Metabolic diseases like obesity and type 2 diabetes affect millions of people worldwide, presenting significant health challenges and economic burdens. A key issue in these conditions is the loss of metabolic flexibility—the body's ability to adapt to changes in nutrient availability, such as food scarcity or abundance. Throughout evolution, adapting to varying nutrient levels has been crucial for the survival of species. However, the precise molecular mechanisms that link this adaptive flexibility to disease and evolutionary processes have remained unknown. Here, we investigated how energy conservation states influence gene organization and contribute to resilient traits associated with rapidly evolving genomic regions called human accelerated regions (HARs).

We discovered that adipocytes (fat cells) help organisms adapt to nutrient stress through specific genomic hubs involving metabolic HARs (mHARs) and cooperative regulators. During nutrient scarcity, these mHAR genomic hubs become active, forming transcriptional compartments that co-regulate genes involved in lipid metabolism. This activity reduces harmful fat accumulation and oxidative stress, preserving cellular flexibility and resilience.

By combining functional genomics, computational modeling, and experimental biology, we identified key genomic strategies that enhance metabolic plasticity and resilience to nutrient stress. Our findings suggest that energy conservation operates within evolutionary constraints, possibly through selection on conserved molecular mechanisms fine-tuned across species.

Our computational and experimental framework uncovers the molecular principles that govern adaptive strategies in biological systems. By decoding the genomic adaptations that have been honed by evolution, scientists can pave the way for innovative treatments for metabolic diseases and enhance our overall understanding of human biology. Our study offers new insights into the interplay between energy conservation, genomic regulation, and specialization of function, providing potential therapeutic targets for metabolic diseases and bioengineering.

Importantly, our work identifies principles computing adaptive strategies for resilience in cells which can be generalize to other complex systems. By understanding how cells compute and implement these adaptive responses, we can inform the development of computational systems that mimic these natural processes.