Machine Learning(ML)-based force fields are attracting ever-increasing interest due to their capacity to span spatiotemporal scales of classical interatomic potentials at quantum-level accuracy.They can be trained bas...Machine Learning(ML)-based force fields are attracting ever-increasing interest due to their capacity to span spatiotemporal scales of classical interatomic potentials at quantum-level accuracy.They can be trained based on high-fidelity simulations or experiments,the former being the common case.However,both approaches are impaired by scarce and erroneous data resulting in models that either do not agree with well-known experimental observations or are under-constrained and only reproduce some properties.Here we leverage both Density Functional Theory(DFT)calculations and experimentally measured mechanical properties and lattice parameters to train an ML potential of titanium.We demonstrate that the fused data learning strategy can concurrently satisfy all target objectives,thus resulting in a molecularmodel of higher accuracycompared to the models trained with a single data source.The inaccuracies of DFT functionals at target experimental properties were corrected,while the investigated off-target properties were affected only mildly and mostly positively.Our approach is applicable to any material and can serve as a general strategy to obtain highly accurate ML potentials.展开更多
While the forward and backward modeling of the process-structure-property chain has received a lot of attention from the materials’ community,fewer efforts have taken into consideration uncertainties.Those arise from...While the forward and backward modeling of the process-structure-property chain has received a lot of attention from the materials’ community,fewer efforts have taken into consideration uncertainties.Those arise from a multitude of sources and their quantification and integration in the inversion process are essential in meeting the materials design objectives.The first contribution of this paper is a flexible,fully probabilistic formulation of materials’ optimization problems that accounts for the uncertainty in the process-structure and structure-property linkages and enables the identification of optimal,high-dimensional,process parameters.We employ a probabilistic,data-driven surrogate for the structure-property link which expedites computations and enables handling of non-differential objectives.We couple this with a problem-tailored active learning strategy,i.e.,a self-supervised selection of training data,which significantly improves accuracy while reducing the number of expensive model simulations.We demonstrate its efficacy in optimizing the mechanical and thermal properties of two-phase,random media but envision that its applicability encompasses a wide variety of microstructure-sensitive design problems.展开更多
Graph Neural Network(GNN)potentials relying on chemical locality offer near-quantum mechanical accuracy at significantly reduced computational costs.Message-passing GNNs model interactions beyond their immediate neigh...Graph Neural Network(GNN)potentials relying on chemical locality offer near-quantum mechanical accuracy at significantly reduced computational costs.Message-passing GNNs model interactions beyond their immediate neighborhood by propagating local information between neighboring particles while remaining effectively local.However,locality precludes modeling long-range effects critical to many real-world systems,such as charge transfer,electrostatic interactions,and dispersion effects.In this work,we propose the Charge Equilibration Layer for Long-range Interactions(CELLI)to address the challenge of efficiently modeling non-local interactions.This novel architecture generalizes the classical charge equilibration(Qeq)method to a model-agnostic building block for modern equivariant GNN potentials.Therefore,CELLI extends the capability of GNNs to model longrange interactions while providing high interpretability through explicitly modeled charges.On benchmark systems,CELLI achieves state-of-the-art results for strictly local models.CELLI generalizes to diverse datasets and large structureswhile providing high computational efficiency and robust predictions.展开更多
In this paper,we propose a novel flexible optimization pipeline for determining the optimal adsorption sites,named AUGUR(Aware of Uncertainty Graph Unit Regression).Our model combines graph neural networks and Gaussia...In this paper,we propose a novel flexible optimization pipeline for determining the optimal adsorption sites,named AUGUR(Aware of Uncertainty Graph Unit Regression).Our model combines graph neural networks and Gaussian processes to create a flexible,efficient,symmetry-aware,translation,and rotation-invariant predictor with inbuilt uncertainty quantification.This predictor is then used as a surrogate for a data-efficient Bayesian Optimization scheme to determine the optimal adsorption positions.This pipeline determines the optimal position of large and complicated clusters with far fewer iterations than current state-of-the-art approaches.Further,it does not rely on hand-crafted features and can be seamlessly employed on any molecule without any alterations.Additionally,the pooling properties of graphs allow for the processing of molecules of different sizes by the same model.This allows the energy prediction ofcomputationally demanding systemsby a model trained on comparatively smaller and less expensive ones.展开更多
Infrared(IR)spectroscopy is a pivotal analytical tool as it provides real-time molecular insight into material structures and enables the observation of reaction intermediates in situ.However,interpreting IR spectra o...Infrared(IR)spectroscopy is a pivotal analytical tool as it provides real-time molecular insight into material structures and enables the observation of reaction intermediates in situ.However,interpreting IR spectra often requires high-fidelity simulations,such as density functional theory based ab-initio molecular dynamics,which are computationally expensive and therefore limited in the tractable system size and complexity.In this work,we present a novel active learning-based framework,implemented in the open-source software package PALIRS,for efficiently predicting the IR spectra of small catalytically relevant organic molecules.PALIRS leverages active learning to train a machine-learned interatomic potential,which is then used for machine learning-assisted molecular dynamics simulations to calculate IR spectra.PALIRS reproduces IR spectra computed with ab-initio molecular dynamics accurately at a fraction of the computational cost.PALIRS further agrees well with available experimental data not only for IR peak positions but also for their amplitudes.This advancement with PALIRS enables high-throughput prediction of IR spectra,facilitating the exploration of larger and more intricate catalytic systems and aiding the identification of novel reaction pathways.展开更多
Transforming CO_(2)into methanol represents a crucial step towards closing the carbon cycle,with thermoreduction technology nearing industrial application.However,obtaining high methanol yields and ensuring the stabil...Transforming CO_(2)into methanol represents a crucial step towards closing the carbon cycle,with thermoreduction technology nearing industrial application.However,obtaining high methanol yields and ensuring the stability of heterocatalysts remain significant challenges.Herein,we present a sophisticated computational framework to accelerate the discovery of thermal heterogeneous catalysts,using machine-learned force fields.We propose a new catalytic descriptor,termed adsorption energy distribution,that aggregates the binding energies for different catalyst facets,binding sites,and adsorbates.The descriptor is versatile and can be adjusted to a specific reaction through careful choice of the key-step reactants and reaction intermediates.By applying unsupervised machine learning and statistical analysis to a dataset comprising nearly 160 metallic alloys,we offer a powerful tool for catalyst discovery.We propose new promising candidates such as ZnRh and ZnPt_(3),which to our knowledge,have not yet been tested,and discuss their possible advantage in terms of stability.展开更多
Imaging offers a fast and accessible means for spatial characterization of halide perovskite photovoltaic materials,yet extracting optoelectrical properties—such as power conversion efficiency(PCE)—remains challengi...Imaging offers a fast and accessible means for spatial characterization of halide perovskite photovoltaic materials,yet extracting optoelectrical properties—such as power conversion efficiency(PCE)—remains challenging.This study presents a deep learning methodology that correlates optical reflective images of perovskite solar cells with their PCE by focusing on image differences rather than absolute visual features.The approach predicts relative changes in PCE by comparing images of the same device in different states(e.g.,before and after encapsulation)or against a reference image.This comparative technique significantly outperforms traditional methods that attempt to directly infer PCE from a single image.Furthermore,it demonstrates high effectiveness in low-data regimes,using only 115 samples.By leveraging convolutional neural networks(CNNs)trained on small datasets,the method offers an adaptable and scalable solution for device characterization.Overall,the comparative approach enhances the accuracy and applicability of machine vision in perovskite solar cell analysis.展开更多
Neural network models for audio tasks,such as automatic speech recognition(ASR)and acoustic scene classification(ASC),are susceptible to noise contamination for real-life applications.To improve audio quality,an enhan...Neural network models for audio tasks,such as automatic speech recognition(ASR)and acoustic scene classification(ASC),are susceptible to noise contamination for real-life applications.To improve audio quality,an enhancement module,which can be developed independently,is explicitly used at the front-end of the target audio applications.In this paper,we present an end-to-end learning solution to jointly optimise the models for audio enhancement(AE)and the subsequent applications.To guide the optimisation of the AE module towards a target application,and especially to overcome difficult samples,we make use of the sample-wise performance measure as an indication of sample importance.In experiments,we consider four representative applications to evaluate our training paradigm,i.e.,ASR,speech command recognition(SCR),speech emotion recognition(SER),and ASC.These applications are associated with speech and nonspeech tasks concerning semantic and non-semantic features,transient and global information,and the experimental results indicate that our proposed approach can considerably boost the noise robustness of the models,especially at low signal-to-noise ratios,for a wide range of computer audition tasks in everyday-life noisy environments.展开更多
Predicting and monitoring battery life early and across chemistries is a significant challenge due to the plethora of degradation paths,form factors,and electrochemical testing protocols.Existing models typically tran...Predicting and monitoring battery life early and across chemistries is a significant challenge due to the plethora of degradation paths,form factors,and electrochemical testing protocols.Existing models typically translate poorly across different electrode,electrolyte,and additive materials,mostly require a fixed number of cycles,and are limited to a single discharge protocol.Here,an attention-based recurrent algorithm for neural analysis(ARCANA)architecture is developed and trained on an ultralarge,proprietary dataset from BASF and a large Li-ion dataset gathered from literature across the globe.ARCANA generalizes well across this diverse set of chemistries,electrolyte formulations,battery designs,and cycling protocols and thus allows for an extraction of data-driven knowledge of the degradation mechanisms.The model’s adaptability is further demonstrated through fine-tuning on Na-ion batteries.ARCANA advances the frontier of large-scale time series models in analytical chemistry beyond textual data and holds the potential to significantly accelerate discovery-oriented battery research endeavors.展开更多
Capacity planning is a very important global challenge in the face of Covid-19 pandemic.In order to hedge against the fluctuations in the random demand and to take advantage of risk pooling effect,one needs to have a ...Capacity planning is a very important global challenge in the face of Covid-19 pandemic.In order to hedge against the fluctuations in the random demand and to take advantage of risk pooling effect,one needs to have a good understanding of the variabilities in the demand of resources.However,Covid-19 predictive models that are widely used in capacity planning typically often predict the mean values of the demands(often through the predictions of the mean values of the confirmed cases and deaths)in both the temporal and spatial dimensions.They seldom provide trustworthy prediction or estimation of demand variabilities,and therefore,are insufficient for proper capacity planning.Motivated by the literature on variability scaling in the areas of physics and biology,we discovered that in the Covid-19 pandemic,both the confirmed cases and deaths exhibit a common variability scaling law between the average of the demand μ and its standard deviationσ,that is,σ ∝ μ^(β),where the scaling parameterμis typically in the range of 0.65 to 1,and the scaling law exists in both the temporal and spatial dimensions.Based on the mechanism of contagious diseases,we further build a stylized network model to explain the variability scaling phenomena.We finally provide simple models that may be used for capacity planning in both temporal and spatial dimensions,with only the predicted mean demand values from typical Covid-19 predictive models and the standard deviations of the demands derived from the variability scaling law.展开更多
The use of three-dimensional self-assembled metallacages(MCgs)as multimodal drug platforms holds great promise.However,the synthesis of MCgs with increased complexity and functionality is a great challenge since under...The use of three-dimensional self-assembled metallacages(MCgs)as multimodal drug platforms holds great promise.However,the synthesis of MCgs with increased complexity and functionality is a great challenge since understanding of the interaction of MCgs with biological targets is still limited.In this context,this work reports on the integration of a gold(III)porphyrin scaffold into a prismatic MCg structure and explores its application for multimodal therapy of cancer in vitro,namely enabling both photodynamic therapy and chemotherapy.Combining experimental approaches with a state-of-the-art metadynamics theoretical study,we discovered that the gold cage shows unprecedented host-vip interaction-driven selective stabilization of guanine-quadruplex(G4)structures-validated anticancer drug targets-disclosing a new mechanism to pursue in the design of supramolecular drugs.展开更多
Metal-organic frameworks(MOF)are an attractive class of porous materials due to their immense design space,allowing for application-tailored properties.Properties of interest,such as gas sorption,can be predicted in s...Metal-organic frameworks(MOF)are an attractive class of porous materials due to their immense design space,allowing for application-tailored properties.Properties of interest,such as gas sorption,can be predicted in silico with molecular mechanics simulations.However,the accuracy is limited by the available empirical force field and partial charge estimation scheme.In this work,we train a graph neural network for partial charge prediction via active learning based on Dropout Monte Carlo.We show that active learning significantly reduces the required amount of labeled MOFs to reach a target accuracy.The obtained model generalizes well to different distributions of MOFs and Zeolites.In addition,the uncertainty predictions of Dropout Monte Carlo enable reliable estimation of the mean absolute error for unseen MOFs.This work paves the way towards accurate molecular modeling of MOFs via next-generation potentials with machine learning predicted partial charges,supporting insilico material design.展开更多
基金funded by the Deutsche Forschungsgemeinschaft(DFG,German Research Foundation)-534045056.
文摘Machine Learning(ML)-based force fields are attracting ever-increasing interest due to their capacity to span spatiotemporal scales of classical interatomic potentials at quantum-level accuracy.They can be trained based on high-fidelity simulations or experiments,the former being the common case.However,both approaches are impaired by scarce and erroneous data resulting in models that either do not agree with well-known experimental observations or are under-constrained and only reproduce some properties.Here we leverage both Density Functional Theory(DFT)calculations and experimentally measured mechanical properties and lattice parameters to train an ML potential of titanium.We demonstrate that the fused data learning strategy can concurrently satisfy all target objectives,thus resulting in a molecularmodel of higher accuracycompared to the models trained with a single data source.The inaccuracies of DFT functionals at target experimental properties were corrected,while the investigated off-target properties were affected only mildly and mostly positively.Our approach is applicable to any material and can serve as a general strategy to obtain highly accurate ML potentials.
基金Funded under the Excellence Strategy of the Federal Government and the Länder in the context of the ARTEMIS Innovation Network.
文摘While the forward and backward modeling of the process-structure-property chain has received a lot of attention from the materials’ community,fewer efforts have taken into consideration uncertainties.Those arise from a multitude of sources and their quantification and integration in the inversion process are essential in meeting the materials design objectives.The first contribution of this paper is a flexible,fully probabilistic formulation of materials’ optimization problems that accounts for the uncertainty in the process-structure and structure-property linkages and enables the identification of optimal,high-dimensional,process parameters.We employ a probabilistic,data-driven surrogate for the structure-property link which expedites computations and enables handling of non-differential objectives.We couple this with a problem-tailored active learning strategy,i.e.,a self-supervised selection of training data,which significantly improves accuracy while reducing the number of expensive model simulations.We demonstrate its efficacy in optimizing the mechanical and thermal properties of two-phase,random media but envision that its applicability encompasses a wide variety of microstructure-sensitive design problems.
基金Funded by the European Union. Views and opinions expressed are, however, those of the author(s) only and do not necessarily reflect those of the European Union or the European Research Council Executive AgencyNeither the European Union nor the granting authority can be held responsible for them. This work was funded by the ERC (StG SupraModel) - 101077842the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) - 534045056 and 561190767.
文摘Graph Neural Network(GNN)potentials relying on chemical locality offer near-quantum mechanical accuracy at significantly reduced computational costs.Message-passing GNNs model interactions beyond their immediate neighborhood by propagating local information between neighboring particles while remaining effectively local.However,locality precludes modeling long-range effects critical to many real-world systems,such as charge transfer,electrostatic interactions,and dispersion effects.In this work,we propose the Charge Equilibration Layer for Long-range Interactions(CELLI)to address the challenge of efficiently modeling non-local interactions.This novel architecture generalizes the classical charge equilibration(Qeq)method to a model-agnostic building block for modern equivariant GNN potentials.Therefore,CELLI extends the capability of GNNs to model longrange interactions while providing high interpretability through explicitly modeled charges.On benchmark systems,CELLI achieves state-of-the-art results for strictly local models.CELLI generalizes to diverse datasets and large structureswhile providing high computational efficiency and robust predictions.
基金funding from the project ProperPhotoMile,supported under the umbrella of SOLAR-ERA.NET Cofund 2 by The Spanish Ministry of Science and Education and the AEI under the project PCI2020-112185 and CDTI project number IDI-20210171the Federal Ministry for Economic Affairs and Energy on the basis of a decision by the German Bundestag project number FKZ 03EE1070B and FKZ 03EE1070A+2 种基金the Israel Ministry of Energy with project number 220-11-031.SOLAR-ERA.NET is supported by the European Commission within the EU Framework Program for Research and Innovation HORIZON 2020(Cofund ERA-NET Action,786483)Further,A.G.acknowledges financial support from TUM Innovation Network for Artificial Intelligence powered Multifunctional Material Design(ARTEMIS)funding in the framework of Deutsche Forschungsgemeinschaft(DFG,German Research Foundation)under Germany’s Excellence Strategy-EXC 2089/1-390776260(e-conversion).Lastly,we wish to express our gratitude to Dr.Inigo Iribarren for creating the flowcharts for this work.
文摘In this paper,we propose a novel flexible optimization pipeline for determining the optimal adsorption sites,named AUGUR(Aware of Uncertainty Graph Unit Regression).Our model combines graph neural networks and Gaussian processes to create a flexible,efficient,symmetry-aware,translation,and rotation-invariant predictor with inbuilt uncertainty quantification.This predictor is then used as a surrogate for a data-efficient Bayesian Optimization scheme to determine the optimal adsorption positions.This pipeline determines the optimal position of large and complicated clusters with far fewer iterations than current state-of-the-art approaches.Further,it does not rely on hand-crafted features and can be seamlessly employed on any molecule without any alterations.Additionally,the pooling properties of graphs allow for the processing of molecules of different sizes by the same model.This allows the energy prediction ofcomputationally demanding systemsby a model trained on comparatively smaller and less expensive ones.
基金funding from Horizon Europe MSCA Doctoral network grant n.101073486, EUSpecLab, funded by the European UnionO.K. and P.R. have received funding from the European Union – NextGenerationEU instrument and are funded by the Research Council of Finland (grant numbers 348179, 346377, and 364227). We acknowledge CSC, Finland for awarding access to the LUMI supercomputer, owned by the EuroHPC Joint Undertaking, hosted by CSC (Finland) and the LUMI consortium through CSC, Finland, extreme-scale project ALVS. The authors also gratefully acknowledge the additional computational resources provided by CSC – IT Center for Science, Finland, and the Aalto Science-IT project.
文摘Infrared(IR)spectroscopy is a pivotal analytical tool as it provides real-time molecular insight into material structures and enables the observation of reaction intermediates in situ.However,interpreting IR spectra often requires high-fidelity simulations,such as density functional theory based ab-initio molecular dynamics,which are computationally expensive and therefore limited in the tractable system size and complexity.In this work,we present a novel active learning-based framework,implemented in the open-source software package PALIRS,for efficiently predicting the IR spectra of small catalytically relevant organic molecules.PALIRS leverages active learning to train a machine-learned interatomic potential,which is then used for machine learning-assisted molecular dynamics simulations to calculate IR spectra.PALIRS reproduces IR spectra computed with ab-initio molecular dynamics accurately at a fraction of the computational cost.PALIRS further agrees well with available experimental data not only for IR peak positions but also for their amplitudes.This advancement with PALIRS enables high-throughput prediction of IR spectra,facilitating the exploration of larger and more intricate catalytic systems and aiding the identification of novel reaction pathways.
基金funding from the European Union – NextGenerationEU instrument and the Research Council of Finland's AICon project (grant number no. 348179). The authors gratefully acknowledge CSC – IT Center for Science, Finland, and the Aalto Science-IT project for generous computational resources.
文摘Transforming CO_(2)into methanol represents a crucial step towards closing the carbon cycle,with thermoreduction technology nearing industrial application.However,obtaining high methanol yields and ensuring the stability of heterocatalysts remain significant challenges.Herein,we present a sophisticated computational framework to accelerate the discovery of thermal heterogeneous catalysts,using machine-learned force fields.We propose a new catalytic descriptor,termed adsorption energy distribution,that aggregates the binding energies for different catalyst facets,binding sites,and adsorbates.The descriptor is versatile and can be adjusted to a specific reaction through careful choice of the key-step reactants and reaction intermediates.By applying unsupervised machine learning and statistical analysis to a dataset comprising nearly 160 metallic alloys,we offer a powerful tool for catalyst discovery.We propose new promising candidates such as ZnRh and ZnPt_(3),which to our knowledge,have not yet been tested,and discuss their possible advantage in terms of stability.
基金EXC 2089:e-conversion DFG-cluster of excellence-TUM innovation network,Technical University of Munich,funded through the German Excellence Initiative and the state of Bavaria(A.G.,M.H.).Support by the project ProperPhotoMile is also gratefully acknowledged(A.G.,I.V-F.),under the umbrella of SOLAR-ERA.NET Cofund 2 by The Spanish Ministry of Science and Education and the AEI under the project PCI2020-112185 and CDTI project number IDI-20210171the Federal Ministry for Economic Affairs and Energy on the basis of a decision by the German Bundestag project number FKZ03EE1070B and FKZ 03EE1070A+2 种基金and the Israel Ministry of Energy with project number 220-11-031SOLAR-ERA.NET is supportedby the European Commission within the EU Framework Programme for Research and Innovation HORIZON 2020(Cofund ERA-NETAction,N 786483)D.K.K.is grateful for the Blaustein postdoctoral fellowship at BGU.R.K.G.is grateful for the Swiss Inst.of Dryland Environmental and Energy Research postdoctoral fellowship at BGU.The authors are also grateful for partial support by the Israel Ministry of Energy,project number 222-11-081.
文摘Imaging offers a fast and accessible means for spatial characterization of halide perovskite photovoltaic materials,yet extracting optoelectrical properties—such as power conversion efficiency(PCE)—remains challenging.This study presents a deep learning methodology that correlates optical reflective images of perovskite solar cells with their PCE by focusing on image differences rather than absolute visual features.The approach predicts relative changes in PCE by comparing images of the same device in different states(e.g.,before and after encapsulation)or against a reference image.This comparative technique significantly outperforms traditional methods that attempt to directly infer PCE from a single image.Furthermore,it demonstrates high effectiveness in low-data regimes,using only 115 samples.By leveraging convolutional neural networks(CNNs)trained on small datasets,the method offers an adaptable and scalable solution for device characterization.Overall,the comparative approach enhances the accuracy and applicability of machine vision in perovskite solar cell analysis.
基金supported by the Affective Computing&HCI Innovation Research Lab between Huawei Technologies and the University of Augsburg,and the EU H2020 Project under Grant No.101135556(INDUX-R).
文摘Neural network models for audio tasks,such as automatic speech recognition(ASR)and acoustic scene classification(ASC),are susceptible to noise contamination for real-life applications.To improve audio quality,an enhancement module,which can be developed independently,is explicitly used at the front-end of the target audio applications.In this paper,we present an end-to-end learning solution to jointly optimise the models for audio enhancement(AE)and the subsequent applications.To guide the optimisation of the AE module towards a target application,and especially to overcome difficult samples,we make use of the sample-wise performance measure as an indication of sample importance.In experiments,we consider four representative applications to evaluate our training paradigm,i.e.,ASR,speech command recognition(SCR),speech emotion recognition(SER),and ASC.These applications are associated with speech and nonspeech tasks concerning semantic and non-semantic features,transient and global information,and the experimental results indicate that our proposed approach can considerably boost the noise robustness of the models,especially at low signal-to-noise ratios,for a wide range of computer audition tasks in everyday-life noisy environments.
基金This work contributes to the research performed at CELEST(Center for Electrochemical Energy Storage Ulm-Karlsruhe)and was partly funded by the German Research Foundation(DFG)under Project ID 390874152(POLiS Cluster of Excellence)This project also received funding from the European Union’s Horizon 2020 research and innovation program under grant agreement No.957189(BIG-MAP)+1 种基金funded by the European Union’s Horizon 2020 research and innovation program under Grant Agreement No.957213HSS acknowledges funding from the German Research Foundation(DFG)under Project ID 390776260(eConversion Cluster of Excellence).
文摘Predicting and monitoring battery life early and across chemistries is a significant challenge due to the plethora of degradation paths,form factors,and electrochemical testing protocols.Existing models typically translate poorly across different electrode,electrolyte,and additive materials,mostly require a fixed number of cycles,and are limited to a single discharge protocol.Here,an attention-based recurrent algorithm for neural analysis(ARCANA)architecture is developed and trained on an ultralarge,proprietary dataset from BASF and a large Li-ion dataset gathered from literature across the globe.ARCANA generalizes well across this diverse set of chemistries,electrolyte formulations,battery designs,and cycling protocols and thus allows for an extraction of data-driven knowledge of the degradation mechanisms.The model’s adaptability is further demonstrated through fine-tuning on Na-ion batteries.ARCANA advances the frontier of large-scale time series models in analytical chemistry beyond textual data and holds the potential to significantly accelerate discovery-oriented battery research endeavors.
基金This research was supported in part by the National Natural Science Foundation of China(72042015,72091211,72031006 and 71722006).
文摘Capacity planning is a very important global challenge in the face of Covid-19 pandemic.In order to hedge against the fluctuations in the random demand and to take advantage of risk pooling effect,one needs to have a good understanding of the variabilities in the demand of resources.However,Covid-19 predictive models that are widely used in capacity planning typically often predict the mean values of the demands(often through the predictions of the mean values of the confirmed cases and deaths)in both the temporal and spatial dimensions.They seldom provide trustworthy prediction or estimation of demand variabilities,and therefore,are insufficient for proper capacity planning.Motivated by the literature on variability scaling in the areas of physics and biology,we discovered that in the Covid-19 pandemic,both the confirmed cases and deaths exhibit a common variability scaling law between the average of the demand μ and its standard deviationσ,that is,σ ∝ μ^(β),where the scaling parameterμis typically in the range of 0.65 to 1,and the scaling law exists in both the temporal and spatial dimensions.Based on the mechanism of contagious diseases,we further build a stylized network model to explain the variability scaling phenomena.We finally provide simple models that may be used for capacity planning in both temporal and spatial dimensions,with only the predicted mean demand values from typical Covid-19 predictive models and the standard deviations of the demands derived from the variability scaling law.
文摘The use of three-dimensional self-assembled metallacages(MCgs)as multimodal drug platforms holds great promise.However,the synthesis of MCgs with increased complexity and functionality is a great challenge since understanding of the interaction of MCgs with biological targets is still limited.In this context,this work reports on the integration of a gold(III)porphyrin scaffold into a prismatic MCg structure and explores its application for multimodal therapy of cancer in vitro,namely enabling both photodynamic therapy and chemotherapy.Combining experimental approaches with a state-of-the-art metadynamics theoretical study,we discovered that the gold cage shows unprecedented host-vip interaction-driven selective stabilization of guanine-quadruplex(G4)structures-validated anticancer drug targets-disclosing a new mechanism to pursue in the design of supramolecular drugs.
基金the TUM Innovation Network“Artificial Intelligence Powered Multifunctional Material Design”(ARTEMIS)funded through the German Excellence Initiative and the state of Bavariathe“EXC 2089:e-conversion”DFG-cluster of excellence[project number:390776260]。
文摘Metal-organic frameworks(MOF)are an attractive class of porous materials due to their immense design space,allowing for application-tailored properties.Properties of interest,such as gas sorption,can be predicted in silico with molecular mechanics simulations.However,the accuracy is limited by the available empirical force field and partial charge estimation scheme.In this work,we train a graph neural network for partial charge prediction via active learning based on Dropout Monte Carlo.We show that active learning significantly reduces the required amount of labeled MOFs to reach a target accuracy.The obtained model generalizes well to different distributions of MOFs and Zeolites.In addition,the uncertainty predictions of Dropout Monte Carlo enable reliable estimation of the mean absolute error for unseen MOFs.This work paves the way towards accurate molecular modeling of MOFs via next-generation potentials with machine learning predicted partial charges,supporting insilico material design.
