AlphaProteo generates novel proteins for biology and health research
AlphaProteo, developed by Google DeepMind, creates novel protein binders for targeted research, outperforming existing methods in binding affinity, validated through testing, with ongoing improvements and biosecurity considerations.
Read original article
AlphaProteo, a new AI system developed by Google DeepMind, is designed to create novel proteins that can bind to specific target molecules, significantly advancing research in biology and health. This system builds on previous protein structure prediction tools like AlphaFold, which provided insights into protein interactions but did not generate new proteins. AlphaProteo can produce high-strength protein binders that enhance drug development, disease understanding, and biosensor creation. It has demonstrated superior performance, achieving binding affinities 3 to 300 times better than existing methods for seven tested target proteins, including those related to cancer and viral infections. The system was validated through experimental testing, confirming its ability to design effective binders, such as those that inhibit SARS-CoV-2. Despite its successes, AlphaProteo faced challenges with certain targets, indicating areas for future improvement. The development of this technology is accompanied by a commitment to responsible use, with ongoing collaboration with experts to address biosecurity concerns. The potential applications of AlphaProteo span various fields, including diagnostics and environmental sustainability, as the team continues to refine its capabilities and expand its range of applications.
- AlphaProteo generates novel protein binders for targeted biological research.
- It outperforms existing methods in binding affinity and success rates.
- The system has been validated through experimental testing with significant results.
- Ongoing improvements aim to address challenges with difficult protein targets.
- Responsible development and biosecurity considerations are integral to its deployment.
Related
Are AlphaFold's new results a miracle?
AlphaFold 3 by DeepMind excels in predicting molecule-protein binding, surpassing AutoDock Vina. Concerns about data redundancy, generalization, and molecular interaction understanding prompt scrutiny for drug discovery reliability.
How AI Revolutionized Protein Science, but Didn't End It
Artificial intelligence, exemplified by AlphaFold2 and AlphaFold3, revolutionized protein science by accurately predicting protein structures. Despite advancements, AI complements rather than replaces biological experiments, highlighting the complexity of simulating protein dynamics.
AI Revolutionized Protein Science, but Didn't End It
Artificial intelligence, exemplified by AlphaFold2 and its successor AlphaFold3, revolutionized protein science by predicting structures accurately. AI complements but doesn't replace traditional methods, emphasizing collaboration for deeper insights.
Creating the largest protein-protein interaction dataset in the world
A-Alpha Bio, founded in 2017, is creating a large protein-protein interaction dataset using its AlphaSeq method, which employs genetically engineered yeast cells to measure interactions, enhancing drug development and biotechnology.
Show HN: An open-source implementation of AlphaFold3
AlphaFold3, an open-source project by Ligo Biosciences, aims to improve biomolecular structure prediction for single-chain proteins, with future expansions planned. It is currently in early development and not production-ready.
- Some commenters question the novelty of AlphaProteo, citing existing work in protein binders and expressing skepticism about its impact.
- Concerns are raised about the practical applications and safety of engineered proteins, particularly regarding off-target effects and biosecurity.
- Several users express interest in the potential for AlphaProteo to create highly specific protein structures for drug development.
- There is a call for more transparency regarding the validation of generated proteins and their actual performance.
- Overall, the discussion reflects a mix of optimism for AI's role in biotechnology and caution regarding its ethical and practical implications.
23 commentsThat being said, as others have commented, my hopes are that all these advancements lead finally to reliable design methods for novel biocatalysts, an area that has been stalling for decades, compared to protein folds and binders.
In the whitepaper they mention that they are novel compared to other in silico design techniques, but to my knowledge other binders to VEGF and Covid spike protein exist and would already be found in the PDB database that Deepmind trained the model on.
This is not to minimize the results- if the history of ML is anything to go by, even if AlphaProteo does not currently beat the best affinity found by in vitro screens, I do not doubt that it soon will!
[0] - https://storage.googleapis.com/deepmind-media/DeepMind.com/B...
Working in this area might also be good test of their technological approach, as small-molecule binding can be somewhat challenging, and even evolved biological systems can struggle to achieve high specificity.
I'm very interested in my research at the moment in pleiotropy, namely mapping pleiotropic effects in as many *omics/QTL measurements and complex traits as possible. This is really helpful for determining which genes / proteins to focus on for drug development.
The problem with drugs is in fact pleiotropy! A single protein can do quite a lot of things in your body, either through a causal downstream mechanism (vertical pleiotropy), or seemingly independent processes (horizontal). This limits a lot of possible drug target as the side-effect / detrimental effect may be too large.
So, if these tools can create ultra specific protein structures that somehow only bind in the areas of interest, then that would be a truly massive breakthrough.
Related
Are AlphaFold's new results a miracle?
AlphaFold 3 by DeepMind excels in predicting molecule-protein binding, surpassing AutoDock Vina. Concerns about data redundancy, generalization, and molecular interaction understanding prompt scrutiny for drug discovery reliability.
How AI Revolutionized Protein Science, but Didn't End It
Artificial intelligence, exemplified by AlphaFold2 and AlphaFold3, revolutionized protein science by accurately predicting protein structures. Despite advancements, AI complements rather than replaces biological experiments, highlighting the complexity of simulating protein dynamics.
AI Revolutionized Protein Science, but Didn't End It
Artificial intelligence, exemplified by AlphaFold2 and its successor AlphaFold3, revolutionized protein science by predicting structures accurately. AI complements but doesn't replace traditional methods, emphasizing collaboration for deeper insights.
Creating the largest protein-protein interaction dataset in the world
A-Alpha Bio, founded in 2017, is creating a large protein-protein interaction dataset using its AlphaSeq method, which employs genetically engineered yeast cells to measure interactions, enhancing drug development and biotechnology.
Show HN: An open-source implementation of AlphaFold3
AlphaFold3, an open-source project by Ligo Biosciences, aims to improve biomolecular structure prediction for single-chain proteins, with future expansions planned. It is currently in early development and not production-ready.