Digital discovery最新文献

{"title":"pyRheo: an open-source Python package for complex rheology†","authors":"Isaac Y. Miranda-Valdez, Aaro Niinistö, Tero Mäkinen, Juha Lejon, Juha Koivisto and Mikko J. Alava","doi":"10.1039/D5DD00021A","DOIUrl":"https://doi.org/10.1039/D5DD00021A","url":null,"abstract":"<p >Mathematical modeling is a powerful tool in rheology, and we present pyRheo, an open-source package for Python designed to streamline the analysis of creep, stress relaxation, small amplitude oscillatory shear, and steady shear flow tests. pyRheo contains a comprehensive selection of viscoelastic models, including fractional order approaches. It integrates model selection and fitting features and employs machine intelligence to suggest a model to describe a given dataset. The package fits the suggested model or one chosen by the user. An advantage of using pyRheo is that it addresses challenges associated with sensitivity to initial guesses in parameter optimization. It allows the user to iteratively search for the best initial guesses, avoiding convergence to local minima. We discuss the capabilities of pyRheo and compare them to other tools for rheological modeling of soft matter. We demonstrate that pyRheo significantly reduces the computation time required to fit high-performance viscoelastic models.</p>","PeriodicalId":72816,"journal":{"name":"Digital discovery","volume":" 4","pages":" 1075-1082"},"PeriodicalIF":6.2,"publicationDate":"2025-03-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://pubs.rsc.org/en/content/articlepdf/2025/dd/d5dd00021a?page=search","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143809072","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Robotic integration for end-stations at scientific user facilities†","authors":"Chandima Fernando, Hailey Marcello, Jakub Wlodek, John Sinsheimer, Daniel Olds, Stuart I. Campbell and Phillip M. Maffettone","doi":"10.1039/D5DD00036J","DOIUrl":"https://doi.org/10.1039/D5DD00036J","url":null,"abstract":"<p >The integration of robotics and artificial intelligence (AI) into scientific workflows is transforming experimental research, particularly at large-scale user facilities such as the National Synchrotron Light Source II (NSLS-II). We present an extensible architecture for robotic sample management that combines the Robot Operating System 2 (ROS2) with the <em>Bluesky</em> experiment orchestration ecosystem. This approach enabled seamless integration of robotic systems into high-throughput experiments and adaptive workflows. Key innovations included a client-server model for managing robotic actions, real-time pose estimation using fiducial markers and computer vision, and closed-loop adaptive experimentation with agent-driven decision-making. Deployed using widely available hardware and open-source software, this architecture successfully automated a full shift (8 hours) of sample manipulation without errors. The system's flexibility and extensibility allow rapid re-deployment across different experimental environments, enabling scalable self-driving experiments for end stations at scientific user facilities. This work highlights the potential of robotics to enhance experimental throughput and reproducibility, providing a roadmap for future developments in automated scientific discovery where flexibility, extensibility, and adaptability are core requirements.</p>","PeriodicalId":72816,"journal":{"name":"Digital discovery","volume":" 4","pages":" 1083-1091"},"PeriodicalIF":6.2,"publicationDate":"2025-03-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://pubs.rsc.org/en/content/articlepdf/2025/dd/d5dd00036j?page=search","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143809073","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"SurfPro – a curated database and predictive model of experimental properties of surfactants†","authors":"Stefan L. Hödl, Luc Hermans, Pim F. J. Dankloff, Aigars Piruska, Wilhelm T. S. Huck and William E. Robinson","doi":"10.1039/D4DD00393D","DOIUrl":"https://doi.org/10.1039/D4DD00393D","url":null,"abstract":"<p >Despite great industrial interest, modeling the physical properties of surfactants in water based on their molecular structure remains a challenge. A significant part of this challenge is in obtaining sufficient amounts of high-quality data. Experimentally determined properties such the critical micelle concentration (CMC) and surface tension at CMC (<em>γ</em><small>_CMC</small>) have been reported for many surfactants. However, surfactant data are scattered across many literature sources, and reported in a manner which is often unsuitable as input for predictive models. In this work, we address this limitation by compiling the SurfPro database of surfactant properties. SurfPro consists of 1624 surfactant entries curated from 223 literature sources, containing 1395 CMC values, 972 <em>γ</em><small>_CMC</small> values and more than 657 values for <em>Γ</em><small>_max</small>, <em>C</em><small>₂₀</small>, π<small>_CMC</small> and <em>A</em><small>_min</small>. However, only 647 structures have all reported properties, and for most surfactants multiple properties are missing. We trained a previously reported graph neural network architecture for single- and multi-property prediction on these incomplete data of all surfactant types in the database to accurately predict pCMC (−log<small>₁₀</small>(CMC)), <em>γ</em><small>_CMC</small>, <em>Γ</em><small>_max</small> and p<em>C</em><small>₂₀</small>. We achieved state-of-the-art performance of these four properties using an ensemble of AttentiveFP models trained on ten different folds of the training data in the multi-property setting. Finally, we leveraged the predictions and uncertainties of the ensemble model to impute all missing properties for all 977 surfactants with an incomplete set of properties. We make our curated SurfPro database, proposed test split and training datasets, the imputed database, as well as our code publicly available.</p>","PeriodicalId":72816,"journal":{"name":"Digital discovery","volume":" 5","pages":" 1176-1187"},"PeriodicalIF":6.2,"publicationDate":"2025-03-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://pubs.rsc.org/en/content/articlepdf/2025/dd/d4dd00393d?page=search","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143943996","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"DOPtools: a Python platform for descriptor calculation and model optimization","authors":"Said Byadi, Philippe Gantzer, Timur Gimadiev and Pavel Sidorov","doi":"10.1039/D4DD00399C","DOIUrl":"https://doi.org/10.1039/D4DD00399C","url":null,"abstract":"<p >The DOPtools (Descriptors and Optimization tools) platform is a Python library for the calculation of chemical descriptors, hyperparameter optimization, and building and validation of QSPR models. In addition to the Python code that can be integrated in custom scripts, it provides a command line interface for the automatic calculation of various descriptors and for eventual hyperparameter optimization of statistical models, enabling its use in server applications for QSPR modeling. It is especially suited for modeling reaction properties <em>via</em> functions that calculate descriptors for all reaction components. While a variety of existing tools and libraries can calculate various molecular descriptors, their output format is often unique, which complicates their integration with standard machine learning libraries. DOPtools provides a unified API for the calculated descriptors as input for the scikit-learn library. The modular nature of the code allows easy addition of algorithms if required by the end user. The code for the platform is freely available at GitHub and can be installed through PyPI.</p>","PeriodicalId":72816,"journal":{"name":"Digital discovery","volume":" 5","pages":" 1188-1198"},"PeriodicalIF":6.2,"publicationDate":"2025-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://pubs.rsc.org/en/content/articlepdf/2025/dd/d4dd00399c?page=search","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143944046","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Measurements with noise: Bayesian optimization for co-optimizing noise and property discovery in automated experiments†","authors":"Boris N. Slautin, Yu Liu, Jan Dec, Vladimir V. Shvartsman, Doru C. Lupascu, Maxim A. Ziatdinov and Sergei V. Kalinin","doi":"10.1039/D4DD00391H","DOIUrl":"https://doi.org/10.1039/D4DD00391H","url":null,"abstract":"<p >We have developed a Bayesian optimization (BO) workflow that integrates intra-step noise optimization into automated experimental cycles. Traditional BO approaches in automated experiments focus on optimizing experimental trajectories but often overlook the impact of measurement noise on data quality and cost. Our proposed framework simultaneously optimizes both the target property and the associated measurement noise by introducing time as an additional input parameter, thereby balancing the signal-to-noise ratio and experimental duration. Two approaches are explored: a reward-driven noise optimization and a double-optimization acquisition function, both enhancing the efficiency of automated workflows by considering noise and cost within the optimization process. We validate our method through simulations and real-world experiments using Piezoresponse Force Microscopy (PFM), demonstrating the successful optimization of measurement duration and property exploration. Our approach offers a scalable solution for optimizing multiple variables in automated experimental workflows, improving data quality, and reducing resource expenditure in materials science and beyond.</p>","PeriodicalId":72816,"journal":{"name":"Digital discovery","volume":" 4","pages":" 1066-1074"},"PeriodicalIF":6.2,"publicationDate":"2025-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://pubs.rsc.org/en/content/articlepdf/2025/dd/d4dd00391h?page=search","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143809047","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"GraphXForm: graph transformer for computer-aided molecular design†","authors":"Jonathan Pirnay, Jan G. Rittig, Alexander B. Wolf, Martin Grohe, Jakob Burger, Alexander Mitsos and Dominik G. Grimm","doi":"10.1039/D4DD00339J","DOIUrl":"https://doi.org/10.1039/D4DD00339J","url":null,"abstract":"<p >Generative deep learning has become pivotal in molecular design for drug discovery, materials science, and chemical engineering. A widely used paradigm is to pretrain neural networks on string representations of molecules and fine-tune them using reinforcement learning on specific objectives. However, string-based models face challenges in ensuring chemical validity and enforcing structural constraints like the presence of specific substructures. We propose to instead combine graph-based molecular representations, which can naturally ensure chemical validity, with transformer architectures, which are highly expressive and capable of modeling long-range dependencies between atoms. Our approach iteratively modifies a molecular graph by adding atoms and bonds, which ensures chemical validity and facilitates the incorporation of structural constraints. We present GraphXForm, a decoder-only graph transformer architecture, which is pretrained on existing compounds and then fine-tuned using a new training algorithm that combines elements of the deep cross-entropy method and self-improvement learning. We evaluate GraphXForm on various drug design tasks, demonstrating superior objective scores compared to state-of-the-art molecular design approaches. Furthermore, we apply GraphXForm to two solvent design tasks for liquid–liquid extraction, again outperforming alternative methods while flexibly enforcing structural constraints or initiating design from existing molecular structures.</p>","PeriodicalId":72816,"journal":{"name":"Digital discovery","volume":" 4","pages":" 1052-1065"},"PeriodicalIF":6.2,"publicationDate":"2025-03-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://pubs.rsc.org/en/content/articlepdf/2025/dd/d4dd00339j?page=search","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143809046","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Ligand design for 227Ac extraction by active learning and molecular topology†","authors":"Jeffrey A. Laub and Konstantinos D. Vogiatzis","doi":"10.1039/D5DD00007F","DOIUrl":"https://doi.org/10.1039/D5DD00007F","url":null,"abstract":"<p >Targeted α-therapy (TAT) is a promising radiotherapeutic technique for the treatment of various cancers due to the high linear energy transfer and low penetration depth of α-particles. Unfortunately, one of the major hindrances in the use of TAT is the accessibility of acceptable α-emitting radioisotopes. Of the acceptable radioisotopes, <small>²²³</small>Ra, <small>²²⁴</small>Ra, <small>²²⁵</small>Ra, and <small>²²⁵</small>Ac can all originate from <small>²²⁷</small>Ac. Being able to selectively isolate <small>²²⁷</small>Ac is crucial for aiding in increasing the accessibility of α-emitting radioisotopes for TAT. Some of the more successful ligands used for the selective separation of trivalent actinides are the 6,6′-bis(1,2,4-triazin-3-yl)-2,2′-bipyridine (BTBP)-based ligand family. Current ligand performance screening is accomplished by using a trial-and-error-based method which is expensive and based primarily on chemical intuition and previous studies. In this study, effective computer-aided ligand screening has been accomplished by generating <strong>CyMe<small>₄</small>–BTBP</strong>-based ligands and predicting stability constants for <small>²²⁷</small>Ac extraction of each using scalar relativistic density functional theory (DFT) followed by supervised machine learning (ML). DFT was used to compute stability constants from a 2 : 1 stoichiometric ratio of BTBP to <small>²²⁷</small>Ac with three nitrate ions for charge balancing as demonstrated by experimental analysis. The computed stability constants coupled with the vectorized information from the optimized BTBP molecular geometries were used for the training of ML workflows. The performance of each algorithm was determined by the validation set and the outcomes compared to the DFT stability constants. This methodology can aid radiochemists in synthesizing targeted ligands for selective isolation of <small>²²⁷</small>Ac.</p>","PeriodicalId":72816,"journal":{"name":"Digital discovery","volume":" 4","pages":" 1100-1112"},"PeriodicalIF":6.2,"publicationDate":"2025-03-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://pubs.rsc.org/en/content/articlepdf/2025/dd/d5dd00007f?page=search","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143809075","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"BitBIRCH: efficient clustering of large molecular libraries†","authors":"Kenneth López Pérez, Vicky Jung, Lexin Chen, Kate Huddleston and Ramón Alain Miranda-Quintana","doi":"10.1039/D5DD00030K","DOIUrl":"10.1039/D5DD00030K","url":null,"abstract":"<p >The widespread use of Machine Learning (ML) techniques in chemical applications has come with the pressing need to analyze extremely large molecular libraries. In particular, clustering remains one of the most common tools to dissect the chemical space. Unfortunately, most current approaches present unfavorable time and memory scaling, which makes them unsuitable to handle million- and billion-sized sets. Here, we propose to bypass these problems with a time- and memory-efficient clustering algorithm, BitBIRCH. This method uses a tree structure similar to the one found in the Balanced Iterative Reducing and Clustering using Hierarchies (BIRCH) algorithm to ensure <em>O</em>(<em>N</em>) time scaling. BitBIRCH leverages the instant similarity (iSIM) formalism to process binary fingerprints, allowing the use of Tanimoto similarity, and reducing memory requirements. Our tests show that BitBIRCH is already &gt;1000 times faster than standard implementations of the Taylor–Butina clustering for libraries with 1 500 000 molecules. BitBIRCH increases efficiency without compromising the quality of the resulting clusters. We explore strategies to handle large sets, which we applied in the clustering of one billion molecules under 5 hours using a parallel/iterative BitBIRCH approximation.</p>","PeriodicalId":72816,"journal":{"name":"Digital discovery","volume":" 4","pages":" 1042-1051"},"PeriodicalIF":6.2,"publicationDate":"2025-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11912344/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143665477","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Digital discovery and the new experimental frontier","authors":"S. Hessam M. Mehr","doi":"10.1039/D5DD00029G","DOIUrl":"https://doi.org/10.1039/D5DD00029G","url":null,"abstract":"<p >The digitisation of chemistry has had a profound effect on the field by boosting the efficiency of information retrieval and data recording, and by automating repetitive laboratory operations. Increasingly complex molecules — both known and <em>de novo</em> — can be rapidly accessed with unprecedented speed and reproducibility. Despite progress as measured by these quantitative productivity metrics, a qualitative transformation in the design and structure of experimentation has yet to materialise. Here, we explore digitisation's role in a larger paradigm shift in experimental chemistry not just as a means of automated execution of procedures but dynamically sensing, interpreting, and manipulating chemical processes in real-time. This paradigm shift is characterised by transitioning from single-point measurements to continuous observation; from homogeneous to spatially organised systems; and from fixed linear experimental procedures to dynamic, branched “programs” that can unfold based on real-time feedback. This shift will enable new types of objectives in experimental chemistry, such as responsiveness, adaptability and persistence, expanding beyond static quantities like product structure, yield and purity. We explore the innovations needed to enable these transitions; the open questions they raise; and how digitisation can catalyse chemistry's evolution beyond its existing confines.</p>","PeriodicalId":72816,"journal":{"name":"Digital discovery","volume":" 4","pages":" 892-895"},"PeriodicalIF":6.2,"publicationDate":"2025-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://pubs.rsc.org/en/content/articlepdf/2025/dd/d5dd00029g?page=search","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143809076","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Predicting performance of object detection models in electron microscopy using random forests†","authors":"Ni Li, Ryan Jacobs, Matthew Lynch, Vidit Agrawal, Kevin Field and Dane Morgan","doi":"10.1039/D4DD00351A","DOIUrl":"https://doi.org/10.1039/D4DD00351A","url":null,"abstract":"<p >Quantifying prediction uncertainty when applying object detection models to new, unlabeled datasets is critical in applied machine learning. This study introduces an approach to estimate the performance of deep learning-based object detection models for quantifying defects in transmission electron microscopy (TEM) images, focusing on detecting irradiation-induced cavities in TEM images of metal alloys. We developed a random forest regression model that predicts the object detection <em>F</em><small>₁</small> score, a statistical metric used to evaluate the ability to accurately locate and classify objects of interest. The random forest model uses features extracted from the predictions of the object detection model whose uncertainty is being quantified, enabling fast prediction on new, unlabeled images. The mean absolute error (MAE) for predicting <em>F</em><small>₁</small> of the trained model on test data is 0.09, and the <em>R</em><small>²</small> score is 0.77, indicating there is a significant correlation between the random forest regression model predicted and true defect detection <em>F</em><small>₁</small> scores. The approach is shown to be robust across three distinct TEM image datasets with varying imaging and material domains. Our approach enables users to estimate the reliability of a defect detection and segmentation model predictions and assess the applicability of the model to their specific datasets, providing valuable information about possible domain shifts and whether the model needs to be fine-tuned or trained on additional data to be maximally effective for the desired use case.</p>","PeriodicalId":72816,"journal":{"name":"Digital discovery","volume":" 4","pages":" 987-997"},"PeriodicalIF":6.2,"publicationDate":"2025-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://pubs.rsc.org/en/content/articlepdf/2025/dd/d4dd00351a?page=search","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143809092","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}