Artificial intelligence design and expression system construction method and system of heparan sulfate-alpha-glucoside N-acetyltransferase

Abstract

The invention relates to the technical field of artificial design of enzymes, in particular to an artificial intelligence design and expression system construction method and system for lysosomal membrane protein type N-acetyltransferase with 11 transmembrane regions, and the method comprises the following steps: collecting reaction rate information of an enzyme and a substrate, associating a structure model with rate parameters, and comparing three-dimensional difference of residues, according to the method, by collecting the reaction rate correlation structure conformation, dynamic recognition of the key conformation state is achieved, the accuracy of three-dimensional space difference analysis is improved, host expression optimization factors are fused in the construction process, and the construction efficiency is improved. According to the method, the expression efficiency is improved, conformation function screening and structural stability parallel evaluation are carried out, the screening accuracy of efficient catalysis and stable expression is enhanced, three links of recognition, screening and construction are broken through, the target enzyme obtaining efficiency and expression quality are improved, and enzyme engineering is promoted to be developed towards high-throughput systematization.

Description

Technical Field

[0001] This invention relates to the field of artificial enzyme design technology, and in particular to the artificial intelligence design and expression system construction method and system for heparan sulfate-α-glucosidase N-acetyltransferase. Background Technology

[0002] The field of artificial enzyme design technology involves the rational analysis and targeted modification of enzyme molecular structure and function to endow them with new catalytic properties or improve their original performance. Core aspects include molecular modeling, active site analysis, mutation site screening, and expression optimization. This technology primarily employs methods such as computational simulation, structural prediction, site-directed mutagenesis, and high-throughput screening to modify the structure and function of enzymes, thereby meeting the application needs of specific industrial, pharmaceutical, or scientific research purposes. Among these, the artificial intelligence design and expression system construction method for traditional heparan sulfate-α-glucosidase N-acetyltransferase refers to engineering methods for enhancing the function and optimizing the expression of key enzyme species required during the modification of heparan sulfate sugar derivatives. The technical issue addressed by this patent is how to effectively obtain N-acetyltransferases with specific substrate recognition capabilities and high transfer efficiency. Traditionally, the structural basis is determined through homology modeling combined with amino acid sequence alignment, enzyme activity analysis is used as a screening method, and further functional residues are modified using site-saturated mutagenesis. Heterologous expression and purification are then performed using eukaryotic expression systems suitable for membrane protein expression (such as yeast or mammals) for functional verification.

[0003] In current technologies, the screening of key enzymes still mainly relies on static structural models and homologous sequence alignment, lacking analysis and utilization of dynamic structural changes during the reaction process. This leads to residue function identification depending on single-point enzyme activity assays, resulting in gaps in spatiotemporal coverage. During construction, expression optimization strategies are often disconnected from functional design, frequently employing single-site mutations to modify functional residues, neglecting transcriptional regulation and transport efficiency in the expression host, easily leading to problems such as insufficient protein expression or folding errors. The configuration of regulatory fragments, transport signals, and induction conditions lacks systematic integration, often requiring multiple rounds of experimental screening, resulting in high resource and time costs. For example, if a system unsuitable for expressing multi-membrane proteins is used, the complex structure of the target protein and the presence of multiple transmembrane regions can easily lead to misfolding, expression failure, or inclusion body formation, resulting in distorted subsequent enzyme activity assessment results and affecting the judgment and optimization of functional residue modification pathways. Furthermore, existing processes fail to form a complete closed loop, lacking an automatic mapping relationship between conformational assessment and construction recommendations, leading to frequent manual intervention and correction of recommended schemes, limiting efficiency and accuracy, and making it difficult to support the rapid and efficient development of membrane protease engineering applications. Summary of the Invention

[0004] To address the technical problems existing in the prior art, embodiments of the present invention provide a method for constructing an artificial intelligence design and expression system for heparan sulfate-α-glucosidase N-acetyltransferase, comprising the following steps: To achieve the above objectives, the present invention employs the following technical solution: a method for constructing an artificial intelligence design and expression system for heparan sulfate-α-glucosidase N-acetyltransferase, comprising the following steps: S1: Collect information on the binding rate of enzyme and substrate reaction process, associate time node structural model number with rate parameter, construct the basis for structural comparison and screening index, and generate a conformation rate number lookup table. S2: Extract the associated structures according to the conformation rate number lookup table, compare the three-dimensional coordinate difference of the same residue under the conformation in the substrate binding region, screen the residue numbers that meet the set threshold, combine the participation range with the regional attribute labeling function, and generate a list of key sites for conformational changes. S3: Based on the residue numbers of the list of key sites for conformational changes, design combinations of replacement amino acids, select regulatory fragments and transport signal sequences with reference to the codon arrangement in the host, and generate a list of sequence component assembly templates; S4: Based on the fragment combination of the assembly template list of sequenced components, summarize the regulatory segments, temperature control requirements, transport signals and auxiliary gene sequences, combine them into an executable process flow, and generate a fragment assembly condition configuration table. S5: Based on the fragment assembly condition configuration table, identify the functional arrangement of the substrate binding region and the catalytic site, and combine structural stability and regional integrity to generate a target enzyme construction sequence design set.

[0005] As a further aspect of the present invention, the conformation rate numbering table includes structural model number, time node identifier, binding rate parameter, and alignment index key; the list of key sites for conformational changes includes residue number, three-dimensional coordinate difference grading threshold, functional participation range label, and regional attribute identifier; the sequence component assembly template list includes replacement amino acid combination scheme, codon optimization scheme, template number, and host reference codon library; the fragment assembly condition configuration table includes regulatory segment, induction temperature control parameter, transport signal fragment, helper gene sequence, assembly process flow number, and operating condition threshold; and the target enzyme construction sequence design set includes recommended construction scheme number, functional arrangement annotation, catalytic site annotation, structural stability grading, reaction region integrity level, and target sequence finished product set.

[0006] As a further aspect of the present invention, the specific steps of S1 are as follows: S101: Collect information on the binding rate of enzyme and substrate at time nodes, extract reaction state data at corresponding nodes, identify number pairs that correspond to rate changes and spatial conformation changes, establish the correspondence between time nodes and structural conformation numbers, and obtain a set of conformation rate pairing numbers. S102: Based on the structure number in the conformation rate pairing number set, extract the main chain and side chain parameters of the corresponding structure model, calculate the structural overlap rate between models, and summarize the degree of structural difference between all conformation models to obtain the conformation structure overlap rate matrix. S103: Call the conformational structure coincidence rate matrix and the conformational rate pairing number set, filter the data point pairs whose rate change magnitude and structure coincidence rate decrease value both exceed the conformational stability limit value, establish the pairing relationship between the conformation number and the corresponding rate parameter, and obtain the conformational rate number lookup table.

[0007] As a further aspect of the present invention, the specific steps of S2 are as follows: S201: Extract the structural data associated with the conformation number according to the conformation rate number lookup table, obtain the list of residues corresponding to the substrate binding region, extract the spatial position parameters of all residues in the region in each conformation state, record the corresponding three-dimensional coordinate information, and generate a set of substrate region residue coordinates. S202: Based on the three-dimensional coordinates of the same residues in the substrate region residue coordinate set under different conformational states, calculate the spatial position difference between corresponding coordinates of residues, filter residue numbers whose spatial position difference exceeds the residue displacement difference threshold, and obtain a position offset residue number table. S203: Call the residue numbers selected from the position offset residue number table, extract the attribute labels of the local region where the number is located based on the position characteristics in the substrate binding region, identify the residue numbers of the reactive functional region, summarize the correspondence between residue numbers and regional attributes, and generate a list of key sites for conformational changes.

[0008] As a further aspect of the present invention, the specific steps of S3 are as follows: S301: Based on the residue numbers in the list of key sites for conformational changes, select multiple replaceable amino acids for the corresponding positions, design permutation and combination methods, generate data corresponding to the combinations and residue numbers, and obtain a residue substitution combination data table. S302: Call the combination generated in the residue replacement combination data table, screen the codon combinations that can be used for expression regulation according to the codon arrangement in the host for yield enhancement, extract the codon sequence that meets the expression benchmark, and obtain the codon matching arrangement value set; S303: Based on the codon sequences in the codon matching permutation value set, select the corresponding control fragments and transport signal sequences, integrate the fragments to form template content that can be directly used for construction, and generate a sequence component assembly template list.

[0009] As a further aspect of the present invention, the specific steps of S4 are as follows: S401: Based on the fragment combinations in the sequence component assembly template list, extract the corresponding regulatory segments, induction temperature control requirements, transport signal fragments and auxiliary gene sequences, establish data mapping relationships for the corresponding fragment types under the combination, and generate a fragment composition structure mapping table. S402: Based on the type characteristics of the segment combination in the segment composition structure mapping table, the temperature value, start-up condition and response time parameters in the induced temperature control requirements are collected, and combined with the start position sequence information of the corresponding control section, the required temperature control matching coefficient is calculated to obtain the temperature control configuration value set. S403: Based on the temperature control coefficients combined in the set of temperature control configuration values, call the corresponding regulatory segments, signal segments and auxiliary gene sequences in the fragment composition structure mapping table, combine them into executable process parameter items according to the type classification rules, and establish a fragment assembly condition configuration table.

[0010] As a further aspect of the present invention, the specific steps of S5 are as follows: S501: Based on the combination content in the fragment assembly condition configuration table, identify the substrate binding region in the conformational model, extract the structural spatial information, and label the hydrophobic sites, polar residues and charge distribution characteristics in the functional arrangement to generate a substrate binding arrangement parameter set. S502: Call the structural information and residue distribution data in the substrate combination arrangement parameter set, calculate the structural stability value and reaction region integrity value of the conformation, screen fragment combinations that meet the screening criteria, and obtain the conformation screening matching coefficient; S503: Based on the conformation screening matching coefficients, extract the corresponding fragment sequences, reconstruct the nucleic acid coding sequences in functional order, and establish a target enzyme construction sequence design set.

[0011] As a further embodiment of the present invention, the structural model number is a three-dimensional protein model identification code generated by a structural prediction tool; The rate parameter is the binding rate constant or dissociation rate constant measured by a surface plasmon resonance device during the enzyme-substrate binding reaction. The three-dimensional coordinate difference is the difference in spatial position of the same residue in different conformational states, expressed as the amount of coordinate change of the main chain atom or the center of the side chain.

[0012] As a further aspect of the present invention, the host is a biological carrier for producing the target enzyme, which is a mammalian cell expression system; The regulatory fragment is a nucleotide sequence that controls the direction and rate of sequence synthesis, and can be a promoter or terminator; The transport signal sequence is a target short peptide sequence located at the amino terminus; The regulatory region is a nucleotide sequence that includes gene expression regulation functions, and its function is similar to that of the regulatory fragment. The accessory gene sequence is a nucleotide fragment encoding a chaperone protein or a folding-promoting factor; The functional arrangement is the spatial distribution order of active residues around the substrate channel within the catalytic region.

[0013] An artificial intelligence-based design and expression system for heparan sulfate-α-glucosidase N-acetyltransferase, including: The structure rate acquisition module collects binding rate information during the enzyme-substrate reaction process, extracts the structure model number generated at the reaction time point, associates the rate parameters with the corresponding model number, and generates a conformation rate number lookup table. The conformation comparison and screening module extracts the spatial position of residues in the substrate binding region according to the model number listed in the conformation rate number lookup table, compares the differences in the three-dimensional coordinates of the same residues in the conformation, screens residue numbers whose position offset meets the standard, and generates a list of key sites for conformational changes by combining the structural region type identifier. The sequence component assembly module selects the alternative arrangements of corresponding amino acids based on the residue numbers in the list of key sites for conformational changes, combines them with known codon sequences in the host that can enhance expression levels, matches regulatory fragments and transport signal sequences, forms a set of fragments that can be used for sequence synthesis, and generates a list of templates for sequence component assembly. The process configuration module summarizes the required control segment information, expression induction conditions, transport signal sequences and auxiliary segment numbers according to the fragment combination content in the sequence component assembly template list, and combines them to form an execution list for expression construction, and generates a fragment assembly condition configuration table. The target sequence construction module identifies the functional arrangement of substrate binding regions and catalytic sites in the conformational model according to the configuration content in the fragment assembly condition configuration table, extracts information related to structural stability and reaction region layout, performs combination and classification analysis, screens construction schemes, and generates a target enzyme construction sequence design set.

[0014] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, by collecting reaction rate-related structural conformations, dynamic identification of key conformational states is achieved, improving the accuracy of three-dimensional spatial difference analysis. The construction process integrates host expression optimization factors to improve expression efficiency. Conformation function screening and structural stability are evaluated in parallel, enhancing the screening accuracy of efficient catalysis and stable expression. By connecting the three major links of identification, screening, and construction, the efficiency of target enzyme acquisition and expression quality are improved, promoting the development of enzyme engineering towards high-throughput systematization. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic diagram of the steps of the present invention; Figure 2 This is a detailed schematic diagram of S1 of the present invention; Figure 3 This is a detailed schematic diagram of S2 of the present invention; Figure 4 This is a detailed schematic diagram of S3 of the present invention; Figure 5 This is a detailed schematic diagram of S4 of the present invention; Figure 6 This is a detailed schematic diagram of S5 of the present invention; Figure 7 This is a system module diagram of the present invention. Detailed Implementation

[0017] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0018] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0019] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning.

[0020] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same. The construction of expression systems includes the sequence design of expression elements, optimization of the combination of regulatory fragments, configuration of expression parameters and output of target construction schemes, but does not necessarily include actual molecular cloning and in vitro expression steps.

[0021] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0022] Please see Figure 1 This invention provides a method for constructing an artificial intelligence design and expression system for heparan sulfate-α-glucosidase N-acetyltransferase, comprising the following steps: S1: Collect information on the binding rate of enzymes and substrates during the reaction process, associate the structural model number and rate parameters generated at the time point, construct the index basis for structural comparison and selection, and generate a conformation rate number lookup table. The structural model number is a three-dimensional protein model identification code generated by a structural prediction tool, used to distinguish differential conformation models, including those that can be generated through the AlphaFold or Chimera platforms. Rate parameters are binding rate constants or dissociation rate constants measured by a surface plasmon resonance device during the enzyme-substrate binding reaction, and are used to characterize the reaction kinetics. S2: Extract associated structures based on the conformation rate number lookup table, extract the spatial positions of residues in the conformational state in the substrate binding region, compare the three-dimensional coordinate differences of the same residues one by one, screen the residue numbers that meet the set position difference standards, and identify them as functional participation ranges according to the attributes of the region they are in, and generate a list of key sites for conformational changes. Conformational state is the three-dimensional spatial structural state of a protein during differential function or reaction stages, used to reflect changes in molecular morphology; Three-dimensional coordinate difference is the difference in spatial position of the same residue in different conformational states. It is expressed as the amount of coordinate change of the main chain atom or the center of the side chain of the residue and is used to assess the degree of structural change. Regional attributes are the functional categories of differentially structured fragments within a substrate-binding region, used to distinguish whether a fragment belongs to a substrate-binding pocket, catalytic site, or hydrophobic region. S3: Based on the residue numbers in the list of key sites for conformational changes, design multiple arrangements of replacement amino acids, refer to the codon arrangements in known hosts that can increase yield, select corresponding regulatory fragments and transport signal sequences, integrate them into a template set for sequence synthesis, and generate a template list for assembling sequence components. The host is a biological carrier used to produce the target enzyme, which is a mammalian cell expression system; Regulatory fragments are nucleotide sequences that control the direction and rate of sequence synthesis. They are promoters or terminators used to regulate the initiation and termination of transcription of target genes. The transport signal sequence is a target short peptide sequence located at the amino terminus, which is used to guide the target protein into the target subcellular compartment or to be secreted into the extracellular space after synthesis; S4: Based on each fragment combination in the sequence component assembly template list, summarize the corresponding regulatory segments, induction temperature control requirements, transport signal fragments and auxiliary gene sequences, and combine them into an executable assembly process flow to generate a fragment assembly condition configuration table. Regulatory regions are nucleotide sequences that include gene expression regulation functions. Similar to regulatory fragments, they are used to control the temporality of transcription and translation processes and can be uniformly defined as regulatory sequences. Induction temperature control requirements are external condition parameters used to regulate the production of target proteins, involving temperature changes or control of chemical inducer concentrations, used to start or stop the protein synthesis process; Helper gene sequences are nucleotide fragments that encode chaperone proteins or folding-promoting factors, which enhance the correct folding and stability of target proteins in the host. S5: Based on the combination content in the fragment assembly condition configuration table, identify the functional arrangement and catalytic site of the substrate binding region in the conformational model, screen and classify according to structural stability and reaction region integrity, determine the recommended enzyme construction scheme, and generate the target enzyme construction sequence design set; Functional arrangement is the spatial distribution order of active residues around substrate channels within the catalytic region, used to describe the conformational characteristics of the enzyme catalytic center. The construction scheme is a technical implementation scheme formed by combining the target enzyme mutant sequence and the corresponding structural template, which is used to guide the vector assembly and recombinant product generation.

[0023] The conformation rate identification table includes structural model number, time node identifier, binding rate parameter, and alignment index key; the list of key sites for conformational changes includes residue number, three-dimensional coordinate difference grading threshold, functional participation range label, and regional attribute identifier; the sequence component assembly template list includes replacement amino acid combination scheme, codon optimization scheme, template number, and host reference codon library; the fragment assembly condition configuration table includes regulatory segment, induction temperature control parameter, transport signal fragment, helper gene sequence, assembly process flow number, and operating condition threshold; and the target enzyme construction sequence design set includes recommended construction scheme number, functional arrangement annotation, catalytic site annotation, structural stability grading, reaction region integrity level, and target sequence finished product set.

[0024] Please see Figure 2 The specific steps of S1 are as follows: S101: Collect information on the binding rate of enzyme and substrate at time nodes, extract reaction state data at corresponding nodes, identify number pairs that correspond to rate changes and spatial conformation changes, establish the correspondence between time nodes and structural conformation numbers, and obtain a set of conformation rate pairing numbers. To collect information on the binding rate of enzyme and substrate at specific time points, a fluorescently labeled enzyme and a fluorescence-quenched substrate must first be introduced into a reaction system with detection capabilities. Fluorescence signal change curves are recorded at set time intervals, for example, recording changes every 0.5 seconds. The recorded fluorescence intensity changes are converted into binding rates. Referring to the conversion factor calibrated by the microplate reader, the rate value at each time point is obtained. Derivative analysis of the rate change trend identifies mutation points; for example, if the rate decreases from 0.18 to 0.11 at 2 seconds, this point is the node to be labeled. Combined with the conformational data at that time point, the three-dimensional structure of the complex is obtained using cryo-electron microscopy or crystal diffraction. Parameters such as the spatial distance, angle, and orientation between key residues and the substrate are extracted to confirm the conformational number at that moment. Simultaneously, a corresponding label is established between time and structure; for example, 2 seconds is labeled as P03. This gradually forms a paired label set consisting of time points and conformational numbers, containing corresponding records of multiple conformational rate changes.

[0025] S102: Based on the structure number in the conformation rate pairing number set, extract the main chain and side chain parameters of the corresponding structure model, calculate the structural overlap rate between models, and summarize the degree of structural difference between all conformation models to obtain the conformation structure overlap rate matrix. Based on the conformation numbers in the conformation rate pairing number set, the corresponding three-dimensional structural files are extracted from the conformation data model. The spatial coordinate parameters of the main chain and side chains in the model are analyzed. The main chain is identified based on the coordinates of the α carbon atom, while the side chains are marked according to their respective functional groups. The spatial overlap of all structural models is compared pairwise. Different models are aligned by rotation and translation operations to obtain their structural differences. The similarity is judged by the degree of structural overlap. If the overall positional change of residues between a conformation pair is less than 1.2 Å, it is considered to have a high degree of overlap, which is converted into an overlap rate of approximately 66%. After comparing all conformation model combinations, a matrix containing the comparison data between all conformations is formed. The value in each cell represents the degree of structural overlap between the two models. For example, it is 66% between P02 and P03, and 48% between P03 and P05. The main chain part expresses the difference in the form of average residue displacement, while the side chain part is evaluated based on the degree of overlap of functional group volume. All results are summarized to form a conformational structure overlap rate matrix.

[0026] S103: Call the conformational structure coincidence rate matrix and the conformational rate pairing number set, filter the data point pairs whose rate change magnitude and structure coincidence rate decrease value both exceed the conformational stability limit value, establish the pairing relationship between the conformation number and the corresponding rate parameter, and obtain the conformational rate number lookup table. By combining the conformational structure overlap rate matrix with the conformational rate pairing number set, the rate value corresponding to each conformation pair is extracted, and the rate change magnitude is calculated. For example, the binding rates of conformations P02 and P03 are 0.18 and 0.11, respectively, with a rate change magnitude of 0.07; the structural overlap rate between them is 66%, which is about 29 percentage points lower than the average overlap rate between reference conformation groups (e.g., 95%). Based on conformational stability limits, the critical value for rate change is 0.05, and the critical value for the decrease in structural overlap rate is 25–30%. These thresholds are determined based on existing literature data, simulation statistics, or experimental standard deviations. When a conformation pair exceeds a set threshold in both rate and conformational overlap, it is identified as a key change combination and enters the subsequent screening process. For example, if P02 and P03 meet both conditions, the conformation pair is marked as a key change combination. A corresponding rate parameter table is established for each combination. For example, the pairing rates of P02 and P03 are recorded as 0.18 and 0.11, respectively, forming a conformation rate number comparison list containing multiple conformation number pairs and their respective rate parameters.

[0027] Please see Figure 3 The specific steps of S2 are as follows: S201: Extract the structural data associated with the conformation number based on the conformation rate number lookup table, obtain the list of residues corresponding to the substrate binding region, extract the spatial position parameters of all residues in the region in each conformation state, record the corresponding three-dimensional coordinate information, and generate a set of substrate region residue coordinates. Based on the conformation rate number lookup table, the structural data associated with each conformation number is extracted. First, each conformation number group (e.g., P02 and P03) must be read from the table. The corresponding structural models are then called to extract complete three-dimensional structural information. The substrate-binding region is located in the model, and the residue numbers involved in this region are confirmed using structural annotations or experimental data, such as residues 45, 67, 89, and 114. Subsequently, for each conformational state, the main chain and side chain coordinates of these residues are read. Each residue contains multiple atomic information; its three-dimensional coordinate parameters are recorded, such as the α-carbon and amino nitrogen atoms in the main chain, and the... For carboxyl oxygen atoms or amino nitrogen atoms, coordinate values ​​are extracted and categorized based on the arrangement of atoms in the structure. For example, all atomic positions of residue 45 in conformation P02 are numbered and organized to obtain a dataset containing multiple ternary coordinates. Then, the same operation is performed on the same residue in P03 to construct a dataset of the spatial positions of all residues in the substrate binding region under different conformations. This dataset is hierarchically organized according to conformation number, residue number, and atom category, with a unified format and field naming rules, which facilitates subsequent calling and comparison. After all data is integrated, a set of substrate region residue coordinates is generated.

[0028] S202: Based on the three-dimensional coordinates of the same residues in the substrate region residue coordinate set under different conformational states, calculate the spatial position difference between the corresponding coordinates of the residues, screen the residue numbers whose spatial position difference exceeds the residue displacement difference threshold, and obtain the position offset residue number table. Based on the spatial positions of residues with the same index in different conformations within the substrate region residue coordinate set, the coordinates of main chain and side chain atoms are calculated separately. By analyzing the three-dimensional positional changes of the same atom in the two conformations, the degree of positional shift is obtained. For example, the coordinates of the α carbon of the main chain of residue 45 are (5.2, 6.8, 3.5) in PO2 and (6.5, 7.1, 4.0) in PO3. This indicates that the atom has shifted in all three directions. Based on the changes of all atoms in the three-dimensional coordinates, the overall displacement of the residue is summarized. When the coordinate shifts of multiple atoms in a residue reach a significant level, it can be determined that the residue is in a different conformation. Significant spatial shifts occurred, and all residues were analyzed accordingly. A threshold for residue displacement differences was set, for example, by adding twice the standard deviation to the average shift across multiple datasets. If the average shift was 0.3 and the standard deviation was 0.25, the threshold was set to 0.8. When the shift of a residue exceeded this threshold, its number was recorded in the offset residue numbering table. For example, residues 45 and 89 showed spatial shifts of 1.4 and 1.2 respectively, both exceeding the set standard, and were therefore included in the offset residue numbering table. This table compiles all residues with significant positional changes, providing a basis for structural-functional analysis.

[0029] S203: Call the residue numbers selected from the position offset residue number table, extract the attribute labels of the local region where the number is located based on the position characteristics in the substrate binding region, identify the residue numbers of the reactive functional region, summarize the correspondence between residue numbers and regional attributes, and generate a list of key sites for conformational changes. The residue numbers in the position offset residue numbering table are retrieved one by one to find the spatial position of each residue in the substrate binding region structural model. By comparing their relative coordinates and regional distribution in the three-dimensional structure, the local functional regions where each residue is located are identified, such as catalytic active sites, hydrophobic embedding regions, or polar recognition bands. If residue number 45 is located in the core region of the catalytic triplet in the model, and residue number 89 is located at the edge of the hydrophobic recognition surface, then they are assigned catalytic and hydrophobic labels respectively. Then, the correspondence between these residue numbers and their assigned functional regions is summarized and organized in numerical order. For example, residue number 45 is labeled as catalytic, 89 as hydrophobic, and 114 as polar recognition region. All data are recorded using a unified coding method to construct a list of key sites for conformational changes. This list is based on position offset residues and combines their structural and functional attributes to form a complete record of important positional changes in the enzyme-substrate recognition process, providing a structural basis for further model optimization and experimental design.

[0030] Please see Figure 4 The specific steps of S3 are as follows: S301: Based on the residue numbers in the list of key sites for conformational changes, select multiple replaceable amino acids for the corresponding positions, design permutation and combination methods, generate data on the correspondence between combinations and residue numbers, and obtain a residue substitution combination data table. Based on the residue numbers in the list of key sites for conformational changes, amino acid properties are analyzed for each residue indicated by each number. Combining this with physicochemical characteristics such as hydrophobicity, charge state, and molecular volume from the database, multiple amino acid types suitable for substitution are selected. For example, glutamic acid (number 67) is negatively charged, and potential substitution types include aspartic acid (also acidic but smaller), glutamine (uncharged but polar), and alanine (nonpolar). Two to four substitution types are assigned to each site. All possible substitution combinations are then arranged. If there are three sites, corresponding to 3, 4, and 2 substitution types, the number of combinations generated is 24. The combination records the amino acid type at each site after the substitution. For example, in combination one, site 67 is replaced with alanine, site 89 with leucine, and site 114 with phenylalanine. In combination two, site 67 is replaced with glutamine, site 89 with threonine, and site 114 with tyrosine. Each combination is assigned a unique number and organized into a traceable data record. The data items include the combination number, the amino acid type before and after the substitution, the corresponding residue number, and the physicochemical properties of the substituted amino acid, such as isoelectric point, hydrophobic parameter, and relative molecular mass. All combinations, their residue numbers, and substitution information are systematically organized and output as a residue substitution combination data table.

[0031] S302: Call the combinations generated in the residue substitution combination data table, screen the codon combinations that can be used for expression regulation based on the codon arrangement in the host for yield enhancement, extract the codon sequences that meet the expression benchmark, and obtain the codon matching arrangement value set; The system retrieves information from the residue substitution combination data table, extracts the codons corresponding to the substituted amino acids one by one, and selects codons with higher expression frequencies for matching based on the codon usage preferences in the host system. Different host expression systems exhibit different codon usage frequencies for various amino acids. For example, glutamic acid often preferentially uses GAA, alanine tends to use GCG, and tyrosine often uses TAC. The expression benchmark is set as an average usage frequency of all codons in the combination not less than 50%, which is used to determine whether the expression regulation requirements are met. For the substitution scheme with combination number 1, if the three codons are GAA, GCG, and TAC, its average frequency in the target host system is approximately 53%, meeting the benchmark condition, and the combination is retained. If a combination uses low-frequency codons, resulting in an average frequency below 40%, it is considered an undesirable expression combination and is excluded. During screening, a weighted average method is used to evaluate codon frequencies. The set expression benchmark value is derived from the median distribution of codon usage rates for 20 amino acids in the host system. All codon combinations that meet the expression benchmark are output and recorded in the form of combination number, matching codon sequence, and corresponding expression frequency data, generating a codon matching permutation value set.

[0032] S303: Based on the codon sequence in the codon matching permutation value set, select the corresponding regulatory segment and transport signal sequence, integrate the segments to form template content that can be directly used for construction, and generate a sequence component assembly template list; Based on the codon sequences in the codon matching set, corresponding regulatory fragments and transport signal sequences are selected, and the fragments are integrated to form template content that can be directly used for construction, generating a list of sequence component assembly templates. Regulatory regions commonly used in eukaryotic expression systems are extracted using standard construction modules, including highly efficient promoter sequences adapted to host systems (such as yeast or mammalian cells), regulated enhancer elements, and the 5'UTR region required for eukaryotic expression. The selection of transport signals is based on the transmembrane structural characteristics of the target protein and the requirements for lysosomal subcellular localization, prioritizing eukaryotic signal peptide fragments suitable for endoplasmic reticulum guidance and membrane anchoring. During integration, it is confirmed that the sequence lengths and linking orders of each fragment do not overlap or conflict; for example, the length of the regulatory region is 35 bases, the optimized codon coding region is 96 bases, and the transport signal fragment is 48 bases, for a total length of 179 bases. Linking elements such as restriction enzyme sites, translation initiation sites, and non-coding regulatory regions are inserted between fragments according to construction specifications to ensure the continuity and controllability of transcription and translation. The overall template indicates the start and end positions of the control area, encoding area and signal area and the specific sequence composition. Each template has a unique identifier number and corresponds one-to-one with the codon combination. The template information includes the splicing order, functional segment annotations and combination number. After all template entries are sorted and output, a sequence component assembly template list is formed.

[0033] Please see Figure 5 The specific steps of S4 are as follows: S401: Based on the fragment combinations in the sequence component assembly template list, extract the corresponding regulatory segments, induction temperature control requirements, transport signal fragments and auxiliary gene sequences, establish data mapping relationships for the corresponding fragment types under the combinations, and generate a fragment composition structure mapping table; Based on the combinations listed in the sequence component assembly template list, each template number is sequentially decomposed to extract four types of fragment information: regulatory region, induction temperature control requirements, transport signal fragment, and helper gene sequence. The regulatory region is located at the beginning of the sequence and is mostly within 35 bases in length. The induction temperature control requirements field includes the temperature setpoint, textual description of the initiation conditions, and numerical response time. The transport signal fragment is identified according to standard labels such as SEC or TAT. The helper gene sequence is labeled with names such as GFP and KanR. During the extraction process, the template number is used as an index to identify the four types of fragment information and establish a mapping structure. The script reads the position of each segment in the template entry and extracts its corresponding content to form four mapping records with the template number as the key and the regulatory region sequence, temperature control field, signal fragment label, and helper gene label as the value. Each item in the mapping record has a unique pointer to ensure that the mapping relationship between different template combinations does not overlap. The recording method adopts a plain text structure, and the content is organized as a combination of number, fragment type, fragment label, and sequence segment. All combination data are uniformly organized, and the output is a clearly structured and well-classified fragment composition structure mapping table.

[0034] S402: Based on the type characteristics of fragment combination in the fragment composition structure mapping table, the temperature value, start-up condition and response time parameters in the induced temperature control requirements are collected. Combined with the start position sequence information of the corresponding control section, the required temperature control matching coefficient is calculated to obtain the set of temperature control configuration values. The combination number is retrieved one by one from the fragment composition structure mapping table to extract the corresponding induction temperature control field, including the temperature setpoint, initiation condition description, and response time value. The temperature value is between 30°C and 42°C, the initiation condition is an inducer such as lactose or IPTG, and the response time ranges from 1 hour to 4 hours. This is then combined with the initiation base sequence at the front end of the regulatory segment, for example, the initiation sequence is TAATACGACTCACTATAGGG, with a length of 20 bases. This sequence is compared with the promoter standard template to confirm the number of base matches. The degree of matching is expressed as a proportion; for example, if 14 sites are completely identical, the alignment ratio is 70%, and this ratio is used as a reference for the sequence response capability. The numerical values ​​are further combined with temperature setpoints and response time information. The degree of matching between temperature control conditions and the activity of the regulated segment is evaluated by textual comparison. If the temperature is within ±5°C of the recommended value of the standard expression system, the response time does not exceed 2 hours, and the initiation condition is a highly efficient inducer, then the combination can be judged as moderate to high matching. This judgment standard is set as a reference range. For example, the degree of matching is divided into lower than 50% as low, 50% to 75% as medium, and above 75% as high. Based on this, the temperature control response level of each combination is summarized and organized into combination attribute items including combination number, temperature setting, inducer type, response time and sequence matching ratio. The output is a set of temperature control configuration values.

[0035] S403: Based on the temperature control coefficients combined in the temperature control configuration value set, call the corresponding regulatory segments, signal segments and auxiliary gene sequences in the fragment composition structure mapping table, combine them into executable process parameter items according to the type classification rules, and establish a fragment assembly condition configuration table; Based on the temperature control coefficients of each combination in the temperature control configuration set, the corresponding regulatory segments, signal fragments, and auxiliary gene sequences in the fragment composition structure mapping table are retrieved and integrated into executable process parameter items according to the classification rules of eukaryotic expression systems. The classification criteria are based on the type of regulatory element (e.g., strong promoters, regulated enhancers), the category of eukaryotic signal peptides (e.g., endoplasmic reticulum importation signals, multi-transmembrane localization signals), and the functional attributes of auxiliary genes (e.g., stable expression tags or folding-promoting factors). The temperature control response is based on the host-suitable induction conditions, such as the response efficiency of chemically inducible elements, time control modules, or expression enhancers. When the matching ratio is not less than 50% and the response time is within the set range, it is considered to meet the expression configuration requirements and included in the combination range. For each matching combination, its corresponding regulatory element tag, transport or localization signal type, auxiliary gene number, and temperature control parameters are extracted and uniformly archived as fragment process combination items. The combination item information includes template number, functional fragment type, sequence identifier, and temperature control response evaluation results. Each combination item is arranged sequentially to form a complete fragment assembly condition configuration table.

[0036] Please see Figure 6 The specific steps of S5 are as follows: S501: Based on the combination content in the fragment assembly condition configuration table, identify the substrate binding region in the conformational model, extract structural spatial information, and label the hydrophobic sites, polar residues and charge distribution characteristics in the functional arrangement to generate a substrate binding arrangement parameter set. Based on the functional sequence content of the combinations listed in the fragment assembly condition configuration table, the amino acid sequence information of the combinations is loaded one by one in the 3D conformational modeling platform to construct the spatial structure model of the protein. The loading method can be achieved by modeling the translated target sequence using a structure prediction tool to obtain the protein folding state. After modeling, the generated spatial conformation is subjected to structure energy minimization processing to identify pocket structural regions with typical substrate binding capabilities in the model. By calculating the volume, entrance width, and residue configuration features of the surface depressions, binding regions are identified, and the structural information inside these regions is extracted, including the spatial coordinates and physicochemical properties of amino acid residues. Hydrophobic residues clustered in the pocket regions are identified using a hydrophobicity scoring method, and records are made such as Val, Leu, Ile, etc. The distribution locations of residues such as Phe are identified, and the distribution density and spatial spacing of polar residues such as Ser, Thr, Asn, and Gln, as well as charged residues such as Lys, Arg, Asp, and Glu, are also marked. The hydrophobicity, polarity, and charge distribution within the region are labeled, and a set of spatial attributes of the local environment is constructed in the three-dimensional model. This set can be composed of multiple parameters, such as a binding region volume of 550 ų, a hydrophobic residue ratio of 60%, a number of polar residues of 8, and a positive to negative charge residue ratio of 1:1. These parameters are used as the core content of the substrate binding arrangement parameter set. Each fragment combination is assigned a unique number and structural identifier, and all structural spatial information is recorded for subsequent matching analysis, generating a substrate binding arrangement parameter set containing functional characteristics.

[0037] S502: Call the structural information and residue distribution data in the substrate binding arrangement parameter set, calculate the structural stability value and reaction region integrity value of the conformation, screen fragment combinations that meet the screening criteria, and obtain the conformation screening matching coefficient; The structural parameters recorded in the substrate binding alignment parameter set are read, and a dual evaluation of structural stability and reaction region integrity is performed for each conformation. Structural stability is assessed based on the number of hydrogen bonds within the binding region, the density of hydrophobic cores, and the contact status of non-bonded interactions. In the model analysis, the total number of hydrogen bonds, the hydrophobic aggregation degree of residues around the binding pocket, and the number of spatial contact points between residues are statistically analyzed. For example, a structure might have 140 hydrogen bonds, a hydrophobic core density of 0.85, and 500 contact points. These three data points are assigned weights and combined to form the structural stability value. Reaction region integrity is evaluated based on the degree of surface closure and the coverage of key residues in the binding region, such as the closure area of ​​the binding region. The value was 380 Ų, and the retention rate of key catalytic residues was 92%. All data were weighted and summarized into a reaction integrity value. The structural stability value and the reaction integrity value were used as a dual-index evaluation system to score all combinations. The screening benchmark was set to both structural stability and reaction integrity values ​​not lower than 200 points. If a combination scored 235 points, it would be included in the list of qualified combinations. All combinations completed this process and were summarized into a set of conformational screening matching coefficients. The combination number and the scoring result were recorded. In the scoring range, less than 180 was defined as the lower range, 180 to 200 as the middle range, and more than 200 as the higher range. Only combinations in the higher range were retained for subsequent sequence reconstruction.

[0038] S503: Based on the conformational screening matching coefficients, extract the corresponding fragment sequences, reconstruct the nucleic acid coding sequences in functional order, and establish the target enzyme construction sequence design set; Based on the combination numbers identified as meeting the criteria in the conformational screening matching coefficient set, the corresponding fragment sequences in the fragment assembly condition configuration table are extracted retrospectively. These fragments include regulatory region sequences, signal fragment coding regions, and auxiliary gene functional segments. Each sequence is reassembled into a continuous nucleic acid coding structure according to its functional order. The reordering process must follow a standard sequence template, for example, starting with the promoter sequence, followed by the transport signal coding segment and the enzyme core coding segment, and finally connecting the auxiliary gene coding segment. For example, the sequence could be T7 promoter sequence, TAT signal peptide sequence, target enzyme coding region, and GFP tag gene. Before assembly, each sequence... Codon optimization is performed by selecting a high-frequency codon table adapted to the target expression host and replacing low-frequency codons in the original coding to improve expression efficiency. For example, the low-frequency CTA codon is replaced with TTG or CTG. The integrity of all spliced ​​sequences is checked to confirm whether the open reading frames are continuous and whether there are any premature stop codons. After all nucleic acid sequences pass the preliminary verification, they are uniformly classified according to the combination number to form the target enzyme construction sequence design set. Each design includes the reconstructed sequence, the order of functional modules, and the suggested expression system type. After the design set is formed, it can be used for the subsequent actual construction and synthesis process.

[0039] Please see Figure 7 An artificial intelligence design and expression system for heparan sulfate-α-glucosidase N-acetyltransferase, including: The structure rate acquisition module collects binding rate information during the enzyme-substrate reaction process, extracts the structure model number generated at the reaction time point, associates the rate parameters with the corresponding model number, and generates a conformation rate number lookup table. The conformation comparison and screening module extracts the spatial position of residues in the substrate binding region based on the model number listed in the conformation rate numbering table, compares the differences in the three-dimensional coordinates of the same residues in the conformation, screens residue numbers whose position offsets meet the standards, and generates a list of key sites for conformational changes by combining the structural region type identifier. The sequence component assembly module selects alternative arrangements of corresponding amino acids based on the residue numbers in the list of key sites for conformational changes, combines them with known codon sequences in the host that can enhance expression levels, matches regulatory fragments and transport signal sequences, forms a set of fragments that can be used for sequence synthesis, and generates a list of templates for sequence component assembly. The process configuration module summarizes the required control segment information, expression induction conditions, transport signal sequences and auxiliary segment numbers based on the fragment combination content in the sequence component assembly template list, and combines them to form an execution list for expression construction, generating a fragment assembly condition configuration table. The target sequence construction module identifies the functional arrangement of substrate binding regions and catalytic sites in the conformational model based on the configuration content in the fragment assembly condition configuration table, extracts information related to structural stability and reaction region layout, performs combination and classification analysis, screens construction schemes, and generates a target enzyme construction sequence design set.

[0040] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for constructing an artificial intelligence design and expression system for heparan sulfate-α-glucosidase N-acetyltransferase, characterized in that, Includes the following steps: S1: Collect information on the binding rate of enzyme and substrate reaction process, associate time node structural model number with rate parameter, construct the basis for structural comparison and screening index, and generate a conformation rate number lookup table. S2: Extract the associated structures according to the conformation rate number lookup table, compare the three-dimensional coordinate difference of the same residue under the conformation in the substrate binding region, screen the residue numbers that meet the set threshold, combine the participation range with the regional attribute labeling function, and generate a list of key sites for conformational changes. S3: Based on the residue numbers of the list of key sites for conformational changes, design combinations of replacement amino acids, select regulatory fragments and transport signal sequences with reference to the codon arrangement in the host, and generate a list of sequence component assembly templates; S4: Based on the fragment combination of the assembly template list of sequenced components, summarize the regulatory segments, temperature control requirements, transport signals and auxiliary gene sequences, combine them into an executable process flow, and generate a fragment assembly condition configuration table. S5: Based on the fragment assembly condition configuration table, identify the functional arrangement of the substrate binding region and the catalytic site, and combine structural stability and regional integrity to generate a target enzyme construction sequence design set.

2. The method for constructing an artificial intelligence design and expression system for heparan sulfate-α-glucosidase N-acetyltransferase according to claim 1, characterized in that, The conformation rate numbering table includes structural model number, time node identifier, binding rate parameter, and alignment index key. The list of key sites for conformational changes includes residue number, three-dimensional coordinate difference grading threshold, functional participation range label, and regional attribute identifier. The sequence component assembly template list includes replacement amino acid combination scheme, codon optimization scheme, template number, and host reference codon library. The fragment assembly condition configuration table includes regulatory segment, induction temperature control parameter, transport signal fragment, helper gene sequence, assembly process flow number, and operating condition threshold. The target enzyme construction sequence design set includes recommended construction scheme number, functional arrangement annotation, catalytic site annotation, structural stability grading, reaction region integrity level, and target sequence finished product set.

3. The method for constructing an artificial intelligence design and expression system for heparan sulfate-α-glucosidase N-acetyltransferase according to claim 1, characterized in that, The specific steps of S1 are as follows: S101: Collect information on the binding rate of enzyme and substrate at time nodes, extract reaction state data at corresponding nodes, identify number pairs that correspond to rate changes and spatial conformation changes, establish the correspondence between time nodes and structural conformation numbers, and obtain a set of conformation rate pairing numbers. S102: Based on the structure number in the conformation rate pairing number set, extract the main chain and side chain parameters of the corresponding structure model, calculate the structural overlap rate between models, and summarize the degree of structural difference between all conformation models to obtain the conformation structure overlap rate matrix. S103: Call the conformational structure coincidence rate matrix and the conformational rate pairing number set, filter the data point pairs whose rate change magnitude and structure coincidence rate decrease value both exceed the conformational stability limit value, establish the pairing relationship between the conformation number and the corresponding rate parameter, and obtain the conformational rate number lookup table.

4. The method for constructing an artificial intelligence design and expression system for heparan sulfate-α-glucosidase N-acetyltransferase according to claim 3, characterized in that, The specific steps of S2 are as follows: S201: Extract the structural data associated with the conformation number according to the conformation rate number lookup table, obtain the list of residues corresponding to the substrate binding region, extract the spatial position parameters of all residues in the region in each conformation state, record the corresponding three-dimensional coordinate information, and generate a set of substrate region residue coordinates. S202: Based on the three-dimensional coordinates of the same residues in the substrate region residue coordinate set under different conformational states, calculate the spatial position difference between corresponding coordinates of residues, filter residue numbers whose spatial position difference exceeds the residue displacement difference threshold, and obtain a position offset residue number table. S203: Call the residue numbers selected from the position offset residue number table, extract the attribute labels of the local region where the number is located based on the position characteristics in the substrate binding region, identify the residue numbers of the reactive functional region, summarize the correspondence between residue numbers and regional attributes, and generate a list of key sites for conformational changes.

5. The method for constructing an artificial intelligence design and expression system for heparan sulfate-α-glucosidase N-acetyltransferase according to claim 4, characterized in that, The specific steps for S3 are as follows: S301: Based on the residue numbers in the list of key sites for conformational changes, select multiple replaceable amino acids for the corresponding positions, design permutation and combination methods, generate data corresponding to the combinations and residue numbers, and obtain a residue substitution combination data table. S302: Call the combination generated in the residue replacement combination data table, screen the codon combinations that can be used for expression regulation according to the codon arrangement in the host for yield enhancement, extract the codon sequence that meets the expression benchmark, and obtain the codon matching arrangement value set; S303: Based on the codon sequences in the codon matching permutation value set, select the corresponding control fragments and transport signal sequences, integrate the fragments to form template content that can be directly used for construction, and generate a sequence component assembly template list.

6. The method for constructing an artificial intelligence design and expression system for heparan sulfate-α-glucosidase N-acetyltransferase according to claim 5, characterized in that, The specific steps of S4 are as follows: S401: Based on the fragment combinations in the sequence component assembly template list, extract the corresponding regulatory segments, induction temperature control requirements, transport signal fragments and auxiliary gene sequences, establish data mapping relationships for the corresponding fragment types under the combination, and generate a fragment composition structure mapping table. S402: Based on the type characteristics of the segment combination in the segment composition structure mapping table, the temperature value, start-up condition and response time parameters in the induced temperature control requirements are collected, and combined with the start position sequence information of the corresponding control section, the required temperature control matching coefficient is calculated to obtain the temperature control configuration value set. S403: Based on the temperature control coefficients combined in the set of temperature control configuration values, call the corresponding regulatory segments, signal segments and auxiliary gene sequences in the fragment composition structure mapping table, combine them into executable process parameter items according to the type classification rules, and establish a fragment assembly condition configuration table.

7. The method for constructing an artificial intelligence design and expression system for heparan sulfate-α-glucosidase N-acetyltransferase according to claim 6, characterized in that, The specific steps of S5 are as follows: S501: Based on the combination content in the fragment assembly condition configuration table, identify the substrate binding region in the conformational model, extract the structural spatial information, and label the hydrophobic sites, polar residues and charge distribution characteristics in the functional arrangement to generate a substrate binding arrangement parameter set. S502: Call the structural information and residue distribution data in the substrate combination arrangement parameter set, calculate the structural stability value and reaction region integrity value of the conformation, screen fragment combinations that meet the screening criteria, and obtain the conformation screening matching coefficient; S503: Based on the conformation screening matching coefficients, extract the corresponding fragment sequences, reconstruct the nucleic acid coding sequences in functional order, and establish a target enzyme construction sequence design set.

8. The method for constructing an artificial intelligence design and expression system for heparan sulfate-α-glucosidase N-acetyltransferase according to claim 1, characterized in that, The structural model number is a three-dimensional protein model identification code generated by a structural prediction tool; The rate parameter is the binding rate constant or dissociation rate constant measured by a surface plasmon resonance device during the enzyme-substrate binding reaction. The three-dimensional coordinate difference is the difference in spatial position of the same residue in different conformational states, expressed as the amount of coordinate change of the main chain atom or the center of the side chain.

9. The method for constructing an artificial intelligence design and expression system for heparan sulfate-α-glucosidase N-acetyltransferase according to claim 1, characterized in that, The host is a biological carrier used to produce the target enzyme, which is a mammalian cell expression system; The regulatory fragment is a nucleotide sequence that controls the direction and rate of sequence synthesis, and can be a promoter or terminator; The transport signal sequence is a target short peptide sequence located at the amino terminus; The regulatory region is a nucleotide sequence that includes gene expression regulation functions, and its function is similar to that of the regulatory fragment. The accessory gene sequence is a nucleotide fragment encoding a chaperone protein or a folding-promoting factor; The functional arrangement is the spatial distribution order of active residues around the substrate channel within the catalytic region.

10. An artificial intelligence design and expression system for heparan sulfate-α-glucosidase N-acetyltransferase, characterized in that, The system is used to implement the artificial intelligence design and expression system construction method for heparan sulfate-α-glucosidase N-acetyltransferase according to any one of claims 1-9, the system comprising: The structure rate acquisition module collects binding rate information during the enzyme-substrate reaction process, extracts the structure model number generated at the reaction time point, associates the rate parameters with the corresponding model number, and generates a conformation rate number lookup table. The conformation comparison and screening module extracts the spatial position of residues in the substrate binding region according to the model number listed in the conformation rate number lookup table, compares the differences in the three-dimensional coordinates of the same residues in the conformation, screens residue numbers whose position offset meets the standard, and generates a list of key sites for conformational changes by combining the structural region type identifier. The sequence component assembly module selects the alternative arrangements of corresponding amino acids based on the residue numbers in the list of key sites for conformational changes, combines them with known codon sequences in the host that can enhance expression levels, matches regulatory fragments and transport signal sequences, forms a set of fragments that can be used for sequence synthesis, and generates a list of templates for sequence component assembly. The process configuration module summarizes the required control segment information, expression induction conditions, transport signal sequences and auxiliary segment numbers according to the fragment combination content in the sequence component assembly template list, and combines them to form an execution list for expression construction, and generates a fragment assembly condition configuration table. The target sequence construction module identifies the functional arrangement of substrate binding regions and catalytic sites in the conformational model according to the configuration content in the fragment assembly condition configuration table, extracts information related to structural stability and reaction region layout, performs combination and classification analysis, screens construction schemes, and generates a target enzyme construction sequence design set.

Images