The Dawn of AI-Designed Life: From Synthetic Genomes to Functional Viruses
The quest to create synthetic life has reached a new milestone. Researchers are now leveraging the power of artificial intelligence to not just edit existing genomes, but to design entirely new ones. This builds on decades of operate, beginning with the 2008 creation of the first synthetic genome of Mycoplasma genitalium, and subsequent efforts to “reboot” these genomes within living cells.
Evo2: The ‘ChatGPT Moment’ for Synthetic Genomics
A new AI model, dubbed Evo2, is at the forefront of this revolution. Trained on a staggering 9 trillion DNA base pairs from across all domains of life, Evo2 can predict the functional impacts of genetic variations and even generate entirely new genomic sequences. Genome engineer Patrick Yizhi Cai of the University of Manchester, UK, describes this as the “ChatGPT moment” for synthetic genomics, enabling scientists to “start writing things that never existed in nature.”
Beyond Tinkering: Designing Life From Scratch
Previous genome editing efforts were largely focused on “tinkering” – modifying existing genetic code. This included editing chapters of existing genomes or removing commas, as Cai puts it. More recent work has involved recoding genomes, like that of Escherichia coli, at thousands of locations. Evo2 and similar DNA language models, still, open the door to creating synthetic life forms that are fundamentally different from anything currently found in nature.
From Phages to Full Genomes: Early Successes
In 2025, researchers used earlier Evo models to design genomes for phages – viruses that infect bacteria. Remarkably, 16 out of 285 of these AI-generated designs resulted in functional viruses capable of killing bacteria. The team led by Hie and Hsu has now extended this capability to create designs for the entire genome of a Mycoplasma genitalium-inspired organism, as well as those of human mitochondria and a yeast chromosome.
The Challenges Ahead: Functionality and Complexity
Despite these advances, significant hurdles remain. Computer predictions suggest that approximately 70% of the genes in the M. Genitalium-inspired sequences appear realistic. However, even a single missing or poorly modeled essential gene can render the entire genome non-functional. As one researcher notes, you can’t design life at 70% efficiency.
the order of genes within a genome is crucial. Simply including all the necessary genes isn’t enough; their arrangement must also support life’s complex processes. The complexity gap between designing a few thousand DNA letters for a phage and creating a functional bacterial genome – even a small one – is vast.
Mycoplasma: A Key Organism in Synthetic Biology
Mycoplasma species, including Mycoplasma genitalium and Mycoplasma pneumoniae, continue to play a central role in synthetic biology research. Their relatively small genomes and lack of a cell wall craft them attractive targets for genome manipulation. Research into Mycoplasma genomes has revealed numerous repeated sequences that are important for their evolution, and molecular typing methods are being developed to track their spread and understand antimicrobial resistance.
Future Trends and the Path Forward
The future of AI-designed life hinges on scaling up genome synthesis and testing, and on developing more sophisticated models that can accurately predict genome functionality. Expect to see:
- Increased automation in genome synthesis: Reducing the cost and time required to build AI-designed genomes.
- More accurate predictive models: AI models that can better anticipate the complex interactions within a genome.
- Focus on minimal genomes: Continued research on organisms with the smallest possible genomes, like Mycoplasma, to simplify the design process.
- Integration with other technologies: Combining AI-designed genomes with advances in synthetic cells and metabolic engineering.
FAQ
Q: What is Evo2?
A: Evo2 is an AI model trained on trillions of DNA base pairs that can predict the function of genetic variations and generate new genomic sequences.
Q: Has synthetic life been created?
A: Researchers have created synthetic genomes and “rebooted” them in cells, but creating a fully synthetic, self-sustaining life form remains a significant challenge.
Q: Why are Mycoplasma species important in this research?
A: Mycoplasma have small genomes and lack a cell wall, making them easier to manipulate genetically.
Q: What are the biggest challenges in creating synthetic life?
A: Ensuring that AI-designed genomes are functional, accurately predicting gene interactions, and scaling up genome synthesis are major hurdles.
Did you understand? The first synthetic genome was created for Mycoplasma genitalium in 2008, marking a pivotal moment in synthetic biology.
Pro Tip: Understanding genome plasticity and the role of repeated sequences, as studied in Mycoplasma, is crucial for designing stable and functional synthetic genomes.
Seek to learn more about the latest breakthroughs in synthetic biology? Subscribe to our newsletter for regular updates and in-depth analysis.
