Scientists have discovered a new capability of AI in biology. For the very first time, scientists have created viruses designed by artificial intelligence (AI), and are capable of hunting down and killing strains of Escherichia coli (E. coli).
Explaining this invention, Brian Hie, a computational biologist at Stanford University, said that this is the first time AI systems are able to write coherent genome-scale sequences and that the next step is AI-generated life.
Adding to this, Samuel King, his colleague, said that a lot more experimental advances are needed in order to design an entire living organism.
The study is posted on the preprint server bioRxiv on September 17, but is not yet peer reviewed. But the authors confirmed that it shows the potential of AI to design biotechnological tools and therapies for treating bacterial infections.
Hie said that strategies like this could complement present phage–therapy strategies and someday augment the therapeutics to target pathogens of concern.
Till now, AI models have been used to generate single proteins, DNA sequences and multi-component complexes. But it is much more complex to design complex interactions between genes, gene replication and regulating processes like that in a whole genome.
AI can now help scientists to manipulate extremely complex biological systems like a whole genome.
How did they design the viral genome?
Scientists used Evo 1 and Evo 2 AI models that could analyse and generate RNA, DNA and protein sequences with realistic genetic architectures and desirable host tropism by using the lytic phage ΦX174 as a design template.
Phage ΦX174 infects E.coli and was the first ever genome to be sequenced completely in 1977, and the first one to be synthesised from scratch in 2003, now the first to be designed by AI.
Results:
- After testing experimentally, the AI-generated genome produced 16 viable phages with substantially evolutionarily distant DNA packaging proteins within its capsid.
- Many phages demonstrated a higher fitness than ΦX174 in growth competitions and in their lysis kinetics.
- A mixture of the generated phages instantly overcomes ΦX174-resistance in strains of E. coli, indicating the potential use of this new approach for designing phage therapy to deal with rapidly evolving bacterial pathogens.
Applications of phage design:
- Phage therapeutics
- Studying evolution
- Synthetic biology
- Modeling methods
This approach can help design a variety of synthetic bacteriophages and, more broadly, initiate the generative design of useful living systems at the genome scale.
