Optimal labeling patterns can significantly enhance OGM accuracy and yield for clinical and research applications.
• This approach holds promise for advancing epigenomic profiling and pathogen identification.