503results about "Chemical structure search" patented technology

The invention provides a metal matrix composite microstructure strengthening simulation method and system, and relates to the technical field of composite mechanical property simulation, and the method comprises the following steps: establishing a polycrystalline aluminum matrix atomic model and a spherical silicon nitride particle strengthening phase atomic model unit; embedding the spherical silicon nitride particles into the polycrystalline aluminum matrix according to the target embedding position coordinates, and integrating to generate an atomic scale structure model; according to the method, an embedded atom potential function is adopted to describe interaction among atoms in a polycrystalline aluminum matrix, a Tersoft potential function is adopted to describe covalent bond interaction among atoms in spherical silicon nitride particles, and a Lennard-Jones potential function is adopted to describe interaction among heteroatoms at a Si3N4 / Al interface; based on the interaction potential between atoms, molecular dynamics simulation is adopted for conducting uniaxial stretching simulation on the atomic scale structure model, mechanical property parameters of the particle reinforced metal matrix composite are obtained through calculation, and a visualization method is combined for analyzing the microcosmic strengthening mechanism of the particle reinforced metal matrix composite.

The invention relates to the technical field of metal plastic forming, and discloses a titanium alloy radial forging reduction rate collaborative optimization method based on material testing, which comprises the following steps of: acquiring dynamic characteristic data in a forging process in real time by acquiring multi-dimensional material performance parameters, and constructing a multi-target collaborative optimization model; and intelligent collaborative configuration of the reduction rate parameters is realized by adopting a self-adaptive weight distribution strategy. And a thermal coupling factor normalization equation and a phase field free energy functional normalization model are established, and the technical problems that in a traditional forging process, due to empirical configuration of reduction rate parameters, the structure is uneven, and residual stress exceeds the standard are solved. A lightweight model is deployed through edge computing nodes, millisecond-level dynamic correction of an optimization scheme is realized in combination with a 5G industrial private network, and the grain size qualification rate and the size precision of the titanium alloy forgings are improved. The method improves the yield of the grain size of the forge piece, ensures that the residual stress of the forge piece is reduced, stably controls the size precision, and improves the batch stability of high-end titanium alloy parts.

The present application relates to the technical field of metal plastic forming, and discloses a titanium alloy radial forging reduction rate collaborative optimization method based on material testing, which acquires dynamic characteristic data of a forging process in real time through multi-dimensional material performance parameter collection, constructs a multi-objective collaborative optimization model, and realizes intelligent collaborative configuration of the reduction rate parameter by using an adaptive weight distribution strategy. A thermal coupling factor normalization equation and a phase field free energy functional normalization model are established to solve the technical problems of uneven structure and excessive residual stress caused by the empirical configuration of the reduction rate parameter in the traditional forging process. A lightweight model is deployed through an edge computing node, and millisecond-level dynamic correction of the optimization scheme is realized in combination with a 5G industrial private network, thereby improving the grain size qualification rate and size accuracy of titanium alloy forgings. The present application improves the grain size qualification rate of forgings, ensures stable control of the size accuracy while reducing the residual stress of forgings, and improves the batch stability of high-end titanium alloy components.

The invention discloses a multi-component catalyst active site prediction system and method fused with quantum embedding, and relates to the technical field of catalysis and material informatics, and the system comprises a structure and site enumeration module which generates candidate sites; the adaptive quantum embedding calculation module obtains key reaction microcosmic parameters; the unified site fingerprint and feature engineering module constructs and fuses standard site fingerprints; the physical consistency machine learning module predicts adsorption energy and other parameters and uncertainty thereof; the active learning and sample selection module selects a high-value sample optimization model; the microdynamics evaluation module calculates index values such as activity; and the multi-objective optimization and sorting module generates an optimization sorting list. According to the method, the unification of calculation precision and efficiency is realized, the problem of non-unification of locus characterization is solved, the model interpretability and extrapolation reliability are improved, and the comprehensive evaluation and optimization sorting of multi-target performance are completed.

The invention discloses a construction method and system for volatile organic compound release behavior characterization, and belongs to the technical field of air analysis and risk assessment. According to the method, feature analysis is carried out based on multi-source information such as a molecular structure, a release carrier and environmental conditions, modeling is carried out on a release process by utilizing a mapping operator, and a three-dimensional representation vector V = [I, S, E] reflecting release intensity, environmental sensitivity and time evolution is constructed. The representation is output in a behavior vector or digital label mode, and unified quantitative benchmarking of different substances in a cross-scene mode is achieved. According to the method, depth characteristics such as molecular polarity and desorption activation energy are introduced, so that the dependence on long-period environmental chamber experiments is reduced, the generalization prediction capability on unknown pollutants is improved, the logic connection problem between a bottom-layer physical mechanism and an upper-layer intelligent management and control system is solved, and the reliability of the system is improved. And standardized and explainable behavior input is provided for prospective risk research and judgment.

The invention belongs to the technical field of biological information, and relates to an acetylcholin esterase inhibitor prediction method based on Stacking ensemble learning and molecular feature fusion, which comprises the steps of data collection and preparation, data annotation and optimization, feature extraction and analysis, construction of a Stacking model, result verification and feedback and construction of a prediction platform. The molecular fingerprints and the property descriptors are used as features, and an acetylcholin esterase inhibitor classifier is successfully constructed by adopting a Stacking algorithm. According to the method, the problems that the efficiency of finding the acetylcholin esterase inhibitor by a traditional experimental method is low, and a common quantitative structure-function relationship method is high in complexity and poor in generalization ability can be solved, the new drug finding speed is increased, experimental candidates are accurately positioned, and resource waste is reduced.

The invention belongs to the technical field of traditional Chinese medicine information processing, and provides a traditional Chinese medicine disease prescription data base construction method and system.The method comprises the steps that five types of data including ancient literatures, modern medical records, prescription data, traditional Chinese medicine resources and medical record manuscript images are accessed through an extensible adapter frame; a traditional Chinese medicine term service platform is constructed by using multi-source term resources, and standardized data is generated from the analyzed data; the method comprises the following steps: extracting diseases, syndromes, prescriptions and traditional Chinese medicine entities by adopting a rule engine and a domain self-adaptive entity recognition model, on the basis of a public traditional Chinese medicine standard knowledge graph, aggregating neighbor node information through GAT, and introducing a contrast learning optimization entity relationship classifier to obtain an enhanced entity vector and classify a semantic relationship; constructing a knowledge graph based on the extracted semantic relationship; and carrying out data quality and algorithm performance monitoring and tracing on data access analysis, standardization processing, knowledge extraction and graph construction. Uniform access and standardized integration of multi-source heterogeneous traditional Chinese medicine data are realized.

The invention relates to the field of in-situ pH value prediction, and discloses an in-situ pH value prediction model construction method and device in an oil and gas field pipeline, equipment and a medium, which can obtain the value ranges of a plurality of in-situ pH influence parameters in the oil and gas field pipeline, and generate a plurality of environmental condition data according to the value ranges of the plurality of in-situ pH influence parameters. And performing in-situ pH value simulation on each piece of environmental condition data to obtain a corresponding in-situ pH simulation value, taking each piece of environmental condition data and the corresponding in-situ pH simulation value as training data, and performing symbolic regression analysis based on all the training data to generate an in-situ pH value prediction model of the oil and gas field pipeline. The in-situ pH value prediction model can be used for predicting the in-situ pH value of the oil and gas field pipeline under different working conditions, the model construction cost is low, the reliability is high, the prediction condition is simple and feasible, the prediction process is efficient and accurate, and the in-situ pH value prediction model can be widely applied to the in-situ pH value prediction of the oil and gas field pipeline under different environment conditions.

The invention discloses a compound smell prediction method based on an isotropic graph neural network. The method comprises the following steps: (1) constructing molecular graph representation; (2) constructing an isovariant graph neural network model; (3) preparing training data; (4) training the model; (5) evaluating the model; and (6) predicting the odor of the compound. The invention further discloses a system of the compound smell prediction method based on the isotropic graph neural network. The method provided by the invention has relatively high prediction accuracy and relatively strong generalization ability.

Disclosed are systems and methods for identifying a target molecule by receiving a molecule dataset from a library of molecules comprising chemical-structural representations of candidate molecules for a material application; determining one or more target molecular property values derived from a set of predictive molecular descriptor values determined from a candidate molecule of the molecular dataset applied to a trained machine learning model, wherein the trained ML model is configured to output the set of predictive molecular descriptor values for a given molecule data input, and wherein the trained ML model was trained for a set of training data and the set of molecular descriptors generated from a molecular descriptor modeling application; and outputting, via a report or to a data store, the one or more target molecular property values for each of the molecular dataset.

The invention belongs to the technical field of high voltage and insulation, and relates to an XLPE breakdown field strength evaluation method and system based on gel content and preparation process, and the method comprises the following steps: S1, sampling, weighing, shearing and bagging; s2, extracting, weighing and calculating the gel content; s3, constructing an evaluation model; s4, predicting the breakdown field intensity; according to the technical scheme, the physical model fusing the cross-linking enhancement effect and the thermal damage effect is constructed, the breakdown field strength is predicted based on the physical model, accurate and rapid evaluation of the XLPE insulation performance under different process conditions can be achieved only through a small amount of breakdown test data calibration model, dependence on a traditional destructive electrical test is remarkably reduced, and the reliability of the XLPE insulation performance is improved. The test cost is greatly reduced, the research and development period is greatly shortened, and an effective tool is provided for cable insulation material performance analysis and production process optimization.

To provide a creation method of initial experimental data and an experiment planning method for creating proper initial experimental data, while reducing calculation costs.SOLUTION: A creation method of initial experimental data includes an initial experimental data creation step for creating initial experimental data 34 by extracting data with which combinations have high adaptability from experiment candidate data 32. In the initial experimental data creation step, the initial experimental data 34 is created by repeating: a selection step in which N pieces of data are extracted from data 33 in process, after the experiment candidate data 32 has become the data 33 in process, and M pieces of data (where M<N) with which combinations have high adaptability are selected from the extracted N pieces of data, by using a genetic algorithm; and a data update step for updating the data 33 in process by combining the data selected in the selection step, and data remaining of the data 33 in process that was not extracted in the selection step.SELECTED DRAWING: Figure 1

A method for machine learning aided modeling of two interacting structures may include: (a) receiving an input structure comprising an interaction region; (b) generating a plurality of candidate structures using a first differentiable machine learning model; (c) docking one or more candidate structures of the plurality of candidate structures at the interaction region of the input structure using a second differentiable machine learning model to predict a docking geometry; (d) ranking the one or more candidate structures of the plurality of candidate structures docked in (c) using a third differentiable machine learning model to predict a score; and (e) backpropagating the score to (i) the first differentiable machine learning model to update the plurality of candidate structures or (ii) the second differentiable machine learning model to update the docking geometry.

A method performed by at least one processor, includes receiving a query comprising at least one of a product, a reactant, a reagent, and a reaction condition; generating a query embedding vector corresponding to the query by inputting the query into a neural network model; extracting a candidate embedding vector from among a plurality of embedding vectors based on a similarity between the query embedding vector and the plurality of embedding vectors corresponding to reaction records stored in a database; and outputting a synthesis recipe corresponding to the query by retrieving a candidate reaction record corresponding to the candidate embedding vector in the database.

The invention belongs to the technical field of biology, and discloses a non-natural pathway generation method and system based on a continuous biological template, and the method comprises the following steps: obtaining an enzymatic reaction data set, and preprocessing the enzymatic reaction data set; constructing a directed reaction diagram on the basis of the preprocessed enzymatic reaction data set, performing similar compound search on a target compound by utilizing the directed reaction diagram, and obtaining a continuous group conversion template of the similar compound in the natural enzymatic pathway by taking the searched similar compound as a starting point and combining a depth-first search mode; the continuous group conversion template is applied to a target compound in a molecular linear input specification form, and an inverse synthesis process of the target compound is simulated to generate a non-natural synthesis pathway based on the continuous group conversion template. The method not only enhances the biological feasibility and construction success rate of the approach, but also has universality, expandability and engineering potential, and provides a generalizable normal form for the field of synthetic biology.

The invention discloses a multi-target joint optimized antibiotic molecule design system and method applied to the field of biological medicine, and the method comprises the steps: obtaining FabI enzyme and AcrAB-TolC efflux pump structure data, extracting double-target space and physicochemical characteristics, and constructing a joint pharmacophore model containing a FabI inhibition domain and an efflux pump blocking domain; screening candidate molecules meeting a double-domain space matching threshold, screening high-affinity double-target molecules through conformation sampling and free energy calculation, and forming a lead compound library through multi-layer filtration of metabolic stability, membrane penetrability, quantum chemical properties and synthesis feasibility; carrying out electron effect, chain length or heteroatom fine tuning on dominant molecules through in-vitro bacteriostasis and efflux inhibition experiment feedback, and carrying out iterative optimization until MIClt is met; 2 [mu] g / mL and the efflux inhibition rate is gt; according to the present invention, the antibacterial effect and the drug resistance inhibition ability are significantly improved, the molecular synthesizability and the biological safety are ensured, and the engineering design new path is provided for the Gram-negative bacterium drug resistance crisis.

The invention relates to the technical field of fluorescent probes, and provides a fluorescent probe luminescence brightness quantification method and system for coupling excitation wavelength detuning. The fluorescence probe luminescence brightness quantification method for coupling excitation wavelength detuning comprises the following steps: constructing an initial structure of a probe molecule, and optimizing a geometric structure of a ground state of the probe molecule to obtain a stable configuration of the ground state of the probe molecule; based on the stable configuration of the ground state of the probe molecule, the excitation states are calculated, and the excitation energy of the lowest n excitation states of the probe molecule and the absorption transition dipole moments of the probe molecule in the X, Y and Z directions are obtained; based on the stable configuration of the ground state of the probe molecule, optimizing the geometric structure of the lowest excited state of the probe molecule to obtain the emission energy of the lowest excited state of the probe molecule and the emission transition dipole moments of the probe molecule in the X, Y and Z directions; and calculating a luminance index in combination with the energy parameter and the transition dipole moment parameter so as to select an optimal probe. Therefore, accurate and rapid prediction of the luminescence brightness of the probe is realized.

System, methods, apparatuses, and computer program products are disclosed for searching a chemical structure database based on centroids. A query is executed against a chemical structure database by first vectorizing a molecular representation associated with the query into a query feature vector. The query feature vector is compared to centroid feature vectors to determine a centroid feature vector associated with a chemical structure representation that satisfies a similarity condition with the query feature vector. A first subset of the chemical structure database that includes chemical structures of the chemical structure database associated with the determined centroid feature vector is searched to determine whether the molecular representation associated with the query is present in the first subset of the chemical structure database.

The invention discloses an explanation method for damage threshold improvement of a fused quartz material in a femtosecond laser processing / welding process. By establishing a fused quartz electron-lattice heat transfer model under femtosecond laser irradiation, a dissociation model of environment molecules and interstitial molecules under the action of femtosecond laser and a defect-containing fused quartz molecule dynamic reaction model based on a ReaxFF reaction force field, atomic structure changes in a system before and after a reaction are analyzed; and finally, obtaining a chemical reaction mechanism of the fused quartz containing the defects under the action of the femtosecond laser, and explaining the phenomenon that the damage threshold of the fused quartz is increased after the fused quartz is irradiated by the low-flux femtosecond laser. According to the method, the research in the current fused quartz femtosecond laser processing / welding process is perfected, and meanwhile, the theoretical blank that the damage threshold lifting mechanism is not clear in the fused quartz glass femtosecond laser processing / welding process is filled up.

Systems and methods for querying a combinatorial synthesis library comprising a plurality of compounds and representing a plurality of reaction types, where each reaction type maps to a plurality of reactants, and each reactant maps to a plurality of synthons, accepts a query in the form of a single graph into a molecular encoder model, thereby obtaining a query vector. The query vector is inputted into a reaction query generator model thereby obtaining a first reaction type and a first plurality of reactants. A synthon is determined for each reactant by inputting the reactant into a synthon query generator model. A set of synthons is therefore determined, each corresponding to a reactant in the first plurality of reactants. A molecular structure in the combinatorial synthesis library is identified that includes the set of synthons arranged in accordance with a synthesis rule associated with the first reaction type.

The invention relates to the field of computational energy chemistry and material preparation, and discloses a method for screening a chemical looping gasification oxygen carrier through machine learning. Comprising the steps of system configuration and initialization, sample database construction, model loading and training, to-be-screened oxygen carrier data set construction, data preprocessing, oxygen carrier performance prediction, multi-condition screening, result output and visualization and the like. According to the method, the screening efficiency of the chemical-looping gasification oxygen carrier is greatly improved, the screening cost of the chemical-looping gasification oxygen carrier is remarkably reduced, and the method also has the characteristics of multi-index collaborative optimization, high expansibility and result visualization.

A method for computational drug design includes defining a population of a plurality of compounds. Each compound includes one or more molecular properties. The method includes defining a training set of compounds from the population for which one or more biological properties are known. The method includes selecting, from the population, a subset of one or more compounds that are not in the training set. The method includes determining a subset score of the selected subset based on molecular properties of the one or more compounds in the selected subset, and evaluating the selected subset based on the determined subset score. The subset score is determined based on a frequency of the molecular properties in the population and on a frequency of the molecular properties in a sampled set comprising the training set and the selected subset.

The invention relates to the technical field of medical artificial intelligence, in particular to a drug discovery task processing method, device and equipment based on large model reasoning, and the method comprises the steps: obtaining a drug discovery task instruction of a user; the drug discovery task instruction is input to the target large model, the target large model outputs an execution result of a drug discovery task, and the reasoning process of the target large model comprises the steps that the drug discovery task instruction is analyzed to obtain an instruction intention and instruction parameters, a target key matched with the instruction intention is selected from a parameterized memory pool based on the instruction intention, and the target key is input to the target large model; the instruction parameter is mapped into a target key value, a target calling tool format is generated based on the target key and the target key value, and a corresponding target discovery tool is called based on the target calling tool format to execute a drug discovery task; and sending an execution result of the drug discovery task to the user. According to the method, the target large model automatically executes the reasoning process, various tools are automatically called, and intelligent drug discovery is achieved.

The invention relates to an energy-absorbing bionic lattice structure design method, which belongs to the technical field of bionics, and is characterized in that a bionic tendon wave lattice structure is designed by referring to an ostrich metatarsophalangeal joint tendon collagenous fiber bundle microstructure in nature; the energy absorption characteristic of the structure is analyzed through tests and simulation, and optimal structure parameters are obtained in combination with a genetic algorithm; the superelastic material and the bionic tendon wave lattice structure are selected to be combined, the diversified requirements of different products for mechanical properties can be effectively met, and in multiple application occasions, impact energy can be efficiently dispersed and absorbed, movement impact force can be effectively dispersed, and a high-quality damping effect can be provided.

The invention discloses a training method, device and equipment of a hydrogen terminal diamond structure-spectrum-effect relationship prediction model and a readable storage medium, and relates to the technical field of material science. Comprising the following steps: firstly, obtaining structure parameters of the hydrogen terminal diamond, and carrying out first-principle high-throughput calculation on the structure parameters to obtain property parameters of the hydrogen terminal diamond; constructing a relevance data set between the structure parameters and the property parameters; then carrying out infrared spectrum calculation on the structure parameters of the hydrogen terminal diamond to obtain spectrum data of the hydrogen terminal diamond; constructing a mapping data set between the structural parameters and the spectral data; and finally, based on the relevance data set and the mapping data set, training a hydrogen terminal diamond structure-spectrum-effect relationship prediction model. According to the scheme, the high efficiency and accuracy of hydrogen terminal diamond structure-spectrum-effect relationship prediction are remarkably improved.

Examples are disclosed that relate to forming embeddings comprising vector representations of chemical structures, and performing chemical similarity searches based on embeddings. One example provides a method (200) comprising receiving (202) a query comprising chemical structure information for a target chemical object, and, based on the query, inputting (214) the chemical structure information into a trained neural network configured to form embeddings of chemical structures. The method further comprises receiving (218) an embedding for the target chemical object from the trained neural network, the embedding comprising a vector representation of the target chemical object, based at least on a similarity score between the embedding for the target chemical object and each of one or more embeddings stored in a vector database, retrieving (220) query results from the vector database, the query results comprising a set of embeddings and metadata for a corresponding set of chemical objects, and outputting (226) the query results.

Provided is a technique for predicting the presence or absence of aroma properties or olfactory receptor activation properties in a substance. In this technique, the presence or absence of the aforesaid target properties in a test substance is predicted on the basis of the maximum similarity in three-dimensional chemical structure between the test substance and a control substance.

The invention provides a three-dimensional porous medium generation method based on a random generation method, and belongs to the technical field of three-dimensional structure generation. The method comprises the following steps of: firstly, specifying porous medium parameters needing to be generated by a user; secondly, placing a growth core in the porous medium domain, and growing until the required porosity is met; and thirdly, smoothing the edge of the porous medium by using a filter. And finally, modeling the porous medium data. Compared with a computed tomography technology and a focused ion beam scanning electron microscope technology, the method has the advantages that the three-dimensional porous medium structures with similar quality can be obtained without professional instruments, and the three-dimensional porous medium structures meeting requirements can be generated in a very short time; a user can obtain a required three-dimensional porous medium structure by adjusting different parameters.

The invention relates to the technical field of environmental monitoring and risk prevention and control, in particular to a target pollutant and fingerprint coordinated water pollution quantitative traceability technology and system, and establishes a water pollution traceability technology coordinated with multi-dimensional fingerprint information of a target pollutant and an environmental matrix thereof. Further constructing databases such as a target pollutant database, a water chemical characteristic fingerprint database, a stable isotope fingerprint database and a soluble organic matter fingerprint database, drawing a multi-dimensional pollution source information puzzle taking the target pollutant as an anchor point, and realizing accurate identification and contribution quantification of the water environment pollution source; the water environment pollution tracing system comprising a data acquisition layer, a feature extraction layer, an information analysis layer and a result output layer is constructed based on the established pollution source information puzzle concept, and the water environment pollution tracing system is complete, simple in process, efficient in output, reliable in result and capable of meeting the water pollution tracing and environmental litigation evidence requirements of various complex environments.

The invention discloses a method for calculating and predicting a defect diffusion behavior of a ferrite martensitic stainless steel oxide layer. The method belongs to the technical field of new materials, and comprises the following operation steps: respectively constructing crystal structure models of a ferrite martensitic stainless steel oxide layer with different defect sites by using MS software, and converting a structure data file into a. Vasp file format readable by VASP software by using VESTA software; adopting VASP software to calculate defect formation energy of different defect sites; screening to form stable defects with low energy, and constructing an initial state model and a final state model of defect diffusion; using a CI-NEB method to calculate diffusion energy barriers of the defect in different paths, and using an approximation formula after approximation optimization to calculate a diffusion rate; and further analyzing the behavior and influence of the defect through a state density map and an electronic local area map. According to the method, the diffusion energy barrier and diffusion rate of the defect between different paths are effectively predicted through a theoretical calculation method, and required data support can be provided for phase field simulation and molecular dynamics simulation.