Digital Discovery

Do materials foundation models learn actual physics, or just hallucinate structures? 🤔 SCALAR benchmark tracks what happens when models go from bulk crystals to 18,000-atom nanoparticles. Spoiler: high accuracy often hides severe physical inconsistencies #AI4Science 🔗 doi.org/10.1039/d6dd...

Looking for some weekend reading?🔥Check out this new hot article introducing an automated in situ heating X-ray diffraction robotic system accelerating materials discovery in self-driving laboratories. 👉 doi.org/10.1039/d6dd... #automation #selfdrivinglab #labautomation #openaccess #openscience

How should we model chemical disorder in the AI era? This Review connects conventional and AI-assisted atomistic approaches, offering guidance on method selection and validation to bridge the experiment–simulation gap for predictive, disorder-aware materials design 🔗doi.org/10.1039/d6dd...

📢Our new themed collection for AI in Drug Discovery at ICANN2025 is now published online! This collection reflects cutting-edge contributions at the workshop in the rapidly evolving field of AI-driven drug discovery. Read this collection at the link below: pubs.rsc.org/dd/coll...

Can machine learning make #virtualscreening more reliable for ion channels? TRPM8 benchmark shows that training on diverse inactive molecules improves early hit enrichment and generalization, revealing a critical lesson for target-specific ML scoring functions. 👉Read it now: doi.org/10.1039/d6dd...

Can we harness homolytic bond energetics to steer inverse radical design? This #RSCDigital article introduces a framework for generating radical fragment pairs and considers how reaction-relevant thermodynamic quantities can be incorporated in generative molecular design. 👉 doi.org/10.1039/d6dd...

Can you get a molecule's 3D shape back from its IR spectrum? IR-GeoDiff is an SE(3)-equivariant latent diffusion model that recovers 3D geometries directly from IR spectra, and its attention lands on the same functional-group bands that chemists read. 👉Read the full story: doi.org/10.1039/d6dd...

Don't miss the RSC BMCS Hot Topics online symposium on "AI in Drug Discovery"! This half-day conference, running 1300-1700 UTC on November 3, features an excellent line-up of speakers exploring how AI tools can assist in drug discovery, spanning industry and academia. www.rscbmcs.org/events/hotto...

🔥New #HOT article online in #RSCDigital now 🔥 Discover the Battery Imaging Library: an open source repository uniting multi-modal synchrotron and laboratory battery imaging data across multiple length scales. 🔎Explore the full open access paper to learn more: doi.org/10.1039/d6dd...

Catch up with the latest news from Digital Discovery in newsletter, available to read now our blog post below. Our next instalment includes new articles, themed collections, upcoming events, and much more. Sign up for newsletter emails: rsc.li/alerts blogs.rsc.org/dd/digital-d...

Don't miss this! New research by Zarghami-Dehaghani et al. combines high-throughput molecular simulations and machine learning to identify promising covalent organic frameworks (COFs) for urea removal from water towards wearable artificial kidney technologies. Read it here: doi.org/10.1039/d6dd...

New #RSCDigital Perspective highlighting that sustainable machine learning requires not only more efficient algorithms, but also transparent reporting, reusable datasets and models, informed benchmarking, and resource-aware community practices. Full story 👉 doi.org/10.1039/D6DD...

We’re delighted to announce Professor Robert House as the winner of the 2026 EES Lectureship, which recognises exceptional emerging talent in energy science! Learn more 👉 https://rsc.li/47dQZeE @kebleoxford.bsky.social @ox.ac.uk @oxfordmaterials.bsky.social @raeng.org.uk #ChemSky

📢New #RSCDigital Advance Article available now! DOS2BandNet predicts unfolded band structures directly from DOS and direction-resolved diffraction spectra, enabling transferable spectral-to-band reconstruction across bulk, defect, and CDW systems. 👉 Read it here: doi.org/10.1039/d6dd...

We're delighted to welcome Dr Vassiliki-Alexandra (Vanda) Glezakou to the Advisory Board of Digital Discovery. Dr Glezakou is a a Distinguished Scientist and Section Head of Chemical Transformations in Oak Ridge National Laboratory, Chemical Sciences Division. Read more blogs.rsc.org/dd/glezakou-...

【Meet the Editor セミナー】 10月20日CSJ化学フェスタにおいて、ランチョンセミナー（2階F会場）を開催いたします。Digital Discoveryのアドバイザリーボードメンバーである武田征士博士（IBM Research）をお招きし、AI駆動型材料科学の最新動向についてご講演いただきます🤖 @digital-discovery.rsc.org また、出版におけるAI活用に関するRSCの最新ポリシーについてもご紹介いたします。 学生の皆様は勿論、研究者や教員の皆様のご参加も歓迎いたします👇 onsite.gakkai-web.net/chemistry/pr...

We're pleased to announce that Digital Discovery is sponsoring the 2027's Gordon Research Conference on Self-Driving Labs, "Integrating AI, Automation, and Experimental Science Across Domains". Check the web site for an update on the program in October. www.grc.org/self-driving...

New HOT article in #RSCDigital🔥Introducing 'AUTOMAT', an LLM-based agent for materials-property prediction! Matteo Cobelli and Stefano Sanvito implement an autoresearch framework to the automated design of chemical-composition descriptors for materials property prediction.🔥doi.org/10.1039/d6dd...

The deadlines for RSC FIRST 2026: AI in Chemistry have now been extended! Don't miss your chance to join us in Xiamen on 19-21 November: Abstract Submission: 15 September 2026 Early-Bird Registration: 15 October 2026 Standard Registration: 15 November 2026 www.rscfirst.org.cn/en/web/index...

"Do humans and large language models agree on the quality of synthesis plans?" 🤖 This new #RSC Advance Article puts 17 expert chemists and 4 LLMs to the test evaluating the feasibility of 50 reactions to explore when and why they agree, or not. 👉 Get the full details now: doi.org/10.1039/d6dd...

📢We are sponsoring a poster prize at the RSC-CICAG Quantum Computing in Chemistry meeting! 🏅 #Submit now - poster abstract deadline 9 October 📍Burlington House, London, UK 🗓️Thursday, 19 November 2026 🔗registrations.hg3con... #quantumcomputing #conference #RSC RSC-CICAG

Can how we represent molecular structure enhance the success of materials design?✨From traditional approaches to modern #machinelearning embeddings, this #RSCDigital Review explores how advances in representation learning accelerate #materialsdiscovery. 👉 Read it here: doi.org/10.1039/d6dd...

Can AI accelerate the path from molecular design to drug discovery?✨A new review from Sahar N explores foundation models, generative design, and high-throughput computational workflows, highlighting opportunities and challenges for translation. 👉 Read it now: doi.org/10.1039/D6DD...

Did you miss this highly cited #RSCDigital article from 2025? Catch it now! "Democratizing self-driving labs: advances in low-cost 3D printing for laboratory automation" a review from S DOLOI and colleagues at NTU School of Materials Science and Engineering 👉Full article: doi.org/10.1039/d4dd...

Can AI automate colloidal packing simulations? ColPackAgent framework runs Monte Carlo simulations of colloidal packing by a Model Context Protocol tool server, operating interactively with human feedback, autonomously from end-to-end or in autoresearch mode 👉Get the details doi.org/10.1039/d6dd...

Looking for an accelerated framework for extrapolative #alloydiscovery? 🔎 By utilizing non-interacting electron density, P Ray et al. bypass costly #DFT calculations to achieve <2% error with <0.2% of dataset, enabling zero-shot transfer in HEAs. 👉 Read it here: doi.org/10.1039/d6dd...

🔎Looking to compare methods for crystal structure prediction? New #RSCDigital article provides a systematic benchmark and quantitative comparison of eight crystal structure prediction approaches, from traditional methods to generative machine learning models. 👉Full story: doi.org/10.1039/d6dd...

Latest addition to 2026 Emerging Investigators collection from Li-Cheng Xu at the Shanghai Academy of Artificial Intelligence for Science CatEmb, a stereoelectronic-aware catalyst embedding learned through contrastive alignment of 2D and 3D molecular graph representations 🔗doi.org/10.1039/d6dd...

Exciting research from Sung Ho Jeon, Satish Nimse and colleagues at Hallym University, South Korea identifies Diphylin & Justicidin B show promise against influenza! 🌿 AI modeling reveals they target a conserved viral protein (RdRp) ➡️Read the full story: doi.org/10.1039/d6dd...

Interested in understanding how artificial intelligence can support the drug discovery process? Don't miss the RSC BMCS Hot Topics online meeting on "AI in Drug Discovery" taking place 1300-1700 UTC on 3 November 2026. Find out more and register here: www.rscbmcs.org/events/hotto...

Looking for "on-demand" epitaxy of van der Waals stacked moiré devices?✨Target a moiré twist angle. Let the AI find how to grow it. 🪴The autonomous workflow hits custom MoS₂ bilayer targets in under 20 iterations using HPC-driven Bayesian ensemble sampling. 👉Full story: doi.org/10.1039/d6dd...

Introducing "CrystalCV: a computer vision system for analysis of crystallization experiments" Nicholas Sandor and Makhsud Saidaminov @uvic.ca present an accessible, Python-based computer vision system for monitoring crystallization processes with three insightful case studies doi.org/10.1039/d6dd...

Our new Digital Discovery themed collection on ‘Quantum Computing for Chemistry, Material Science and Biotechnology’ has now been published online! Read this collection at the link below: blogs.rsc.org/dd/ddquant25/