English
Arabic
Chinese (Simplified)
French
German
Hebrew
Italian
Portuguese
Russian
Spanish
Programming evolution
Evolution could be programmed using a hierarchy of hot-spot mutations
When we watch fireworks, we see a small speckle of white light transforming into bright yellow strings, then dissipating, then coming back to life as alternating sparkles, then exploding into a flickering rainbow ball before dying completely out. For a child, it looks like that small speckle was evolving. But, we know that this evolution is an illusion and the sequence of light was programmed by sequentially placed charges.
What about the fireworks of the evolutionary tree?
Just like a firework is programmed with sequential charges, evolution can be programmed with hot-spot mutations.
It is not controversial that genetic changes underlie evolution. These changes must occur in a certain order. Each known hot-spot sequence has a very specific frequency (probability) of taking place. Look at the below table of frequency of hot-spot onco-mutations:
If you have a sufficient number of hot-spot mutations with known frequencies you can insert them into a DNA sequence in such a way that it will evolve into a specific order over time not unlike the firework charge.
In theory, entire evolution can be programmed in this fashion, explaining its increasing complexity and variety of species. In this context, Darwinian evolution takes its appropriate place by being merely responsible for small horizontal changes without changing the complexity – survival of the fittest.
Programming evolution
Programming Evolution (GPT-4o)
Desired Sequence:
- AAG CCT GGA TTG CCA GAT
Choosing Hot-Spot Mutations with Appropriate Frequencies:
- TGA to AAG – 1 hour
- CGG to CCT – 2 hours
- TGC to GGA – 2 hours
- AAC to TTG – 1 hour
- GGA to CCA – 3 hours
- CTA to GAT – 3 hours
Create Required Starting Sequence:
- TGA CGG TGC AAC GGA CTA
Evolution Process:
- Start: TGA CGG TGC AAC GGA CTA
- After 1 Hour: AAG CGG TGC AAC GGA CTA (TGA mutated to AAG)
- After 2 Hours: AAG CCT TGC AAC GGA CTA (CGG mutated to CCT)
- After 3 Hours: AAG CCT GGA AAC GGA CTA (TGC mutated to GGA)
- After 4 Hours: AAG CCT GGA TTG GGA CTA (AAC mutated to TTG)
- After 5 Hours: AAG CCT GGA TTG CCA CTA (GGA mutated to CCA)
- After 6 Hours: AAG CCT GGA TTG CCA GAT (CTA mutated to GAT)
Based on the above calculation and the average number of codons per gene, you could theoretically program the evolution of approximately 20 genes within the constraints of GPT-4’s token limits.
Potential Timeline
Near Term (5-10 years):
- AI systems will likely become highly proficient at predicting and implementing specific mutations for targeted gene editing.
- Significant progress in cataloging hotspot mutations and understanding their roles in various organisms.
Mid Term (10-20 years):
- AI could potentially handle more complex evolutionary programming tasks, possibly managing the evolution of simple organisms or specific pathways within more complex organisms.
- Improved integration of AI with experimental biology, leading to faster and more precise genetic manipulations.
Long Term (20-50 years):
- AI might be capable of programming the evolution of entire organisms, given a comprehensive understanding of all necessary mutations and their interactions.
- Potential for creating synthetic life forms or extensively modifying existing organisms to exhibit desired traits.
Very Long Term (50+ years):
- Full-scale programming of evolution, possibly extending to multi-generational evolutionary trajectories.
- Advanced AI systems might simulate and implement evolutionary processes in silico before applying them in vivo.
Another important point is that hot-spot mutations we see today are probably just remaining and evolutionary inconsequential mutations because those responsible for evolution have already happened and are all gone: we see the smoldering coals of what once was a raging fire.
Hot-point mutations we see today
Hot-spot mutations billion years ago