67 results about "Atomic coordinates" patented technology

Systems and methods for characterizing an interaction between a compound and a polymer include obtaining a plurality of sets of atomic coordinates. Each set of atomic coordinates comprises the compound bound to the polymer in a corresponding pose in a plurality of poses. Each respective set of atomic coordinates, or an encoding thereof, is sequentially inputted into a neural network, to obtain a corresponding initial embedding as output, thereby obtaining a plurality of initial embeddings. Each initial embedding corresponds to a set of atomic coordinates in the plurality of sets of atomic coordinates. An attention mechanism is applied to the plurality of initial embeddings, in concatenated form, to obtain an attention embedding. A pooling function is applied to the attention embedding to derive a pooled embedding. The pooled embedding is inputted into a model to obtain an interaction score of the interaction between the compound and the polymer.

The invention belongs to the field of glass material research, and discloses a method for regulating and controlling rare earth ion clusters in laser glass based on molecular dynamics simulation and application of the method. Based on given glass components, density and atomic number, an initial amorphous structure model M0 is constructed; based on a molecular dynamics theory, performing high-temperature melting quenching simulation on the initial amorphous structure model M0, and constructing a glass structure model M; calculating glass microstructure parameters and designing rare earth ion cluster parameters based on atomic coordinate information of the glass structure model M; establishing a glass component-microstructure mathematical model and a microstructure-rare earth ion cluster mathematical model based on the glass components, the microstructure parameters and the rare earth ion cluster parameters; based on the model, regulation and control on the agglomeration degree of the rare earth ions in the glass are realized. According to the method, the problem that the rare earth ion clusters in the laser glass cannot be quickly and effectively regulated and controlled due to the fact that the rare earth ion clusters in the glass are difficult to quantitatively describe by experimental means is solved.

The invention relates to the technical field of computers, in particular to a multi-scale automatic analysis method for an amorphous material atomic structure. Comprising the following steps: S1, taking a reactant as an initial unit, and obtaining a cluster unit of a reaction product based on first principle molecular dynamics simulation; s2, carrying out unit cell construction on the cluster unit, and optimizing unit cell parameters and atomic coordinates of the cluster unit through calculation according to a first principle to obtain unit cell units; s3, spatial stacking is carried out on the unit cell units, unit cell parameters and atomic coordinates of the unit cell units are optimized through calculation according to a first principle, and supercell units are obtained; s4, performing first principle molecular dynamics simulation on the supercell unit to obtain an accurate machine learning potential function; s5, carrying out cell expansion construction on the supercell unit, carrying out molecular dynamics simulation by using an accurate machine learning potential function, and obtaining an amorphous material atomic structure; and S6, packaging the steps S1 to S5 into an automatic workflow. The method is high in efficiency and small in error.

The invention relates to the technical field of semiconductor manufacturing, and discloses a wafer doping control method based on intelligent feedback adjustment, and the method comprises the steps: carrying out the atomic scale imaging of the surface of a wafer, so as to recognize a target doping site, and generating an atomic coordinate mapping graph, secondly, inputting the atomic coordinate mapping graph into a multi-scale simulation model to generate an initial process parameter set, and processing the initial process parameter set by combining a transformer model with a reinforcement learning algorithm to obtain a dynamic process data matrix; performing optimization processing on the dynamic process data matrix through a cross-material process migration graph neural network to obtain a process parameter migration mapping relation, comparing measured data with a multi-scale model prediction result through closed-loop feedback iteration, and outputting a corrected process parameter migration mapping relation and a multi-scale model updating instruction; according to the wafer doping method, the stability, the consistency and the yield of the wafer doping process are improved, and high accuracy and high uniformity of wafer doping are realized.

The invention discloses a method for identifying a molecular structure and constructing a reaction network based on atomic coordinates, and belongs to the technical field of material simulation. According to the method for recognizing the molecular structure based on the atomic coordinates, by obtaining trajectory data, atoms outside a box are converted into atoms inside the box through minimum mirror agreement; grids are divided according to a space truncation radius, potential bonding atoms are accurately screened, and redundancy of distance calculation between every two atoms of the whole system is avoided; atomic clusters are quickly divided in combination with depth-first search, and molecular structure recognition is achieved. The method only depends on basic data such as atomic coordinates and does not depend on a simulation software post-processing module, and the limitation that only a single simulation system is adapted in the prior art is broken through. According to the method for constructing the reaction network, molecular structure description is compared frame by frame, and invalid reactions are offset and filtered through positive and negative reactions to obtain net reactions; counting the net reaction times to determine the reaction probability so as to quantify the tendency of the reaction; and constructing a reaction network by taking the molecular structure as a node and the reaction frequency as an edge weight.

Embodiments of the present application provide a three-dimensional molecular structure generation method, device, equipment and storage medium, at least applied to the field of artificial intelligence and the field of drug synthesis, wherein the method comprises: obtaining atomic type noise and atomic coordinate noise at a current time step; constructing a full connection adjacency matrix based on the atomic type noise; performing three-dimensional isometry processing on the atomic type noise, the atomic coordinate noise and the full connection adjacency matrix to obtain a molecular structure distribution average value at the current time step; generating a molecular structure of the molecule iteratively based on the molecular structure distribution average value at the current time step to obtain a molecular representation of the molecule; and constructing the three-dimensional molecular structure through the molecular representation. Through the present application, the efficiency of three-dimensional molecular structure generation can be accelerated and cumulative errors can be avoided, so as to accurately generate a three-dimensional molecular structure of an effective molecule.

The invention relates to the field of artificial intelligence, in particular to a molecular property prediction method, system, device and equipment, and the method comprises the steps: firstly obtaining 3D molecular structure data containing atomic coordinate information; according to the set radius alpha and the filling value sigma, zero-dimensional, one-dimensional and two-dimensional simplex structures are extracted, and substructure position coding information is calculated; then, taking one-dimensional and two-dimensional simplex structures as atomic feature indexes, and fusing atomic topology and atomic features; on the basis of zero-dimensional simplex structure and substructure position coding, obtaining pairing position coding information of an atom-simplex pair; inputting the fused atomic features and pairing position coding information into a multi-head self-attention module to generate global molecular features; and finally, obtaining a molecular property prediction result according to the global molecular characteristics. According to the method, the defect of processing fine-grained substructure information in the prior art can be effectively solved, and the 3D molecular property prediction performance is greatly improved.

The present invention relates to the fields of chemistry and materials science, and discloses a visualization system for large-scale quantum chemistry, which calculates the energy information of each structure in the calculation system, and finds the minimum-energy structure in the system as the optimized structure by means of gradient descent, and displays the structures before and after optimization in the form of an atomic coordinate matrix for easy visualization; at the same time, it can also obtain information such as charge density, differential charge density tensor, difference tensor, etc. according to the energy information of the structure, and draw visual images by using these information.

The present invention discloses a method, device, terminal and storage medium for generating molecular motion trajectories, and relates to drug discovery technology. The method comprises: using a pre-trained generative model, generating the atomic coordinate information of several subsequent frames frame by frame for the atomic category sequence of the target protein-ligand complex and the atomic coordinate information of the first frame; the pre-trained generative model is trained based on the escape trajectory of an existing protein-ligand complex; and generating the molecular motion trajectory of the target protein-ligand complex according to the atomic category sequence of the target protein-ligand complex and the atomic coordinate information of all frames. The present invention simulates the molecular motion trajectory by adopting a paradigm of generating coordinates frame by frame by a generative model, and learns the inter-frame coordinate evolution law of the ligand in the process of escaping the protein pocket through the existing escape trajectory. In the inference stage, it is only necessary to input the atomic category sequence and the atomic coordinate information of the initial state of the first frame, and the atomic coordinate information of the subsequent frames can be iteratively generated quickly and accurately.

The application relates to the technical field of semiconductor manufacturing, and discloses a wafer doping control method based on intelligent feedback adjustment, which comprises the following steps: atomic scale imaging is performed on the surface of a wafer to identify target doping sites and generate an atomic coordinate mapping diagram, the atomic coordinate mapping diagram is input into a multi-scale simulation model to generate an initial process parameter set, a dynamic process data matrix is obtained by combining a transformer model with a reinforcement learning algorithm to process the initial process parameter set, then, the dynamic process data matrix is optimized by a cross-material process migration graph neural network to obtain a process parameter migration mapping relationship, and measured data and multi-scale model prediction results are compared through closed-loop feedback iteration and output of a corrected process parameter migration mapping relationship and a multi-scale model update instruction. The application improves the stability, consistency and yield of wafer doping processes, and realizes high precision and high uniformity of wafer doping.

The invention discloses a method for screening local vibration modes of an atomic system. The method comprises the following steps of: 1, acquiring an atomic structure file, a target atomic coordinate file and a feature vector file containing a real part and an imaginary part of vibration displacement of each atom; 2, analyzing the coordinate file to extract lattice basis vectors, atomic coordinates and element information; 3, fusing the real parts of the feature vectors to generate a total vibration vector file of each atom in a frequency domain; step 4, generating visual vibration data based on the atomic structure and the total vibration vector; 5, setting an incident light direction, extracting a vibration component in the direction, and generating a transitional vibration vector in combination with the coordinate file; step 6, selecting frequency and fusing transitional vectors to generate Raman / infrared spectrogram data; and step 7, optimizing the spectrogram through Gaussian broadening to complete screening. According to the invention, vector synthesis can be carried out on the displacement of each atom in each direction under different frequencies and visual display is carried out, so that atom vibration can be observed more visually.

The disclosure relates generally to methods and systems for implementation of density functional theory on quantum computing through meta generalized gradient approximation (meta GGA). Conventional methods implement DFT simulations on classical processors which is time consuming. The present disclosure provides a quantum circuit for computing the direct matrix efficiently on a quantum processor. Initially, a plurality of atomic coordinates of each atom of a chemical compound whose one or more properties are to be extracted, are obtained. Electron integrals, a core Hamiltonian, and a collocation matrix are computed from the plurality of atomic coordinates. The core Hamiltonian is diagonalized to obtain an initial density matrix of the chemical compound. The initial density matrix is further updated iteratively until a convergence criteria is satisfied using the meta GGA to obtain a final density matrix. The final density matrix is used to extract the one or more properties of the chemical compound.

The present application relates to the technical field of computers, and relates to a kind of amorphous material atomic structure multiscale automatic analysis method. Including: step S1: with reactant as initial unit, based on first principle molecular dynamics simulation obtains the cluster unit of reaction product;Step S2: the cluster unit is constructed, and its cell parameters and atomic coordinates are optimized using first principle calculation, to obtain cell unit;Step S3: the cell unit is stacked in space, and its cell parameters and atomic coordinates are optimized using first principle calculation, to obtain supercell unit;Step S4: the supercell unit is simulated by first principle molecular dynamics, to obtain accurate machine learning potential function;Step S5: the supercell unit is constructed by cell expansion, and the molecular dynamics simulation is carried out using accurate machine learning potential function, to obtain amorphous material atomic structure;Step S6: steps S1 to S5 are packaged into automation workflow. The method is efficient, and the error is small.

The invention provides a method for reconstructing a three-dimensional atomic configuration of a special interface, and the method comprises the steps: combining actual representation data with theoretical calculation, generating a theoretical template, and carrying out the matching with an actual image, thereby improving the precision of atomic coordinate extraction; the position of each atom is calibrated through Gaussian fitting, and it is ensured that the precision reaches the sub-angstrom scale; atomic phase information in a two-dimensional transmission electron microscope image is utilized, and a direction matrix and coordinate transformation are combined to realize reconstruction from a two-dimensional atomic projection image to a three-dimensional crystal structure.

The present invention describes a molecular dynamics simulation method for tensile testing of a refractory CuNb amorphous alloy with heterogeneous components. The specific steps are: establishing a CuNb crystal alloy model with uniform component distribution; using an EAM potential function to describe the atomic interaction between Cu and Nb; simulating the process of rapid heating and rapid cooling of the model to obtain a uniform CuNb amorphous alloy model, and outputting atomic coordinates; shearing and combining uniform amorphous alloy models with different atomic ratios to establish a series of amorphous alloy tensile models with different component distributions; performing tensile simulation on the amorphous alloy tensile model to output atomic information under different strains; visualizing the atomic information under different strains to output the deformation morphology, strain cloud map, stress cloud map and atomic cluster distribution map of the amorphous alloy tensile model under different strains; revealing the plastic deformation mechanism of the refractory CuNb amorphous alloy and the influence mechanism of component distribution on the tensile deformation of the refractory CuNb amorphous alloy.

The application relates to a multi-descriptor and coordinate reversible mapping atomic clustering analysis method and system, which comprises the following steps: inputting a set of multiple descriptors representing atomic arrangement characteristics; selecting single descriptor data or comprehensive descriptor data obtained by multi-descriptor weighted combination according to system characteristics by a user; performing clustering analysis on the selected descriptor data, and generating descriptor data grouping and a sample index set corresponding to each group; extracting the Coulomb matrix of each sample in each group from the set of multiple descriptors based on the sample index set of each group, forming a Coulomb matrix subset of each group and respectively calculating the Coulomb matrix mean value; and applying a multi-dimensional scaling algorithm to reduce the Coulomb matrix mean value of each group to a three-dimensional space to generate corresponding three-dimensional atomic coordinates of each group. Compared with the prior art, the application improves the analysis capability of atomic or molecular arrangement characteristics in the data of simulation outputs such as molecular dynamics.

The application relates to a training method and device of a crystal property prediction model and a computer device. The method comprises the following steps: obtaining a sample crystal and a property label corresponding to the sample crystal; the sample crystal comprises a plurality of sample atoms; determining respective atomic coordinates of each sample atom, and determining orbital characteristics of the sample atoms according to the atomic coordinates; determining respective atomic potential energies of the sample atoms according to the orbital characteristics, and determining a predicted property corresponding to the sample crystal according to the atomic potential energies; training the crystal property prediction model according to the difference between the predicted property and the property label, stopping when a first training stop condition is reached, and obtaining a trained crystal property prediction model; and the trained crystal property prediction model is used to determine the property of the crystal. The method can improve the accuracy of predicting the property of the crystal.

The application discloses a kind of covalent crystal structure intelligent prediction method based on molecular unit topological assembly, by regarding molecular unit as rigid body and using hybrid gene coding system, topological connection and geometric pose decoupling are realized, compared with traditional atomic coordinate random search, the application can keep the internal configuration of molecule stable during evolution, avoid the collapse of geometric configuration of complex molecule during search process, and improve the calculation efficiency of large-scale, non-spherical molecular system;The method does not depend on specific empirical parameters, and can cover the structure prediction of fullerene polymer, covalent organic framework and complex porous material through general topological assembly logic, and has strong cross-field universality.

The application provides an electron density information acquisition method, system, medium, terminal and program product based on machine learning. The method comprises the following steps: first, constructing a real differential density according to a real electron density and a reference electron density; inputting atomic species, atomic coordinates, cell information and uniform grid definition information of a target structure into a machine learning model for training, and taking the real differential density as a supervision target during the training to construct a differential density prediction model; deploying the constructed differential density prediction model to output a predicted differential density of a to-be-tested complex structure according to to-be-tested complex structure information; and performing a recovery operation on the output predicted differential density to obtain total electron density information of the to-be-tested complex structure. The application can acquire electron density information of a complex structure with high precision and high efficiency, and stably support downstream calculations such as photoelectric properties and quantum transport of heterostructures, interface structures and device structures.

The invention relates to a protein binding conformation prediction method and system. The protein binding conformation-based prediction method comprises the following steps: acquiring a static structure file of a target protein, and extracting atomic information to construct a residue index table; performing standardization processing on the atomic coordinates in the residue index table, and calculating a distance matrix between the residue pairs based on the standardized atomic coordinates; based on the conformation sampling data and the residue index table, obtaining a residue motion freedom degree through adaptive weighting calculation and correlation correction; a dynamic distance threshold value is obtained based on the residue motion freedom degree; converting the distance matrix into a binary contact graph based on the dynamic distance threshold, obtaining the contact frequency of the residue pair, and constructing a contact frequency matrix; extending the residue motion degree of freedom into a symmetric matrix, and fusing the contact frequency matrix and the symmetric matrix to obtain an initial dynamic probability matrix; performing normalization processing and probability propagation on the initial probability matrix in sequence to obtain a dynamic probability matrix; protein binding conformations are predicted based on the dynamic probability matrix.

The invention discloses a molecular motion trail generation method and device, a terminal and a storage medium, and relates to a drug discovery technology. The method comprises the following steps: generating atomic coordinate information of a plurality of subsequent frames frame by frame according to an atomic category sequence of a target protein-ligand compound and atomic coordinate information of a first frame through a pre-trained generation model; the pre-trained generation model is obtained by training based on the escape trajectory of the existing protein-ligand compound; and according to the atom category sequence of the target protein-ligand compound and the atom coordinate information of all frames, generating a molecular movement track of the target protein-ligand compound. According to the method, a normal form of generating coordinates frame by frame is adopted to simulate a molecular motion track, and an inter-frame coordinate evolution rule of a ligand in an escape protein pocket process is learned through an existing escape track. In the reasoning stage, the atomic coordinate information of the subsequent frames can be rapidly and accurately iteratively generated only by inputting the atomic category sequence and the atomic coordinate information of the initial state of the first frame.

This invention discloses a molecular dynamics calculation method and system based on quantum simulation. The method first applies a discrete quantum Fourier transform to the atomic coordinates of the molecular system, mapping the original plane wave basis to a system containing only O(N) atoms. 2 The invention constructs a plane-wave bibasic quantum mechanics equation; then, it constructs and solves the electronic structure Hamiltonian on a quantum computer to obtain the ground state energy. This quantum energy is combined with the classical empirical potential energy according to predetermined weighting coefficients to calculate the total synthesized energy, from which the atomic forces are determined. The integration time step is then adaptively adjusted based on a comparison of the quantum energy change between consecutive time steps with a preset threshold. This adaptive mechanism shortens the time step to ensure accuracy when quantum effects are significant, and lengthens the step to improve efficiency when changes are gradual. This invention significantly reduces Hamiltonian complexity, balancing quantum accuracy and classical efficiency, and is applicable to various fields such as drug screening, materials design, and catalytic reaction simulation.

The present invention discloses a method for analyzing glass phase separation based on molecular dynamics simulation. The method comprises the following steps: constructing an amorphous initial structure model of the glass; selecting a potential function for the interaction forces between the atoms of the glass according to the types of the initial glass atoms, and performing correction processing to obtain a corrected potential function; using the corrected potential function to evolve the amorphous initial structure model into a glass structure, and collecting the atomic coordinate information of the glass structure; inputting the atomic coordinate information into the visualization software VMD to obtain a glass phase separation structure diagram, and further obtaining the microscopic structure information of the glass phase separation. Based on the molecular dynamics theory, the present invention obtains the structural phase separation information at the atomic level of the glass through the molecular dynamics simulation method, makes up for the deficiency that the experimental research cannot achieve the atomic-level structure analysis, helps to more accurately understand the glass structure phase separation behavior, and regulates the glass phase separation.

The invention discloses a method for constructing a molecular dynamics model of monolayer molybdenum disulfide containing point defects. First, an atomic coordinate is obtained by constructing a molecular dynamics model of defect-free monolayer molybdenum disulfide through the open-source software LAMMPS. Secondly, the coordinate relationship between defective atoms is obtained by analyzing the atomic defect forms of six kinds of point-defect molybdenum disulfide, and a program for processing the atomic coordinates of six kinds of point-defect molybdenum disulfide is obtained through programming according to the coordinate relationship between atoms. Finally, the atomic coordinates of defect-free monolayer molybdenum disulfide are imported into the program for operation, and a molecular dynamics model of monolayer molybdenum disulfide with specific defects and specific defect concentrations is obtained, providing a reference for the modeling of point-defect molybdenum disulfide in the study of various properties of monolayer defective molybdenum disulfide at the nanoscale.

The invention discloses a method for exploring the influence of electromagnetic waves on a protein structure directly based on molecular dynamics simulation. The method comprises the following steps: acquiring target protein from a protein database; constructing a simulation system; adding a protein force field and a water molecule model; solvent molecules and ions are added to neutralize system charges, and energy minimization treatment is carried out; pre-balancing is carried out under the NVT ensemble and the NPT ensemble; performing molecular dynamics simulation on the simulation system, and storing an atomic coordinate trajectory; and performing post-processing on the track file, extracting distance time sequence data between atoms or residues, screening out atom pairs or residue pairs meeting conditions, performing Fourier transform analysis, and judging whether the possibility of resonance with the electromagnetic wave with the specific frequency exists or not. According to the method, the interaction between the protein and the electromagnetic field can be revealed under the atomic resolution without consumption of experimental materials, and an efficient and reliable molecular-level theoretical basis is provided for evaluating the biological effect of environmental electromagnetic radiation.

The invention discloses a wettability-adjustable porous quartz atomic-scale model generation method based on CT scanning data, and relates to the technical field of petroleum engineering and molecular simulation, and the method comprises the following steps: obtaining template coordinates; generating a quartz substrate; carrying out atom screening and hollowing; performing surface atom cleaning and surface hydroxylation treatment on the initial atomic model; on the basis of a wettability field obtained through CT scanning, the functional group type of a hydroxylation area on the pore surface is automatically adjusted, and a target atom model is obtained; and exporting the atomic coordinates of the target atomic model as a standard atomic structure file. According to the method, an atomic-scale model with geometric fidelity, chemical stability and adjustable wettability can be automatically constructed from CT scanning data, so that the physical consistency and prediction precision of molecular simulation on real rock core wetting behaviors are improved, and a high-quality initial structure is provided for molecular dynamics simulation and oil and gas flow research.

The application discloses a three-dimensional molecular structure generation method and model, relates to the technical field of the cross of computational chemistry and machine learning, and the three-dimensional molecular structure generation method comprises the steps of: generating an initial three-dimensional conformation by adopting a preset distance geometry algorithm in combination with a preset molecular force field optimization based on molecular topological information; constructing three SE(3) isovariant geometric graphs according to the initial three-dimensional conformation and the molecular topological information, and generating isovariant node representations of the graphs based on the SE(3) isovariant geometric graphs; fusing the isovariant node representations of the SE(3) isovariant geometric graphs to obtain a molecular embedding; predicting atomic coordinate offsets according to the molecular embedding and preset DFT true value coordinates; and generating a target three-dimensional molecular structure according to the initial three-dimensional conformation and the atomic coordinate offsets. The application can realize high-precision, high-efficiency and strong-robustness three-dimensional conformation prediction under limited computing resources and data scale.

The invention discloses an execution method and device of a molecular dynamics constraint algorithm, a chip and a medium. The execution method comprises the steps that a GPU chip sequentially determines a molecule as a current molecule in a plurality of substance molecules to be processed; obtaining an atomic coordinate and an atomic velocity corresponding to the current molecule, and executing a molecular dynamics constraint algorithm operation on the current molecule according to the atomic coordinate, the atomic velocity and a plurality of predetermined target parameters; the multiple target parameters comprise atomic position correlation parameters and PV values; judging whether operation on all the molecules is completed or not; and if not, returning to execute the operation of sequentially determining one molecule as the current molecule in the plurality of to-be-processed substance molecules until the operation on all the molecules is completed. According to the technical scheme of the embodiment of the invention, the execution efficiency of the GPU chip on the molecular dynamics constraint algorithm can be improved.

The invention discloses an AI crystal analysis method established based on a ChatGLM model, and the method is characterized in that the method comprises the steps: obtaining a training data set, and the data comprises a crystal name, a chemical formula, corresponding lattice parameters, atomic coordinates, a space group, an element type, the position of a crystal in a unit cell, and the size and shape of the crystal; a ChatGLM model is trained; and analyzing the crystal structure by adopting the trained ChatGLM model. The method has a great reference effect on cross application of the field of artificial intelligence and the field of material chemistry.

This invention discloses a heterogeneous node electrostatic force calculation device for molecular dynamics simulations, comprising: a CPU node for constructing molecular topology, calculating Gaussian width parameters, initializing short-range force lookup tables, and outputting atomic trajectories and system energy after simulation; an FPGA node connected to the CPU node via a PCIe bus for selecting short-range atom pairs based on truncation radius and atomic coordinates, and calculating short-range electrostatic forces in parallel using a lookup table; and a GPU node connected to both the FPGA and CPU nodes via the same PCIe bus for performing long-range electrostatic force calculations on all atoms in the system, and updating atomic positions and velocities after reducing short-range and long-range forces. The heterogeneous node electrostatic force calculation device for molecular dynamics simulations provided by this invention, through reasonable task allocation, fully utilizes the advantages of GPUs and FPGAs to achieve asynchronous parallel computing between GPUs and FPGAs, thereby improving the overall speed of electrostatic force calculations.