38 results about "Molecular Fingerprinting" patented technology

The invention belongs to the field of wastewater treatment and industrial automatic control, and particularly relates to an intelligent dosing control method based on water quality online monitoring. The method comprises the following steps: capturing a water body spectrum fingerprint spectrum by using a full-spectrum scanning module to obtain a characteristic signal sequence; a chemical informatics molecular algorithm is introduced for preprocessing and feature enhancement, and molecular fingerprint information is recognized; a convolutional neural network model is constructed to decouple chemical oxygen demand, ammonia nitrogen, total phosphorus and water toxicity indexes from the spectral signals, and multi-parameter parallel inversion is achieved; and based on the inversion index and the spectral dynamic trend, the dosage is calculated through an intelligent decision-making algorithm, and accurate dosing of the medicament is driven. According to the invention, through fusion of full spectrum scanning and deep learning, high-precision anti-interference water quality perception is realized, potential risks can be identified and preventive intervention can be carried out, and the intelligent level and economical efficiency of dosing control are improved.

The application discloses a high-sweetness-amino-acid-content Flammulina velutipes variety and a MNP molecular identification method and application thereof. The Flammulina velutipes variety has excellent properties, the cap is not easy to open, the content of sweet amino acids is high, the diversified market demand is met, the factory annual bottle cultivation is suitable, and the Flammulina velutipes variety has a good application and popularization prospect. The MNP molecular fingerprint of the Flammulina velutipes variety 'Shangyan A111' is comprehensive multiple amplification and sequencing technology, sequence analysis of all marker sites of multiple samples can be performed at one time, compared with ISSR, RAPD, SSR and other molecular markers and mushrooming test, the MNP molecular fingerprint has the advantages of high throughput, multi-target, high sensitivity and high precision. The MNP molecular fingerprint of the Flammulina velutipes variety 'Shangyan A111' has specificity and specificity in identifying the Flammulina velutipes strain 'Shangyan A111'.

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 application relates to an intelligent screening system of a mitochondrion effect molecule, a construction method and application thereof, and belongs to the technical field of molecular biology. The construction method of the intelligent screening system of the mitochondrion effect molecule comprises the following steps: 1, establishing a target protein library; 2, obtaining a data set of the mitochondrion effect molecule; 3, adopting Morgan molecular fingerprint to characterize the mitochondrion effect molecule in the data set, carrying out deduplication and decontamination processing, and then carrying out molecular similarity processing to obtain an input set of a model; and 4, taking accuracy and AUC value as evaluation indexes to construct a support vector machine model. The application utilizes the support vector machine model to carry out prediction in a large amount of molecular data set, and gives a molecule with a high probability score which may have an effective effect on mitochondria, the model is helpful for researchers in the field of mitochondria to reduce parameter adjustment time and improve work efficiency.

The application relates to the technical field of electric digital data processing, and discloses a molecular design constraint condition generation method, system, device and medium based on an interpretable artificial intelligence, which comprises the following steps: acquiring molecular structure data marked with active / inactive labels, and extracting molecular property features and / or molecular fingerprint structure features; training a molecular activity prediction model by using a machine learning algorithm; applying the model to a to-be-tested molecule, analyzing a prediction result by using interpretability analysis, and generating a molecular design constraint condition. The application can convert the prediction result of the artificial intelligence model into specific and executable molecular design guidance, and breaks through the limitation that a traditional model can only output "whether active" but cannot explain "why active" and "how to design". The application can be flexibly adapted to various molecular feature input modes, can provide the most comprehensive design prompt, significantly reduces the dependence on expert experience, and improves the efficiency and success rate of molecular design.

This invention discloses a novel molecular fingerprinting algorithm for assisting in the design of MOFs for acid gas separation, comprising the following steps: S1, using drawing software to draw 17 possible ortho, meta, and para positions of benzene rings, five-membered rings, and six-membered rings, obtaining their .mol ​​files; S2, converting the .mol ​​files to .smart format and writing a separate .py file for the new fingerprint; S3, calling the original fingerprint in the newly written .py file, adding the two results to obtain a new 184-bit fingerprint; S4, using the new fingerprint to test multiple CoRE-MOFs using various machine learning methods. This invention can determine whether the N-bond in a five-membered or six-membered ring of a macromolecule is ortho, meta, or para, which is beneficial for identifying differences in molecular structure between molecules with different performance characteristics, thereby accelerating the screening of high-performance materials, effectively saving time and costs, and shortening the development cycle.

This application discloses a method for predicting the quantitative activity of endocrine disruptors, relating to the field of virtual screening of endocrine disruptors. The method includes: acquiring in vitro experimental data of nuclear receptors and removing duplicate data, as well as removing compound sets that do not contain the simplified molecular linear input canonical (SMILES) representation; using a molecular fingerprinting method to extract the primary, secondary, and tertiary structural features of the compounds. For each compound cluster, a quantitative prediction model based on machine learning or quantitative read-across is constructed to predict the quantitative activity value of the compound. Addressing the low efficiency of existing methods for predicting the quantitative activity of endocrine disruptors, this application extracts multi-level structural features of the compounds and constructs corresponding quantitative prediction models for compound structural clusters of different sizes. Through molecular docking and molecular dynamics simulations, the interaction mechanism between endocrine disruptors and nuclear receptors is studied from a structural biology perspective, thus improving efficiency.

The present disclosure provides compositions comprising sequence-specific endoribonucleases and methods for their use in RNA analysis, RNA synthesis, and fingerprinting of RNA molecules. In particular, the present disclosure relates to compositions and samples comprising ToxN endoribonucleases that recognize and cleave single-stranded RNA, as well as optimal conditions for obtaining cleavage.

This invention provides SSR molecular markers for Hibiscus rosa-sinus and their applications, belonging to the field of molecular marker technology. The invention comprises 15 molecular markers, with corresponding primer sequences shown in SEQ ID NO. 1~30. Testing revealed that this series of markers exhibits a 100% polymorphism rate, excellent polymorphism information content, marker index, and resolution, and strong genetic stability. The molecular marker primers of this invention can efficiently distinguish Hibiscus rosa-sinus germplasm materials, clearly revealing the population's genetic structure and phylogenetic relationships. They are suitable for Hibiscus rosa-sinus germplasm resource identification, genetic diversity analysis, molecular fingerprinting construction, and assisted breeding, possessing the characteristics of high specificity, high resolution, and wide applicability.

The invention relates to the cross technical field of artificial intelligence and bioinformatics, in particular to a multi-modal enzyme kinetic parameter prediction method of an adaptive protein language model. Aiming at the problems that an existing processing method is lack of enzymatic reaction dynamic mechanism modeling and insufficient in three-dimensional structure utilization, the invention provides the following technical scheme: step 1, acquiring and preprocessing multi-modal data; 2, protein language model and molecular fingerprint feature extraction; 3, carrying out substrate recognition feature fusion based on cross attention; step 4, extracting conformation adaptive features based on the hybrid expert network; 5, correcting enzyme-substrate distribution alignment characteristics; step 6, kinetic parameter regression prediction; step 7, constructing a multi-objective loss function; and step 8, model training and parameter optimization. Through an enzyme reaction bridging adapter and an enzyme-substrate distribution alignment technology, high-precision prediction of enzyme kinetic parameters is realized, and cross-data-set accuracy and robustness are remarkably improved.

The application provides a lake dissolved organic matter quantitative tracing method based on molecular fingerprint and application thereof, and relates to the technical field of environmental analysis and water ecological management. The method comprises the following steps: collecting end member samples such as lake water and sediments, surface soil, plant litter, chemical fertilizer, livestock and poultry manure, tail water of a sewage treatment plant, tail water of aquaculture, etc., and enriching DOM through solid phase extraction; then obtaining molecular formula data through mass spectrometry; screening stubborn inert molecules that satisfy H / C < 1.5, conversion times of 0 and exist only in a single end member, and verifying the ecological conservation thereof through -2 < betaNTI < 2 and -0.95 < RCBray < 0.95; adopting SHAP algorithm to optimize a high-contribution molecular feature subset, and inputting the high-contribution molecular feature subset into a MixSIAR model to output the contribution proportion of each end member and the 95% confidence interval. The method guarantees the reliability of the tracer from the mechanism, breaks through the bottleneck of spectral overlap, and realizes accurate and reliable quantitative disassembly of endogenous and exogenous DOM.

This invention belongs to the field of bioinformatics and relates to a molecular property prediction method based on multimodal gating and contrastive learning, including techniques such as contrastive learning, graph neural networks, cross-modal alignment, and gated attention. First, data standardization and graph construction are performed, and molecular fingerprint embeddings are extracted. Second, a heterogeneous dual-channel graph encoding architecture is adopted, with one channel capturing short-range atomic interactions through an attention mechanism, and the other integrating the global molecular structure and long-range dependencies to generate complementary molecular representations. Subsequently, a cross-modal attention mechanism is introduced to achieve bidirectional association between graph and fingerprint features, and modal weights are adaptively and dynamically allocated via a gated fusion module. Finally, a contrastive pre-training strategy is employed to construct sample pairs using the molecular graph and fingerprint, learning discriminative molecular representations on unlabeled data. This method significantly improves the accuracy of molecular property prediction, providing an efficient and reliable computational tool for virtual drug screening and lead compound optimization.

The invention relates to the field of artificial intelligence, in particular to a chemical reaction yield optimization method, system and device and a medium, and the method comprises the steps: obtaining a training sample and a yield label of a target chemical reaction; generating a molecular fingerprint of each training sample as an initial representation, and determining substructures contained in the training samples; the representation mapping model to be trained is trained with the yield label as supervision, the feature similarity after mapping is made to be consistent with the yield similarity, and a representation mapping model is obtained; mapping the initial representation through a representation mapping model to obtain a reaction implicit representation; inputting the reaction implicit representation into a yield prediction model to obtain a predicted yield; calculating the contribution degree of each substructure according to the influence of the corresponding codes of the covering substructures on the prediction yield; and optimizing the target chemical reaction according to the contribution degree. According to the present invention, the specific substructure capable of promoting or inhibiting the yield can be accurately positioned, and the favorable substructure is specifically introduced or the adverse substructure is avoided so as to improve the target chemical reaction yield.

The present invention relates to a multimodal non-invasive in vivo system and related method for the characterization of the tissue microenvironment. The system comprises: a multispectral imaging (MSI) camera for superficial spectral screening; a multispectral optoacoustic tomography (MSOT) module for dynamic vascular and perfusion analysis; a Mueller Matrix Polarimetry (MMP) module for extracellular matrix anisotropy assessment; a Raman spectroscopy module for molecular fingerprinting of fibroblast-associated proteins; an optical coherence tomography (OCT) module for morphological and stratigraphic evaluation; an optional Elastic Scattering Spectroscopy (ESS) module for subcellular scattering biomarkers; and a diachronic analysis module for longitudinal monitoring. Data are integrated by an artificial intelligence processing unit to generate quantitative biomarkers of vascularization, extracellular matrix features, immune activity (Langerhans cells) and fibroblast phenotype, distinguishing physiological myofibroblasts from CAF. The invention enables non-nvasive, biopsy-free evaluation of scars, melanocytic lesions, tumor stroma, wound healing and surgical site monitoring.

The invention belongs to the field of vegetable molecular technology breeding, and particularly relates to a method for analyzing germplasm genetic diversity and constructing a fingerprint spectrum of balsam pear based on a simple sequence repeat (SSR) marker. By screening high-polymorphism SSR core markers, genetic diversity analysis is carried out on 26 germplasm parts of the balsam pear, and an SSR molecular fingerprint spectrum is constructed. Core SSR markers (MC0569594, MC0450530, MC0687314 and MC10146038) are used for carrying out PCR (Polymerase Chain Reaction) amplification and banding pattern analysis, the polymorphism information content (PIC) is calculated, and the germplasm of the balsam pear is clustered into three main groups through clustering analysis. By combining a fingerprint database, efficient identification and information storage of germplasm of the balsam pear are realized. The method is easy and convenient to operate and high in accuracy, and important technical support is provided for protection, utilization and breeding of germplasm resources of the balsam pear. By means of the method, the variety of the large-top bitter gourds on the market can be rapidly identified, materials with excellent characters can be effectively screened out in the breeding process, the breeding efficiency is improved, and the method has wide application prospects.

This invention discloses a dual-objective solvent screening paradigm based on machine learning, belonging to the fields of chemical engineering and machine learning technology. The paradigm includes the following steps: constructing a dual-objective solvent screening paradigm with solubility and flammability as screening objectives; collecting and organizing solubility data and flammability risk data; constructing differentiated feature engineering for solubility prediction and flammability prediction respectively; constructing a multi-dimensional hybrid feature set integrating molecular fingerprints, physicochemical descriptors, and temperature for solubility prediction, and using multi-dimensional molecular descriptors for flammability prediction; optimizing the model using different hyperparameter optimization methods for different prediction tasks; and employing a Pareto ranking mechanism to output a Pareto-optimal solvent set with the objectives of maximizing solubility and minimizing risk, thus achieving dual-objective collaborative screening. This method achieves accurate prediction of solubility and flammability, possesses strong adaptability and generalization ability, and provides an effective tool for high-throughput solvent screening in crystallization process design.

The invention provides a metabolic stability prediction method based on a pharmacophore-oriented multi-mode representation framework and contrast hypergraph learning, and belongs to the technical field of biological information, and the method comprises a pharmacophore extraction and representation module which is used for extracting hydrogen bond donor and acceptor functional groups and converting the functional groups into numerical features; the multi-mode molecular encoder module is composed of a molecular fingerprint encoder, a molecular map encoder, a pharmacophore encoder and a text encoder; the hypergraph comparative learning module is used for constructing and executing comparative learning based on the molecular graph and the pharmacophore hypergraph; and the attention fusion and prediction module is used for integrating multi-modal features through a multi-head attention mechanism and outputting a prediction result and pharmacophore importance analysis. According to the method, by integrating multi-modal molecular representation and hypergraph contrast learning, the accuracy and generalization ability of compound metabolism stability prediction are effectively improved, and specific structure guidance can be provided for compound optimization in drug research and development.

The invention discloses a terahertz biomolecular fingerprint detection chip based on strong coupling resonance, and relates to the technical field of terahertz molecular fingerprint sensing and metasurface. The terahertz sensing metasurface sensing chip based on the quasi-continuous domain bound state mode and the local electric field and analyte space matching mechanism comprises a substrate layer and a metal layer which are integrally compounded from bottom to top, and the metal layer is composed of a hollow oval structure array arranged in a square lattice. The two may be integrated by a bonding operation. According to the terahertz sensing metasurface sensing chip, broadband coverage of a harmonic peak of the terahertz sensing metasurface sensing chip can be realized by adopting a structural parameter change strategy, experimental observation of strong vibration coupling and terahertz fingerprint identification can be realized by filling an analyte into a hot spot region, and high-sensitivity qualitative and quantitative detection of small molecular substances such as lactose and the like can be realized; the method is simple, convenient, effective, high in cost performance and wide in application prospect.

The invention provides a drug-drug interaction prediction method PG-DDI based on pre-trained drug characteristics. According to the method, DDI prediction is converted into a multi-source feature fusion classification task: drug molecular map structural features are extracted by using GAT and GIN of freezing parameters, molecular fingerprint features of three fingerprints of MACCS and the like subjected to full connection layer processing are combined, and MolCLR pre-training features are superposed; self-adaptively capturing feature interior and cross-feature association through a double-attention module (self-attention and cross-attention); the model introduces a pre-training feature branch and combines with parameter freezing to realize knowledge migration. On the basis of the reference data set, the PG-DDI is evaluated according to indexes such as ACC, AUC and F1, the performance of the PG-DDI is superior to that of an existing advanced method, the accuracy and generalization of DDI prediction are improved, and the method is suitable for the medication safety evaluation scene of drug research and development.

The invention belongs to the technical field of water ecology evaluation and water environment analysis, and discloses a dissolved organic matter chemical diversity-based water ecology integrity degradation diagnosis method, which analyzes DOM molecular fingerprints in a water body through an ultrahigh-resolution mass spectrum, realizes fine characterization of DOM molecular composition complexity and change thereof, and improves the water ecology integrity degradation diagnosis accuracy. The limitation that the traditional water ecology evaluation method is difficult to identify the molecular scale recessive degradation is broken through; by constructing a chemical diversity index and introducing an interpretable data driving model, a quantitative corresponding relation between a molecular structure parameter threshold value and water body chemical integrity degradation is established, and a degradation signal of a DOM (Document Object Model) library converted from complex stability to simple activity in a water ecosystem can be stably identified; and a molecular-scale technical support is provided for lake and river water ecological integrity degradation identification and management decision.

The invention relates to the technical field of patent analysis and drug research and development decision support, in particular to an LLM-based medical patent infringement risk multi-dimensional early warning method, which comprises the following steps of: analyzing a user attention level, configuring a risk dimension weight in combination with a decision scene, obtaining medical patent technical characteristics and a target product, and establishing a medical patent infringement risk model; the method comprises the following steps: analyzing claims by utilizing BERT to generate a multi-dimensional sub-structure map, quantifying technical characteristics by combining atomic stability and an attention matrix, evaluating avoidance feasibility by calculating a candidate site bond energy disturbance value and distribution density, fusing the technical characteristics with multi-source intelligence, constructing a multi-dimensional risk index set, and establishing a multi-dimensional risk index map; according to the multi-dimensional risk early warning method, the BERT is utilized to construct a substructure map, the contribution weight is quantified in combination with the atomic stability, the key energy is calculated to evaluate the avoidance feasibility, the molecular fingerprint is fused with multi-source data mapping, the weighted aggregation is executed according to the scene, the subjective deviation is eliminated, and the microscopic difference is mined. And accurate quantitative early warning of the medical patent infringement risk is realized.

The invention belongs to the field of guiding air-gap membrane distillation to recycle organic solvents, and discloses a method and a system for membrane material screening and process parameter optimization based on machine learning, and the method comprises the following steps: collecting pervaporation document data and air-gap membrane distillation experimental data, cleaning and aligning, and constructing a heterogeneous data set; generating an SMILES sequence code and a molecular fingerprint of the solvent, and carrying out standardization processing on process parameters to generate multi-modal input data; based on a deep learning regression model of a Transform architecture, executing a pre-training task based on pervaporation data and a transfer learning fine tuning task based on air gap type membrane distillation data; and performing high-throughput prediction on the chemical molecule library, and outputting an optimal solvent list and recommended process parameters. According to the method, historical data prediction and reverse optimization are utilized through machine learning, so that the experiment cost and trial and error times of membrane distillation organic solvent separation process optimization are remarkably reduced.

This invention discloses a method for predicting the function of bioactive peptides based on multi-view, multimodal representation learning. The method includes: extracting amino acid sequence information of peptides using multi-scale dilated convolutional CNN and bidirectional LSTM; extracting structural and functional features of peptide molecules using an ESM-2 model; processing molecular fingerprint information using convolutional CNN and Mamba structures; extracting topological information of the peptide molecular graph using traditional convolutional CNN, and processing node features using graph convolutional neural networks. All these multi-view features are ultimately fused into an aggregated feature representation, which is then passed through a fully connected layer and a sigmoid function is applied for multi-label classification. By concatenating and fusing the extracted features, a comprehensive peptide molecule feature representation is formed to predict various bioactive properties of the peptide. This multi-view, multimodal feature integration method not only enhances the model's predictive ability but also improves its flexibility and accuracy when processing complex biological data.

This invention discloses a method and system for predicting drug IC50 based on molecular structure and gene expression. By comprehensively characterizing the drug's molecular structure, chemical structure and property information are extracted using molecular fingerprinting and molecular mapping, and then fused with the cell's gene expression matrix to establish a unified feature expression system. A gene regulatory network reflecting potential regulatory relationships between genes is inferred based on a variational autoencoder (VAE). Furthermore, a multi-layer feature extraction mechanism combining global and local information is constructed. Global information is used to learn the overall regulatory structure between genes in the entire regulatory network through a graph neural network, while local information is captured by dividing and analyzing subgraphs to identify local interactions between drugs and key regulatory factors. This invention can more accurately characterize the mechanism of drug influence on cellular systems, significantly improve the prediction accuracy of IC50 values ​​and the interpretability of the model, and has good adaptability and broad application prospects.

The invention provides a polymorphic genome SSR molecular marker combination for identifying cotton varieties and application of the polymorphic genome SSR molecular marker combination in developing a molecular fingerprint spectrum and a two-dimensional code recognition system, and belongs to the technical field of molecular markers and bioinformatics. The invention provides a polymorphic genome SSR (Simple Sequence Repeat) molecular marker combination for identifying cotton varieties. The polymorphic genome SSR molecular marker combination for identifying the cotton varieties is NBRIHQ526730, NBRIHQ527820, HAU3071a, NAU3736b, NAU3913, NAU5172 and MONDPL0893. A core primer is screened from known cotton SSR primers positioned on a chromosome, and the polymorphic genome SSR molecular marker combination for identifying the cotton variety and the primers of the polymorphic genome SSR molecular marker combination are provided. And then constructing a DNA fingerprint spectrum and two-dimensional code recognition system of the Xinjiang cotton variety by using the SSR molecular marker combination.

The invention relates to a method for predicting fasting state drug solubility through a graph gating attention network based on meta-learning and application of the method. The method comprises the steps that molecular characterization information is integrated to form a framework model through multi-modal mixed molecular feature engineering, and model-independent meta-learning sample training is adopted; the multi-modal mixed molecular feature engineering comprises the following steps: inputting an SMILES character string of a to-be-tested compound to obtain a molecular graph, a molecular fingerprint vector and a normalized descriptor vector, constructing the molecular graph into a molecular graph vector by adopting an attention gating graph network, splicing and fusing the molecular fingerprint vector and the normalized descriptor vector, and obtaining the multi-modal mixed molecular feature engineering of the to-be-tested compound. And fusing with a molecular graph vector, and inputting into a multi-layer perceptron to obtain a predicted value. According to the method, the solubility of the candidate drugs in various key biological related media can be rapidly predicted, the solubility characteristic of the drugs is accurately evaluated in the early stage of research and development, the research and development period of new drugs is shortened, and huge research and development cost is saved.

The present application relates to the technical field of pollutant screening. Disclosed are a method and system for identifying organic pollutants in soil on the basis of a combination of AI and high-throughput screening. The method comprises: extracting substance peaks from high-resolution mass spectrometry data of a soil sample; constructing an organic pollutant mass spectrometry database, extracting spectrum features and structural features of compounds from the organic pollutant mass spectrometry database, constructing a molecular fingerprint prediction model, and establishing a spectrum-to-structure mapping relationship; on the basis of the extracted substance peaks, constructing spectrum vectors to predict molecular fingerprints, and acquiring candidate chemical structures by searching the organic pollutant mass spectrometry database; and scoring the candidate chemical structures by means of the predicted molecular fingerprints, selecting chemical structures that meet a preset standard, constructing an identification basis on the basis of the selected chemical structures, and acquiring an organic pollutant identification result for the soil sample. In the present application, non-targeted intelligent analysis is performed by means of the synergistic integration of artificial intelligence and mass spectrometry analysis, thereby realizing the rapid and accurate identification of organic pollutants in soil.

The invention relates to a molecular property prediction method based on topological perception generative molecular characterization, which comprises the following steps of: sampling molecular data from a public unlabeled molecular database, analyzing the molecular data into a molecular graph, and extracting a topological unit set; encoding the molecular graph by using a graph convolutional neural network and a cross-scale attention module, and obtaining a latent variable for subsequent generation; through a dynamic generation mechanism, atoms and chemical bonds are reconstructed in combination with Laplacian position codes, a molecular fingerprint prediction task is introduced, and a loss function is jointly constructed to pre-train a model; and carrying out fine tuning on the pre-trained model on a downstream reference data set MoleculeNet for specific molecular property prediction. By adopting the method disclosed by the invention, the problems of'feature inhibition ', 'shortcut learning' and the like of the existing self-supervised learning method can be solved, and the high calculation overhead of the traditional generative method is avoided, so that the molecular representation with strong discrimination is learned, and the accuracy, robustness and efficiency of molecular property prediction are improved.

The present application relates to the field of artificial intelligence, in particular to a chemical reaction yield optimization method, system, device and medium, comprising: obtaining training samples and yield labels of a target chemical reaction; generating molecular fingerprints of each training sample as initial representation, and determining substructures contained in the training samples; training a to-be-trained representation mapping model supervised by the yield labels, so that the similarity of the mapped features is consistent with the yield similarity, and obtaining a representation mapping model; mapping the initial representation to a reaction hidden representation through the representation mapping model; inputting the reaction hidden representation into a yield prediction model to obtain a predicted yield; calculating the contribution degree of each substructure by masking the influence of the corresponding code of the substructure on the predicted yield; and optimizing the target chemical reaction according to the contribution degree. The present application can accurately locate specific substructures that promote or inhibit the yield, and introduce beneficial substructures or avoid adverse substructures, thereby improving the yield of the target chemical reaction.

The present application provides a model for predicting endocrine disruptors by integrating pharmacology and toxicology profiles, and a construction method and application thereof. Specifically, the present application constructs an endocrine disruptor prediction model based on pharmacology and toxicology data and in combination with molecular fingerprints of substructures of compounds. The endocrine disruptor prediction model of the present application can more accurately and efficiently evaluate whether a to-be-tested compound is an endocrine disruptor. The present application also develops an endocrine disruptor prediction system based on the endocrine disruptor prediction model.