TC-HUR: A Tri-Phase Cauchy-Assisted Hunger Games Search and Unified Runge-Kutta Optimizer for Robust DNA Data Storage

Area of Science:

• Bioinformatics
• Computational Biology
• Synthetic Biology

Background:

• DNA data storage offers high theoretical density but faces challenges from biochemical constraints.
• Conventional methods struggle with computational cost and low encoding efficiency.

Purpose of the Study:

• To develop a hybrid metaheuristic algorithm (TC-HUR) for optimizing DNA sequence design.
• To simultaneously address information density and critical biophysical constraints like homopolymer length, GC content, melting temperature, and reverse-complement similarity.

Main Methods:

• Proposed the TC-HUR hybrid algorithm combining Cauchy jump-enhanced Hunger Games Search (HGS) and Runge-Kutta (RUN) operators.
• Incorporated an adaptive intensive mutation mechanism for nucleotide-level constraint refinement.
• Evaluated algorithm performance on an 1853-nucleotide dataset under various noise conditions.

Main Results:

• TC-HUR demonstrated superior average fitness compared to RUN (2.5% increase) and HGS (16.7% increase).
• Achieved reduced reverse-complement similarity (19.10%) while maintaining homopolymer length near the ideal threshold.
• Under high noise, TC-HUR achieved a normalized edit distance of 0.1290, reducing indel errors by ~14%.

Conclusions:

• The TC-HUR algorithm effectively generates biophysically synthesizable DNA codes.
• The proposed method enhances noise resilience, crucial for practical DNA data storage applications.