|
|
By Jeff Brown, Editor, The Bleeding Edge |
|
|
In August 2006, a Japanese researcher named Shinya Yamanaka published some groundbreaking research out of the Nara Institute of Science and Technology. |
|
Yamanaka demonstrated that adult human cells – somatic cells – could be exposed to four different kinds of factors (proteins), which would induce the adult cells into pluripotent stem cells, hence induced pluripotent stem cells (iPSCs). |
|
Source: BioPharma Trend |
This was a remarkable development, because iPSCs act much like embryonic stem cells – in that they can become any kind of human cell or tissue. |
The four natural factors (proteins) were Oct3/4, Sox2, c-Myc, and Klf4. |
They have since become known as the Yamanaka factors. |
Clearing the Path to Reversing Aging |
Yamanaka's research was groundbreaking for two reasons: |
- It meant that iPSCs could be created with each patient's own cells, and the use of embryonic stem cells was no longer necessary. It therefore eliminated the ethical concerns and issues with harvesting and using embryonic stem cells for therapeutic applications.
- It provided a clear path forward towards reversing aging, specifically by rejuvenating cells and replacing damaged cells and tissues.
|
For his work, Yamanaka won the Nobel Prize in Medicine in 2012. |
|
Source: Nobel Prize |
There is no greater human ambition than reversing aging and extending human longevity. |
Long thought to be impossible, we have now proven that it is not only possible, but it can be scaled and applied to both humans and animals. |
But there was one major challenge with Yamanaka's discovery… |
Using the Yamanaka factors proved that adult cells could be transformed into immature pluripotent stem cells,but the process is extremely inefficient. |
The process is so inefficient… typically, less than 0.1% of adult cells are successfully converted into iPSCs using the Yamanaka factors. |
Making matters worse, one of the factors (proteins), namely c-Myc, is linked to cancer. |
This helps explain why not much has happened since the research was published in 2012. |
That is, until just this last month. |
It appears that the challenge in Yamanaka's approach has just been overcome, thanks to the help of a highly unlikely company with no background in life sciences whatsoever… |
OpenAI. |
|
Can AI Biohack Human Longevity? |
Human longevity and the reversal of aging have long been a passion of the tech elite in Silicon Valley. |
Regardless of their background, there has been an extraordinary level of interest in investing in and funding any promising approach to reverse aging and extend human longevity. |
It's completely logical. There is clearly a lot of money to be made in the pursuit, and an extended lifespan gives builders and inventors the ability to pursue even more ambitious projects in their lifetime. What more motivation would you need? |
|
Not much was known about Retro, other than its focus on age-related diseases and desire to extend the human lifespan using its therapies. |
Altman's connection between the two companies – OpenAI and Retro – led to the development of an entirely new AI model designed to develop new proteins that could potentially solve the challenge of the inefficiency of the Yamanaka factors. |
OpenAI took its GPT-4o large language model, scaled it down, and then trained the new model – GPT-4b micro – on a large dataset comprised of protein sequences, biological and life sciences data, and 3D structure data. |
|
GPT-4b micro was designed with one critical objective: To re-engineer the Yamanaka factors (proteins) into something that could radically improve conversion efficiency to iPSCs. |
The results announced in the last month are simply astonishing. |
A 50X Improvement |
GPT-4b created newly engineered factors, RetroSOX and RetroKLF, that demonstrated a 50X improvement in stem cell reprogramming to iPSCs. |
Not 5%, not 50%, a 50X improvement. |
The picture below shows how dramatic this breakthrough really is. |
|
Source: OpenAI |
In the pictures above, the increase in the round white cells is an indication of the reprogramming of cells to iPSCs. The image on the left is the tissue to be treated. |
In the middle, we can see the tissue with cells that have been exposed to traditional Yamanaka factors with low efficiency of reprogramming. We can see visually that there hasn't been much reprogramming of the cells. |
And on the far right, we can see what happens when the tissue is exposed to the AI-designed RetroSOX and RetroKLF factors. |
What a radical improvement, and it was all done with a custom AI model developed by OpenAI. |
What was most interesting about what GPT-4b micro did was how radical its solution was. The Yamanaka factors SOX2 and KLF4 contain 317 and 513 amino acids, respectively. That means that the number of possible variants that could be engineered for those factors is on the order of 101000. That's 1 followed by 1,000 zeros, an impossible number of variants to explore for any human team of researchers. |
Historically, human researchers would make only a few changes to the proteins for analysis, which meant that the proteins were largely similar to the original Yamanaka factors. But the AI wasn't bound in its thinking, and it created radically different proteins by changing more than 100 amino acids, on average. |
What GPT-4b micro could accomplish by "thinking" entirely out of the box has reignited the possibilities of reversing aging – using our own cells to create iPSCs – and for addressing age-related diseases. |
Better yet, OpenAI and Retro's research demonstrated that the use of RetroSOX and RetroKLF reduces any DNA damage in the process, which was another challenge with the original Yamanaka factors. |
To say that this breakthrough is life-changing would be an understatement. |
Retro Biosciences, which is still private, is now at the forefront of iPSC research. And its breakthrough will certainly reinvigorate other companies working in this space. I absolutely expect investment levels to increase as a result. |
Who's interested in living longer and healthier lives? |
Jeff |
|