Enzymatic reactions play an emerging role in a broad spectrum of scientific and industrial applications. The inherent complexity of enzymes, such as their substrate specificity, conformational flexibility, and the vast diversity of reactions involved, poses substantial challenges for the advanced computational prediction of enzymatic reactions with desirable accuracy. Moreover, existing approaches are mostly tailored for a specific sub-task, such as substrate prediction or binding site annotation, which limits their applicability. In this study, we introduce ERAM, a task-agnostic multimodal learning framework capable of addressing a broad range of downstream applications with both accuracy and efficiency. ERAM aligns pre-trained molecular representations from Protein Language Model with the knowledge of enzyme catalysis by modeling enzymatic reactions as multi-relational data. In enzyme retrieval tasks, ERAM achieves an improvement of 28.31% in mean average precision compared with the state-of-the-art (SOTA) method, CREEP. In substrate prediction tasks, ERAM outperforms the SOTA method ESP, achieving average improvements of 35.53% and 22.97% in Matthews correlation coefficient across two datasets. Additionally, ERAM exhibits commendable interpretability by assigning higher attention weights to binding sites, resulting in lower false-positive rates (42.36%) and higher overlap scores (70.59%) in the unsupervised binding site prediction task compared to RXNAA Mapper. By learning embeddings of substrates, enzymes, and products within a unified knowledge graph latent space, ERAM demonstrates its potential as a versatile and effective tool for enzyme catalysis research.
Biotoxins, mainly produced by venomous animals, plants, and microorganisms, exhibit high physiological activity and unique effects such as lowering blood pressure and analgesia. A number of venom-derived drugs are already available on the market, with many more candidates currently undergoing clinical and laboratory studies. However, drug design resources related to biotoxins are insufficient, particularly because of a lack of accurate and extensive activity data. To fulfill this demand, we developed the Biotoxins Database (BioTD). BioTD is the largest open-source database for toxins, offering open access to 14,607 data records (8,185 activity records), covering 8,975 toxins sourced from 5,220 references and patents across over 900 species. The activity data in BioTD are categorized into five groups: Activity, Safety, Kinetics, Hemolysis, and other physiological indicators. Moreover, BioTD provides data on 1,532 mutants, refines the whole sequence and signal peptide sequences of toxins, and annotates disulfide-bond information. All of the data in the database can be downloaded for free. Given the importance of biotoxins and their associated data, this new database is expected to attract broad interest from diverse research fields in drug discovery. BioTD is freely accessible at http://biotoxin.net/.
Gaoang Wang, Hang Wu, Yang Liao et al.· Journal of Chemical Informat...· 0 citations
Targeting the intrinsically disordered N-terminal domain of the androgen receptor (AR-NTD) represents a promising strategy to overcome resistance in prostate cancer. However, its inherent lack of a stable tertiary structure and highly dynamic conformational ensemble pose formidable challenges for rational drug design. This study introduces an integrated computational workflow that combines enhanced sampling techniques and machine learning collective variables to identify druggable conformations of the AR-NTD and elucidate the binding mechanism of its modulator, EPI-002. We characterize nine metastable states of the Tau-5 region and reveal that ligand recognition is driven by π–π stacking and structured water-mediated hydrogen bonds. Leveraging these insights, we perform structure-based virtual screening based on the identified druggable conformations and identify K53, a rationally designed AR-NTD antagonist, which exhibits potent anti-proliferative activity in enzalutamide-resistant prostate cancer cells. K53 directly binds the AR-NTD, suppresses AR transcriptional activity, and demonstrates high selectivity for cancer cells. This work provides a rational design paradigm for targeting intrinsically disordered proteins and offers a therapeutic candidate for resistant prostate cancer. In this work, the authors develop a machine learning–based enhanced sampling workflow to target the intrinsically disordered AR-NTD, identifying druggable conformations and enabling transferable modeling of ligand binding for rational drug discovery.
Kai Zhu, Huating Wang, Jintu Zhang et al.· Nature Communications· 0 citations
Spin-torque diodes based on magnetic tunnel junctions have emerged as a promising paradigm for microwave detection and neuromorphic computing due to their intrinsic nonlinear electrical characteristics. Traditional artificial intelligence image processing technology based on Von Neumann architecture faces the bottlenecks of power consumption and storage wall. Therefore, we propose and demonstrate hardware-inspired artificial neurons featuring a rectified linear unit (ReLU) activation function by leveraging injection-locked spin-torque diodes. Taking full advantage of the above properties, we construct a supervised autoencoder architecture for image denoising. The proposed architecture is validated on the MNIST dataset with additive Gaussian noise. Experimental results show that its denoising performance is on par with traditional software-implemented neural networks, featuring similar convergence performance and stable high-quality image reconstruction. Notably, even under heavy noise intensity of 0.3, the system consistently maintains a peak signal-to-noise ratio above 20 dB and a structural similarity index exceeding 0.85 across all scenarios. This study provides a viable hardware-native technical pathway toward edge artificial intelligence applications.
Jianling Hu, Zhenhao Liu, Bin Fang et al.· AIP Publishing· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
Multi-tenant Software-as-a-Service (SaaS) platforms rely on webhook change notifications from cloud APIs such as Microsoft Graph to synchronize calendar and resource-booking data in near-real time. Empirical analysis reveals a structural failure in this model, which we term webhook event amplification: a single state mutation triggers a burst of unordered notifications carrying no deduplication token. Controlled measurement shows the provider delivering approximately nine times the notifications the subscription model warrants, reaching a mean of 53 notifications per mutation in the most severe configuration. Left unmitigated, this exhausts downstream compute, degrades observability, and introduces race conditions that compromise transactional idempotency. Because providers offer no native suppression and delta queries target periodic synchronization rather than real-time deduplication, remediation must occur at the subscriber edge. This paper introduces the Distributed Notification Broker (DNB), which interposes a four-stage filtering pipeline between the webhook source and downstream consumers, combining tenant-isolated partition-keyed streaming, delta-validated event propagation, and content-hash business-change detection so that only verified, business-relevant changes propagate. Empirical evaluation over a 78-day observation period demonstrates an 88% reduction in downstream processing volume, a substantial reduction in concurrency-induced race conditions restoring system stability, an 89% reduction in volume-attributable infrastructure cost, and a 68% reduction in total infrastructure cost.
Aayush Pandey· Zenodo (CERN European Organi...· 0 citations
Industrial Internet Technology Application and Practice is compiled by the team of experts from Wenzhou Vocational Secondary School and fully meets the vocational talent training requirements of modern industries, laying a solid foundation for cultivating versatile and innovative technical talents in the era of Industry 4.0 and smart manufacturing.This book, as a practical companion textbook for vocational and technical training, is grounded in the laws of hands-on teaching and the characteristics of real-world industrial applications. Closely aligned with the talent training objectives of the Industrial Internet, it systematically integrates core content such as digital twin simulation, IoT networking, PLC programming, cloud platforms, and edge computing. It covers key projects including the Installation and Commissioning of Smart Logistics Systems, Smart Energy Consumption Systems, Smart Production Workshops, and Intelligent Warehousing Systems, helping students and practitioners build a complete and systematic knowledge framework of industrial Internet technologies and their practical applications.Author/Editor-in-Chief: Li Jiang, Wu Jie, Wang Bailiang
Li jiang, Wu Jie, Bailang Wang· Figshare· 0 citations
Driven by the escalating global agricultural workforce shortage and the urgent need to meet the “Zero Hunger” mandate, the automation of harvest–transport workflows has emerged as a cornerstone of Agriculture 4.0. This paper highlights the latest research progress in multi-robot collaborative logistics scheduling across highly unstructured farming environments, underpinned by cutting-edge spatial perception and digital twin frameworks. Initially, we summarize the technological leap from conventional 2D geometric mapping to multi-modal semantic 3D reconstruction—fusing light detection and ranging (LiDAR), unmanned aerial vehicle (UAV) imagery, and spatial data—to enable high-fidelity forward-looking predictions. The discussion then transitions to algorithmic advancements, emphasizing the shift from traditional centralized operations research to decentralized, data-driven approaches such as Multi-Agent Reinforcement Learning (MARL). We also explore micro-kinematic predictive control mechanisms and the growing integration of ecological sustainability metrics into routing models. To demonstrate practical engineering progress, multi-agent implementations are analyzed across three typical spatial settings: high-throughput continuous relays in open fields, global navigation satellite system (GNSS)-denied discrete routing in dense orchards, and close-proximity human–robot collaboration (HRC) in smart greenhouses. Finally, we identify the remaining barriers to the large-scale commercialization of Agricultural Cyber-Physical Systems (ACPS), such as the “Sim-to-Real” gap restricted by edge-computing capacities, unclosed economic loops, and HRC ethical dilemmas, offering a forward-looking roadmap for next-generation resilient agricultural networks.
Jun Zhang, Tiantian Jing, Ziqi Tian et al.· Sensors· 0 citations
This paper proposes a novel approach to computing, termed Dynamic Semantic Network Neuro-Morphic Computing, which leverages the principles of biological neural networks to achieve parallel, adaptive learning, and reasoning for complex data structures. The core idea is to mimic the dynamic connectivity and synaptic plasticity mechanisms found in biological neurons, creating a programmable neuro-morphic architecture. This architecture utilizes dynamically adjustable neural network connections and synaptic strengths based on input data and learning algorithms. A temporal signal processing and feedback mechanism simulates the dynamic behavior of biological neural networks, combined with Graph Neural Networks (GNNs) to construct and update dynamic semantic networks for representing and inferring relationships within data. This approach addresses the limitations of existing neuro-morphic computing, which primarily focuses on static hardware architectures, and traditional GNNs facing inefficiencies in handling large-scale, dynamic semantic networks. The resulting system aims to provide real-time learning and reasoning capabilities for complex data relationships, with potential benefits of low power consumption and high parallelism. The system is formally defined as follows: Let *S* = {*s*1, *s*2, ..., *s*K} be a set of nodes representing data elements. Let *E* = {*e*1, *e*2, ..., *e*N} be a set of edges representing relationships between nodes. Let *W* = {*w*ij} be a matrix representing the connection weights between nodes *s*i and *s*j, where *w*ij ∈ ℝ. Let *θ* = {*θ*ij} be a matrix representing the synaptic strengths between nodes *s*i and *s*j, where *θ*ij ∈ ℝ. Let *a*i ∈ ℝd be the activation value of node *s*i at time *t*. Let *l*i ∈ ℝ be the learning rate for node *s*i at time *t*. Let *h*i ∈ ℝd be the hidden state of node *s*i at time *t*. The dynamic update rule for node activation is given by: *a*i(t+1) = σ(*∑*j (*w*ij*h*j(t+1)) + *θ*ij *a*i(t+1)) where σ is an activation function (e.g., sigmoid, ReLU). The dynamic update rule for hidden state is given by: *h*i(t+1) = *f*(*a*i(t+1)) where *f* is a function that transforms the activation value into a hidden state. The learning rule updates the connection weights and synaptic strengths as follows: *w*ij(t+1) = *w*ij(t) + *l*i *∑*k (*w*ik(t+1) (*a*k(t+1)) ) *θ*ij(t+1) = *θ*ij(t) + *l*i *∑*k (*w*ik(t+1) (*a*k(t+1)) ) where *l*i is the learning rate. The system's performance is evaluated based on metrics such as accuracy, convergence time, and energy consumption. The core architecture will be implemented using a neuromorphic hardware platform, potentially utilizing spiking neural networks (SNNs) for efficient temporal processing.
Jincheng Zhang· Zenodo (CERN European Organi...· 0 citations
This paper introduces a novel approach to computation security leveraging the principles of spectral theory. Traditional computational security relies heavily on cryptographic methods, which are increasingly vulnerable to advancements in computing power and algorithmic attacks. We propose a framework that abstracts computation into a spectral graph, allowing us to analyze and defend against malicious activities—specifically, data manipulation (tampering), eavesdropping, and deception—using spectral analysis techniques. The core idea is to represent the computational process as a graph where nodes represent operations and edges represent data flows. Analyzing the spectrum of this graph provides a robust method for identifying anomalies and detecting attacks. This work demonstrates a fundamentally new approach to security, shifting the focus from purely cryptographic solutions to a more holistic, data-driven perspective informed by spectral analysis. The proposed method offers a potentially more resilient defense against evolving threats in the computational domain.
Jincheng Zhang· Zenodo (CERN European Organi...· 0 citations
This study employs a Graph Neural Network (GNN) to identify predictive associations between channel integration structures and brand equity, with a particular focus on heterogeneous nodes and multi-relation edges in multichannel commercial data. A Multi-Relation Integrated Graph Neural Network (MRI-GNN) is developed by incorporating node feature fitting, relation weight learning, and ensemble message passing into a unified optimization process. In the generalization evaluation, MRI-GNN achieves an accuracy of 93.54% and a recall of 92.39% for node classification on the Digital Bibliography & Library Project (DBLP) dataset, and an accuracy of 90.87% and a recall of 90.84% on the Association for Computing Machinery Citation Network dataset. When only 10% of the DBLP training data is used, the model still achieves an accuracy of 87.68%. In the brand equity prediction task based on multi-source commercial data, MRI-GNN achieves a root mean square error of 15.23, a mean absolute error of 12.11, and a coefficient of determination (R 2 ) of 0.81. Ablation results indicate that node feature fitting, information consistency, and price coordination make relatively substantial contributions to prediction performance. The results demonstrate that MRI-GNN can jointly represent node attributes and multi-relation structures, providing data-driven support for analyzing the association between channel integration and brand equity.
Haotian Xue, Jongbin Park· International Journal of Pat...· 0 citations
Abstract The integration of Internet of Things (IoT) and Digital Twin (DT) technologies provides a transformative paradigm for intelligent monitoring and predictive maintenance in industrial environments. Traditional environmental control systems rely on reactive architectures, leading to potential equipment failure before corrective actions are deployed. This paper presents a low-cost, real-time Environmental Digital Twin architecture designed to shift facility management from reactive to predictive. Utilizing a dual-node edge computing architecture—featuring an ATmega2560 for real-time spatial mapping and physical actuation, paired with an ESP32 for wireless network bridging—the physical system streams high-fidelity telemetry and proximity data. Simultaneously, a full-stack software architecture featuring a Python-driven backend and a React-based WebSocket dashboard provides low-latency visualization and algorithmic threshold monitoring. The system actively logs historical temperature and humidity data to forecast critical thermal breaches, proving that enterprise-grade predictive analytics can be achieved using accessible, scalable hardware.
Hayatullah Abdulwahab· Zenodo (CERN European Organi...· 0 citations
This paper presents the architectural blueprint of the OMEGA INFINITY KAORU processor, a computing substrate that solves the Boolean circuit satisfiability problem (Circuit-SAT)---the canonical NP-complete problem---in practical constant time, in strictly literal $O(\log n)$ time, and in $O(n)$ space. The architecture couples a conductive GRID, realized as a two-dimensional lattice of interconnect, to a digital Circuit-SAT instance. The positive terminal of a source is connected to the midpoint of the left edge of the GRID, while the right edge is interfaced to the Boolean inputs $v_1, v_2, \dots, v_n$ of the Circuit-SAT instance. The GRID concurrently explores all admissible conduction states; ambient physical variation (noise), which is discrete in nature, steers the current toward the path consistent with a satisfying assignment, in accordance with the principle of least action. The GRID can therefore be regarded as an enormous macroscopic, noise-resilient analogue of a qubit---a hypercomputational element that is not subject to the limitations of the BQP class. The satisfying assignment is recovered either by thresholded voltage measurement at the inputs $v_1, v_2, \dots, v_n$ or by the standard search-to-decision reduction, which becomes practical when the Circuit-SAT stage is implemented as a programmable processor rather than as a fixed lithographic pattern. Fabrication is fully viable with present-day photolithography, either as a single-use, instance-specific device or as a recommended programmable variant in which a conventional processor drives arbitrary SAT formulae into the GRID. Because the architecture resolves an NP-complete problem in practical constant (strictly, logarithmic) time and linear space, it establishes, in practice, $O(1)=\log\text{-time}=P=NP$. Since cryptographic constructions---RSA, elliptic-curve systems, and post-quantum schemes alike---reduce to SAT instances, they are solvable within the same practical constant time. The implications extend to artificial intelligence, optimization, logistics and the distribution of goods, automated mathematical reasoning, and drug discovery.
Kaoru Aguilera Katayama· Zenodo (CERN European Organi...· 0 citations
What if pathology foundation models could do more with less? GigaPath-Flash and GigaTIME-Flash cut computational demands while maintaining strong performance, opening the door to larger studies and broader exploration. The post GigaPath-Flash and GigaTIME-Flash: Toward population-scale discovery with efficient pathology foundation models appeared first on Microsoft Research.
MIT News · Artificial Intelligence· news.mit.eduAug 31, 2026
With millions of users across the world, Julia has been used to conduct cutting-edge research and to design new drugs, jet engines, heat pumps, and more.
MIT News · Artificial Intelligence· news.mit.eduAug 27, 2026
A new machine-learning framework aims to improve the success rate of computational protein design while moving away from results that reproduce sequences found in nature.
