Ancestral Genome Reconstruction Analysis Based on Artificial Intelligence and Evolutionary Algorithms

Abstract Ancestral genome reconstruction is a critical area of research for understanding evolutionary processes and genomic adaptations. This study presents a novel evaluation framework leveraging the Improved Whale Optimization Algorithm-Deep Belief Network (IWOA-DBN) to assess the performance of ancestral genome reconstruction. As a evolutionary algorithm, the IWOA algorithm enhances the optimization of initial parameters for the DBN by integrating advanced techniques such as nonlinear convergence mechanisms, chaotic disturbance, and improved population diversity strategies. These enhancements improve the DBN's ability to process complex genomic data and extract deep features, ensuring more accurate and reliable performance evaluations. The IWOA-DBN model combines the robust feature learning capabilities of Deep Belief Networks with the adaptive optimization strengths of the IWOA, forming a comprehensive solution for analyzing reconstruction outcomes. Systematic experiments were conducted to evaluate the reconstruction accuracy and computational efficiency of the proposed method compared to traditional approaches. The results demonstrate that IWOA-DBN significantly improves the reliability and precision of performance evaluations, highlighting its potential as a powerful tool for analyzing ancestral genome structures and evolutionary relationships. This work provides an effective strategy for addressing the challenges of genome reconstruction evaluation using artificial intelligence and evolutionary algorithm techniques.