55 results about "Protein design" patented technology

Disclosed herein are predictive models for T-cell epitope prediction, B-cell epitope prediction, and protein design wherein a method is implemented for generating a protein variant amino acid sequence of a target protein having one or more modified properties, the method comprising: (a) iteratively sampling an input amino acid sequence of the target protein, and (b) sampling the individual protein score of at least one weighted relative contribution of the single residue mutant input amino acid sequence to the at least one target property across a plurality of other single residue mutant input amino acid sequences to generate a combined protein score, wherein the combined protein score corresponds to the protein variant comprising one or more amino acid mutations of the single residue mutant input amino acid sequences.

The invention relates to germline-targeting designs, stabilization designs, and / or combinations thereof, of proteins designed with modified surfaces helpful for immunization regimens, other protein modifications and / or development of nanoparticles, methods of making and using the same, and to (a) germline-targeting priming or boosting / shepherding immunogens to initiate or guide maturation of VRC01-class responses (b) PCT64 / PG9-germline-targeting designs (c) BG18-germline-targeting designs or boosting / shepherding immunogens to initiate or guide maturation of BG18-like responses, and / or (d) trimer stabilization and presentation in a membrane-bound format.

The application discloses a kind of fowl secretory bacterium surface proteins and its screening method and application, belong to biotechnology field.The application can be soluble high-efficiency expression in escherichia coli by screening protective antigen from fowl secretory bacterium heparin binding protein, combined with antigen prediction, soluble analysis and structure-oriented protein design, successfully obtained protein A0A3Q9GGY1, A0A3S9QKJ0 and six-site mutant PLO that can be soluble high-efficiency expression in escherichia coli, has good immunoprotective efficiency, and has accumulated experience for the screening and design of high-yield antigen, and has laid a foundation for the research and development of fowl secretory bacterium vaccine.

A biological programming specification that identifies at least one protein design condition in accordance with a biological programming language is received. A machine learning model is used to convert the biological programming specification to a model input format version for a biological reasoning model. The model input format version is used as a conditioning input for the biological reasoning model to generate a protein design having the at least one protein design condition.

The present disclosure relates to a compound of Formula (I) or a salt, enantiomer, diastereomer, isotopically enriched analogue, solvate, prodrug or polymorph thereof, and the use thereof. Further provided in the present disclosure are a pharmaceutical composition comprising, as an active ingredient, the compound of Formula (I) or a salt, enantiomer, diastereomer, isotopically enriched analogue, solvate, prodrug or polymorph thereof, and the use thereof. A series of compounds designed and synthesized in the present disclosure can effectively prevent and / or treat diseases or disorders associated with cereblon protein.

The application provides a method and system capable of end-to-end protein design and analysis verification, and can generate protein or protein complex structure and sequence according to specific requirements, and the method comprises the following steps: performing loss calculation on the three-dimensional structure of the initialized protein sequence obtained through the structure prediction algorithm independently developed by the company based on confidence, stability, target correlation; continuously improving the designed protein sequence and structure based on the Markov chain Monte Carlo algorithm or gradient regression according to the obtained loss; and putting the designed protein structure into a sequence design model to optimize the side chain, and obtaining the optimal candidate protein through screening. The method can train the model to be specific and automatic for protein design, and can generate a new artificial protein which is completely different from natural protein in structure and sequence.

A training set may be generated to include a plurality of noisy sample sequences. Each noisy sample sequence in the training set may be generated by adding noise to a corresponding sample sequence from a data distribution. A protein design computation model may be trained by at least applying the protein design computation model to generate one or more output sequences, and adjusting the protein design computation model to reduce a difference between the one or more output sequences and the plurality of noisy sample sequences in the first training set. The trained protein design computation model may be applied to generate an output sequence by at least modifying an input sequence.

The invention discloses a leaf and branch compost cutinase link and application thereof, and belongs to the technical field of enzyme molecule construction. The leaf-branch compost cutinase kink is obtained by adjusting the connection sequence of fragments in leaf-branch compost cutinase, and comprises a fragment II, a connecting peptide fragment I, a fragment I, a connecting peptide fragment II and a fragment III which are connected in sequence from the N end to the C end; wherein the fragment I, the fragment II and the fragment III respectively correspond to 37th to 79th amino acids, 81st to 143rd amino acids and 150th to 293rd amino acids or homologous sequences thereof of the leaf branch compost cutinase. According to the method, protein topology engineering and an artificial intelligence assisted protein design technology are organically combined, knot topology transformation is performed on the leaf branch compost cutinase, sequence optimization is performed on a connecting peptide fragment of the knot, and the leaf branch compost cutinase knot with good biological activity is obtained; the expression quantity and the stability of the leaf and branch compost cutinase kink are improved, the problem of kinetic barriers existing in protein kink combination is solved, and the method has important application value.

This invention discloses a method for designing transmembrane regulatory proteins based on cueing strategies and generative models, belonging to the field of bioinformatics. It includes a target survey stage and a closed-loop design stage. In the target survey stage, a generative diffusion model is used to generate virtual probes targeting the membrane protein. A set of complex conformations is obtained by combining sequence design and structure prediction models, and binding hotspot regions are identified based on the spatial distribution density of the probes, overcoming the dependence on manually specified binding sites. In the closed-loop design stage, based on the identified hotspot regions, an initial backbone is generated using a generative diffusion model. A structure cueing strategy guides the sequence design and structure prediction models to perform closed-loop iterative optimization, generating sequences that selectively bind to the transmembrane domains of membrane proteins and regulate their functions. This invention achieves automated, function-guided design of transmembrane regulatory proteins, effectively expanding the range of designable targets and significantly improving the stability and functional specificity of the designed products in the membrane environment.

The invention relates to the technical field of recombinant protein, in particular to self-assembled calcium chelate keratin RK35DE and a preparation method thereof. According to the invention, the self-assembled calcium-chelated keratin RK35DE is obtained through protein design, whole-gene synthesis, vector construction and escherichia coli expression. The keratin RK35DE is subjected to self-assembly research, and the TEM image of the keratin RK35DE shows that the keratin RK35DE has self-assembly capability. In a calcium chelating ability experiment, compared with that before recombination, the self-assembled calcium-chelated keratin RK35DE disclosed by the invention has a remarkable chelating effect on calcium ions. In the prior art, keratin with self-assembly capability lacks effective coordination capability of forming stable chelates with calcium ions in hydroxyapatite, so that stable compounds are difficult to form and exert effects. The self-assembled calcium chelate keratin researched and developed by the scheme provides a new thought for researching a new hydroxyapatite tooth restoration material, and has an ideal popularization and application prospect.

The present application relates to the technical field of protein design, and discloses a protein design knowledge graph construction method based on multi-agent, comprising the following steps: obtaining a natural language research topic input by a user; obtaining entity terms and intention objects by using a large language model, and generating a query strategy for a multi-source heterogeneous database based on the standardized entity terms and intention objects; obtaining original data from the multi-source heterogeneous database by using the query strategy; processing the parsed original data uniformly, and storing the processed data into a relational database; automatically extracting knowledge triples from the relational database and unstructured data, performing knowledge verification and deduplication, and storing the knowledge triples into a graph database; simultaneously, using a graph embedding algorithm and link prediction to perform knowledge completion and reasoning, and forming a domain-enhanced protein design knowledge graph; the present application solves the problems of low efficiency and poor generality of the existing protein design knowledge graph construction method.

The invention discloses arcanobacterium pyogenes surface protein as well as a screening method and application thereof, and belongs to the technical field of biology. According to the invention, a protective antigen which can be soluble and efficiently expressed in escherichia coli is screened from arcanobacterium pyogenes heparin binding protein, and antigen prediction, solubility analysis and structure-oriented protein design are combined; proteins A0A3Q9GGY1, A0A3S9QKJ0 and a six-site mutant PLO which can be soluble and efficiently expressed in escherichia coli and have good immune protection efficacy are successfully obtained, experience is accumulated for screening and design of high-yield antigens, and a foundation is laid for research and development of arcanobacterium pyogenes vaccines.

A biological programming specification that identifies at least one protein design condition in accordance with a biological programming language is received. A machine learning model is used to convert the biological programming specification to a model input format version for a biological reasoning model. The model input format version is used as a conditioning input for the biological reasoning model to generate a protein design having the at least one protein design condition.

Provided herein are systems and methods for protein sequence design. Also provided herein are engineered antibody sequences.

The invention relates to the technical field of protein design, and discloses a multi-agent-based protein design knowledge graph construction method, which comprises the following steps of: obtaining a natural language research theme input by a user; obtaining entity terms and intention objects by using a large language model, and generating a query strategy for the multi-source heterogeneous database based on the standardized entity terms and intention objects; acquiring original data from the multi-source heterogeneous database by utilizing a query strategy; performing alignment processing on the analyzed original data, and storing the processed data in a relational database; knowledge triples are automatically extracted from the relational database and the unstructured data, knowledge verification and duplicate removal are carried out, the knowledge triples are stored in a graph database, meanwhile, a graph embedding algorithm and link prediction are used for knowledge completion and reasoning, and a domain-enhanced protein design knowledge graph is formed; the problems that an existing protein design knowledge graph construction method is low in efficiency and poor in universality are solved.

Disclosed in the present invention are a protein for inhibiting the conserved helix of TDP-43 and the use thereof. The protein contains an amino acid sequence as shown in SEQ ID NO: 1 or SEQ ID NO: 2. Further disclosed in the present invention are the use of the protein in the preparation of a drug for diagnosing, preventing and / or treating neurodegenerative diseases, and the use in the preparation of an inhibitor of a target protein containing the conserved helix of TDP-43 LCD. In the present invention, by means of using artificial intelligence-assisted protein design techniques, a high-affinity non-natural protein that precisely binds to the conserved helix of a TDP-43 protein is designed, which is used for inhibiting the participation of the conserved helix region in the formation of a β-sheet aggregation core, is mainly used in the treatment of diseases targeting the TDP-43 protein, such as amyotrophic lateral sclerosis, and is used as a research tool for TDP-43 protein phase separation experiments.

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.

This invention discloses a method for identifying thermosensitive UDG constraint anchor points based on dual-path evolutionary analysis. The invention employs two paths for site identification. The first path constructs a two-layer multi-sequence alignment within the sequence space defined by the UDG family type 1 gate, consisting of a whole-family layer and a cold-adapted subset layer. This is combined with column-level conservation, branch dissimilarity, and conserved functional motif neighborhood window constraints to obtain a set of functional motif constraint anchor points. The second path constructs homologous multi-sequence alignments centered on the seed sequence, fusing alignment statistics with a sequence profile hidden Markov model for joint scoring to obtain a set of seed homologous core anchor points. Finally, the results from the two paths are fused in a union within the seed sequence residue coordinate system, outputting the site index and corresponding amino acid type. This can be directly used as fixed-site constraint input for thermosensitive UDG protein engineering or generative protein design.

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 invention relates to a protein design method, an electronic device and a storage medium, based on a structure file of a protein and a structure file of an interaction component, docking of the protein and the interaction component is carried out, and a first important site set of the protein is determined according to a docking result; analyzing a second important site set related to the function, structural stability, interaction or application value of the protein based on a preset knowledge base; designing the protein according to the first important site set and the second important site set, obtaining a design data set of the protein, and determining at least one piece of target design data in the design data set according to the screening model. According to the technical scheme, the standardized protein design process can be achieved, a set of standardized process solution is provided, the tool splitting problem of novel protein research is solved, the novel protein generation efficiency is improved, and the research cost is reduced.

This disclosure provides protein mutants generated using the large language program EVOLVE-Pro, which is designed for a wide range of applications. Examples of protein mutants include miniature CRISPR nucleases. EVOLVE-Pro substantially enhances the efficiency and effectiveness of in silico protein evolution, surpassing current state-of-the-art methods and yielding proteins with more than 100-fold improvement of desired properties. EVOLVE-Pro demonstrates the necessity for protein engineering models to zoom in on desired functional properties rather than predicted fitness, paving the way for broader applications of AI-guided protein engineering in biology and medicine.

The present disclosure provides an engineered delivery system comprising bacterial membrane vesicles derived from bacteria and one or more non-bacterial proteins for anchoring to the bacterial membrane vesicles.

The invention relates to the technical field of protein design, and particularly discloses an Mfp-GPX recombinant protein as well as a preparation method and application thereof, the recombinant protein takes mussel mucin Mfp-3 or Mfp-5 as a skeleton to assist anchoring of glutathione peroxidase GPX4, and a novel recombinant protein is obtained. The recombinant protein has excellent skin surface affinity and the functions of oxidation resistance, whitening, spot fading, inflammation resistance, sensitive skin repair, sunscreen and the like through fusion of the two proteins, and can play a long-acting role in oxidation resistance.

The application discloses a kind of based on CLDN1 protein's derivative peptide, fusion protein and its application, the derivative peptide includes the amino acid sequence as shown in SEQ ID NO:3, the amino acid sequence of the fusion protein is as shown in SEQ ID NO:2.The application finds that several derivative peptides containing CLDN1 protein extracellular loop EL1 peptide segment have significant inhibitory effect on various RNA and DNA envelope viruses.The fusion protein shows significant in vivo antiviral activity, can alleviate lung inflammation caused by RSV infection, and reduce viral nucleic acid level.Therefore, the derivative peptide and fusion protein based on CLDN1 protein designed in the application can be used for antiviral therapy or prophylactic drug research and development, and have wide application prospect.

This invention discloses a protein structure screening method and system based on topological causal auditing. Addressing the problems of poor chiral consistency and low laboratory synthesis success rates caused by deficiencies in statistical probability models in existing protein structure design, this invention proposes to utilize three-dimensional vector field modeling and a topological curl integral algorithm to achieve accurate auditing and screening of AI-generated structures by evaluating the physical causal consistency score of protein structures. Experimental data show that this method can improve the effective folding success rate of protein design by more than three times compared to existing technologies, and significantly reduce the computational power consumption and physical experimental costs in the R&D process. This invention can be widely applied to AI-assisted drug development, protein engineering, and target screening and structure verification of new biomaterials.

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for designing proteins. In one aspect, a method includes generating noisy molecular structure data sampled from a noise distribution, the noisy molecular structure data defining, for each position in an amino acid sequence of the protein, a corresponding initial spatial position for each atom in a predefined set of possible atoms; and processing the noisy molecular structure data using a diffusion model comprising a de-noising neural network to generate de-noised molecular structure data defining a de-noised version of the noisy molecular structure data.

The invention discloses a protein design method and device based on symmetric self-game online reinforcement learning in the technical field of bioinformatics and artificial intelligence. The protein design method based on symmetric self-game online reinforcement learning comprises the following steps: inputting main chain structure information of a target protein, constructing a training set based on a protein structure database, and representing a protein structure as graph data; constructing an isotropic graph neural network EGNN as an infrastructure of a sequence generator, and initializing a first strategy model, a second strategy model and a reference model; and symmetric self-game online training is carried out. According to the protein design method based on symmetric self-game online reinforcement learning, the limitation of existing static supervised learning and single-agent reinforcement learning is overcome, and collaborative optimization and sustainable evolution of a protein sequence under multi-objective constraint are realized.

The invention provides a protein sequence design method and system based on deep learning and 3D features. The method comprises the following steps: acquiring 3D structure data of a target protein; extracting 3D space structure features based on a physical and chemical environment coding framework; obtaining pseudo multi-sequence comparison information of the target protein, and extracting multi-sequence context features from the pseudo multi-sequence comparison information; constructing a memory tuning network with a bidirectional gating mechanism, and performing dynamic weight fusion on the 3D spatial structure features and the multi-sequence context features to generate a fused residue state; performing multiple rounds of iterative refining on the residue state by using a memory tuning network; and determining a final amino acid category of the residue position according to the amino acid category probability distribution corresponding to the maximum prediction confidence obtained by each residue position in all iteration rounds, and obtaining a final protein design sequence of the target protein. According to the method, the sequence design accuracy in the protein reverse folding task can be improved.

The invention relates to the technical field of recombinant protein design and synthesis, and particularly discloses a recombinant spider silk protein and a synthesis method thereof. The recombinant spider silk protein provided by the invention is subjected to N-terminal and C-terminal structural domain replacement of different spider species and different spider silk types, intermediate functional domain repetitive sequence large fragment recombination and other optimization designs, and a series of recombinant spider silk proteins are obtained through a cell-free protein synthesis method and prokaryotic fermentation expression; compared with natural spider dragline silk protein or other existing recombinant spider silk protein, the spider silk protein has good composite mechanical performance and high temperature resistance, meanwhile, the molecular weight is small, the total expression quantity and soluble expression are remarkably improved, and a novel raw material and a design synthesis method of the recombinant spider silk protein are provided for development of high-performance biological materials.