89 results about "Protein function" patented technology

The invention provides a protein methylation site prediction method fusing quantum and region perception, which comprises the following steps: step 1, acquiring a protein sequence as a data source, and respectively constructing a training set and an independent test set; 2, constructing a multi-modal feature for each protein sequence by adopting a three-way nested scattering network, and fusing the multi-modal features to obtain an optimized fusion feature tensor; and step 3, inputting the optimized fusion feature tensor into a RaQMeNet network model, and performing a methylation site prediction task. The performance indexes of the method are greatly superior to those of the prior art, and the method has higher adaptability, stability and interpretability, can be widely applied to a plurality of bioinformatics and biological medicine related fields such as protein function annotation, disease mechanism research and drug target discovery, and has good application prospects and commercial values.

The invention provides application of ancient-Chinese health-preserving essence in preparation of a protein regulating agent, and belongs to the technical field of biological medicines. The invention provides application of an ancient-Chinese health-preserving essence or an ancient-Chinese health-preserving essence extract in preparation of a reagent for regulating protein functions. Protein comprises at least one of NCAM2, NCAM1, MAP2, VGLUT1, CDK5, DCLK1, TUBB3 and apolipoprotein E. The invention also provides an application of the ancient-Chinese health-preserving essence or the ancient-Chinese health-preserving essence extract in preparation of a reagent for regulating protein functions. In ancient and Chinese health preserving essence positive feedback regulation proteins, an important nerve synaptic function network exists and comprises proteins such as NCAM2, NCAM1, MAP2 and the like, and the proteins play an important role in biological processes such as neural development, synaptic shaping, neural signal transmission and the like. In addition, the ancient-Chinese health-preserving essence can regulate the expression of APOE. The application of the invention not only helps to deepen the scientific understanding of the pharmacodynamic effect of the ancient-Chinese health-preserving essence, but also provides a theoretical basis for the modern and precise application of traditional Chinese medicine compounds.

The invention provides a protein function topology design method based on SE (3) isovariant diffusion and ex-situ generation and a product, and relates to the technical field of protein design. According to the method, geometric deep learning and generative artificial intelligence are fused, isovariant generation and optimization of protein function sites under the action of a three-dimensional Euclidean space (SE (3) group) are achieved, and the limitation of traditional protein design on conformation sampling efficiency, function guidance and physical realizability is broken through.

The invention discloses a protein function prediction method based on multi-modal fusion and a dynamic label network, and belongs to the technical field of biological information, an adjacent matrix of a label association network can be smoothly updated in a model training process, and the protein function prediction method can be used for predicting protein functions by extracting various modal information of protein. The redundant relation among multiple modes is removed, the prediction effect of protein functions is improved, meanwhile, a training method combining a protein function association network and a tag association network is used, the influence of multiple tags on protein function prediction is considered, and protein function prediction is achieved by fusing the sequence, structure and structural domain information of protein. According to the method, the complementarity of various data is fully utilized, so that the prediction capability of the model is improved, and compared with a traditional method using a static label relationship, the scheme dynamically updates the label relationship in model training, so that the generalization capability of the model is further improved.

The invention discloses a drug target prediction method based on fragment-level local and global feature fusion, and belongs to the technical field of computational biology and artificial intelligence drug design. Comprising the following steps: acquiring a medicine SMILES character string and a protein amino acid sequence; respectively segmenting the drug SMILES character string and the protein amino acid sequence to obtain a drug structure fragment sequence and a protein functional fragment sequence; and inputting the drug structure fragment sequence and the protein function fragment sequence into a pre-trained drug-target interaction prediction model to obtain a prediction probability of drug-target pair interaction. Compared with the prior art, the method has the advantages that convolution feature extraction, a multi-head attention mechanism and a gating fusion strategy are combined, an end-to-end DTI prediction framework is constructed, and the interaction between drugs and targets can be comprehensively mined.

The invention relates to the technical field of artificial intelligence, and provides a protein function prediction method and device based on multi-modal protein data, and the method comprises the steps: obtaining protein multi-source data, carrying out the feature extraction of a protein sequence in the protein multi-source data, and obtaining a protein sequence feature; constructing a heterogeneous graph based on the protein multi-source data; performing feature coding on the heterogeneous graph by adopting a graph attention mechanism to obtain protein graph features; performing multi-modal fusion on the protein sequence features and the protein map features by adopting a gating fusion mechanism to obtain fusion features; and performing multi-label classification prediction based on the fusion features to obtain a protein function annotation result. The accuracy and robustness of protein function prediction can be improved, and the problems that in the prior art, multi-source protein data cannot be effectively integrated, and the method is sensitive to data noise are solved.

The invention relates to the field of bioinformatics, and provides a protein function prediction method and system based on deep learning. The method comprises the following steps: constructing a multi-modal data set of protein; performing feature extraction on the multi-modal data set through an encoder to obtain multi-modal features; training a neural network through the multi-modal features to obtain a prediction model; and predicting the function of the to-be-detected protein through the prediction model to obtain a prediction result. The protein function prediction accuracy is improved.

The application discloses a recombinant oncolytic virus targeting CD317 gene and application thereof in anti-tumor, and belongs to the technical field of tumor treatment. The recombinant oncolytic virus comprises a CD317 inhibitor and an oncolytic virus, and is formed by integrating the CD317 inhibitor into the oncolytic virus genome. The CD317 inhibitor is a substance capable of inhibiting CD317 gene expression or targeting degradation of CD317 protein function, and is selected from shRNA or siRNA targeting CD317. The application develops the oncolytic virus targeting knockdown of CD317 expression, inhibits tumor cell proliferation by reducing CD317 expression of tumor cells, reduces PD-L1 expression so as to break the immune escape mechanism, simultaneously enhances the killing sensitivity of tumor cells to CD8+ T cells, forms a synergistic effect with the oncolysis of the oncolytic virus, and the recombinant oncolytic virus has stronger in-vivo anti-tumor activity, thereby providing a new potential scheme for CD317-driven tumor treatment.

The invention relates to a method for producing an animal model of heart failure. The method comprises the step of weakening or deleting DDB1 protein function in myocardial cells of the animal model. The present application demonstrates that nuclear DDB1 co-agglomerates with MEF2C to control NAD + biosynthesis as well as ion homeostasis genes in the heart, and the lack of which results in the development of HFpEF. Development of HFpEF in a'double strike 'mouse model can be reversed through AAV-mediated DDB1 overexpression in myocardial cell nucleuses. Therefore, it is detected that DDB1 coordinates NAD + biosynthesis and ion homeostasis to protect the heart from being affected by ejection fraction retention heart failure caused by obesity, and the scheme of the application has therapeutic significance on treatment of obesity / diabetes HFpEF.

The application discloses a kind of heterodimer interchain residue contact prediction method.The present application is aimed at the deficiency of existing method in prediction accuracy, long-range dependence modeling capability and generalization, and proposes a kind of deep neural network integrating multiple features and attention mechanism.Specifically, the application integrates multiple features such as protein language model as network model input, then the network adopts efficient channel attention (ECA) and spatial attention (SA) module and KAN convolution network module, effectively captures the local and global dependence features of heterodimer, so as to predict the interchain residue contact of heterodimer.Experiments show that the prediction accuracy of the application on the benchmark dataset is significantly better than that of existing methods, and the model has high robustness.The application can be widely used in the field of protein heterodimer interchain residue contact prediction and protein structure prediction, further promoting protein function research and protein drug development.

The invention provides a method for inserting an unnatural amino acid and application thereof, the method adopts a pyrrolysine aminoacyl-tRNA synthetase mutant as an orthogonal translation element to introduce the unnatural amino acid into a protein to obtain a protein containing the unnatural amino acid, the invention also provides an in-vitro cell-free protein synthesis system for inserting the non-natural amino acid, and the system can efficiently introduce the non-natural amino acid, especially the lysine analogue non-natural amino acid. The technical bottleneck that a natural translation system is low in non-natural amino acid recognition efficiency is solved, and an efficient and controllable technical tool is provided for protein function research and biological medicine development.

The application discloses a kind of quick site analysis method and detection system of biotin labeled KRAS protein, belong to intelligent analysis technical field;Construct the site fluorescence intensity database under historical detection state, obtain the fluorescence intensity data of each amino acid site;Based on the data, construct the amino acid site-fluorescence intensity coordinate system of each sample, generate fluorescence intensity fluctuation curve;Extract the wave crest and wave trough of curve, construct fluorescence peak site pair set and valley site pair set;Obtain the mode of peak site pair and valley site pair in all samples, as high response site and low response site, calculate its probability of occurrence in new sample, and combine the peak value and valley value of real-time sample with preset threshold, output high response type, low response type or stable type classification result.The application realizes the quick, objective, intelligent discrimination of KRAS protein function state by historical data probabilistic modeling and fluctuation curve feature extraction.

A protein active site prediction method based on geometric graph neural network belongs to the field of bioinformatics and protein structure analysis. First, the original protein data is preprocessed by multi-modal feature extraction and geometric graph construction, and ProtT5 deep embedding and physicochemical properties are fused. Then, a deep ActiveSiteGNN model with explicit geometric perception ability is constructed, and the stacked geometric encoder and geometric edge update layer are used to dynamically capture the micro three-dimensional spatial features. Next, a multi-task collaborative optimization and dynamic threshold search strategy is designed, combined with weighted sampling to solve the serious sample imbalance, and the best decision threshold is selected based on the validation set in real time. The integrated reasoning and graph diffusion smoothing technology is introduced to globally calibrate the prediction probability distribution based on the biological space prior. Finally, the evaluation is carried out on the independent test set. The method has strong structure perception ability and provides a feasible solution for accurate prediction of protein functional sites.

This invention discloses a drug target affinity prediction method that integrates PPI quality and uncertainty. The method constructs a drug molecule map and a multimodal protein structure representation, and extracts multi-source features by combining the local PPI sub-map of the target protein. By calculating the protein's low-frequency level, prediction uncertainty, and PPI quality, a PPI quality-aware gating factor is generated to adaptively adjust the PPI information injection intensity, and a residual enhancement strategy is used to preserve the original protein features. Subsequently, the drug representation and the enhanced protein representation are fused using adaptive gating, and the result is input into a prediction network to output the drug-target affinity. This method effectively integrates protein function and interaction information, improves the prediction stability of low-frequency proteins and the model's generalization ability, and provides an accurate and reliable computational tool for drug screening and candidate molecule selection.

The invention relates to the technical field of prevention and control of agricultural and forestry pests, and discloses a synergistic protein ORF76 of insect baculovirus and application of the synergistic protein ORF76. According to the invention, a synergistic protein (ORF76) is obtained from an insect biocontrol resource, i.e., a Pinus fumosa nuclear polyhedrosis virus, and protein function verification shows that the protein has a remarkable synergistic effect on a fall webworm nuclear polyhedrosis virus, an apocheima cinerarius nuclear polyhedrosis virus and a Pinus fumosa nuclear polyhedrosis virus; the compound can be used as a synergistic factor to be added into insect viruses, and has important significance on prevention and control of major forestry pests (fall webworms, spring inchworm and smoke wing pine bees).

The invention discloses a method for creating a male sterile line and a maintainer line of maize based on a CRISPR-Cas12i. 3 system and application of the method. The invention belongs to the field of genetic breeding, and particularly relates to a method for creating a male sterile line and a maintainer line of maize based on a CRISPR-Cas12i. 3 system and application of the method. The method for preparing the transgenic corn comprises the following steps: mutating an Ms26 gene of a receptor corn genome to cause Ms26 protein function deletion to obtain the transgenic corn, and the transgenic corn has at least one of the following characteristics: (1) compared with the receptor corn, the tassel of the transgenic corn shows compact spikelet, tight glume protection and incapability of normal pollen scattering, and the transgenic corn has the characteristics that the Ms26 protein function deletion is caused by the mutation of the Ms26 gene of the receptor corn genome; no pollen is exposed; and (2) compared with the receptor corn, the transgenic corn is completely aborted. By establishing an efficient backcross transformation method, the genic male sterile line without transgenic ingredients is rapidly obtained, the breeding period is greatly shortened, and the application cost is reduced.

The invention relates to the technical field of medical detection, and particularly discloses a biomarker combination for identifying cervical cancer molecular subtypes, a kit and application. The biomarker combination comprises a protein CDH13, a protein TP53BP1, a protein NNMT and a protein HSPB1, and molecular subtype correlation analysis is carried out on a sample by detecting relative expression characteristics of the proteins in a cervical cancer in-vitro sample. Based on the relative expression characteristics of each protein, a cervical cancer sample can be divided into at least one of an epithelial-mesenchymal transition related subtype, a proliferation related subtype, an immune response related subtype and an epithelial differentiation related subtype. The invention also provides a detection kit and an analysis system for realizing cervical cancer molecular subtype identification. According to the technical scheme, the cervical cancer can be subjected to molecular typing from the protein function execution level, a reliable technical means is provided for molecular typing research, prognosis evaluation and accurate treatment related research of the cervical cancer, and the application prospect is good.

A protein function identification method, a system, a terminal, and a storage medium. The method comprises: acquiring a bulk protein sequence and a bulk protein function corresponding to the bulk protein sequence, inputting the bulk protein sequence into multiple deep learning pre-trained models, respectively, and outputting multiple multi-dimensional feature vectors; performing feature fusion processing on the multiple multi-dimensional feature vectors to obtain a fused feature vector, and training a bi-directional long short-term memory model on the basis of the fused feature vector and the bulk protein function to obtain a protein function identification model; and acquiring a protein sequence to be identified, inputting said protein sequence into the protein function identification model, and outputting a protein function identification result. Feature vectors of a protein sequence are extracted by using multiple different deep learning models, and a bi-directional long short-term memory model is trained to obtain a protein function identification model, such that rapid and accurate identification of protein functions can be realized.

This disclosure describes a machine learning (ML) framework, including a graph convolutional neural network (GCN), for identifying gene expression signatures associated with cancer driver events. The model is trained to identify the TP53 mutation status of cancer samples from gene expression, utilizing a comprehensive, curated graph structure of gene interactions. Quantitative scores are generated to rank the severity of driver events in each sample. Very high AUC results for unknown data across several tumor types are achieved in this method. A strong correlation with protein function exists. The Signature in Transcriptome Associated with Mutant Proteins (STAMP) model can also predict driver events in many combinations of key oncogenes / pathways and several tumor types, based on well-established annotations from the literature. Thus, the STAMP model can identify and quantify driver events, which may lead to improved targeted therapy selection and prioritization in cancer patients.

The application discloses a kind of based on protein-based line and relationship field reasoning's dog tumor risk prediction method and device.Step includes: data acquisition and uniform warehousing, obtain the sample data containing dog metadata and multiple protein quantitative values;Numerical pre-processing and canine context coding, logarithmic smooth transformation is carried out to protein value, and context base function vector is constructed;Canine individualized protein-based line reconstruction;Protein function module division and module abstract construction;Protein relationship field construction, based on standardized residual, form the relationship channel of protein, generate original evidence intensity;Inflammation confounding stripping, obtain tumor-specific evidence;Multi-branch evidence accumulation and hierarchical risk determination, calculate cross-module consistency and multiple branch scores, output risk stratification conclusion through gate threshold system.The application breaks through the limitation of uniform reference interval, effectively eliminates inflammation interference, the reasoning process is strong in explainability, significantly reduces the misjudgment rate and guarantees clinical safety, and AUC is stable and can reach 0.84.

The invention discloses a PRRSV (Porcine Reproductive and Respiratory Syndrome Virus) N protein monoclonal antibody as well as a preparation method and application thereof, belongs to the technical field of biological materials, and solves the problem that in the prior art, the stability and specificity of a PRRSV monoclonal antibody screening result are not ideal enough. The method comprises the following steps: constructing a gene of a PRRSV N protein into a plasmid, and performing expression and purification to obtain a recombinant N protein; uniformly mixing the recombinant N protein, a Modifer reagent, a Quencher reagent and a DR-B reagent, centrifuging, taking supernate, re-suspending the supernate in an SB reagent, and separating out biotinylated N protein through a membrane; and screening the positive B cells combined with the biotinylated N protein by using a Hypercell high-throughput single cell screening platform to obtain the monoclonal antibody. The monoclonal antibody aiming at the PRRSV N protein is prepared by a single B cell screening technology, and a foundation is laid for PRRSV diagnosis and related research of N protein functions.

The present application discloses a protein marker for early pregnancy detection of mammals, and particularly relates to a hidden marker protein serum amyloid A of early pregnancy test paper. Protein sequencing used in the present application adopts liquid chromatography-mass spectrometry (LC-MS) technology, sample correlation is compared through Venn, sample correlation, PCA and PLS-DA analysis; protein function annotation is carried out through GO, KEGG, COG and Pfam, and subcellular localization prediction is carried out; difference protein, protein set and difference metabolite are calculated and counted; correlation analysis of proteome and transcriptome is carried out, difference protein of sheep serum in early pregnancy is obtained, and a key marker protein of early pregnancy of sheep is screened, which provides a basis for unknown early pregnancy detection of sheep.

The invention refers to the regulation of function of proteins through insertion of a peptide into the selected protein and its interaction with a regulatory peptide that interacts with the inserted peptide. The invention can be used to activate or inactivate the function of different selected proteins and therefore to regulate their properties and processes, useful for pharmacological, therapeutic, diagnostic, sensing, biotechnological and other industrial applications.

This invention discloses a protein sequence spatial compression and functional optimization method based on a large-scale model, belonging to the fields of artificial intelligence and proteomics. This invention mines potential amino acid substitution sites in consensus sequences and then controls the sequential substitution process using a large protein language model, thereby maintaining the functional stability of proteins during sequence substitution and subsequently screening for substitution combinations that effectively enhance protein function. Introducing a large protein model transforms protein sequences into embedding vectors representing protein structure, function, and physicochemical properties. By analyzing the embedding vectors during the substitution process, it is possible to prevent new proteins from deviating from their original function and basic structure due to substitution. This invention combines consensus substitution identification with large-scale model analysis, effectively compressing the sequence space of amino acid substitutions, thereby significantly improving the efficiency of protein design and modification.

The invention provides a protein sequence structure joint design method based on a natural language, and relates to the technical field of artificial intelligence, and the method comprises the following steps: constructing natural language information representing protein functions, an amino acid secondary structure and an amino acid contact diagram according to protein design requirements; processing the natural language information through a text encoder to generate condition features; meanwhile, the triangular perception encoder is used for processing the amino acid secondary structure and the amino acid contact diagram information, and condition features are generated; a protein map structure is constructed through multiple pieces of modal information, and local and global context features of protein are captured; iteratively updating the amino acid sequence and three-dimensional structure coordinates of the protein by using an equivariant decoder; according to the method, invariant cross entropy loss, invariant frame alignment loss and contrast loss are utilized to perform joint optimization on generation of protein sequences and structures, and after multiple rounds of iteration, protein amino acid sequences and three-dimensional structures with'sequence-structure-function 'consistency are output.

The application provides a renaturation tag for purifying inclusion body proteins, and a gene-derived product and application thereof, and belongs to the technical field of protein purification. The application provides a renaturation tag P67, and the amino acid sequence is at least one of SEQ ID NO:1-SEQ ID NO:3, wherein the 33th amino acid is lysine, glutamic acid or proline. The renaturation tag P67 is fused with target proteins for expression in prokaryotes, so that the recombinant expressed inclusion body (insoluble inclusion body) is successfully renatured and has biological activity. The renaturation tag P67 establishes a foundation for subsequent protein function research and application transformation.

The invention relates to an anthrax protein nanopore system and application thereof in protein sequencing, and belongs to the technical field of nanopore detection. The invention aims to solve the problems of low flux, insufficient accuracy and limited nanopore resolution in the existing protein sequencing technology. According to the invention, an anthrax protein PA63 heptamer nanopore electrochemical detection device is constructed, an asymmetric pH buffer system is adopted, Trypsin, Calpain-1, Pancreative Elatase, CarboxyceptidaseA enzyme and the like are added step by step for polypeptide enzyme digestion, voltage is applied to drive amino acid to pass through nanopores, current signals are collected, amplitude, retardation time and frequency characteristics are analyzed, and polypeptide amino acid sequence detection is realized. The nanopore is small in opening current (fA level) and high in resolution ratio, detection accuracy and sensitivity are enhanced by combining multi-enzyme step-by-step cutting and carving, and a direct and efficient sequencing means is provided for protein function research.

This invention belongs to the field of biotechnology, and particularly relates to the use of the SPON2 gene or its protein as a target in the diagnosis or treatment of endometriosis. Supported by multi-omics and single-cell level data and tested with clinical samples, this invention, for the first time, elucidates the correlation between SPON2 and endometriosis (EM), revealing that high expression of SPON2 in eutopic and ectopic endometrial fibroblasts in endometriosis patients enhances their proliferation, invasion, and migration abilities. Inhibiting SPON2 gene expression or inhibiting SPON2 protein function reduces the proliferation and invasion abilities of endometrial tissue, as well as lesion migration, invasion, and fibrosis, thus improving the condition of endometriosis. This suggests that the SPON2 gene or its protein can serve as a diagnostic or therapeutic target, providing new ideas and directions for the clinical diagnosis and targeted therapy of endometriosis.

The invention relates to a Contig perception-based protein function prediction method and system. The method comprises the following steps: obtaining a protein amino acid sequence, a nucleotide sequence corresponding to Contig and arrangement information of CDS on the Contig; splicing the protein-level features corresponding to the CDS sequence on the same Contig with the k-mer frequency vector of the Contig nucleotide sequence to generate enhanced protein-level features; setting the length of a sliding window, acquiring a plurality of fragments with fixed window lengths, which have protein function tags at the central positions of enhanced protein-level features according to a CDS sequence on the same Contig, taking one fragment as a sample, and training a function prediction network consisting of a bidirectional long-short-term memory network BiLSTM and a multilayer perceptron, and obtaining a target protein prediction probability value of each fragment. And the accuracy of protein function annotation is obviously improved.