Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

67 results about "Protein Databases" patented technology

Protein structure database. In biology, a protein structure database is a database that is modeled around the various experimentally determined protein structures. The aim of most protein structure databases is to organize and annotate the protein structures, providing the biological community access to the experimental data in a useful way.

Method for Sequence-Based Prediction of Controlled Terms and Generating Protein Sequences from Controlled Terms using Enhanced Large Language Models

The present invention relates to a method for enhancing the creativity of a generative pre-trained Large Language Model (LLM) in protein sequence generation and predicting controlled terms from protein sequences. The method includes incorporating 22 novel names representing the 22 amino acids into the vocabulary of the pre-trained LLM, conducting self-supervised learning using protein sequences encoded with the novel names to improve the LLM's comprehension and generation of coherent protein sequences, performing supervised learning using protein sequences to refine the LLM's ability to predict controlled terms based on protein sequences, and performing supervised learning to refine the LLM's ability to generate protein sequences based on controlled terms. The method includes generating the novel names either through a computer program or manually and utilizing datasets of protein sequences and their corresponding controlled terms from a protein database. The self-supervised learning employed is a masked language model (MLM). Additionally, an alternative method is disclosed, which involves identifying a set of 22 amino acid names from the original vocabulary of the pre-trained LLM and proceeding with self-supervised and supervised learning steps using the selected names. The two methods can be used independently or in combination to enhance the creativity of the LLM in achieving protein sequence generation and predicting controlled terms from protein sequences.
Owner:HO SARAH WENHSIA

Method for predicting antibacterial peptide in bacteria based on deep learning

The invention provides a method for predicting antibacterial peptides in bacteria based on deep learning, and relates to the technical field of antibacterial peptide prediction of microorganisms, antibacterial peptide sequences are collected from a plurality of antibacterial peptide databases to serve as a positive data set, and protein sequences are collected from a protein database to serve as a negative data set for training; all amino acid sequences in the positive data set and the negative data set are converted into vector sequences according to a digital-amino acid conversion table to serve as model input; and constructing a multi-layer combination model, inputting the vector sequence, and outputting a prediction probability value so as to judge the antibacterial peptide sequence with antibacterial activity.
Owner:OCEAN UNIV OF CHINA

Protein toxicity prediction method and system based on sequence information

The invention provides a protein toxicity prediction method and system based on sequence information. The method comprises the following steps: acquiring a plurality of protein sequences from a protein database; performing feature calculation on all the protein sequences to obtain six types of protein feature vectors; splicing the six types of feature vectors according to columns to obtain a first feature vector; performing dimension reduction screening on the first feature vector to obtain a second feature vector of the target dimension; training a neural network model based on graph attention by using the second feature vector, and taking the trained neural network model as a protein toxicity prediction model; and inputting the new protein sequence into the protein toxicity prediction model to obtain a toxicity prediction result output by the protein toxicity prediction model. The method can adapt to different types of protein sequences, and the protein toxicity is accurately predicted through the learned universal sequence mode.
Owner:SUZHOU UNIV

Kit for enriching low-abundance protein in blood at high flux and application of kit

The invention relates to a kit for high-flux enrichment of low-abundance protein in blood and application of the kit, the surface of a PSA group modified nano magnetic bead is modified with a functional group containing an N-propylethylenediamine structure, the functional group is-Si-O-Si-(OR) n-CH2CH2CH2NHCH2CH2NH2, n is independently 0, 1 or 2, and n is independently 0, 1 or 2. In the formula,-OR is-OCH3,-OCH2CH3, or-OCH2CH2CH3, and-OR is-OCH3,-OCH2CH3, or-OCH2CH2CH3. Compared with existing commercially available MAX strong negative ion group modified nano magnetic beads, the number of low-abundance proteins in enriched plasma proteins is increased by 308%, high-flux enrichment can be achieved, low-abundance proteins with the concentration lower than 10 ng / L can be enriched, the sensitivity is high, the nano magnetic beads can be used for high-flux enrichment of blood low-abundance proteins, and the application prospect is wide. Particularly, low-abundance proteins which are ranked from 2000 to 5000 in a plasma protein database are enriched.
Owner:PUDU ZHONGHE (WUHAN) LIFE TECH CO LTD

Universal HLA antigen presentation prediction method and system based on protein language model and multi-modal neural network

The invention discloses a universal HLA antigen presentation prediction method based on a protein language model and a multi-modal neural network, which comprises the following steps: firstly, extracting verified HLA binding peptide fragment sequences from an immune epitope database, generating equivalent non-epitope peptide fragment sequences in combination with a protein database, and constructing an HLA-I class and HLA-II class balanced data set; then, extracting sequence embedding characteristics and contact graph structure information of the peptide fragment sequence by utilizing a protein language model; the method comprises the following steps: processing a protein map constructed by a contact map through a map neural network to obtain global structure features, and meanwhile, carrying out regional convolution and residual convolution processing on sequence embedding by adopting a one-dimensional convolutional neural network (1DCNN) to extract local context sequence features; the method effectively fuses sequence semantics and space structure information, improves the accuracy and generalization ability of HLA antigen presentation prediction, and is suitable for immune recognition modeling tasks under various HLA subtypes.
Owner:WUHAN HUADA ZHIYAN TECHNOLOGY CO LTD +1

Protein implicit binding site prediction method based on deep learning and application thereof

The invention relates to a protein implicit binding site prediction method based on deep learning. The method comprises four stages of data acquisition, feature engineering, model training and verification application. The method comprises the following steps: firstly, acquiring initial data from a protein database, performing data screening and sequence redundancy elimination, performing standardization processing on a PDB structure file, and generating a sample; then, constructing a composite feature space containing basic protein features and evolutionary conservative features; thirdly, feature dimension reduction is achieved through a hierarchical feature conversion and integrated feature selection method, and a model is constructed; finally, verification and application show that the method obtains 99.44% prediction accuracy on 954 non-redundant protein structures, ROC-AUC and PR-AUC both reach 0.9998, and the overlapping rate with a BioLiP database is 44.4%. The method can accurately identify potential drug binding sites, especially shows good generalization ability for drug targets difficult to prepare, and has important application prospects.
Owner:THE NAT CENT FOR NANOSCI & TECH NCNST OF CHINA

Recombinant plasmid vector, preparation method of enzyme for hydrolyzing organophosphate, carbamate and pyrethroid insecticides and preparation using enzyme as functional component

The invention relates to a process for producing three recombinases capable of hydrolyzing organophosphorus, carbamate and pyrethroid insecticides, and preparations containing these enzymes. The present invention provides recombinant plasmid DNA vectors in which the protein secretion system has been modified to improve the ability of recombinant proteins to secrete into the space around the plasmid or culture medium. A nucleotide sequence for coding a PelB (butyric acid lyase B) signal sequence is deleted and substituted by a nucleotide sequence for coding a LamB (maltodextrin or phage lambda receptor) signal sequence, so that the secretion efficiency of the recombinant protein is higher. The invention discloses three recombinant plasmid DNA (deoxyribonucleic acid) vectors which are constructed by utilizing the improved plasmid DNA vectors and are used for expressing recombinant proteins. The recombinant protein comprises (a) a recombinant organophosphate hydrolase having at least 80% sequence identity with the amino acid sequence of methyl parathion hydrolase (MPH) (Protein Database (PDB) ID: 1P9E) as shown in SEQ ID NO: 1, wherein the recombinant organophosphate hydrolase has hydrolytic activity on organophosphate insecticides; (b) a recombinant carbamate hydrolase having at least 80% sequence identity with the amino acid sequence of L-aminopeptidase-D-Ala-esterase / amidase (aryl amidase) (MAH) (PDB ID: 1B65), as shown in SEQ ID NO: 2, wherein the recombinant carbamate hydrolase has hydrolytic activity on carbamate insecticides; and (c) a recombinant pyrethroid hydrolase having at least 80% sequence identity with the amino acid sequence of the pyrethroid hydrolytic carboxylesterase (EstA) (PDB ID: 4CCY) as shown in SEQ ID NO: 3, wherein the recombinant pyrethroid hydrolase has hydrolytic activity on a pyrethroid insecticide. The embodiment of the invention further provides three recombinant bacterial strains, an escherichia coli strain BL21 (DE3) is used as a host cell, and each recombinant plasmid DNA vector is used for transformation. The invention also discloses culture conditions of the recombinant strain, wherein the culture conditions comprise culture medium composition, induction conditions and enzyme harvesting and enzyme activity detection methods. In addition, the embodiment of the invention also provides a preparation which contains the three recombinases as functional components.
Owner:BIOM GMBH

Method for directionally mining glutamine transaminase based on structural analysis

The invention discloses a method for directionally mining glutamine transaminase based on structural analysis, which comprises the following steps of: 1, retrieving a bioinformatics database, downloading a sequence containing transglutaminase annotation information, and performing duplicate removal processing; step 2, obtaining a protein structure file corresponding to the sequence through an AlphaFoldDB database by using a programming language; 3, performing structural similarity analysis on the protein structure files and known TGase locally in batches by using a programming language through DaliLite software; 4, screening to obtain a potential novel TGase sequence and structure with high similarity with the known TGase; and 5, carrying out activity verification so as to confirm the novel Tgase. According to the bioinformatics database mining method suitable for glutamine transaminase, redundant information in a protein database can be processed in batches, the workload of traditional manual data processing and experimental screening is reduced, rapid and accurate database resource mining for target industrial enzyme is achieved, and universality is achieved.
Owner:JIANGNAN UNIV

Preparation method for botulinum toxin type a and use of same

PCT designated stage expiredWO2024250799A1Senses disorderPeptide/protein ingredientsProtein DatabasesBotulinum toxin type
Provided is a preparation method for recombinant botulinum toxin type A (BoNT / A), the method comprising: designing a BoNT / A amino acid sequence in a P0DP10 protein database; by using a homologous recombination method, inserting an His tag and an SUMO tag sequence into the N terminal of the BoNT / A amino acid sequence; and then performing cleavage between GG-M by means of a SUMO tag protease. A result shows that no amino acid residue of any tag and the SUMO sequence is present, so that the safety is remarkably improved. Comparative experiments and animal experiments show that the recombinant botulinum toxin type A polypeptide which is prepared, purified and activated on the basis of a bacterium body centrifugal supernatant has higher virulence, so that the effective dose during practical use is reduced, thus remarkably reducing side effects and stimulation and achieving higher safety.
Owner:SUZHOU REJUVO PHARMACEUTICAL CO LTD

Prediction of the selection of the T-cell receptor repertoire using a physical model augmented by pseudomarkers

Systems and methods for predicting T-cell receptor (TCR) peptide interactions, including training a deep learning model for predicting TCR peptide interactions by determining a multiple sequence alignment (MSA) for TCR peptide pairs from a dataset of TCR peptide pairs using a sequence analyzer, constructing TCR structures and peptide structures using the MSA and corresponding structures from a Protein Data Bank (PDB) using a modeler, and generating an extended TCR peptide training dataset based on docking energy scores determined by peptide docking to TCRs using physical modeling based on the TCR structures and peptide structures constructed using the modeler.TCR peptide pairs are classified and labeled as positive or negative pairs by using pseudo-labels based on the docking energy scores, and the deep learning model is iteratively retrained based on the extended TCR peptide training dataset and the pseudo-labels until it converges.
Owner:NEC LABORATORIES AMERICA INC

Protein complex model global quality evaluation method based on deep fusion network

A protein complex model global quality evaluation method based on a deep fusion network belongs to the field of bioinformatics and computer application, and comprises the following steps: firstly, setting specific thresholds for maximum sequence redundancy, resolution and residue length in a protein database RCSB PDB to screen out a natural protein structure; for each protein structure, generating a disturbance structure model by using an HDck and xTrimoMultimer docking algorithm and combining a chain orientation disturbance strategy; then, extracting evolutionary features, physical features and geometric features based on each protein structure model, and constructing a deep fusion network fusing a deep convolutional neural network, an axial attention mechanism network and a graph attention network so as to fully capture local and global topological information of a protein compound; and finally, a global quality score is obtained through prediction of a decoding module. According to the method, the global topological information of the protein complex model can be effectively captured, so that the quality of the protein complex model is evaluated.
Owner:ZHEJIANG UNIV OF TECH

A molecular design method for improving the hydrophobicity of the catalytic center of a bacillus subtilis protease

The application belongs to the field of bioengineering. The application provides a molecular modification design method for improving the hydrophobicity of the catalytic center of bacillus subtilis protease, which comprises the following steps: searching for a resolved bacillus subtilis protease crystal structure in a protein database PDB, generating an enzyme-substrate complex by docking a hexapeptide substrate through a Glide mode of software, calculating the polar charge of each amino acid residue of polar atoms N and O and H combined thereof within a 10 angstrom range of the catalytic center amino acid of the bacillus subtilis protease by using a PPC force field, calculating the absolute value sum of the polar charge, selecting a potential mutation site of the bacillus subtilis protease, virtually mutating the potential mutation site into a large steric hindrance nonpolar amino acid, calculating the absolute value sum of the polar charge of polar atoms N, O and H within a 10 angstrom range of the catalytic center amino acid of the single-point mutant by using the PPC force field, selecting a virtual bacillus subtilis protease single-point mutant, expressing the mutant and iteratively mutating, re-expressing, obtaining a modified bacillus subtilis protease mutant, and verifying the hydrophobicity. The molecular modification design method can quickly and effectively reduce water molecules in the catalytic center of the bacillus subtilis protease and improve the hydrophobicity thereof. It has been verified that the enzyme modification method is effective and reliable.
Owner:EAST CHINA NORMAL UNIV +1

Universal method for screening protein cancer biomarkers and application thereof in nasopharyngeal carcinoma

The invention belongs to the technical field of biology, and particularly relates to a universal method for screening protein cancer biomarkers and application thereof. The universal method for screening the protein cancer biomarkers comprises the following steps: (1) preparing a specific enrichment material to carry out peptide fragment enrichment on an actual biological sample, and carrying out primary data processing through LC-MS (Liquid Chromatography-Mass Spectrometer); and (2) further retrieving the data obtained by the primary processing through a protein database, and screening target proteins with research potential for subsequent qualitative and quantitative research. The method disclosed by the invention can be used for simply, conveniently and quickly screening potential protein type biomarkers, and the research efficiency of the biomarkers is improved.
Owner:EAST CHINA UNIV OF SCI & TECH

Homologous protein searching method based on scale invariant feature transformation and pre-filtering

The invention relates to the technical field of bioinformatics, in particular to a homologous protein searching method based on scale invariant feature transformation and pre-filtering, and the method comprises the steps: obtaining # imgabs0 #-carbon atom coordinates in a to-be-compared protein structure, calculating a corresponding distance matrix, extracting a feature description vector matrix of the distance matrix by using an SIFT (scale invariant feature transform) algorithm, and obtaining a feature description vector matrix of the to-be-compared protein structure; the matrix is mapped to a high-dimensional spherical surface, candidate protein structures meeting pre-filtering conditions are rapidly screened out from a large-scale protein database through a spherical surface partition pre-filtering algorithm, and the to-be-compared protein structure and the feature description vector matrix of the candidate protein structures are accurately compared through GPU acceleration. And generating a protein structure comparison result, and further determining the homologous protein corresponding to the to-be-compared protein. Therefore, the comparison speed is increased while the comparison precision is kept, the defect that efficiency and precision are difficult to consider in the prior art is overcome, and the method is suitable for rapid search and efficient comparison of a large-scale protein database.
Owner:TSINGHUA UNIVERSITY

Oligomeric protein de novo design method, system and equipment and storage medium

The invention discloses an oligomeric protein de novo design method and system, terminal equipment and a computer readable storage medium. The method comprises the steps that functional characteristics of target protein are acquired, and a first design rule is determined based on the functional characteristics; based on the first design rule, extracting a first protein sequence and a second protein sequence from a natural protein database, and determining a second design rule; generating a candidate mutation sequence library according to the first design rule and the second design rule; performing structure prediction and stability calculation on the sequences in the candidate mutation sequence library, and screening out optimized sequences meeting preset requirements; performing functional characteristic simulation verification on the optimized sequence, and if the verification is passed, outputting a final target protein sequence; otherwise, the verification result is fed back to the protein sequence optimization model for iterative optimization design. The problem that the efficiency is low when the oligomeric protein is designed is solved, and the effect of efficiently designing the oligomeric protein from the beginning is achieved.
Owner:SHENZHEN INST OF ADVANCED TECH CHINESE ACAD OF SCI

Preparation method of polypeptide with antioxidant activity

The invention discloses a preparation method of polypeptide with antioxidant activity, and relates to the technical field of biology. The preparation method comprises the step of preparing polypeptide with an amino acid sequence shown as SEQ ID NO.3 through a solid-phase synthesis method. According to the invention, a plurality of antioxidant active peptides are obtained by using separation and purification and mass spectrum identification technologies and by means of a saury protein database; multiple rounds of screening are carried out by utilizing the DPPH free radical scavenging rate, reducing power, water solubility, biotoxicity and other properties of the active peptide; the interaction between the antioxidant active peptide and the receptor MPO is explored by combining molecular docking, and the in-vitro antioxidant activity of the polypeptide is verified through solid-phase synthesis and cell experiments. Results show that the active peptide obtained through screening has in-vitro antioxidant activity, has good water solubility and is free of biotoxicity.
Owner:BOHAI UNIV

Discovery method of heterogeneous entanglement elements and application of heterogeneous entanglement elements in topological protein synthesis

The invention discloses a discovering method of a heterogeneous entanglement element and application of the heterogeneous entanglement element in topological protein synthesis. Through automatic screening and manual verification of a protein structure database, a plurality of protein heterogeneous entanglement elements are systematically found, the efficiency of mediating protein isohydrocarbon formation is identified, and a richer tool library is provided for protein mechanical coupling. The protein heterogeneous entanglement element not only can efficiently form single-structural-domain Soxhlet hydrocarbon, but also can tolerate fusion of functional proteins, does not affect the folding structure and activity of the protein, and can be used for designing a mechanical coupling fluorescence resonance energy transfer fluorescent protein pair and a bispecific antibody analogue to realize four-level structure regulation and control and multifunctional integration.
Owner:PEKING UNIV

A screening method for molecularly mimicking bisphenol A-specific peptides

This invention proposes a method for screening bisphenol A-specific peptides using molecular simulation, belonging to the field of food safety technology. The method includes the following steps: 1) Through protein stacking and molecular docking, the molecular structure and interactions of bisphenol A-protein cocrystal compounds in a protein structure database are comprehensively analyzed to obtain the receptor protein and the original parent chain; 2) Based on the original parent chain obtained in step 1), the parent peptide is obtained by truncation of the parent peptide segment by analyzing the main active amino acids in the molecular docking results; 3) By performing virtual amino acid mutations on the parent peptide obtained in step 2), a peptide library specifically binding to bisphenol A is constructed, and molecular dynamics is used for preliminary screening of the peptides; 4) The peptide chains screened in step 3) are used as specific recognition elements, and peptides labeled with fluorescein isothiocyanate are used to form specific recognition probes. The specific recognition probes are mixed with bisphenol A, and the peptide sequences with the strongest specific binding ability to bisphenol A are screened based on the different fluorescence differences. This invention is mainly applied to the screening of specific recognition elements for toxic and harmful small molecules, active ingredients, and functional factors. This method avoids the complex preparation steps of antibodies and the cumbersome screening process of aptamers, achieving high-throughput design of specific recognition elements.
Owner:ANHUI GUJING DISTILLERY CO LTD +2

Self-supervised learning pre-training method and system based on protein size prompt

The invention discloses a self-supervised learning pre-training method and a self-supervised learning pre-training system based on protein size prompt. The self-supervised learning pre-training method comprises the following steps: firstly, collecting unlabeled protein data from a protein database; encoding the protein size information into a prompt vector by using a size prompt adapter, and embedding the prompt vector into a protein encoder; then, based on the graph or point cloud representation of the protein, executing a mask prediction task to complete pre-training; and finally, performing fine adjustment on the pre-training model in a downstream task, and optimizing the specific task performance of the pre-training model. According to the method, the problem of inconsistent data distribution between the pre-training and the downstream task is remarkably reduced through protein size prompt, so that the universality of the pre-training model and the performance of the downstream task are improved. The method has high data efficiency, model universality and expansibility, and is suitable for various tasks such as protein function prediction and binding site detection.
Owner:FUDAN UNIVERSITY

Construction method of jellyfish toxin metalloproteinase substrate peptide library

The invention discloses a construction method of a jellyfish toxin metalloproteinase substrate peptide library, and relates to the technical field of biology. The construction method comprises the following steps: inducing mice with different dermatitis degrees by using a metalloproteinase inhibitor and jellyfish toxin; skin tissues with different dermatitis degrees are collected for proteome identification, and substrate membrane proteins of jellyfish toxin metalloproteinase are analyzed and screened through bioinformatics; screening by utilizing a protein database to obtain a potential substrate sequence in the substrate membrane protein; and screening sequences from the potential substrate sequences according to application conditions to construct a peptide library, and synthesizing the jellyfish toxin metalloproteinase substrate peptide library. The sequence contained in the jellyfish toxin metalloproteinase substrate peptide library constructed by the method is clear, the effect that the prepared polypeptide is hydrolyzed by toxin is remarkable, and the application prospect of the peptide library in jellyfish stings research is further guaranteed.
Owner:INST OF OCEANOLOGY - CHINESE ACAD OF SCI

Cyclopeptide for specifically recognizing sucralose and computer-aided optimization design method thereof

According to the cyclopeptide capable of specifically recognizing sucralose and the computer-aided optimization design method thereof, through computer simulation, molecular docking, molecular dynamics, deep learning and other technologies, the combination free energy of molecular docking score and molecular dynamics simulation is used as a judgment basis, and the cyclic peptide capable of specifically recognizing sucralose is obtained without a wet test. And the cyclopeptide with high affinity to target small molecules such as sucralose can be rapidly and efficiently designed. Comprising the following steps: extracting a high-quality protein structure from a protein database, splitting the high-quality protein structure into oligopeptide fragments and carrying out geometric fingerprint calculation; through molecular docking with a target small molecule, an optimal peptide fragment is selected and combined for recombination, and the space structure is optimized; and reconstructing a cyclic peptide sequence through information of the affinity layer, the adjacent layer and the shell layer obtained through calculation to obtain the high-affinity cyclic peptide for targeted recognition of the target small molecules. The design cost is greatly reduced, the design success efficiency is improved, and the method has a wide prospect when being used for identifying and detecting sucralose in small molecules such as environmental sewage.
Owner:JIANGNAN UNIV

Soybean sensitization protein epitope detection method based on co-immunoprecipitation

The invention discloses a soybean sensitization protein epitope detection method based on co-immunoprecipitation, and belongs to the technical field of immunodetection. The detection method comprises the following steps: performing enzymolysis on soybean meal; mixing the enzymatic hydrolysate with human or animal serum allergic to the soybean allergenic protein, and incubating overnight to obtain a mixed sample; preparing agarose beads coupled with the anti-IgE antibody and the protein A; fully mixing the soybean allergenic protein peptide with a mixed sample, and incubating overnight to obtain soybean allergenic protein peptide-IgE antibody-anti-IgE antibody-protein A agarose beads; eluting, and collecting the eluent to obtain the soybean allergenic protein epitope peptide; and carrying out mass spectrum sequencing, comparing the sequencing result with a protein sequence in a soybean allergenic protein database, and identifying the soybean allergenic protein epitope. The invention provides a soybean allergenic protein epitope detection method based on co-immunoprecipitation, the detection method is efficient, accurate and simple and convenient to operate, is suitable for the fields of food or feed quality control, allergenic mechanism research, food or feed safety guarantee and the like, and can effectively improve the detection efficiency of soybean allergenic protein epitopes.
Owner:SHENYANG AGRI UNIV

Method for exploring influence of electromagnetic waves on protein structure directly based on molecular dynamics simulation

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

Peptide sequence generation model and method utilizing intermediate sequence MSA and diffusion masking mechanism

This invention relates to a peptide sequence generation model and method utilizing intermediate sequence semantic masking (MSA) and diffusion masking mechanisms, specifically a peptide sequence AI model training method incorporating evolutionary information. It innovatively introduces intermediate sequences to address the problem that peptide sequences are too short to directly extract evolutionary information. First, an MSA is constructed for the peptide dataset. For sequences where an effective MSA cannot be constructed, their sequences are compared with a protein database to select intermediate sequences with high homology and appropriate length. Then, MSA data of these intermediate sequences is constructed to guide the AI ​​model in capturing evolutionary features. Model training employs a diffusion language masking mechanism and an MSA Transformer. All regions are randomly masked in the peptide MSA, and peptide mapping regions are masked in the intermediate MSA, enabling embedding modeling and generation of peptides. This method can capture sparse peptide evolutionary information, achieving semantic modeling and efficient generation of peptides, and is applicable to peptide sequence optimization and synthesis in fields such as drug design and protein engineering.
Owner:LIANGZHU LAB

Protein toxicity prediction method and system based on sequence information

The present invention provides a method and system for predicting protein toxicity based on sequence information. The method comprises obtaining multiple protein sequences from a protein database; performing feature calculation on all protein sequences to obtain protein feature vectors for six categories; concatenating the six feature vectors by column to obtain a first feature vector; performing dimensionality reduction screening on the first feature vector to obtain a second feature vector of the target dimension; using the second feature vector to train a graph attention-based neural network model, and using the trained neural network model as a protein toxicity prediction model; inputting a new protein sequence into the protein toxicity prediction model to obtain a toxicity prediction result output by the protein toxicity prediction model. The present invention is adaptable to different types of protein sequences and accurately predicts protein toxicity by learning common sequence patterns.
Owner:SUZHOU UNIV

A method for identifying a breakable cross-linked peptide segment, a method for evaluating and screening the identification result

The application provides a method for identifying a breakable cross-linked peptide segment, and the method comprises the following steps: S1, extracting characteristic double peaks with a mass difference corresponding to a fixed mass difference of a cross-linking agent in a spectrum, taking the mass corresponding to the characteristic double peaks as the mass of an alpha peptide segment, obtaining a candidate mass pair composed of the mass of the alpha peptide segment and the mass of a beta peptide segment based on the parent ion mass corresponding to the breakable cross-linked peptide segment, and the mass of the beta peptide segment is complementary to the mass of the alpha peptide segment; wherein the parent ion mass is obtained by calibrating a primary spectrum of the breakable cross-linked peptide segment; S2, according to the candidate mass pair obtained in step S1, screening a plurality of candidate peptide segments respectively equal to the mass of the alpha peptide segment and the mass of the beta peptide segment from a protein database according to the mass of the alpha peptide segment and the mass of the beta peptide segment in the candidate mass pair, respectively forming paired candidate cross-linked peptide segments by combining all candidate peptide segments of the alpha peptide segment with all candidate peptide segments of the beta peptide segment to form a candidate cross-linked peptide segment set, matching and scoring the cross-linked peptide segments in the candidate cross-linked peptide segment set according to a preset fragment ion, and selecting the cross-linked peptide segment with the highest score.
Owner:INST OF COMPUTING TECH CHINESE ACAD OF SCI

A Cathelicidin family polypeptide derived from an aquatic mammal, and its preparation method and use

The present invention discloses a Cathelicidin family polypeptide derived from aquatic mammals. The polypeptide is obtained by mining the proteome database of aquatic mammals. The amino acid sequence of the polypeptide is as shown in SEQ ID NO: 1, SEQ ID NO: 2, or an amino acid sequence having more than 90% homology with SEQ ID NO: 1 or 2. The present invention intercepts the mature peptide segments of the Cathelicidin family by sequence alignment and mining of the protein database of aquatic mammals, and obtains the polypeptides BM-CATH and LV-CATH. The antibacterial polypeptides BM-CATH and LV-CATH designed and synthesized by the present invention have a broad antibacterial spectrum, a fast bactericidal speed, low hemolytic activity against sheep red blood cells, and low cytotoxicity to mammalian cells. They can be applied to the treatment of common bacterial infections and drug-resistant bacterial infectious diseases, and are excellent alternative drugs or adjuvant drugs for existing anti-infection treatments.
Owner:CHINA PHARM UNIV

Greedy approach to identifying peptides with multiple post-translational modifications

A greedy approach to identifying peptides with multiple post-translational modification is provided. A method includes extracting tags from input data and reducing information indicative of a protein database. The extracting includes converting peaks in tandem mass spectra of the input data into a weighted directed graph, resulting in extracted tags. The tags represent sequential amino acids. The reducing includes determining respective coverages of proteins in the protein database using the extracted tags. Further, the method includes locating a selected tag in an indexed database configured for protein candidate retrieval and scoring ones of the proteins that comprise the selected tag. The selected tag is selected from the extracted tags. Further, the method includes using a greedy approach process that characterizes post-translational modification patterns of the selected tag based on the scoring. The method also includes, based on a result of the greedy approach process, implementing a quality control process.
Owner:THE HONG KONG UNIV OF SCI & TECH

Homologous protein search method based on scale-invariant feature transformation and pre-filtering

The present application relates to the field of bioinformatics technology, and in particular to a method for searching homologous proteins based on scale-invariant feature transformation and pre-filtering, wherein the method comprises: obtaining the carbon-atom coordinates in the protein structure to be compared, calculating the corresponding distance matrix, and extracting the characteristic description vector matrix of the distance matrix using the SIFT algorithm, mapping the matrix to a high-dimensional sphere, and quickly screening out candidate protein structures that meet the pre-filtering conditions from a large-scale protein database through a pre-filtering algorithm of spherical partitioning, and performing GPU-accelerated precise comparison of the characteristic description vector matrix of the protein structure to be compared with the candidate protein structure to generate a protein structure comparison result, and then determining the homologous protein corresponding to the protein to be compared. As a result, the present application improves the comparison speed while maintaining the comparison accuracy, overcomes the insufficiency of the prior art that efficiency and accuracy are difficult to take into account, and is suitable for rapid search and efficient comparison of large-scale protein databases.
Owner:TSINGHUA UNIVERSITY

Methods and systems for predicting drug binding using synthetic data

A method for predicting drug-target binding using synthetically enhanced data involves generating multiple phantom ligands for multiple proteins in a protein structure database; generating multiple DTI features for proteins and ligands in a drug-target interaction (DTI) database using the multiple phantom ligands; generating a machine learning model using the multiple DTI features; and using the machine learning model to predict the likelihood of interaction for a combination of a query protein and a query ligand.
Owner:CYCLICA INC