Maurizio Morri – Science Blog

A new study from Google DeepMind is pushing the boundaries of how artificial intelligence can understand biology. Scientists have trained a model called AlphaFold3 that does not just predict protein structure like its predecessor, but can now also simulate how proteins interact with DNA, RNA, and small molecules. This brings researchers a step closer to modeling complex molecular machinery inside cells.

The innovation lies in how AlphaFold3 combines physical principles with deep learning, using a single unified architecture to simulate biomolecular complexes. Instead of relying on existing templates, the model generates predictions from raw sequence data. This means it can handle previously unseen combinations of biological components, giving scientists the ability to explore novel interactions that were hard or impossible to study before.

The model has already demonstrated major improvements in accuracy over existing tools. It predicts not just the shapes of individual proteins but also how they bind to other molecules, which is essential for understanding how cells work and how diseases begin. This is also critical for designing new drugs that target specific biological pathways.

While the full version of AlphaFold3 is not yet publicly released, DeepMind has launched a service that allows researchers to use the tool for non commercial purposes. This opens the door to new research in structural biology, drug discovery, and synthetic biology, where understanding how biological parts fit and move together is key.

By giving scientists access to predictions about protein interactions on a scale never before possible, this new AI model could accelerate discoveries across medicine and biotechnology.

https://www.deepmind.com/blog/alphafold-3-a-unified-model-of-the-protein-universe

A new deep learning model called MutSig AtlasAI is changing how we understand cancer. Built by researchers at the Broad Institute and MIT, it looks at tumor DNA from thousands of patients and learns to tell which mutations actually help the tumor grow, the so called drivers, and which ones are harmless.

Most mutations in cancer are just along for the ride. The hard part is figuring out which ones matter. This tool makes that easier. It uses not just how often a mutation appears, but also where it is in the gene, how it affects protein structure, and how evolution has treated that part of the genome. It also works across different cancer types, including rare and understudied ones.

In tests, it found new driver mutations that older methods missed, especially in cancers like bile duct and soft tissue sarcomas. The model is open source and ready for other scientists to build on, making it a potentially powerful step forward in precision cancer treatment.

https://www.broadinstitute.org/news/machine-learning-tool-pinpoints-cancer-driver-mutations

In a significant advancement for prostate cancer care, researchers from the US, UK, and Switzerland have developed an artificial intelligence (AI) tool capable of predicting which men with high-risk, non-metastatic prostate cancer will benefit from the drug abiraterone. This development, unveiled at the American Society of Clinical Oncology’s annual meeting, promises to tailor treatment plans more effectively, potentially improving outcomes and reducing unnecessary side effects.

Abiraterone, often described as a “gamechanger” for advanced prostate cancer, has been shown to significantly reduce the risk of death. However, its use in earlier stages of the disease has been limited due to potential side effects, including increased risks of heart issues and diabetes. The new AI tool analyzes biopsy images to identify a biomarker indicating likely benefit from abiraterone. In a study involving over 1,000 men, the AI identified that approximately 25% had biomarker-positive tumors. For these individuals, abiraterone halved the five-year mortality rate from 17% to 9%. Conversely, for those without the biomarker, the benefit was negligible.

This breakthrough could lead to more personalized treatment strategies, ensuring that patients receive therapies most likely to benefit them while avoiding unnecessary exposure to potential side effects. It also has implications for healthcare systems, potentially prompting revisions in treatment guidelines and drug approval policies.

Experts have lauded the development, emphasizing its potential to optimize personalized cancer care and improve patient outcomes.

For more details, read the full article here:

https://www.theguardian.com/society/2025/may/30/new-ai-test-can-predict-which-men-will-benefit-from-prostate-cancer-drug

In a major step forward for synthetic biology, scientists at the Center for Genomic Regulation in Barcelona have used artificial intelligence to design brand-new DNA sequences that can control gene activity in living mammalian cells. This study, published on May 8, 2025, shows that AI can create synthetic regulatory elements—small pieces of DNA that decide when and where a gene is turned on.

The AI model developed by the team doesn’t rely on copying natural DNA. Instead, it generates new sequences from scratch that are customized to work in specific cell types. For example, the model can design a sequence that activates a gene only in stem cells that are on their way to becoming red blood cells, while leaving other cells alone.

To test their work, the researchers built these sequences in the lab and inserted them into mouse blood cells. The results confirmed that the AI-made DNA successfully activated the right genes in the right cells. That kind of control is incredibly valuable for future treatments where you want to change gene activity only in a targeted group of cells.

This kind of precision could lead to safer and more effective gene therapies, where unwanted side effects are avoided because the treatment only works where it’s needed.

AI is now helping us go beyond what nature provides, letting scientists write DNA code that behaves exactly how we want it to. That changes the game for genetics, medicine, and our ability to design life at the molecular level.

https://www.sciencedaily.com/releases/2025/05/250508112324.htm

In a groundbreaking development, researchers at the University of Missouri have unveiled an advanced artificial intelligence tool capable of predicting the three-dimensional structure of chromosomes within individual cells. This innovation offers unprecedented insights into gene organization and function, marking a significant leap forward in genetic and biomedical research.

Traditionally, studying the 3D architecture of chromosomes at the single-cell level has been challenging due to noisy and incomplete data. The newly developed AI tool addresses these challenges by effectively interpreting weak patterns and estimating chromosomal structures even when some information is missing. Its ability to recognize biological structures accurately, regardless of their orientation, sets it apart from previous methods.

Compared to earlier deep learning approaches, this AI tool demonstrates more than double the accuracy in analyzing human single-cell data. By making the software freely available to the global scientific community, the researchers aim to facilitate a deeper understanding of gene function, disease mechanisms, and the development of targeted treatments.

Looking ahead, the team plans to enhance the tool further, aiming to reconstruct high-resolution structures of entire genomes. Such advancements could provide scientists with the most detailed view yet of the genetic blueprint within our cells, potentially revolutionizing our approach to studying and treating various diseases.

For more details, you can read the full article here:

https://www.sciencedaily.com/releases/2025/05/250528131824.htm

In the last two weeks, artificial intelligence has made major waves in biology.

One of the biggest stories comes from a biotech startup called EvolutionaryScale. Their new AI model, ESM3, was trained on 770 billion protein sequences. It didn’t just predict proteins—it created one. The AI simulated half a billion years of evolution and designed a completely new fluorescent protein called esmGFP. Scientists tested it, and it actually worked, glowing green in the lab. This could open the door to designing entirely new biological tools using AI.

In India, researchers at IIT Indore built a new system that uses quantum nanotechnology and AI to detect genetic mutations, including cancer-linked ones. It works by analyzing electrical signals from DNA and could make early diagnosis cheaper and more accurate.

Meanwhile in Australia, AI helped scientists uncover hidden patterns in our genes—tiny variations that could explain why some people are more likely to develop certain diseases. These patterns were previously invisible without AI’s help.

But there’s also a warning: researchers in the UK are concerned that some AI-generated science is low quality. Tools like ChatGPT can help write papers, but if they’re used poorly, the research could be misleading or flawed.

To help guide good science, the AI company Anthropic just launched a new “AI for Science” program. They’re giving free access to advanced AI tools for researchers working on real-world biology problems.

AI and biology are merging faster than ever. From creating proteins to spotting disease, this is just the beginning.

Sources:

• https://www.technologyreview.com/2024/05/17/1091690/this-ai-just-invented-a-new-protein-that-shines-green/
• https://timesofindia.indiatimes.com/city/indore/iit-i-develops-quantum-ai-nanotechnology-to-aid-in-early-detection-of-genetic-mutations/articleshow/120942599.cms
• https://medicalxpress.com/news/2025-05-ai-uncovers-hidden-patterns-genes.html
• https://www.eurekalert.org/news-releases/1083476
• https://www.anthropic.com/news/ai-for-science-program

Biology and AI are no longer just neighbors in the science department. They are in a full-blown situationship. Not the cute kind. The kind that keeps producing breakthroughs at 3am with too much caffeine and not enough ethical review.

We are past predicting proteins. That was fun. Now we are generating them. Feeding DNA into language models. Letting diffusion models dream up enzymes. Training neural nets not to write poetry but to build life. The lab bench has a new assistant and it runs on CUDA.

Forget years of trial and error. Forget million-dollar screening campaigns. Now you can simulate thousands of cell states in hours. Test mutations before they happen. Build new molecular tools the way you build software modules. Edit life with version control.

Biologists are retraining as coders. Coders are moonlighting as synthetic biologists. Suddenly it matters if your transformer understands exon skipping or promoter leakage. Suddenly models hallucinating is not just a bug, it is a feature. Some of those hallucinations might cure cancer.

The future of biology is not just wet. It is digital. It is probabilistic. It is learned from data and optimized on clusters. If you think this sounds like science fiction, good. That means you are paying attention. Because the next big discovery might come from a GitHub repo. Not a journal.

NASA Discovers Building Blocks of Life on Asteroid Bennu

In a stunning breakthrough, NASA has confirmed that samples returned from the near-Earth asteroid Bennu contain all five nucleobases — the essential ingredients of DNA and RNA. This discovery, announced in late March 2025, marks a monumental step forward in our understanding of the origins of life in the universe.

The OSIRIS-REx mission, which collected and returned samples from Bennu, has provided scientists with a rare window into the chemistry of the early solar system. Within the dust and rock fragments, researchers identified adenine, guanine, cytosine, thymine, and uracil — molecules that form the genetic code of every living organism on Earth. Alongside these, the samples also revealed signs of brine and minerals that form only in the presence of liquid water.

These findings suggest that Bennu, or the body it originated from, once harbored water and facilitated prebiotic chemistry — the kind that could eventually lead to life. More importantly, it supports the long-standing hypothesis that key ingredients for life were delivered to early Earth by asteroids and comets.

The discovery is not only a triumph for planetary science and astrobiology, but also for the future of space exploration. It reinforces the idea that life may not be unique to Earth and that the seeds of biology might be scattered throughout the cosmos.

As analysis of the Bennu samples continues, scientists hope to uncover even more about the complex organic molecules preserved on the asteroid. Each new finding brings us one step closer to answering one of the most profound questions of all: how did life begin?

Read more on NASA’s official OSIRIS-REx page.

In a significant development for lunar exploration, a Gloucestershire-based company, Naicker Scientific, has been awarded £150,000 for inventing a device capable of producing clean water from the Moon’s icy soil. The system, known as SonoChem, utilizes a combination of microwave technology, ultrasonic waves, and a biomass feeder to extract and purify water, a crucial step toward establishing a sustainable human presence on the lunar surface. This innovation was recognized as part of the Aqualunar Challenge, a competition funded by the UK Space Agency to encourage novel solutions for water production on the Moon. 

The SonoChem system operates through a multi-step process: it begins by heating lunar soil using microwaves to release water vapor, which is then captured and condensed into liquid form. Subsequently, ultrasonic waves are applied to further purify the water, ensuring it meets the standards necessary for human consumption and use. This method not only addresses the challenge of accessing water on the Moon but also paves the way for utilizing in-situ resources to support long-term lunar missions.

The success of Naicker Scientific’s technology holds promise beyond lunar applications. The team is exploring the potential of their system for terrestrial use, such as portable water purification devices and treatment of industrial effluents. Additionally, the extracted lunar water could be used to produce rocket fuel, facilitating further space exploration endeavors. This achievement underscores the growing synergy between space technology and sustainable solutions on Earth, highlighting the broader impact of innovations developed for extraterrestrial environments.

https://www.theguardian.com/science/2025/mar/27/producing-clean-drinking-water-on-moon-invention-naicker-scientific-gloucestershire-company-prize