155 results about "Molecular descriptor" patented technology

The invention relates to the technical field of drug research and development, and discloses a drug-like molecule screening system and method based on deep learning, and the system comprises a physical and chemical characteristic analysis module which is used for calculating molecular descriptors of compounds and generating a visual characteristic distribution chart; and the drug-like molecule screening module integrates a Lipinski rule, a Veber rule, a Ghose rule and a Muegge rule, and supports a user to customize a screening rule. Compound characteristics are comprehensively analyzed through the physical and chemical characteristic analysis module, flexible screening is achieved through the drug-like molecule screening module, toxicity risks are accurately recognized through the toxicity assessment module, the affinity assessment module is combined to assess the binding capacity in multiple ways, and the synthesis feasibility assessment module plans an efficient and low-cost synthesis path. The dynamic feedback optimization module continuously improves the system performance; all the modules work cooperatively, the problems of scattered tools, insufficient precision, low planning efficiency, optimization deficiency and the like in the prior art are effectively solved, the medicine research and development efficiency is greatly improved, and the research and development risk and cost are reduced.

The invention discloses a molecular property prediction method based on a molecular line graph structure. The method comprises the following steps: firstly, collecting SMILES molecular formulas of drug molecules, and carrying out pretreatment to extract molecular fingerprints and molecular descriptors; then, the molecular line graph structure serves as model input, feature representation of a molecular line graph is calculated through a graph neural network MolGraphTrans based on Transformer, and molecular properties are predicted. According to the method, substructure masks, a multi-head attention mechanism, path coding and distance coding are introduced, so that molecular structure characteristics can be better captured. And finally, the prediction performance is optimized through a low-rank linear layer and a multi-head attention layer. Compared with the prior art, the accuracy and efficiency of molecular property prediction are improved, and the method is suitable for drug screening and optimization.

The invention discloses a method for predicting the thermal expansion coefficient of polyimide based on an interpretable machine learning model, which comprises the following steps: collecting the structure and thermal expansion coefficient information of polyimide from a public database and related literatures, then blocking the two ends of a repetitive unit by using hydrogen atoms, generating corresponding molecular descriptors through an SMILES expression of the repetitive unit, and predicting the thermal expansion coefficient of polyimide. A plurality of most important molecular descriptors are selected; constructing a thermal expansion coefficient prediction model, optimizing the model, evaluating the performance of the model by adopting cross validation, and measuring the model by using a decision coefficient, a mean absolute error and a root-mean-square error in the process; the SHAP is used for explaining the model, factors influencing related properties can be explained, experimental synthesis can be rapidly guided, and the method has the advantages of being high in calculation speed and precision and convenient in structural analysis; the method plays a guiding role in synthesis of novel low-thermal-expansion-coefficient polyimide.

The invention discloses a microemulsion interfacial tension efficient prediction method and system based on active learning and molecular dynamics. According to the method, 217 molecular descriptors corresponding to each molecular structure are calculated by adopting an RDKit software package, and the descriptors are used for representing molecular structure characteristics and serve as input variables of a machine learning model, so that key structure information including molecular branching degree, polarity and the like is transmitted. For an oil-water-surfactant ternary interface system, the oil-water interfacial tension in the presence of a surfactant is simulated and calculated through molecular dynamics, and an IFT value is set as a model prediction target. An active learning mechanism is introduced, and iterative sample labeling in the molecular dynamics simulation process is guided; and integrating the obtained IFT data with the molecular descriptor features, constructing a machine learning data set, and training a random forest model. According to the method, the problem of screening a high-performance surfactant layer by a middle-phase microemulsion system can be solved, and the ultra-low oil-water interfacial tension can be rapidly and efficiently screened.

The invention discloses a rare and endangered organism toxicity prediction method based on a machine learning algorithm and a quantitative structure-activity relationship, which constructs a toxicity prediction model through the machine learning algorithm, can effectively assess the toxicity influence of environmental pollutants on rare and endangered organisms, and provides technical support for rare and endangered organism protection and ecological environment risk assessment. Comprising the following steps: 1, collecting data related to rare and endangered organisms from a public database, and establishing a rare and endangered organism toxicity prediction database based on machine learning; 2, generating molecular descriptors for the chemical substances; step 3, data preprocessing; 4, development of acute and chronic toxicity prediction models of rare and endangered organisms based on multiple machine learning is carried out, and performance evaluation is carried out; 5, performing internal and external verification on the rare and endangered biotoxicity prediction model; step 6, analyzing the importance of the features by using the valuable, rare and endangered biotoxicity prediction model, and finding out the most influential features; and 7, predicting the toxicity value of the pollutants in combination with the optimal machine learning model.

The present disclosure relates to a structure-activity relationship analysis method and electronic device based on multidimensional interpretability, the method comprising: obtaining a series of ligand molecules; calculating target feature data of the series of ligand molecules, the target feature data including molecular descriptors, parent core structures, substituent groups and protein-ligand interactions; obtaining the real molecular activity data of each ligand molecule obtained based on experiments; using the target feature data of the ligand molecule as sample data, using the real molecular activity data as label data of the corresponding ligand molecule, and training the target machine learning model based on the sample data and label data; analyzing the target machine learning model based on the Shapley addition and interpretation algorithm to obtain the Shapley value of each target feature data; determining the influence of each target feature data on the molecular activity based on the Shapley value. The method aims to promote the intelligence and precision of structure-activity relationship analysis based on the interpretability of artificial intelligence models based on multidimensional feature representation and the Shapley addition and interpretation algorithm.

The invention relates to a compound crystal density prediction model construction method and a prediction method. A multi-molecular structure third-order volume characteristic variable description method and numerical values of first-order characteristics, second-order characteristics and third-order characteristics of atomic elements are utilized to form descriptor vectors; according to the method, the crystal density of the novel single-molecule energetic compound can be rapidly predicted on the basis of the molecular SMILES code, verification shows that the method has the advantages of being simple, efficient and accurate in prediction of the crystal density, the method is not limited by the molecular variety of the energetic compound, the universality is good, the average absolute prediction error of the crystal density of various energetic molecules is 0.026 g / cm < 3 >, and the gold standard of 0.030 g / cm < 3 > is broken through. The method is of great significance to design, construction and research of energetic molecule descriptors, and can guide design and synthesis of more energetic compounds with high energy density and high detonation performance.

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 discloses a meta-learning-based small sample pharmacokinetic property prediction method and related equipment, and relates to the field of pharmacokinetics. The method comprises the following steps: acquiring multi-dimensional feature data of candidate compounds, wherein the multi-dimensional feature data comprises atomic-scale features, molecular descriptors and molecular structure features; inputting the multi-dimensional feature data into a preset pharmacokinetic property prediction model to obtain a preliminary prediction result, the pharmacokinetic property prediction model being pre-trained through meta-learning based on small samples, the model comprising a first machine learning model and a second machine learning model, the preliminary prediction result comprises a first prediction result output by the first machine learning model and a second prediction result output by the second machine learning model; and determining a pharmacokinetic property prediction result of the candidate compound according to the first prediction result and the second prediction result. According to the method, the problem of low accuracy of a pharmacokinetic property prediction result can be relieved in a small sample scene of early drug research and development.

The present invention provides a technology for predicting the level of PPO inhibition objectively and with high accuracy using LUMO distribution as a "molecular descriptor," and a technology for selecting a PPO inhibitor based on the predicted level of PPO inhibition. [Solution] The present invention provides a method for predicting PPO inhibitory activity based on the correlation between LUMO distribution and PPO inhibitory activity for each compound, predicting the level of PPO inhibitory activity based on the correlation between the difference in variation in PPO inhibitory activity between different plant species and the level of PPO inhibitory activity, and selecting PPO inhibitors using the predicted PPO inhibitory activity as an index.

The invention provides a method and device for predicting the glass transition temperature of a polymer and electronic equipment, and the method comprises the steps: obtaining a plurality of molecular descriptors of the polymer, obtaining an original database of the characteristic parameters of the polymer, and enabling the molecular descriptors to be used for representing the glass transition temperature distribution of the polymer, the characteristic parameter and the distribution of the polymer glass transition temperature have a preset relationship; according to the feature engineering, removing a plurality of feature parameters in the original database to obtain an optimized database; and constructing a deep learning model at least according to the characteristic parameters in the optimization database, determining a mapping relationship between the characteristic parameters of the polymer and the glass transition temperature according to the deep learning model, the characteristic parameters and an SHAP analysis method, and predicting the glass transition temperature of the polymer according to the mapping relationship. The problem of inaccurate polymer prediction results in the prior art is solved.

The invention discloses a pre-training model and multi-source coding feature fusion-based water pollutant molecular property prediction method and system, and relates to the technical field of environmental science and artificial intelligence crossing. Comprising the steps of collecting pollutant data; the collected pollutant data is preprocessed; taking SMILES as a molecular input basis, and extracting molecular features; according to the molecular features, feature vectors are obtained from an encoder and then input to a vector interaction module for feature fusion; and embedding the comprehensive molecules obtained after feature fusion into an expert mixed structure module to obtain comprehensive representation of an expert mechanism, inputting the comprehensive representation of the expert mechanism into a preset prediction head, and completing prediction of molecular properties through the prediction head. According to the method, the molecular language model features, the molecular graph neural network features and the molecular descriptor features are integrated, and feature fusion is carried out by utilizing an attention mechanism and residual connection, so that high-precision prediction of the molecular properties of pollutants is realized.

The invention discloses a method for predicting molecular solvation free energy based on machine learning, relates to the technical field of computational chemistry, and solves the technical problem that a molecular dynamics method consumes a large amount of computing resources and time. The method comprises the following steps: using SMILES formulas of solute molecules and solvent molecules as input data; converting the SMILES formula into a one-dimensional descriptor, a two-dimensional descriptor and a three-dimensional descriptor of a molecule based on an RDKit tool; based on an open source experiment data set, training a machine learning model through an association relationship between a one-dimensional descriptor, a two-dimensional descriptor and a three-dimensional descriptor of a molecule and experiment solvation free energy to obtain a prediction model; and inputting molecular descriptors of solute molecules and solvent molecules into the prediction model to obtain a result of predicting solvation free energy of the solute molecules. Existing molecular dynamics simulation is replaced by the machine learning model, and meanwhile, model training is performed by adopting a machine learning algorithm and experimental data, so that the accuracy and the stability of a prediction result are improved.

The invention discloses a reverse molecular design method and system of an organic framework material COFs based on a machine learning and quantitative calculation technology, and relates to the technical field of material science and computer simulation. The photoelectric property of the COFs material can be accurately predicted, and a new material can be directionally synthesized under guidance. A COFs structure database is established, Gaussian is used for representing the fragment structure and electronic characteristics of the COFs, a molecular descriptor and a machine learning algorithm are used for establishing a structure-performance mapping relation, quantitative calculation software is used for obtaining an important target quantity for describing the photoelectric characteristics of the COFs, SHAP is used for carrying out importance analysis, and finally a machine learning model for predicting the photoelectric properties of the COFs is obtained. According to the model, through theoretical prediction of a novel COFs structure, a mapping relation between the COFs structure and photoelectric specificity can be established, and a novel COFs hypothetical structure can be reset according to important molecular fragments or word structures of the important molecular fragments; theoretical guidance can be provided for application in the fields of photoelectric catalysts and the like.

The invention relates to the technical field of new pollutant risk identification, in particular to a new pollutant multi-medium PNEC predictive analysis system and method based on machine learning, and the method comprises the steps: collecting a compound SMILES structure and fresh water PNEC, BCF and Koc data thereof; using tools such as RDKit, rcdk and the like to automatically calculate molecular descriptors; key features are screened through redundant feature processing and a random forest algorithm; an RF model, an XGBoost model, a LightGBM model and a CatBoost model are respectively constructed for PNEC prediction, and a BCF model and a Koc model are constructed by adopting the LightGBM; and a visual platform is built based on an R language Shiny framework, so that full-process automation of SMILES input, model prediction and result display is realized. The system provides a convenient tool for multi-medium PNEC prediction, has the advantages of high efficiency, accuracy, low cost and the like, is suitable for large-scale new pollutant environmental risk identification, and has wide application prospects and important environmental protection significance.

The invention discloses a method for predicting liquid chromatography retention time of traditional Chinese medicine active ingredients. The method comprises the following steps: detecting retention time of flavonoid compounds or anthraquinone compounds under different chromatographic conditions through a reverse high performance liquid chromatograph; performing structure optimization on the flavonoid compound or the anthraquinone compound; performing molecular descriptor calculation on the optimized compound structure to obtain a molecular descriptor data set; performing genetic algorithm screening on the molecular descriptor data set to obtain characteristic molecular descriptors; combining the characteristic molecule descriptors with the different chromatographic conditions into a complete data set; the complete data set serves as an input variable, retention time serves as an output variable, and a gradient elevator GB algorithm or a random forest method RF is adopted for modeling to obtain a QSRR model; and predicting the retention time of flavonoid or anthraquinone compounds under different chromatographic conditions by using the constructed QSRR model.

The invention provides a method for predicting the flammability upper limit volume percentage of a pure compound. The method can be used for predicting a mathematical model of the flammability upper limit volume percentage of the pure compound which is composed of 12 or less elements of hydrogen, carbon, nitrogen, oxygen, sulfur, fluorine, chlorine, bromine, iodine, silicon, phosphorus, arsenic and the like and has the atom number of 25 or less (excluding hydrogen) with high accuracy. Wherein the model is obtained as a universal quantitative structure-property relationship model by determining an optimal model from a plurality of multiple linear regression models by means of a step-by-step selection method, and the model takes some of the various molecular descriptors as independent variables, takes flammability upper limit volume percent as a dependent variable, and takes the flammability upper limit volume percent as a dependent variable. The value of the molecular descriptor included in the model can be received and input in a short time, and the flammability upper limit volume percentage can be output, so that only the specific value of the molecular descriptor included in the model is known. And the flammability upper limit volume percentage of the compound purely formed by the molecule can be predicted for any molecule meeting the requirements of the invention.

The invention discloses a molecular antibacterial activity prediction method, system and device and a storage medium, and the method comprises the steps: obtaining molecular antibacterial activity data of multiple sources, the molecular antibacterial activity data comprising molecular chemical formulas and corresponding antibacterial activity labels; converting the molecular chemical formula into a molecular image and a molecular descriptor, and constructing a training data set according to the molecular image, the molecular descriptor and the antibacterial activity tag; inputting the training data set into a pre-constructed multi-modal neural network for training to obtain a trained molecular antibacterial activity prediction model; and obtaining a target image and a target descriptor of the target molecule, inputting the target image and the target descriptor into the molecular antibacterial activity prediction model to obtain a molecular antibacterial activity prediction value of the target molecule, and further generating a visual heat map representing prediction contribution according to the molecular antibacterial activity prediction value. The accuracy and reliability of molecular antibacterial activity prediction are improved, and the method can be applied to the technical field of artificial intelligence.

The techniques disclosed herein enable an autonomous agent to interpret an input dataset and orchestrate a suite of software modules to perform a computational task on a representation of a chemical material. The input dataset includes a prompt defining a computational task to be performed on a chemical material. Moreover, the input dataset includes data defining a chemical included in the chemical material, molecular descriptors describing the chemical and / or the chemical material, and an external variable. The agent analyzes the benefits and drawbacks of each model within the context of the computational task to determine a technique for performing the computational task. Accordingly, the agent formulates a chain of calls invoking the functionality of data processing tools and models to perform the computational task responsive to the prompt.

Provided is a mathematical model that can predict, with high accuracy, the saturated liquid density of a pure compound at 298.15 K, said pure compound comprising 12 or less elements such as hydrogen, carbon, nitrogen, oxygen, sulfur, fluorine, chlorine, bromine, iodine, silicon, phosphorus, and arsenic and having 25 or less atoms (excluding hydrogen). The model is a quantitative structure-property relation model and is obtained by solving an optimal model from a plurality of multiple linear regression models by a step-by-step selection method, and the model takes some molecule descriptors in various molecule descriptors as independent variables, takes saturated liquid density under 298.15 K as a dependent variable, and takes the saturated liquid density under 298.15 K as a dependent variable. The value of the molecule descriptor included in the input model can be received in a short time, and the saturated liquid density at 298.15 K can be output, so that the saturated liquid density at 298.15 K of a compound formed by the molecule singly can be predicted for any molecule meeting the requirements of the invention as long as the specific value of the molecule descriptor included in the model is known.

The invention relates to the technical field of drug research and development, and discloses a drug intermediate database construction and AI intelligent retrieval method. The method comprises the following steps: acquiring a chemical structure and synthesis path data of a target drug intermediate, extracting a molecular descriptor by utilizing a preset feature extraction model, and constructing a structured intermediate database according to the molecular descriptor; and inputting the database into an AI retrieval model based on a graph neural network algorithm to generate a dynamic retrieval result, performing multi-dimensional sorting by using a preset optimization algorithm, and outputting a standardized retrieval list. The method further comprises the steps of feature extraction model training, data preprocessing, retrieval result verification, data security control and the like. According to the method, multi-source data can be efficiently integrated, key information can be accurately extracted, intelligent retrieval is achieved, data safety is guaranteed, the medicine research and development efficiency can be remarkably improved, the research and development cost is reduced, and powerful support is provided for medicine research and development.

The invention relates to the technical field of electrical data processing, and provides an anthocyanin stabilizer screening method and system based on fusion of quantum chemistry calculation and machine learning. The method comprises the following steps: constructing a molecular structure data set; performing geometric structure optimization on the anthocyanin molecules and the candidate polyphenol molecules to obtain a ground state optimization structure; performing multi-dimensional molecule descriptor calculation on the candidate polyphenol molecules to obtain a polyphenol molecule descriptor matrix; performing structure optimization, single-point energy calculation and free energy calculation on the anthocyanin polyphenol compound system to obtain a combined free energy value set; integrating the polyphenol molecule descriptor matrix and the combination free energy value set, constructing a training data set, and performing feature screening; performing model training through a machine learning algorithm to obtain a combined free energy prediction model; and predicting the binding free energy of the anthocyanin polyphenol compound through the binding free energy prediction model to obtain candidate molecules of the anthocyanin stabilizer. The screening efficiency of the anthocyanin stabilizer is improved.

The present invention discloses a drug molecular property prediction and classification method and system based on BiLSTM; wherein the method includes: obtaining a drug molecule to be predicted, and converting the molecular structure of the drug analysis to obtain a drug molecular descriptor and a drug molecular fingerprint; inputting both the drug molecular descriptor and the drug molecular fingerprint into a trained drug molecular property prediction and classification network, and outputting a drug molecular property prediction result; the working principle of the trained drug molecular property prediction and classification network includes: respectively extracting features from the drug molecular descriptor and the drug molecular fingerprint to obtain two drug molecular feature vectors; performing feature fusion on the two drug molecular feature vectors; performing classification prediction on the fused feature vectors, and outputting a drug molecular property prediction result.

To provide an insilico toxicity risk evaluation method in which high prediction accuracy is compatible with the explicitness of a prediction basis.SOLUTION: A machine learning model for predicting the intensity of the toxicity risk of a chemical substance by an information processing system is constructed using a molecular descriptor based on a chemical structure and inchemico or invitro test data, the prediction result of the toxicity risk intensity output by the information processing system by the machine learning model is used as an index of similarity specific to toxicity, and the toxicity risk is evaluated by a Reed-Acros method based on the similarity by the index.SELECTED DRAWING: Figure 1

A formulation graph convolution network (F-GCN) with multiple GCNs assembled in parallel and connected to filters and an external learning architecture is able to predict the effectiveness of a formulation. Input into the multiple GCNs are molecular structures of formulants, which are processed as molecular graphs and output as molecular descriptors. The molecular descriptors are filtered by normalized ratios or fractions of the ingredient molecules in a formulation, such as a battery electrolyte or solvent. A formulation descriptor combines the filtered molecular descriptors to arrive at a predicted performance for the formulation, such as the battery capacity for an electrolyte formulation, by an external learning architecture. F-GCN may use a pre-trained GCN with physico-chemical properties of known molecular structures.

The invention discloses a multi-label smell description prediction method, and relates to the field of compound smell prediction, and the method comprises the steps: obtaining compound identification information, molecular structure descriptors and smell label data, and constructing a multi-label smell data set; generating a molecular structure feature vector through a molecular fingerprint coding technology, and extracting a multi-dimensional descriptor reflecting the physicochemical properties of molecules; compressing the molecular fingerprint features to a low-dimensional space through a dimension reduction algorithm; performing unbalanced data processing on the training set, fusing the dimension-reduced molecular fingerprints with the molecular descriptors to form a joint feature matrix, and configuring a class weight balance mechanism and overfitting suppression parameters by adopting a multi-label classification architecture; independently optimizing a probability threshold for each odor label based on the verification set; and outputting a multi-odor label combination prediction result according to the target molecule identification information. According to the scheme, the multi-odor characteristics of the compound can be accurately depicted, and the combined recognition accuracy of the compound odor is remarkably improved.

The invention discloses a dynamic prediction chloropropanol ester omics analysis method. The method comprises the following steps: S1, constructing a molecular descriptor matrix of glyceride of a chloropropanol ester seed structure and a three-dimensional information base of chloropropanol ester; s2, establishing a computer simulation algorithm of a synthetic reaction of chloropropanol ester in the sample; s3, taking glyceride as a seed structure, and establishing a chloropropanol mass spectrum recognition database; and S4, identifying chloropropanol ester omics in the sample by using a threshold score. According to the method, a liquid chromatography-mass spectrometer is used for constructing a molecular descriptor matrix of different types of chloropropanol esters, and the range of the chloropropanol esters is greatly expanded through a computer simulation algorithm of a synthetic reaction of the chloropropanol esters in a sample. According to the method, basic information units are expanded on a computer in a mechanism-driven manner to realize the identification of chloropropanol esters, mass spectrum information is deeply annotated to re-characterize functional groups, the mass spectrum information is fully utilized, the identification accuracy of chloropropanol esters is ensured, and finally, the identification of different types of chloropropanol esters in a sample is realized.

The present invention discloses a toxicity prediction model for predicting agranulocytosis induced by non-chemotherapeutic drugs. The toxicity prediction model uses the best molecular descriptor combination of the drug to be tested as input information, and inputs it into the integrated model of the trained BalancedBaggingClassifier combined with the gradient boosting decision tree of the base classifier, thereby obtaining the toxicity prediction result of the drug to be tested. The toxicity prediction model can accurately predict the toxicity of agranulocytosis induced by non-chemotherapeutic drugs, reduce the risk of adverse reactions for agranulocytosis in new drug research and development, reduce research and development costs, and accelerate the research and development process; and the toxicity prediction model can evaluate the risk of agranulocytosis caused by existing non-chemotherapeutic drugs, newly listed unknown adverse reaction drugs or preclinical research / clinical research compounds, provide medication warnings for clinicians, help improve patient medication safety, and improve medical quality and patient quality of life.

The invention discloses a new pollutant identification method, system and device, and a storage medium. The new pollutant identification method comprises the following steps: acquiring a multi-modal feature of a to-be-identified pollutant; wherein the multi-modal feature comprises at least one of a first molecular descriptor feature, a first molecular structure image feature and a first text feature; inputting the multi-modal features into a trained modal detection and completion model for modal completion to obtain second molecular descriptor features, second molecular structure image features and second text features; fusing the second molecular descriptor feature, the second molecular structure image feature and the second text feature to obtain a fused feature; and inputting the fused features into the trained new pollutant identification model for new pollutant identification to obtain a classification result of the to-be-identified pollutants, and improving the accuracy of new pollutant type identification by fusing the structured data, the image data and the semantic text data.

The present invention discloses a method for predicting the ADMET properties of ERα antagonists based on the MMS_ResNet_1d model, wherein n ADMET properties of ERα antagonist compounds are used as dependent variables, and m molecular descriptor information of corresponding compounds are used as independent variables to create raw data, and a Micro_Multi_Scale_resnet_1d deep learning model is built and iteratively trained to obtain optimal model parameters for predicting the properties of ADMET properties. The present invention can more quickly and effectively perform statistical modeling and prediction on drug properties, improve the accuracy of prediction, thereby reducing the number of pharmacologically inappropriate compounds in the early stages of drug development, saving time and cost.