Models & Labs · Europe

Agentic cheminformatics with Google Antigravity

Credited to Google DeepMind · deepmind.google · 2h ago

Photograph · Google DeepMind

Zoi Tsangalidou, Ruijie (Kelly) Zhang, Sukhdeep Singh and Michael V. LeVine

Designing new medicines requires curating complex data from charts, figures and literature, training predictive models, and proposing candidate molecules to synthesize — a workflow that typically takes multiple days for a single iteration.

Here we show how Google Antigravity, equipped with our new Cheminformatics Skills, accelerates drug discovery workflows. Using real-world lung cancer treatments, an agent searches through thousands of compounds, performs matched molecular pair analysis, and proposes novel rigidified macrocycles with synthetic routes in just an hour.

Chemistry is one of science’s richest public records — spanning millions of structures across journal articles and vendor catalogs. It provides a common representation that links physics and mathematics to the life and environmental sciences. Molecules and their transformations are ubiquitous and shape the everyday world.

Despite decades of progress in cheminformatics, chemists still spend too much time wrangling data, a process complicated by the fact that molecules can be represented in many ways: from text descriptions to images or data structures. The same molecule can appear as an IUPAC name in a paper, a Markush structure in a patent, or a structure-data file (SDF) in a catalog. Advanced cheminformatics systems must be able to unify these sources and deal with the many ways errors can arise in them. This task has no margin for error – alter a single atom or bond and you have described a different compound altogether, rendering a promising inhibitor or catalyst inert.

To let scientists focus on hypotheses rather than data wrangling, we built a suite of Cheminformatics Skills for Google Antigravity, turning the surface into a collaborative science workbench with full access to local tooling for both computational chemists and experimental bench chemists.

The skills also create a division of labour where the underlying language model handles workflow orchestration, database navigation and scientific reasoning, while deterministic chemistry tools execute molecular calculations. Grounded in RDKit, the foundational open source cheminformatics toolkit developed over two decades by a global community of contributors, alongside connectors to curated resources from ChEMBL and PubChem, the skills encode domain best practices directly into the agent’s workflow.

Rigidifying a floppy molecule

In this technical walkthrough, we highlight a classic medicinal chemistry technique to improve drug potency: rigidification.

Drug molecules can be complex, often linking multiple substructures through rotatable bonds. Those bonds let the molecule twist into many 3D shapes (conformations), making it “floppy” in solution. To bind its target, however, a floppy molecule must lock into a single active shape. Losing that freedom of motion carries an entropic penalty, which weakens binding affinity and lowers potency.

To solve this problem, medicinal chemists often rigidify molecules by adding new bonds. Two substructures linked by a single rotatable bond can twist into many conformations, but adding a second bond to tie them together sharply limits how they can move in three dimensional space. By placing that bond in the right spot, chemists can lock the molecule into the exact shape needed to bind its target. This strategy, often called “paying the entropic cost in advance”, can dramatically boost potency.

Here, we will focus on a well-known case of successful rigidification. The drug crizotinib is a first-generation, small-molecule tyrosine kinase inhibitor (TKI) used to treat certain forms of non-small cell lung carcinoma. In crizotinib, there are five major rotatable bonds. Lorlatinib is a third-generation, small-molecule TKI that resembles crizotinib, but has been cyclized at its end to reduce the effective number of rotatable bonds to zero, leading to higher potency among other effects.

Collaborative Intelligence

We will ask the agent to search through literature for these drugs and their analogues. With the Cheminformatics skills in Antigravity, the agent is able to extract molecules from these documents, associate them with the data reported, and prepare the compounds for downstream analysis.

Within minutes, the agent retrieves thousands of compounds, filters out irrelevant molecules (such as buffers and solvents), and documents their sources in a report. The report also provides an initial analysis of the retrieved compound set: highlighting, for example, that the macrocycles are largely predicted to be brain-penetrant, an important property given that the target lung cancer frequently metastasizes to the brain.

For deeper exploration, the agent launches an interactive viewer displaying the curated compounds, their associated data, and customizable scatter plots. For example, plotting potency against rotatable bonds, colored by macrocycle status, immediately reveals that macrocycles like lorlatinib tend to have fewer rotatable bonds and higher potency. Ultimately, we designed these skills to not only expand the agent’s chemistry capabilities, but also to make user-agent collaboration seamless.

In the interactive viewer, scientists can browse every compound retrieved from the literature, click individual molecules to inspect their metadata, and plot relationships across the dataset. For example, plotting cellular potency versus rotatable bonds,colored by macrocycle status, clearly shows that the highest potency compounds are the most rigid.

After downloading and curating all the data, the agent is able to perform more advanced analysis to support the work of a computational chemist. As a next step, we will perform matched molecular pair analysis, a technique used in medicinal chemistry to isolate the effects of small, discrete changes to a molecule, by comparing pairs of molecules that only differ by that change. The agent had also independently suggested this analysis, as we expect the dataset to include many informative pairs. We will also ask the agent to build a predictive model using its molecular machine learning capabilities.

The matched molecular pair analysis (MMPA) compared many pairs of molecules that differ by whether or not they are cyclized. The results of the analysis show that cyclization reduces the number of rotatable bonds while generally increasing potency, as expected. The agent also built a large number of potency models using varying feature sets and model architectures, and validated them on a held-out set. The skills provide the agent with guidance on how to best build molecular machine learning models - for example, the agent will often use chronological splits, training on older compounds to predict newer ones, when this information is available, but can fall back to scaffold splits to better estimate generalizability.

The agent benchmarks multiple machine learning architectures with differing feature sets across four different experimental read outs, using a scaffold split to better gauge out-of-distribution performance. A combination of statistical and visual outputs allow the scientist to assess model performance and then iterate with the agent.

Finally, we ask the agent to design new, improved macrocycles based on its learnings and it recommends multiple reactions and synthesis routes involving different compounds. The agent can prepare these compounds for further analysis with AlphaFold or physics-based techniques like molecular dynamics simulation.

The skill equips the agent with the tools required to manipulate and validate reactions, generate curved arrow electron pushing diagrams, and look up similar reactions from public databases. The agent generated some interesting designs using its own knowledge combined with the previously developed MMPA and potency prediction models, explained its medicinal chemistry rationale, and suggested potential syntheses routes consistent with documented procedures. It also used the models it trained in the previous turn to predict each design’s potency.

The agent is able to reason about chemistry as it proposes new synthesis plans. Here the agent proposes a cyclization reaction, providing the scientist with electron-pushing diagrams to demonstrate the mechanism.

Once we are done with our analysis, the results can be exported in various formats, ready to be loaded into many industry standard chemistry software packages.

At the same time, it is also important to know what these tools cannot do. Sparse data, inter-lab assay noise, and narrow dynamic ranges limit the predictive precision of even the most powerful models; computational designs are testable hypotheses, not finished drugs. With the Cheminformatics Skills in Antigravity, a drug discovery scientist can generate more strongly-supported design hypotheses faster, but every proposed design must be synthesized and validated experimentally.

The speed at which researchers can make meaningful scientific decisions is accelerating. Not long ago, the workflow demonstrated here would have taken multiple days and required significant computational expertise. Using Antigravity as a science workbench allows both computational and experimental chemists to accelerate their workflows and focus on the questions that really matter.

Acknowledgements

The cheminformatics skills are available on Google Antigravity today. Download the Antigravity App and install the Science plugin to start using these skills. Explore Science use cases here.

You can also get started with cheminformatics on Antigravity today as an Enterprise user, by choosing “Use Business Account” and logging into the app with your Google Cloud credentials.

To further advance the pace of research, we’re offering 10,000 scientists the highest usage tier in Google Antigravity free of charge for 6 months (onboarded on a rolling basis), subject to eligibility, capacity, and selection. Applications are now open.

We would like to acknowledge the contributions of the following team members: Filip Radlinski, Artem Sobolev, Victoria Johnston, Hannah Kiely-Collins, Zachary Wu, Jue Wang, Jack Dunger, Adriana Fernandez Lara, Ngân Vũ, Doug Fritz, Yunhan Xu, Sebastian Nowozin, Pushmeet Kohli.

TechnologyEurope

How AI Is Reshaping High-Capacity Optical Networks: Insights from Openreach and Ciena

Credited to Capacity Media · capacityglobal.com5h ago

Useful

The rapid growth of AI is creating unprecedented demand for bandwidth across global networks, forcing operators to rethink how they build, scale and manage connectivity infrastructure.

In this Capacity TV interview from Capacity Europe 2026, Simon Williams, Head of Optical Products at Openreach, and Joe Marsella, Vice President of International Business Development at Ciena, explore the trends shaping the future of high-capacity networking. The discussion covers the impact of AI training and inferencing workloads, the rise of distributed compute architectures, and the growing importance of fibre, optical innovation and scalable network design.

The pair also examine evolving customer requirements around flexibility and future-proofing, Openreach’s latest high-capacity access developments, and how its long-standing partnership with Ciena is helping to support the next generation of AI-driven connectivity.

Datacloud USA 2027

Apple faces unprecedented memory cost increases as RAM prices continue to surge, manufacturing bill climbs by $33 billion — rising component expenses threaten margins through the holiday quarter and into 2027

Credited to Tom's Hardware · tomshardware.com · 5h ago

Photograph · Tom's Hardware

Apple is about to feel the squeeze from rising memory prices, with the cost of RAM climbing just as the company heads into its busiest shopping season. According to Reuters, Apple expects to pay more for memory in the December quarter, with the CEO, John Ternus, describing the current price increases as “unprecedented. The company has so far managed to soften the blow by using memory bought at lower prices, cutting costs elsewhere and raising prices on some products. However, it is reported that its manufacturing bill has climbed by $33 billion in just a few months.

AI data centers are swallowing the world's memory and storage supply Samsung debuts three next-generation memory technologies for AI data centers The future of DRAM: From DDR5 to future ICs High-bandwidth memory roadmap Inside the history of DRAM price-fixing lawsuits

But that cushion is getting thinner, and Apple expects memory purchases to become even more expensive into 2027. “What we're seeing with memory pricing right now is unprecedented. And as we said earlier in the September quarter, we did see higher memory costs as compared to the June quarter, which were partly offset by the benefit of carrying inventory and lower costs on certain non-memory components. Now for the December quarter, we expect even higher memory costs, partly offset by a few factors. So as you know, we changed the price on many of our products, including Mac and iPad at the end of June and iPhone in September, and these price increases provided some offset.” Ternus said during Apple’s November 2 earnings call.

Memory is used across Apple's hardware lineup, so sustained price increases could affect manufacturing cost for several product categories—although Apple has not announced another round of price increases. Absorbing higher components could, however, indefinitely become difficult. Dave Nicholson, Chief Research Officer at the Futurum Group, also identified memory prices as a growing pressure point for Apple's margins and hardware pricing through the holiday season and into 2027. According to him, Apple is in a different position from companies pouring hundreds of billions of dollars into AI data center infrastructure. The impact of higher component costs is easier to quantify than the eventual returns on massive AI capital expenditure, he said.

The next pressure point for @Apple may not be AI CapEx, but the rising cost of memory across its hardware portfolio.@TheFuturumGroup Chief Research Officer @DaveN007 explained that higher memory prices could put more pressure on Apple’s margins and pricing through the holiday season and into 2027. Even with demand holding up, the question is how much of those higher costs consumers will absorb.Dave also sees Apple in a stronger position than companies committing hundreds of billions to data center infrastructure, because the impact of higher component costs is easier to quantify than the long-term returns on massive AI CapEx.▶️ Watch Dave’s analysis of Apple’s margin outlook, memory costs, and why its AI infrastructure strategy could give the company more flexibility.#Apple #AI #Semiconductors #Futurum November 3, 2026

And the warnings from Apple's suppliers suggest that the cost pressure is unlikely to ease quickly. Last month, Samsung, a major supplier of RAM(DRAM), NAND flash storage and OLED displays for Apple, said that memory demand would far exceed supply as AI infrastructure continues to expand. Its memory division also reported a quarter-on-quarter increase in DRAM average prices during the third quarter, while NAND prices rose by high-teen percentages. Senior executive vice president of its memory business warned the supply crunch could extend to 2028 and beyond. Elsewhere, SK Hynix delivered a similar warning, saying it had secured long-term agreements covering more than Half of its production capacity for2027 and 2028. Micron, the third major supplier for Apple aside from Samsung and SK Hynix, is also benefiting from the surging demand for memory. Its latest published outlook warns that supply shortages in memory and storage could take “considerable time” to improve.

We have previously reported several signs of the pressure now confronting Apple. In January, the company was already scrambling to secure memory supplies amid strong demand, with then CEO Tim Cook warning that rising prices could have a greater impact on earnings. By June, rising component expenses had translated into higher retail prices. Apple increased prices across several Mac and iPad models, including a $400 increase to the starting price of the MacBook Pro, due to the RAM crisis. In July, Cook described the memory market as a “ hundred-year-flood ” and said Apple expected to pay more for memory in the September quarter. The company had also increased its inventory to $11.09 billion as it built a buffer against supply and price pressures.