AlphaFold 3
In 2018, Google subsidiary DeepMind developed a purpose-built AI tool to predict the shapes into which different proteins could fold, called AlphaFold.
The upgraded AlphaFold 2 followed two years later.
Many scientists and technologists acknowledge that these two deep-learning systems have transformed human awareness of protein structures
AlphaFold 3, which can reportedly predict the shapes with nearly 80% accuracy as well as model DNA, RNA, ligands, and modifications to them.
As with the first two AlphaFolds, no. 3 is great for being able to elucidate the folded proteins’ structures in seconds rather than the years humans have required with advanced microscopic techniques.
AlphaFold 3's Release and Controversy
In May 2024, Google DeepMind released AlphaFold 3
This model was built upon previous versions, AlphaFold and AlphaFold 2, both of which were released open-source for broader scientific use.
However, AlphaFold 3 differed as it did not release the full code, restricting scientists’ ability to fully explore and use its capabilities.
DeepMind chose to withhold the full code due to commercial interests, as a spinoff company, Isomorphic Labs, is using the technology to develop drugs.
The decision to limit access was made to protect Isomorphic's competitive advantage while ensuring that AlphaFold 3 could still benefit the scientific community.
Many scientists expressed concerns over the lack of transparency, as it hinders verification and reproduction of the findings.
A group of researchers signed an open letter calling for full access to the code, emphasizing that withholding such crucial information goes against the principles of scientific progress.
Tension Between Science and Commercialization
A fundamental issue is the tension between intellectual property (IP) protection for commercialization and the open, transparent nature of scientific research.
While commercial interests drive secrecy, science thrives when methods and results are reproducible and shared openly for further discovery and improvement.
Universities and research institutions face financial pressure and often encourage researchers to patent their work
This situation puts scientists in a difficult position, as they must balance the pursuit of knowledge with the need for funding and commercial success.
Balancing Transparency and Commercialization
Some researchers advocate for releasing the basic algorithms and code behind their discoveries but keeping more developed, ready-to-use versions that could be commercialized.
Publishing open-source software for scientific use and then sells a more refined, robust version for commercialization is a good option
This approach allows scientists to fulfill their mission of advancing knowledge while also tapping into the financial benefits of commercialization.
There is always the risk that someone else might commercialize the published work, but publishing key algorithms ensures that the scientific community can still verify and build upon the findings.
COMMENTS