RSC - Digital Discovery latest articles

Jeweler-in-the-Loop: Personalized Alloy Color Optimization via Preference-Based BO

Digital Discovery, 2025, Accepted Manuscript
DOI: 10.1039/D6DD00009F, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Chase Katz, Ting Yu Yang, Parker King, Md Shafiqul Islam, Brent Vela, Raymundo Arroyave
Many materials design attributes that are central to adoption, such as aesthetics, perceived quality, or user-specific preferences, are difficult to quantify directly, making preference feedback a practical proxy for optimization....
The content of this RSS Feed (c) The Royal Society of Chemistry

Distilling System Complexity to Enable Unbiased and Predictive Computational Reaction Investigations

Digital Discovery, 2025, Accepted Manuscript
DOI: 10.1039/D5DD00366K, Paper

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

Raphaël Robidas, Claude Y. Legault
Computational modelling is a powerful tool to study chemical reactions. Currently, human guidance is nearly always required to avoid the intractable complexity of all a priori possible reaction steps, which...
The content of this RSS Feed (c) The Royal Society of Chemistry

Coupled fragment-based generative modeling with stochastic interpolants

Digital Discovery, 2026, Advance Article
DOI: 10.1039/D5DD00535C, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Tuan Le, Yanfei Guan, Djork-Arné Clevert, Kristof T. Schütt
Fragment-based generative model using stochastic interpolants for structure-based drug design. Explicit conditional training outperforms inpainting (87% vs. 25%). PLK3 kinase validation confirms diverse fragment generation with key interactions.
To cite this article before page numbers are assigned, use the DOI form of citation above.
The content of this RSS Feed (c) The Royal Society of Chemistry

A universal machine learning model for the electronic density of states

Digital Discovery, 2025, Accepted Manuscript
DOI: 10.1039/D5DD00557D, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

How Wei Bin, Pol Febrer, Sanggyu Chong, Arslan Mazitov, Filippo Bigi, Matthias Kellner, Sergey Pozdnyakov, Michele Ceriotti
In the last few years several "universal" interatomic potentials have appeared, using machine-learning approaches to predict energy and forces of atomic configurations with arbitrary composition and structure, with an accuracy...
The content of this RSS Feed (c) The Royal Society of Chemistry

Quantum simulation of carbon capture in periodic metal–organic frameworks

Digital Discovery, 2026, Advance Article
DOI: 10.1039/D6DD00023A, Paper

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

Dario Rocca, Jérôme F. Gonthier, Joshua Levin, Tobias Schäfer, Andreas Grüneis, Hong Woo Lee, Byeol Kang
We present a quantum computing workflow based on Wannier functions and MP2 natural orbitals to study CO₂ adsorption in Fe–MOF-74, advancing the applicability of quantum algorithms to realistic carbon capture and periodic materials.
To cite this article before page numbers are assigned, use the DOI form of citation above.
The content of this RSS Feed (c) The Royal Society of Chemistry

Artificial intelligence in the discovery and design of molecular semiconductors: a systematic review

Digital Discovery, 2026, Advance Article
DOI: 10.1039/D5DD00552C, Review Article

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Malin Zollner, Yashar Moshfeghi, Tahereh Nematiaram
AI-driven molecular semiconductor discovery: models learn from molecular datasets to predict spectra and electronic structure, guide photoactive and emissive materials design, and accelerate device optimisation for next-generation optoelectronics.
To cite this article before page numbers are assigned, use the DOI form of citation above.
The content of this RSS Feed (c) The Royal Society of Chemistry

Food additive lens: an on-device AI application for real-time science-based consumer education on food additives using retrieval-augmented generation

Digital Discovery, 2026, Advance Article
DOI: 10.1039/D5DD00444F, Paper

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

Yihang Feng, Yi Wang, Xinhao Wang, Bo Zhao, Jinbo Bi, Song Han, Yangchao Luo
Food additive lens is an on-device agentic AI application that classifies food products, identifies additives from ingredient lists, and generates contextual and science‑based explanations for everyday packaged foods.
To cite this article before page numbers are assigned, use the DOI form of citation above.
The content of this RSS Feed (c) The Royal Society of Chemistry

Chat-RFB: a flow battery chat system leveraging knowledge graphs and large language models

Digital Discovery, 2026, Advance Article
DOI: 10.1039/D5DD00494B, Paper

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

Hao-Tian Wang, Xuefeng Bai, Zhiling Zheng, Xin Zhang, Ruipeng Jin, Hao-Tian An, Zheng-He Xie, Xiu-Liang Lv, Jian-Rong Li
Chat-RFB integrates knowledge graphs with large language models to enable accurate, source-verified scientific dialogue and accelerate discovery in redox flow battery research.
To cite this article before page numbers are assigned, use the DOI form of citation above.
The content of this RSS Feed (c) The Royal Society of Chemistry

Using Flory-Huggins-Informed Human-in-the-Loop Bayesian Optimization to Map the Phase Diagram of Polymer Blends

Digital Discovery, 2025, Accepted Manuscript
DOI: 10.1039/D5DD00556F, Paper

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

Justin C. Hughes, Dylan J. York, Kevin G. Yager, Chinedum O. Osuji, Russell J. Composto
Mapping the phase diagram of polymer blends is an essential step in controlling the structure-property relationship of polymer-based material. However, traditional grid-based approaches are inefficient and rely on subjective judgements...
The content of this RSS Feed (c) The Royal Society of Chemistry

Automated reaction transition state search for bimolecular liquid-phase reactions using internal coordinates: a test case for neutral hydrolysis

Digital Discovery, 2026, Advance Article
DOI: 10.1039/D5DD00506J, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Leen Fahoum, Alon Grinberg Dana
Orientation effects hinder automated TS search for bimolecular kinetics. We present a SMILES-to-TS framework using internal-coordinate heuristics, achieving up to 97% success across 91 diverse neutral hydrolysis reactions in liquid phase.
To cite this article before page numbers are assigned, use the DOI form of citation above.
The content of this RSS Feed (c) The Royal Society of Chemistry

Siamese Graph Neural Networks for Melting Temperature Prediction of Molten Salt Eutectics

Digital Discovery, 2025, Accepted Manuscript
DOI: 10.1039/D5DD00480B, Paper

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

Nila Mandal, James Maniscalco, Mark Aindow, Qian Yang
High-throughput screening enabled by structure-property prediction models is a powerful approach for accelerating materials discovery. However, while machine learning of structure-property models have become widespread, its application to mixtures remains...
The content of this RSS Feed (c) The Royal Society of Chemistry

Harnessing generative AI for efficient organic materials discovery in low-data regimes

Digital Discovery, 2026, Advance Article
DOI: 10.1039/D5DD00463B, Paper

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

Jun Hyeong Kim, Kyunghoon Lee, Hyeonsu Kim, MinSoo Kang, Suk-Ku Chang, Yinglan Jin, Dongwook Kim, Woo Youn Kim
A building block-based generative AI combined with DFT screening enables the efficient discovery of TADF materials in low-data regimes. Experimental validation confirms the design of green emitters with external quantum efficiencies up to 11.22%.
To cite this article before page numbers are assigned, use the DOI form of citation above.
The content of this RSS Feed (c) The Royal Society of Chemistry

A mobile robotic process chemist

Digital Discovery, 2026, Advance Article
DOI: 10.1039/D5DD00563A, Paper

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

Emma J. Brass, Satheeshkumar Veeramani, Zhengxue Zhou, Hatem Fakhruldeen, J. Sebastian Manzano, Rob Clowes, Isil Akpinar, Miriam R. Ward, John W. Ward, Andrew I. Cooper
We report a modular automated platform using a mobile robot for late-stage process chemistry, integrating synthesis, work-up, and analysis. The system reproduces human–level reaction performance while supporting extended, unattended operation.
To cite this article before page numbers are assigned, use the DOI form of citation above.
The content of this RSS Feed (c) The Royal Society of Chemistry

MSIGN: A deep learning framework based on multi-scale interaction graph neural networks for predicting binding of synthetic cannabinoids to receptors

Digital Discovery, 2026, Advance Article
DOI: 10.1039/D5DD00317B, Paper

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

Zhenyong Cheng, Dinghao Liu, Yuanpeng Fu, Kewei Sheng, Yan Xing, Yanling Qiao, Shangxuan Cai, Jubo Wang, Peng Xu, Bin Di, Jun Liao
MSIGN leverages multi-scale graph features for affinity prediction. Following domain-specific fine-tuning, the model is effectively applied to synthetic cannabinoid binding assessment, achieving high accuracy as confirmed by experimental assays.
To cite this article before page numbers are assigned, use the DOI form of citation above.
The content of this RSS Feed (c) The Royal Society of Chemistry

Structure-Guided Machine Learning for Efficiency Prediction of Organic Photovoltaics Using Experimentally Informed Molecular Descriptors

Digital Discovery, 2025, Accepted Manuscript
DOI: 10.1039/D5DD00496A, Paper

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

JuHyun Lee, Hyojin Ban, HyunIl Seo, Hang Ken Lee, Fiza Arshad, DongWook Kim
The efficiency of organic photovoltaics was estimated using a machine learning (ML) approach. We used the organic photovoltaics database built in-house by the Korea Research Institute of Chemical Technology. Representative...
The content of this RSS Feed (c) The Royal Society of Chemistry

Can we automate scientific reasoning in closed-loop experiments using large language models?

Digital Discovery, 2026, Advance Article
DOI: 10.1039/D5DD00520E, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Abdoulatif Cissé, Max E. Cooper, Mengjia Zhu, Xenophon Evangelopoulos, Andrew I. Cooper
Reasoning models (LLMs) were used as optimisers for a 10-D chemistry problem and a 7-D physics simulation. The best performance was achieved using iterative closed-loop reasoning with a batch size of one.
To cite this article before page numbers are assigned, use the DOI form of citation above.
The content of this RSS Feed (c) The Royal Society of Chemistry

Advances and perspectives in computer-assisted structure elucidation: a review

Digital Discovery, 2026, Advance Article
DOI: 10.1039/D5DD00438A, Review Article

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Dagny Aurich, Emma L. Schymanski
Computer-Assisted Structure Elucidation (CASE) is a powerful yet underused approach in chemistry to determine molecular structures from experimental data without necessarily being restricted to the contents of chemical databases.
To cite this article before page numbers are assigned, use the DOI form of citation above.
The content of this RSS Feed (c) The Royal Society of Chemistry

Precision fragment addition: domain-specific DeepFrag2 models for smarter lead optimization

Digital Discovery, 2026, Advance Article
DOI: 10.1039/D5DD00425J, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

César R. García-Jacas, Harrison Green, Shayne D. Wierbowski, Jacob D. Durrant
DeepFrag2, a machine-learning tool for lead optimization via fragment addition, is more accurate when trained on fragments with specific sizes, charge states, or aromaticity. Fine-tuning on specific receptor classes further boosts performance.
To cite this article before page numbers are assigned, use the DOI form of citation above.
The content of this RSS Feed (c) The Royal Society of Chemistry

Synthesis Planning in Reaction Space: A Study on Success, Robustness and Diversity

Digital Discovery, 2025, Accepted Manuscript
DOI: 10.1039/D5DD00280J, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Alan Kai Hassen, Helen Lai, Samuel Genheden, Mike Preuss, Djork-Arné Clevert
Computer-aided synthesis planning aims to identify viable synthetic routes from a target compound to readily available building blocks by iteratively decomposing molecules into smaller precursors. Self-play search algorithms, trained with...
The content of this RSS Feed (c) The Royal Society of Chemistry

An automated sampling workflow for parallel long-term membrane diffusion cell testing

Digital Discovery, 2026, Advance Article
DOI: 10.1039/D5DD00523J, Paper

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

Claire Benstead, Maria Politi, David S. Bergsman, Lilo D. Pozzo
The ADT workflow utilizes customized hardware and a liquid handling robot to automate and parallelize long-term membrane permeability testing.
To cite this article before page numbers are assigned, use the DOI form of citation above.
The content of this RSS Feed (c) The Royal Society of Chemistry

A feature-aligned diffusion model for controllable generation of 3D drug-like molecules

Digital Discovery, 2026, Advance Article
DOI: 10.1039/D5DD00440C, Paper

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

Hao Lu, Zhiqiang Wei, Xiancong Hou, Wenzheng Han, Yang Zhang, Hao Liu
We introduce a molecule-design framework that integrates molecular representations with a diffusion model to generate compounds exhibiting improved docking affinity, enabling more effective exploration of structure-based drug candidates.
To cite this article before page numbers are assigned, use the DOI form of citation above.
The content of this RSS Feed (c) The Royal Society of Chemistry

Deep set model for the automated NMR fingerprinting of unknown mixtures

Digital Discovery, 2026, Advance Article
DOI: 10.1039/D5DD00490J, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Jens Wagner, Kerstin Münnemann, Thomas Specht, Hans Hasse, Fabian Jirasek
Deep set model enables the automated elucidation of structural groups in unknown mixtures from standard NMR spectra.
To cite this article before page numbers are assigned, use the DOI form of citation above.
The content of this RSS Feed (c) The Royal Society of Chemistry

A fully differentiable pore network for digital reconstruction of porous media

Digital Discovery, 2026, Advance Article
DOI: 10.1039/D5DD00455A, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Michael McKague, Mohammad Mehrnia, Mohammad Amin Sadeghi, Jeff Gostick
This work highlights a fully differentiable framework for constructing pore networks from porosimetry and anisotropic permeability data. The resulting network can be scaled using a trained Gaussian Process model.
To cite this article before page numbers are assigned, use the DOI form of citation above.
The content of this RSS Feed (c) The Royal Society of Chemistry

Connecting the concepts of quantum state tomography and molecular representations for machine learning

Digital Discovery, 2026, Advance Article
DOI: 10.1039/D5DD00484E, Perspective

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Raul Ortega-Ochoa, Luis Mantilla Calderón, Juan Bernardo Perez Sanchez, Mohsen Bagherimehrab, Abdulrahman Aldossary, Tejs Vegge, Tonio Buonassisi, Alán Aspuru-Guzik
We connect quantum state tomography with molecular machine learning, arguing prediction of many molecular descriptors from a shared representation progressively constrains latent representations toward physically meaningful organization.
To cite this article before page numbers are assigned, use the DOI form of citation above.
The content of this RSS Feed (c) The Royal Society of Chemistry

Bayesian diversity control for batch-based phase diagram determination

Digital Discovery, 2026, Advance Article
DOI: 10.1039/D5DD00486A, Paper

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

Peiheng Zou, Ryo Tamura, Koji Tsuda
We introduce DPP-PDC, a Bayesian method for batch-based phase diagram determination with automatic diversity control.
To cite this article before page numbers are assigned, use the DOI form of citation above.
The content of this RSS Feed (c) The Royal Society of Chemistry

Scientific knowledge graph and ontology generation using open large language models

Digital Discovery, 2026, Advance Article
DOI: 10.1039/D5DD00275C, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Alexandru Oarga, Matthew Hart, Andres M. Bran, Magdalena Lederbauer, Philippe Schwaller
Knowledge graphs (KGs) are powerful tools for structured information modeling, increasingly recognized for their potential to enhance the factuality and reasoning capabilities of Large Language Models (LLMs).
To cite this article before page numbers are assigned, use the DOI form of citation above.
The content of this RSS Feed (c) The Royal Society of Chemistry

Enhancing molecular structure elucidation with reasoning-capable LLMs

Digital Discovery, 2026, Advance Article
DOI: 10.1039/D5DD00359H, Paper

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

Martin Priessner, Richard J. Lewis, Magnus J. Johansson, Jonathan M. Goodman, Jon Paul Janet, Anna Tomberg
We introduce a novel workflow that integrates reasoning-capable language models (LLMs) with chemical analysis tools to enhance molecular structure determination using NMR spectroscopy.
To cite this article before page numbers are assigned, use the DOI form of citation above.
The content of this RSS Feed (c) The Royal Society of Chemistry

Statistics makes a difference: machine learning adsorption dynamics of functionalized cyclooctyne on Si(001) at DFT accuracy

Digital Discovery, 2026, Advance Article
DOI: 10.1039/D5DD00420A, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Hendrik Weiske, Rhyan Barrett, Ralf Tonner-Zech, Patrick Melix, Julia Westermayr
The interpretation of experiments on reactive semiconductor surfaces requires statistically significant sampling of molecular dynamics, but conventional ab initio methods are limited due to prohibitive computational costs.
To cite this article before page numbers are assigned, use the DOI form of citation above.
The content of this RSS Feed (c) The Royal Society of Chemistry

A physics-informed measurement protocol for expectation values of fermionic observables

Digital Discovery, 2026, Advance Article
DOI: 10.1039/D5DD00251F, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Davide Bincoletto, Jakob S. Kottmann
A scalable and practical protocol for estimating the expectation values of fermionic observables is presented. The approach is based on an iterative procedure that measures low-cost operator groups across different orbital bases.
To cite this article before page numbers are assigned, use the DOI form of citation above.
The content of this RSS Feed (c) The Royal Society of Chemistry

Mapping Bloch-Redfield dynamics into a unitary gate-based quantum algorithm

Digital Discovery, 2026, Advance Article
DOI: 10.1039/D5DD00405E, Paper

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

Koray Aydoğan, Maryam Abbasi, Whitney J. Short, Mikayla Z. Fahrenbruch, Timothy J. Krogmeier, Anthony W. Schlimgen, Kade Head-Marsden
Quantum circuit for Bloch–Redfield relaxation dynamics.
To cite this article before page numbers are assigned, use the DOI form of citation above.
The content of this RSS Feed (c) The Royal Society of Chemistry

Automated and robotic sample delivery systems for mass spectrometry and ion-mobility spectrometry

Digital Discovery, 2026, Advance Article
DOI: 10.1039/D5DD00492F, Perspective

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Chikondi Shaba, Decibel P. Elpa, Pawel L. Urban
Mass spectrometry and ion-mobility spectrometry are two complementary tools for molecular analysis. They have been extensively automated opening the doors to exciting applications such as single-cell analysis.
To cite this article before page numbers are assigned, use the DOI form of citation above.
The content of this RSS Feed (c) The Royal Society of Chemistry

MC3D: the materials cloud computational database of experimentally known stoichiometric inorganics

Digital Discovery, 2026, Advance Article
DOI: 10.1039/D5DD00415B, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Sebastiaan P. Huber, Michail Minotakis, Marnik Bercx, Timo Reents, Kristjan Eimre, Nataliya Paulish, Nicolas Hörmann, Martin Uhrin, Nicola Marzari, Giovanni Pizzi
We introduce MC3D, a curated computational database of structures filtered from the COD, ICSD and MPDS databases, optimized using density-functional theory with automated workflows and curated input protocols.
To cite this article before page numbers are assigned, use the DOI form of citation above.
The content of this RSS Feed (c) The Royal Society of Chemistry

When machine learning models learn chemistry I: quantifying explainability with matched molecular pairs

Digital Discovery, 2026, 5,571-582
DOI: 10.1039/D5DD00398A, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Kerrin Janssen, Jan M. Wollschläger, Jonny Proppe, Andreas H. Göller
Explainability methods in machine learning-driven research are increasingly being used, but it remains challenging to assess their reliability without deeply investigating the specific problem at hand.
The content of this RSS Feed (c) The Royal Society of Chemistry

When machine learning models learn chemistry II: applying WISP to real-world examples

Digital Discovery, 2026, 5,583-591
DOI: 10.1039/D5DD00399G, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Kerrin Janssen, Jan M. Wollschläger, Jonny Proppe, Andreas H. Göller
In our previous work, we introduced WISP (Workflow for Interpretability Scoring using Matched Molecular Pairs), which enables users to quantitatively assess the performance of explainability methods for machine learning models.
The content of this RSS Feed (c) The Royal Society of Chemistry

Correction: A case study on hybrid machine learning and quantum-informed modelling for solubility prediction of drug compounds in organic solvents

Digital Discovery, 2026, 5,957-957
DOI: 10.1039/D6DD90005D, Correction

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Weiling Wang, Isabel Cooley, Morgan R. Alexander, Ricky D. Wildman, Anna K. Croft, Blair F. Johnston
The content of this RSS Feed (c) The Royal Society of Chemistry

Introduction to the “Accelerate Conference 2023–2024” themed collection

Digital Discovery, 2026, 5,480-481
DOI: 10.1039/D5DD90057C, Editorial

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Janine George, Claudiane Ouellet-Plamondon, Kristofer Reyes
Janine George, Claudiane Ouellet-Plamondon, and Kristofer Reyes introduce the themed collection on the 2023 and 2024 Accelerate Conferences. (JG Photo: BAM.)
The content of this RSS Feed (c) The Royal Society of Chemistry

FiberForge: enabling high-throughput simulations of the mechanical properties of helical fibrils

Digital Discovery, 2026, 5,919-930
DOI: 10.1039/D5DD00307E, Paper

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

Kieran Nehil-Puleo, Zhongyue John Yang
FiberForge provides an end-to-end platform for molecular modeling and design of amyloid materials, enabling physics-based identification of sequences and polymorphs with targeted mechanical behavior.
The content of this RSS Feed (c) The Royal Society of Chemistry

Accelerating catalytic advancements through the precision of high-throughput experiments & calculations

Digital Discovery, 2026, 5,497-509
DOI: 10.1039/D5DD00524H, Opinion

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

Jenny G. Vitillo, Alán Aspuru-Guzik, Eric Doskocil, Omar K. Farha, Timur Islamoglu, Heather J. Kulik, Peter M. Margl, Stuart Miller, Jordan Reddel, Aayush R. Singh, Varinia Bernales
This Opinion article shows how integrated computational and experimental approaches accelerate catalyst development and highlights key challenges in data quality, model accuracy, and scalability across Technology Readiness Levels.
The content of this RSS Feed (c) The Royal Society of Chemistry

ToPolyAgent: AI agents for coarse-grained bead-spring topological polymer simulations

Digital Discovery, 2026, 5,901-909
DOI: 10.1039/D5DD00471C, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Lijie Ding, Jan-Michael Carrillo, Changwoo Do
We introduce ToPolyAgent, a multi-agent AI framework for performing coarse-grained molecular dynamics (MD) simulations of topological polymers through natural language instructions.
The content of this RSS Feed (c) The Royal Society of Chemistry

Joint embedding predictive architecture for self-supervised pretraining on polymer molecular graphs

Digital Discovery, 2026, 5,819-834
DOI: 10.1039/D5DD00308C, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Francesco Piccoli, Gabriel Vogel, Jana M. Weber
We address the challenge of limited labeled data in polymer ML, using a Joint Embedding Predictive Architecture (JEPA) for self-supervised pretraining on polymer graphs. This improves downstream performance when finetuning on small labelled datasets.
The content of this RSS Feed (c) The Royal Society of Chemistry

Towards utility-scale electronic structure with sample-based quantum bootstrap embedding

Digital Discovery, 2026, 5,945-956
DOI: 10.1039/D5DD00416K, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Joel Bierman, Yuan Liu
The first benchmarking study of the quantum bootstrap embedding (QBE) method on real quantum hardware is performed for the H₈ molecule under an extended realistic basis.
The content of this RSS Feed (c) The Royal Society of Chemistry

Data augmentation in a triple transformer loop retrosynthesis model

Digital Discovery, 2026, 5,653-661
DOI: 10.1039/D5DD00465A, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Yves Grandjean, David Kreutter, Jean-Louis Reymond
To mitigate bias in the USPTO dataset, we generated fictive reactions from USPTO templates and validated them with a triple transformer loop. Retrosynthesis models trained on this data outperform those trained on USPTO alone.
The content of this RSS Feed (c) The Royal Society of Chemistry

Deep learning-enabled discovery of low-melting-point ionic liquids

Digital Discovery, 2026, 5,643-652
DOI: 10.1039/D5DD00282F, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Gaopeng Ren, Austin M. Mroz, Frederik Philippi, Tom Welton, Kim E. Jelfs
A link-prediction model was used to expand an ionic liquid database and construct a robust generative model. Low-melting-point ionic liquids were then identified through melting-point classification and validated using molecular dynamics simulations.
The content of this RSS Feed (c) The Royal Society of Chemistry

DFT meets Bayesian inference: creating a framework for the assignment of calculated vibrational frequencies

Digital Discovery, 2026, 5,592-602
DOI: 10.1039/D5DD00453E, Paper

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

Michael Nicolaou, Hans M. Senn, Emma Gibson, Mario González-Jiménez, Laia Vilà-Nadal
Determination of vibrational modes in aromatic VOCs via DFT and Bayesian inference to match theoretical and experimental spectra.
The content of this RSS Feed (c) The Royal Society of Chemistry

ChemBERTa-3: an open source training framework for chemical foundation models

Digital Discovery, 2026, 5,662-685
DOI: 10.1039/D5DD00348B, Paper

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

Riya Singh, Aryan Amit Barsainyan, Rida Irfan, Connor Joseph Amorin, Stewart He, Tony Davis, Arun Thiagarajan, Shiva Sankaran, Seyone Chithrananda, Walid Ahmad, Derek Jones, Kevin McLoughlin, Hyojin Kim, Anoushka Bhutani, Shreyas Vinaya Sathyanarayana, Venkat Viswanathan, Jonathan E. Allen, Bharath Ramsundar
An overview of ChemBERTa-3, an open-source training and benchmarking framework designed to train and fine-tune large-scale chemical foundation models.
The content of this RSS Feed (c) The Royal Society of Chemistry

Assessing the performance of quantum-mechanical descriptors in physicochemical and biological property prediction

Digital Discovery, 2026, 5,803-818
DOI: 10.1039/D5DD00411J, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Alejandra Hinostroza Caldas, Artem Kokorin, Alexandre Tkatchenko, Leonardo Medrano Sandonas
QUED, a QM/ML framework that combines structural and electronic molecular information to build regression models for physicochemical and biological property prediction. Our work highlights the value of QM data for reliable and interpretable models.
The content of this RSS Feed (c) The Royal Society of Chemistry

DBMLFF: linear scaling machine learning force fields via electron density decomposition for molecular electrolytes

Digital Discovery, 2026, 5,931-944
DOI: 10.1039/D5DD00508F, Paper

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

Jie Shen, Chenyu Wang, Libin Chen, Shaoqin Jiang, Jianhui Chen, Cuilian Wen, Bo Wu, Baisheng Sa, Lin-Wang Wang
DBMLFF enables modular, transferable force fields with ab initio accuracy for complex multi-component systems.
The content of this RSS Feed (c) The Royal Society of Chemistry

OBELiX: a curated dataset of crystal structures and experimentally measured ionic conductivities for lithium solid-state electrolytes

Digital Discovery, 2026, 5,910-918
DOI: 10.1039/D5DD00441A, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Félix Therrien, Jamal Abou Haibeh, Divya Sharma, Rhiannon Hendley, Leah Wairimu Mungai, Sun Sun, Alain Tchagang, Jiang Su, Samuel Huberman, Yoshua Bengio, Hongyu Guo, Alex Hernández-García, Homin Shin
OBELiX is a database of 599 synthesized solid electrolyte materials and their experimentally measured room temperature ionic conductivities gathered from literature and curated by domain experts.
The content of this RSS Feed (c) The Royal Society of Chemistry

Evaluating large language models for inverse semiconductor design

Digital Discovery, 2026, 5,780-792
DOI: 10.1039/D5DD00544B, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Muhammed Nur Talha Kilic, Daniel Wines, Kamal Choudhary, Vishu Gupta, Youjia Li, Sayak Chakrabarty, Wei-Keng Liao, Alok Choudhary, Ankit Agrawal
Large Language Models (LLMs) can enable inverse materials discovery by generating text-encoded crystal structures from target properties.
The content of this RSS Feed (c) The Royal Society of Chemistry

The PPP model – a minimum viable parametrisation of conjugated chemistry for modern computing applications

Digital Discovery, 2026, 5,482-496
DOI: 10.1039/D5DD00445D, Perspective

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Marcel D. Fabian, Nina Glaser, Gemma C. Solomon
The semi-empirical Pariser–Parr–Pople (PPP) Hamiltonian is reviewed as a minimal model for the chemistry of conjugated π-electron systems. Its current applications and limitations are discussed, and we conjecture how using the PPP model in quantum computing could be mutually beneficial.
The content of this RSS Feed (c) The Royal Society of Chemistry

Efficient quantum simulation of non-adiabatic molecular dynamics with precise electronic structure

Digital Discovery, 2026, 5,548-570
DOI: 10.1039/D5DD00433K, Paper

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

Tianyi Li, Yumeng Zeng, Qiming Ding, Zixuan Huo, Xiaosi Xu, Jiajun Ren, Diandong Tang, Xiaoxia Cai, Xiao Yuan
We introduce an quantum-computing adapted surface hopping framework, ensuring numerical stability and parallelizability. Combined with sub-microhartree-accurate electronic structure, it enables practical NAMD quantum simulations on diverse systems.
The content of this RSS Feed (c) The Royal Society of Chemistry

LivePyxel: accelerating image annotations with a Python-integrated webcam live streaming

Digital Discovery, 2026, 5,835-843
DOI: 10.1039/D5DD00421G, Paper

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

Uriel Garcilazo-Cruz, Joseph O. Okeme, Rodrigo A. Vargas-Hernández
LivePyxel is a pixel-level annotation editor designed for the development of image segmentation models. Inspired by popular graphics editors, it supports precise annotations through Bézier spline–based tools and live-feed webcam integration.
The content of this RSS Feed (c) The Royal Society of Chemistry

Autonomous elemental characterization enabled by a low cost robotic platform built upon a generalized software architecture

Digital Discovery, 2026, 5,891-900
DOI: 10.1039/D5DD00263J, Paper

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

Xuan Cao, Yuxin Wu, Michael L. Whittaker
A generalized software architecture based on dual-layer action servers facilitates the development of autonomous experimental systems. An autonomous elemental characterization platform serves as an example of such a system.
The content of this RSS Feed (c) The Royal Society of Chemistry

Context-aware computer vision for chemical reaction state detection

Digital Discovery, 2026, 5,630-642
DOI: 10.1039/D5DD00346F, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Junru Ren, Abhijoy Mandal, Rama El-khawaldeh, Shi Xuan Leong, Jason Hein, Alán Aspuru-Guzik, Lazaros Nalpantidis, Kourosh Darvish
Real-time monitoring of laboratory experiments is essential for automating complex workflows and enhancing experimental efficiency.
The content of this RSS Feed (c) The Royal Society of Chemistry

Applications of modular co-design for de novo 3D molecule generation

Digital Discovery, 2026, 5,754-768
DOI: 10.1039/D5DD00380F, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Danny Reidenbach, Filipp Nikitin, Olexandr Isayev, Saee Gopal Paliwal
A scalable and accurate transformer model unifies diffusion and flow-matching approaches to generate realistic 3D molecules.
The content of this RSS Feed (c) The Royal Society of Chemistry

MOFReasoner: think like a scientist—a reasoning large language model via knowledge distillation

Digital Discovery, 2026, 5,869-877
DOI: 10.1039/D5DD00429B, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Xuefeng Bai, Zhiling Zheng, Xin Zhang, Hao-Tian Wang, Rui Yang, Jian-Rong Li
MOFReasoner is a domain-specific LLM enhanced by chain-of-thought reasoning and knowledge distillation that, when compared to GPT-4.5, shows superior performance in chemical tasks like MOF adsorption prediction.
The content of this RSS Feed (c) The Royal Society of Chemistry

A multi-task learning approach for prediction of missing bioactivity values of compounds for the SLC transporter superfamily

Digital Discovery, 2026, 5,878-890
DOI: 10.1039/D5DD00536A, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Tarik Ćerimagić, Sergey Sosnin, Gerhard F. Ecker
A multi-task deep neural network uses compound descriptors to predict missing bioactivity values for multiple targets. The model learns shared representations to complete the sparse pChEMBL (pC) matrix across the SLC transporter superfamily.
The content of this RSS Feed (c) The Royal Society of Chemistry

A case study on hybrid machine learning and quantum-informed modelling for solubility prediction of drug compounds in organic solvents

Digital Discovery, 2026, 5,716-733
DOI: 10.1039/D5DD00456J, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Weiling Wang, Isabel Cooley, Morgan R. Alexander, Ricky D. Wildman, Anna K. Croft, Blair F. Johnston
Machine learning pipeline integrates COSMO-RS and multiple molecular descriptors to predict and interpret solubility across diverse solute–solvent systems.
The content of this RSS Feed (c) The Royal Society of Chemistry

Molecular dynamics simulations accelerated on FPGA with high-bandwidth memory

Digital Discovery, 2026, 5,844-861
DOI: 10.1039/D5DD00391A, Paper

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

Jing Xiao, Jinfeng Chen, Ye Ding, You Xu, Jing Huang
This work presents a full FPGA implementation of molecular dynamics (MD) simulations with high-bandwidth memory (HBM2), achieving accurate, energy-efficient performance and enabling scalable hardware acceleration for biomolecular systems.
The content of this RSS Feed (c) The Royal Society of Chemistry

A data-driven approach to control stimulus responsivity of functional polymer materials: predicting thermoresponsive color-changing properties of polydiacetylene

Digital Discovery, 2026, 5,862-868
DOI: 10.1039/D5DD00442J, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Risako Shibata, Nano Shioda, Hiroaki Imai, Yasuhiko Igarashi, Yuya Oaki
Stimulus-responsive color-changing properties of layered polydiacetylene were controlled by a data-driven approach based on small experimental data. Such method can accelerate dynamic properties of functional polymer materials.
The content of this RSS Feed (c) The Royal Society of Chemistry

AI-driven robotic crystal explorer for rapid polymorph identification

Digital Discovery, 2026, 5,734-742
DOI: 10.1039/D5DD00203F, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Edward C. Lee, Daniel Salley, Abhishek Sharma, Leroy Cronin
A closed-loop robotic platform with human supervision uses AI-driven vision to detect, classify and discover crystal polymorphs, mapping complex crystallisation landscapes with minimal experiments to reveal hidden regions of polymorph space.
The content of this RSS Feed (c) The Royal Society of Chemistry

An exploration of dataset bias in single-step retrosynthesis prediction

Digital Discovery, 2026, 5,793-802
DOI: 10.1039/D5DD00358J, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Sara Tanovic, Ewa Wieczorek, Fernanda Duarte
We investigate how the size and diversity of training datasets influence various single-step retrosynthesis models. Accuracy is primarily affected by template frequency, and models show limited generalisation, highlighting a fundamental class-imbalance problem in retrosynthesis prediction.
The content of this RSS Feed (c) The Royal Society of Chemistry

Kinetic predictions for SN2 reactions using the BERT architecture: comparison and interpretation

Digital Discovery, 2026, 5,743-753
DOI: 10.1039/D5DD00192G, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Chloe Wilson, María Calvo, Stamatia Zavitsanou, James D. Somper, Ewa Wieczorek, Tom Watts, Jason Crain, Fernanda Duarte
This study introduces a BERT model for predicting experimental log k values of S_N2 reactions, comparing it to a Random Forest model. Both models achieve similar accuracy and identify key chemical features.
The content of this RSS Feed (c) The Royal Society of Chemistry

Computer vision for high-throughput materials synthesis: a tutorial for experimentalists

Digital Discovery, 2026, 5,510-522
DOI: 10.1039/D5DD00384A, Tutorial Review

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

Madeleine A. Gaidimas, Abhijoy Mandal, Pan Chen, Shi Xuan Leong, Gyu-Hee Kim, Akshay Talekar, Kent O. Kirlikovali, Kourosh Darvish, Omar K. Farha, Varinia Bernales, Alán Aspuru-Guzik
Computer vision enables the rapid identification of chemical phases, such as solid metal–organic framework (MOF) materials, from images of sample vials.
The content of this RSS Feed (c) The Royal Society of Chemistry

Explainable active learning framework for ligand binding affinity prediction

Digital Discovery, 2026, 5,769-779
DOI: 10.1039/D5DD00436E, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Satya Pratik Srivastava, Rohan Gorantla, Sharath Krishna Chundru, Claire J. R. Winkelman, Antonia S. J. S. Mey, Rajeev Kumar Singh
Active learning (AL) guides the selection of which compounds to evaluate next for protein–ligand binding affinity when assay or simulation budgets are limited.
The content of this RSS Feed (c) The Royal Society of Chemistry

Adsorb-Agent: autonomous identification of stable adsorption configurations via a large language model agent

Digital Discovery, 2026, 5,617-629
DOI: 10.1039/D5DD00298B, Paper

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

Janghoon Ock, Radheesh Sharma Meda, Tirtha Vinchurkar, Yayati Jadhav, Amir Barati Farimani
Adsorb-Agent: an LLM-powered agent for determining the most stable adsorption configurations using reasoning and prior knowledge.
The content of this RSS Feed (c) The Royal Society of Chemistry

Hierarchical attention graph learning with LLM enhancement for molecular solubility prediction

Digital Discovery, 2026, 5,603-616
DOI: 10.1039/D5DD00407A, Paper

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

Yangxin Fan, Yinghui Wu, Roger H. French, Danny Perez, Michael G. Taylor, Ping Yang
Solubility quantifies the concentration of a molecule that can dissolve in a given solvent.
The content of this RSS Feed (c) The Royal Society of Chemistry

Optimizing data extraction from materials science literature: a study of tools using large language models

Digital Discovery, 2026, 5,698-715
DOI: 10.1039/D5DD00482A, Paper

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

Wenkai Ning, Musen Li, Jeffrey R. Reimers, Rika Kobayashi
Benchmarking five AI tools on materials science literature shows promising capabilities, but performance remains inadequate for large-scale data extraction. Our analysis offers detailed insight and guidance for future methodological improvements.
The content of this RSS Feed (c) The Royal Society of Chemistry

Advancing metal organic framework and covalent organic framework design via the digital-intelligent paradigm

Digital Discovery, 2026, 5,523-547
DOI: 10.1039/D5DD00401B, Review Article

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

Bing Ma, Na Qin, Qianqian Yan, Wei Zhou, Sheng Zhang, Xiao Wang, Lipiao Bao, Xing Lu
AI and machine learning combined with multiscale simulations accelerate framework materials design. This review summarizes AI-assisted strategies for synthesis prediction, condition optimization, and inverse functional design.
The content of this RSS Feed (c) The Royal Society of Chemistry

Efficient symmetry-aware materials generation via hierarchical generative flow networks

Digital Discovery, 2026, 5,686-697
DOI: 10.1039/D4DD00392F, Paper

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

Tri Minh Nguyen, Sherif Abdulkader Tawfik, Truyen Tran, Sunil Gupta, Santu Rana, Svetha Venkatesh
Efficient exploration of crystal structure space using hierarchical generative models optimized for diversity and stability.
The content of this RSS Feed (c) The Royal Society of Chemistry

Redox potential prediction of Fe(II)/Fe(III) complexes: a density functional theory and graph neural network approach

Digital Discovery, 2026, Advance Article
DOI: 10.1039/D5DD00431D, Paper

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

Fakhrul H. Bhuiyan, Hassan Harb, Rajeev Surendran Assary, Álvaro Vázquez-Mayagoitia
This work presents an integrated computational approach combining tight-binding and standard density functional theory with graph neural networks for accurate, high-throughput prediction of redox potentials in iron-based transition metal complexes.
To cite this article before page numbers are assigned, use the DOI form of citation above.
The content of this RSS Feed (c) The Royal Society of Chemistry

Mapping sleep-promoting volatiles in aromatic plants with machine learning: a comprehensive survey of 2300 molecules

Digital Discovery, 2026, Advance Article
DOI: 10.1039/D5DD00173K, Paper

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

Peiqin Shi, Xing Huang, Qinfei Ke, Xingran Kou, Dachuan Zhang
Machine learning and in vivo validation uncover sleep-promoting volatiles and map their distribution across 991 aromatic plants.
To cite this article before page numbers are assigned, use the DOI form of citation above.
The content of this RSS Feed (c) The Royal Society of Chemistry

A simple compound prioritization method for drug discovery considering multi-target binding

Digital Discovery, 2026, Advance Article
DOI: 10.1039/D5DD00464K, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Alžbeta Kubincová, David L. Mobley
Active learning is an emerging paradigm used to help accelerating drug discovery, but most prior applications seek solely to optimize potency, whereas multiple properties influence a compound's utility as a drug candidate.
To cite this article before page numbers are assigned, use the DOI form of citation above.
The content of this RSS Feed (c) The Royal Society of Chemistry

Machine learning-based time-series forecasting prevents electrode corrosion in organic electrochemistry

Digital Discovery, 2026, Advance Article
DOI: 10.1039/D5DD00458F, Paper

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

Josef Tausendschön, Michael Poelzl, Nikola Petrovic, Jason D. Williams, Elisabeth Fink
Real-time sensing and machine learning unite to monitor electrode corrosion and predict reaction dynamics in electrochemical synthesis.
To cite this article before page numbers are assigned, use the DOI form of citation above.
The content of this RSS Feed (c) The Royal Society of Chemistry

Towards an understanding of photoluminescence in lead-free Cs2AgxNa1−xBiyIn1−yCl6 double perovskites by machine learning prediction from density functional theory ground state properties

Digital Discovery, 2026, Advance Article
DOI: 10.1039/D5DD00152H, Paper

Open Access

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Marina S. Günthert, Larry Lüer, Oleksandr Stroyuk, Oleksandra Raievska, Christian Kupfer, Andres Osvet, Bernd Meyer, Christoph J. Brabec
The optoelectronic properties of lead-free halide double perovskites are tuneable through their composition. Combining high-throughput synthesis, DFT, and ML, we identify optimal ion ratios and predict photoluminescence from ground-state data.
To cite this article before page numbers are assigned, use the DOI form of citation above.
The content of this RSS Feed (c) The Royal Society of Chemistry