A method for high-throughput screening of caudipteryx lipid for antidepressant activity evaluation

Abstract

The present application relates to the technical field of biological by-product quality evaluation, and discloses a high-throughput screening method for anti-depression activity evaluation of giant salamander lipid, which solves the problems of difficult identification of active lipid and lack of quality markers by constructing a technical system of "lipidomics detection-anti-depression activity evaluation-feature correlation analysis-key lipid screening-quality marker construction". The system comprises a lipid extraction module, a lipidomics detection platform (ultra-high performance liquid chromatography-high resolution mass spectrometry is used to detect triglycerides, phospholipids, sphingolipids and free fatty acids), an activity evaluation module (combined with in vitro cells and in vivo animal models), a correlation analysis server (Pearson, PCA, PLS-DA and random forest, SVM, neural network algorithms are used for modeling), and a quality marker database. The present application screens key active lipids through a multi-index weighted comprehensive scoring algorithm, and provides a scientific basis for quality control of giant salamander lipid products.

Description

Technical Field

[0001] This invention relates to the field of biological by-product quality assessment technology, and more specifically, to a high-throughput screening method for lipids from giant salamanders used to evaluate antidepressant activity. Background Technology

[0002] In recent years, with the rapid development of bioactive lipid research, functional food development, and precise screening technology for natural products, the development of health products with neuromodulatory functions using active ingredients derived from aquatic organisms has gradually become a research hotspot. Especially in the field of antidepressant active substance development, research directions have gradually shifted from traditional single-component extraction and identification to a combination of lipidomics, high-throughput screening, and mechanism of action analysis. As a biological resource rich in various lipid active ingredients, the lipid components of the giant salamander show potential application value in neurotransmitter regulation, inflammation suppression, and oxidative stress regulation. Therefore, screening and evaluating the antidepressant activity of giant salamander lipids has become an important research direction in related fields.

[0003] Currently, research on bioactive lipids from natural sources primarily employs a combined approach of lipid extraction, chromatographic separation, mass spectrometry, and cell or animal experiments for validation. Qualitative and quantitative analyses of the lipid components in the extracts are performed, and their antidepressant activity is evaluated by incorporating behavioral indicators, biochemical indicators, and changes in neurotransmitter levels. Some studies have also introduced lipidomics techniques for systematic analysis of the lipid composition in samples to improve the efficiency of active ingredient discovery. However, due to the complex composition of naturally derived lipid systems and the significant structural and functional differences among different lipid molecules, the accurate identification and evaluation of bioactive lipids remains a significant challenge.

[0004] However, existing technologies mostly focus on the qualitative and quantitative analysis of lipid components and the overall activity evaluation, lacking research on the precise identification of key active lipids and the construction of quality biomarkers. In actual research, sample quality is often evaluated based solely on total lipid content, fatty acid composition, or some physicochemical indicators, making it difficult to determine the core lipid components that truly exert antidepressant effects. Therefore, even if different batches of raw materials or extracts have similar conventional detection indicators, their actual antidepressant activities may still vary significantly, resulting in a lack of stable correlation between product quality evaluation results and functional activity. Furthermore, because existing technologies have not established a correlation analysis mechanism between the structural characteristics and content changes of active lipids and their antidepressant biological effects, and also lack a quality biomarker system that can reflect functional activity, it is impossible to accurately track changes in key active lipids during quality control. When the source of raw materials, processing conditions, or storage environment changes, some low-abundance lipids with antidepressant effects may degrade, transform, or experience content fluctuations, but conventional quality detection methods are unable to detect such changes in a timely manner, leading to decreased product quality stability, insufficient repeatability of activity evaluation results, and difficulties in unifying quality standards during industrial production.

[0005] Therefore, there is an urgent need for a high-throughput screening method for salamander lipids to evaluate their antidepressant activity, so as to achieve accurate identification, activity evaluation and quality biomarker construction of key active lipids. Summary of the Invention

[0006] This application provides a high-throughput screening method for salamander lipids for evaluating antidepressant activity. By constructing a technical system of "lipomics detection - antidepressant activity evaluation - feature association analysis - key lipid screening - quality marker construction", the method can achieve efficient discovery and accurate evaluation of antidepressant active components in salamander lipids.

[0007] The present application is described below from different aspects. It should be understood that the different implementation methods and beneficial effects described below can be referenced from each other.

[0008] In a first aspect, embodiments of this application provide a high-throughput screening method for salamander lipids for evaluating antidepressant activity. The system includes: a lipid extraction module, a lipidomics detection platform, an activity evaluation module, a correlation analysis server, and a quality biomarker database. The lipid extraction module is used for standardized pretreatment of salamander raw materials and lipid extraction, including organic solvent extraction, supercritical fluid extraction, or enzyme-assisted extraction. The lipidomics detection platform is used to comprehensively detect triglycerides, phospholipids, sphingolipids, free fatty acids and their derivatives in the lipid extract, generating lipid molecular composition information, relative abundance information, and lipid characteristic spectrum data. The activity evaluation module is used to apply the lipid extract to nerve cell models, neuroinflammation models, or stress injury models to obtain activity evaluation data such as cell survival rate, neurotransmitter secretion level, inflammatory factor expression level, and oxidative stress indicators, or to obtain antidepressant phenotypic data such as motor behavior, exploratory behavior, and sucrose preference behavior using animal models such as zebrafish behavioral models and mouse chronic unpredictable stress models. The correlation analysis server is used to couple and analyze lipidomics data and activity evaluation data to construct a correlation model between lipid characteristics and antidepressant activity. The quality marker database is used to store information on key active lipids obtained through screening and their evaluation parameters.

[0009] For example, the lipid extracts described herein may include at least one of the following: triglyceride lipids, phospholipid lipids, sphingolipid lipids, free fatty acid lipids, fat-soluble vitamins and their derivatives. The lipidomics detection here employs ultra-high performance liquid chromatography-high resolution mass spectrometry (UHPLC-MS / MS), with detection modes including positive ion mode and negative ion mode.

[0010] In the above scheme, the lipid extraction module communicates with the lipidomics detection platform, the lipidomics detection platform communicates with the correlation analysis server, and the correlation analysis server communicates with the quality biomarker database. After the lipid extraction module completes lipid extraction, the lipidomics detection platform detects and analyzes the full spectrum of lipids in the extract to obtain the mass spectrometry characteristics of lipid molecules. The correlation analysis server performs coupled analysis of lipidomics data and activity evaluation data to identify candidate lipid molecules that are significantly related to activity changes. The system generates key lipid screening results based on the determined correlation characteristics, replacing the traditional single-indicator evaluation screening mode, and outputs a multi-dimensional antidepressant activity correlation analysis report.

[0011] In one possible implementation, the lipidomics detection platform is used to establish a full-spectrum lipid detection method using ultra-high performance liquid chromatography-high resolution mass spectrometry.

[0012] For example, the chromatographic conditions used were an ACQUITY UPLC HSS T3 column (2.1 × 100 mm, 1.8 μm), a column temperature of 45 °C, and mobile phase A of acetonitrile / water (60:40). It contains 0.1% formic acid and 10 mM ammonium formate, and the mobile phase B is acetonitrile / isopropanol (10:90). The sample contained 0.1% formic acid and 10 mM ammonium formate. The flow rate was set to 0.35 mL / min. The gradient elution program was as follows: 0-2 min to maintain 5% B, 2-15 min to linearly increase from 5% B to 80% B, 15-20 min to maintain 80% B, 20-22 min to decrease from 80% B to 5% B, and 22-25 min to maintain 5% B. Mass spectrometry conditions included an electrospray ionization source with both positive and negative ion modes. The capillary voltage was set to 3.0 kV (positive ion mode) or 2.8 kV (negative ion mode), the cone voltage was set to 30 V, and the collision energy was set to 15-40 eV (dynamically adjusted according to the lipid molecule type). Data-dependent acquisition was used with an acquisition frequency of 10 Hz and a mass scan range of [missing information]. 200-1800.

[0013] In the aforementioned implementation, the lipidomics detection platform achieves highly sensitive detection of the full spectrum of lipids in giant salamander lipid extracts through optimized chromatographic conditions and mass spectrometry parameters. Combined with positive and negative ion mode switching, it covers lipid molecule types with different ionization characteristics, ensuring the comprehensiveness and accuracy of lipidomics data.

[0014] In one possible implementation, the activity evaluation module is used to establish a multi-level antidepressant activity evaluation system by combining in vitro cell models with in vivo animal models.

[0015] For example, the in vitro cell models include a PC12 cell injury model, a SH-SY5Y cell neuroinflammation model, and a microglia activation model; the drug administration concentrations are set at five gradients: 12.5 μg / mL, 25 μg / mL, 50 μg / mL, 100 μg / mL, and 200 μg / mL, with six replicates for each concentration; the detection indicators include cell viability (MTT assay), cell viability (LDH release), neurotransmitter content (5-HT, DA, NE levels, ELISA assay), inflammatory factor levels (IL-1β, TNF-α, ELISA assay), and oxidative stress indicators (SOD activity, MDA content, WST-8 assay). In vivo animal models included a zebrafish depression model (induced by chronic unpredictable stress) and a mouse chronic unpredictable stress depression model (CUMS model). The zebrafish experiment included a blank control group, a model group, a positive control group (fluoxetine, 10 mg / L), and test drug groups (25 mg / L, 50 mg / L, 100 mg / L), with 30 zebrafish in each group. Behavioral indicators included open field tests (center dwell time, number of crossings), suspended log tests (resting time, swimming distance), and sucrose preference tests (sucrose intake rate). The mouse CUMS model experiment included a normal control group, a model group, a positive control group (fluoxetine, 10 mg / kg / d), and test drug groups (25 mg / kg / d, 50 mg / kg / d, 100 mg / kg / d), with 12 mice in each group. Behavioral indicators included sucrose preference tests (sucrose intake rate), suspended tail tests (cumulative immobility time), forced swimming tests (cumulative immobility time), and open field tests (center dwell time, total movement distance).

[0016] In the above implementation, the activity evaluation module achieves a comprehensive assessment of the antidepressant activity of salamander lipids through a multi-level evaluation system combining in vitro cell models and in vivo animal models. In vitro models are used for high-throughput initial screening, while in vivo models are used for confirmatory verification; the combination of both improves the accuracy and reliability of the activity evaluation.

[0017] In one possible implementation, the correlation analysis server is used to introduce machine learning algorithms and multivariate statistical analysis methods to construct a correlation analysis model between lipid characteristics and activity indicators.

[0018] For example, statistical analysis methods include Pearson correlation coefficient analysis (threshold ≥0.6 or ≤-0.6 for initial screening), principal component analysis (PCA, retaining ≥3 principal components, cumulative explained variance ≥70%), and partial least squares discriminant analysis (PLS-DA, VIP value ≥1.0 for screening). Machine learning algorithms include the Random Forest algorithm (the number of decision trees is set to 500, and the number of random samples of feature variables is set to...). The minimum number of leaf node samples is set to 5), and the Support Vector Machine (SVM) algorithm uses the RBF kernel, with a penalty parameter of 5. Set to 1.0-100.0 for optimization via grid search. The parameters are set to 0.01-1.0 (optimized via grid search) and the neural network algorithm is set to (BP neural network, network structure is set to input layer-hidden layer-output layer, number of neurons in hidden layer is set to 8-32, activation function is ReLU, activation function in output layer is Sigmoid, learning rate is set to 0.001, number of iterations is set to 500-1000).

[0019] In the above implementation, the association analysis server, through its multi-algorithm fusion association analysis capabilities, overcomes the limitations of traditional single statistical methods in analyzing complex lipid-activity relationships, achieving accurate identification of key active lipids. Combining feature importance assessment using random forests, classification prediction using SVMs, and nonlinear fitting using neural networks improves the prediction accuracy of the association model.

[0020] In one possible implementation, the association analysis server is also used to construct a key active lipid screening algorithm, which is based on a comprehensive calculation of contribution weight, difference fold, stability index and repeated validation results.

[0021] For example, the screening algorithm uses a multi-index weighted comprehensive scoring algorithm. First, the contribution weights of candidate lipids are calculated using each algorithm (random forest feature importance score normalization, SVM weight vector absolute value normalization, and Garson importance decomposition of neural network weights). Then, the difference fold (between the model group and the control group) is calculated. The stability index was calculated (the absolute value after logarithmic transformation) by taking the standard error of each index based on 1000 Bootstrap resampling cycles; a standard error / mean of ≤0.2 was considered stable. The replication validation results were also calculated (the validation rate of independent batch experiments; a validation rate ≥70% was considered passing). Then, after standardizing each index, a comprehensive score was calculated by weighting the following: contribution weight 0.4, difference fold weight 0.25, stability index weight 0.2, and replication validation result weight 0.15. Finally, a list of key active lipids was output, sorted from highest to lowest comprehensive score, and lipid molecules with a comprehensive score ≥0.7 were included in the quality biomarker library.

[0022] In the above implementation method, the association analysis server uses a scientific multi-index screening mechanism to ensure that key lipids with high correlation to antidepressant activity are selected first, and avoids low-correlation lipids being mistakenly selected as quality markers.

[0023] In one possible implementation, the correlation analysis server is also used to synchronously build a quality biomarker library management module, which automatically matches the corresponding lipid structure information, physicochemical property parameters and activity evaluation results when screening key active lipids, forming a complete quality biomarker profile.

[0024] For example, a quality marker profile may include at least one of the following: lipid molecule name (generic and systematic name), lipid molecule structure, mass spectrometry characteristic ions (qualitative and quantitative ions), retention time, concentration range, activity contribution score, and batch-to-batch coefficient of variation.

[0025] In the above implementation method, the quality marker library management module systematically stores the key active lipid information obtained through screening through a preset file structure, providing a data foundation for subsequent quality control.

[0026] In one possible implementation, the activity evaluation module is also used to construct an activity prediction model based on the quality markers obtained through screening, so as to achieve rapid prediction and quantitative evaluation of the antidepressant activity of the sample.

[0027] For example, the activity prediction model employs a multiple linear regression model or a machine learning regression model. The input parameters are the type, abundance, and proportion of the quality markers, and the output parameter is the predicted antidepressant activity score (calculated with the activity score of the positive control group as 100% for relative activity). Model validation uses leave-one-out cross-validation or 5-fold cross-validation. Model performance evaluation metrics include the correlation coefficient. (Target ≥ 0.85), Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE) (Target ≤ 15%).

[0028] In the above implementation, the activity evaluation module uses an activity prediction model to quickly predict the antidepressant activity of unknown samples, which can estimate the sample activity without conducting a complete activity evaluation experiment, thus improving screening efficiency.

[0029] In one possible implementation, the system is also used to establish a quality control system to dynamically monitor key active lipids in raw materials, extraction processes, intermediate products, and end products, thereby achieving a unified characterization between antidepressant activity and quality indicators.

[0030] For example, quality control indicators include at least one of the following: completeness of key active lipid species (target ≥90% detection of quality markers), content of key active lipids (determined according to the concentration range in the quality marker library), ratio of key active lipids (ratio deviation between quality markers ≤ ±20%), and batch-to-batch coefficient of variation (CV ≤ 25%) of key active lipids. The detection method uses the same ultra-high performance liquid chromatography-mass spectrometry conditions as the lipidomics detection platform for quantitative detection; quality judgment uses a two-level (pass / fail) or graded judgment.

[0031] In the above implementation method, the system establishes a quality control system to achieve comprehensive monitoring of the quality of giant salamander lipid products, ensure the consistency and stability of antidepressant activity between different batches of products, and improve the standardization control capability in the industrial production process.

[0032] In one possible implementation, the system is also used to establish a verification feedback channel based on a dynamic update mechanism of the quality marker library, in collaboration with the R&D team, to collect activity evaluation data and quality detection data from independent verification experiments, and to verify the screening results of key active lipids.

[0033] For example, the verification feedback channel includes an experimental recording system and a data analysis platform, where researchers can evaluate and score the activity contribution of key active lipids and submit supplementary verification experimental data.

[0034] In the above implementation method, the system establishes a verification feedback channel to achieve closed-loop verification between the screening results of quality markers and independent verification experiments, providing data support for subsequent model optimization.

[0035] In one possible implementation, the system is also used to establish a parameter optimization library based on feedback information, and to iteratively adjust the algorithm parameters, screening thresholds and prediction model coefficients of the correlation analysis server, forming a closed-loop mechanism of "model building - verification feedback - parameter optimization - model iteration".

[0036] For example, the parameter optimization library records model performance metrics under different parameter configurations, including key lipid screening accuracy, activity prediction correlation coefficient, and quality control compliance rate; through regular model fine-tuning and retraining, the model's adaptability in the field of evaluating the antidepressant activity of salamander lipids is continuously improved.

[0037] In the above implementation, the system uses a closed-loop optimization mechanism to achieve dynamic iterative evolution of the model, ensuring that the high-throughput screening method always meets the actual needs of lipid activity evaluation in giant salamanders, and improving the practicality, stability and scalability of the technical solution.

[0038] Secondly, embodiments of this application provide a high-throughput screening method for salamander lipids for evaluating antidepressant activity. The system includes: a lipid database management server, a mass spectrometry data processing cluster, a machine learning computing engine, and a result visualization engine. The lipid database management server stores the salamander lipid database and historical analysis archives; the mass spectrometry data processing cluster performs peak detection, qualitative identification, and quantitative analysis of large-scale lipid mass spectrometry data; the machine learning computing engine performs correlation analysis and key lipid screening calculations; and the result visualization engine renders the analysis results into interactive charts. The lipid database management server performs the following operations: receiving lipid mass spectrometry data uploaded from a lipidomics detection platform; performing peak detection and noise removal on the lipid mass spectrometry data; matching and identifying the processed data with the lipid database; and storing the identification results in the salamander lipid database and updating the index structure.

[0039] In the above scheme, the entire process of lipid data processing, from detection to analysis, is realized through signaling interaction between the lipid database management server, the mass spectrometry data processing cluster, and the machine learning computing engine. The mass spectrometry data processing cluster utilizes the identification capabilities of the lipid database to transform complex lipid mass spectrometry data into structured lipid molecular information, providing a data foundation for subsequent correlation analysis and key lipid screening.

[0040] Thirdly, embodiments of this application provide a method for evaluating and screening the antidepressant activity of salamander lipids, applied to a lipidomics detection platform. The method includes: obtaining salamander lipid extract samples; injecting the samples into an ultra-high performance liquid chromatography-high resolution mass spectrometry (UHPLC-MS / MS) system; performing data-dependent acquisition in positive and negative ion modes; acquiring mass spectrometry data of lipid molecules; obtaining a list of characteristic peaks through peak detection and noise reduction; matching and identifying the characteristic peaks with a lipid database; obtaining qualitative and quantitative analysis results of lipid molecules; and outputting a lipidomics data report, including lipid molecular composition information, relative abundance information, and lipid characteristic spectral data.

[0041] Fourthly, embodiments of this application provide a method for evaluating and screening the antidepressant activity of salamander lipids, applied to an association analysis server. The method includes: receiving lipidomics data output from a lipidomics detection platform; receiving activity evaluation data output from an activity evaluation module; preprocessing and standardizing the lipidomics data and activity evaluation data; performing initial screening using Pearson correlation coefficient analysis to select candidate lipids significantly correlated with activity indicators; performing dimensionality reduction analysis using PCA and PLS-DA to identify key principal components; constructing an association analysis model using random forest, SVM, or neural network algorithms; calculating the contribution weight, fold change, stability index, and repeatability results of candidate lipids; screening key active lipids using a multi-indicator weighted comprehensive scoring algorithm; and storing the screening results in a quality biomarker library and generating an analysis report.

[0042] Compared with the prior art, the present invention has the following beneficial effects: This system, utilizing a lipidomics detection platform and a multi-algorithm fusion analysis server for full-spectrum lipid detection and correlation analysis, achieved precise screening and evaluation of antidepressant active ingredients in giant salamander lipids. This screening method, based on the combined use of high-throughput lipidomics and machine learning, replaces the traditional single-indicator evaluation model, improving the identification ability and screening efficiency of active lipids. The key active lipids screened by the system and the constructed quality biomarker library provide a scientific basis for the quality control of giant salamander lipid products and reduce the cost of activity evaluation. Furthermore, the activity evaluation module, through a multi-level evaluation system using in vitro cell models and in vivo animal models, achieved a comprehensive assessment of antidepressant activity. The activity prediction model generated by the system transforms abstract lipidomics data into quantifiable activity scores, assisting researchers in quickly grasping the activity characteristics of samples and improving the efficiency of activity evaluation. The quality control system, based on the dynamic monitoring of key active lipids, achieves a unified characterization between antidepressant activity and quality indicators, improving product quality stability. Combined with a validation feedback channel and a model iteration optimization mechanism, the system achieves a closed-loop operation of "model construction - validation feedback - parameter optimization - model iteration," ensuring that the high-throughput screening method always meets the actual needs of giant salamander lipid activity evaluation. This invention solves the technical problems of difficulty in identifying active lipids in giant salamander lipids, lack of quality markers, and difficulty in uniformly evaluating quality and efficacy, thereby improving the overall level of evaluation of the antidepressant activity of giant salamander lipids. Attached Figure Description

[0043] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 A flowchart illustrating a high-throughput screening method for evaluating the antidepressant activity of lipids in giant salamanders, provided in an embodiment of the present invention. Figure 2 This is a schematic diagram of the structure of a lipid extraction module provided in an embodiment of this application; Figure 3 This is a flowchart of a lipidomics detection platform provided in an embodiment of this application; Figure 4 This is a diagram illustrating a multi-level evaluation system architecture for an activity evaluation module provided in an embodiment of this application. Figure 5 This is a flowchart of a coupling analysis process for an association analysis server provided in an embodiment of this application; Figure 6This is a schematic diagram of a key active lipid screening algorithm provided in an embodiment of this application; Figure 7 This is a schematic diagram of the archive structure of a quality marker library provided in an embodiment of this application; Figure 8 This is a schematic diagram of the construction interface of an activity prediction model provided in an embodiment of this application. Detailed Implementation

[0044] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the disclosure to those skilled in the art. It should be noted that, unless otherwise specified, embodiments and features in the embodiments of the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0045] In this invention, "lipomics detection" and "activity evaluation" represent the flow of data. For example, "sending lipidomics data to the correlation analysis server" can be understood as the destination of the data being the correlation analysis server, which can include direct transmission via a network interface or indirect transmission via a bus from other units or modules. "Receiving lipid extract from the lipid extraction module" can be understood as the source of the lipid extract being the lipid extraction module, which can include direct reception from the lipid extraction module via a network interface or indirect reception from the lipid extraction module via middleware. "Sending" can also be understood as the "output" of a software module, and "receiving" can be understood as the "input" of a software module. In other words, lipidomics detection and activity evaluation can be performed between modules, for example, by sending or receiving data between modules, software modules, or hardware modules via a bus, wiring, or interface.

[0046] The technical solution of this invention can be applied to the evaluation of the antidepressant activity of lipids in giant salamanders, and various lipidomics detection protocols can be used in this scenario. These detection protocols include, but are not limited to, ultra-high performance liquid chromatography-high resolution mass spectrometry (UPLC-HRMS), liquid chromatography-mass spectrometry (LC-MS), gas chromatography-mass spectrometry (GC-MS), or matrix-assisted laser desorption / ionization time-of-flight mass spectrometry (MALDI-TOF MS), etc., and are not listed here.

[0047] The technical solutions of this invention can be applied to communication scenarios between a lipid extraction module and a lipidomics detection platform, as well as between a lipidomics detection platform and a correlation analysis server, and between a correlation analysis server and a quality biomarker database. In this invention, the term "communication" can also be described as "data exchange," "information flow," or "signaling transmission." In this invention, the term "transmission" can also be described as "reporting" and / or "distribution."

[0048] In one possible implementation, the lipid extraction module can be a device with lipid extraction and pretreatment functions, supporting communication via a standard interface and capable of communicating with a lipidomics detection platform. This lipid extraction device can be a complete machine, or a software plug-in or logic circuit installed within a complete machine. Devices with these plug-ins or circuits can implement the methods and functions of this invention under the control of a processor. The lipid extraction module can be deployed in laboratories, sample preparation centers, and mobile sampling terminals. The lipid extraction module can be understood as a bridge connecting the salamander raw material and the lipidomics detection platform, its main function being to extract lipid components from the salamander raw material and convert them into samples suitable for mass spectrometry detection. For example, the lipid extraction module can be a laboratory device with an extraction apparatus or a fully automated sample pretreatment system.

[0049] In one possible implementation, the lipidomics detection platform can be a device with lipidomics analysis capabilities, supporting ultra-high performance liquid chromatography-high resolution mass spectrometry (UHPLC-MS / MS) detection and capable of communicating with a lipid extraction module or a correlation analysis server. This lipidomics detection device can be a complete mass spectrometer, or a software plugin or data processing module installed within the mass spectrometer. The lipidomics detection platform can also be a high-performance computing cluster, a cloud computing platform, or a local server. For example, the lipidomics detection platform can be a quadrupole-time-of-flight mass spectrometer (Q-TOF) equipped with an electrospray ionization (ESI) source, an orbital trap mass spectrometer equipped with a linear ion trap, or a detection subsystem integrated into a lipidomics analysis system.

[0050] In one possible implementation, the correlation analysis server can be a device with machine learning and multivariate statistical analysis capabilities, supporting the execution of correlation analysis algorithms between lipid features and activity indicators, and capable of communicating with lipidomics detection platforms or quality biomarker databases. This device with correlation analysis capabilities can be a full-fledged server, a computing board or parallel processing system installed in a rack-mount device, etc. The correlation analysis server can also be a high-performance computing cluster, an edge computing node, or a cloud virtual machine. For example, the correlation analysis server can be a graphics processing unit (GPU) server supporting multidimensional matrix operations, a stream processing engine supporting real-time analysis, or a correlation analysis subsystem integrated into a lipid analysis system.

[0051] In one possible implementation, the activity evaluation module can be a device with bioactivity evaluation capabilities, supporting in vitro cell model and in vivo animal model evaluation experiments, and capable of communicating with a correlation analysis server or lipid extraction module. This device with activity evaluation capabilities can be a complete laboratory instrument, or a software system or automated operation module installed in an experimental platform. The activity evaluation module can also be an automated workstation, a high-throughput screening system, or a virtual simulation platform. For example, the activity evaluation module can be an ELISA reader system equipped with a microplate reader, a video tracking system supporting behavioral analysis, or an evaluation subsystem integrated into the activity evaluation system.

[0052] In one possible implementation, the quality biomarker database can be a device with data storage and management capabilities, supporting indexing and retrieval of lipid information and activity data, and having the ability to communicate with correlation analysis servers or activity evaluation modules. This database management device can be a full-fledged server, database management software installed in a storage system, or distributed storage nodes. The quality biomarker database can also be a relational database, a distributed file system, or a cloud storage service. For example, the quality biomarker database can be a relational database equipped with an SQL query engine, a search engine supporting full-text search, or a database subsystem integrated into a quality control management system.

[0053] The lipid extracts typically include triglycerides (TG), phospholipids (PL), sphingolipids (SM), free fatty acids (FFA), and their derivatives, which will not be listed here. In this embodiment of the invention, the system including these lipid components can be referred to as the *Gymnocypris saurus* lipid antidepressant activity evaluation system. In practical applications, this invention does not limit the number of modules or detection terminals in the system.

[0054] The system architecture of the embodiments of the present invention will be described below. It should be noted that the system architecture and application scenarios described in the embodiments of the present invention are for the purpose of more clearly illustrating the technical solutions of the embodiments of the present invention, and do not constitute a limitation on the technical solutions provided by the embodiments of the present invention. As those skilled in the art will know, as the system architecture or application scenarios evolve, the technical solutions provided by the embodiments of the present invention are also applicable to similar technical problems.

[0055] The salamander lipid antidepressant activity evaluation and screening system provided by this invention is applicable to various scenarios, including lipidomics analysis, pharmacological evaluation, and product quality control. The system is suitable for different evaluation types, such as in vitro activity screening (e.g., cell model evaluation), in vivo activity verification (e.g., animal model evaluation), and multi-level comprehensive evaluation. The following explanation uses a multi-level comprehensive evaluation scenario as an example.

[0056] Please see, Figure 1 This invention provides a method for evaluating and screening the antidepressant activity of lipids in giant salamanders, specifically comprising: Step S100: Obtain a lipid extract sample from a giant salamander, wherein the lipid extract sample includes at least one of triglyceride lipids, phospholipid lipids, sphingolipid lipids, free fatty acid lipids, and fat-soluble vitamins and their derivatives.

[0057] Step S200: Perform lipidomics detection on the lipid extract sample to generate lipid molecular composition information, relative abundance information, and lipid characteristic spectral data. This lipidomics detection includes: establishing a full-spectrum lipid detection method using ultra-high performance liquid chromatography-high resolution mass spectrometry (UHPLC-HSM), and performing data-dependent acquisition using both positive and negative ion modes. It also includes: using an ACQUITY UPLC HSS T3 column (…). mm, μm), column temperature set to 45℃, mobile phase A is acetonitrile / water (μ ... (v / v) Contains 0.1% formic acid and 10mM ammonium formate, mobile phase B is acetonitrile / isopropanol ( (v / v) Contains 0.1% formic acid and 10mM ammonium formate, with the flow rate set to mL / min, gradient elution program: 0-2 min hold B, 2-15 minutes from B increases linearly to B, hold for 15-20 minutes B, 20-22 minutes from B drops to B, hold for 22-25 minutes B. Also includes: employing an electrospray ionization source, with ionization modes of positive ion mode and negative ion mode respectively, and capillary voltage set to... kV (positive ion mode) or kV (negative ion mode), cone voltage set to kV. V, collision energy set to eV (dynamically adjusted based on lipid molecule type), data-dependent acquisition mode, acquisition frequency set to eV, and acquisition frequency set to eV. Hz, quality scan range set to 200-1800.

[0058] Step S300: Evaluate the antidepressant activity of the lipid extract samples and obtain activity evaluation data. This evaluation includes: obtaining activity evaluation data such as cell viability, neurotransmitter secretion levels, inflammatory factor expression levels, and oxidative stress indicators using an in vitro cell model; obtaining antidepressant phenotypic data such as kinetic behavior, exploratory behavior, and sucrose preference behavior using a zebrafish behavioral model; and obtaining antidepressant phenotypic data such as sucrose preference behavior, tail-hanging behavior, and forced swimming behavior using a mouse chronic unpredictable stress model.

[0059] Step S400: Couple the lipidomics data and activity evaluation data to construct a correlation model between lipid characteristics and antidepressant activity. This coupling analysis includes: initial screening using Pearson correlation coefficient analysis to identify candidate lipids significantly correlated with activity indicators; dimensionality reduction analysis using principal component analysis (PCA) and partial least squares discriminant analysis (PLS-DA) to identify key principal components; constructing a correlation analysis model using random forest, support vector machine, or neural network algorithms; and screening key active lipids using a multi-index weighted comprehensive scoring algorithm.

[0060] Step S500 also includes: obtaining verification opinions on the screening results of key active lipids; recording model performance indicators under different parameter configurations; and iteratively adjusting the algorithm parameters, screening thresholds, and coefficients of the prediction model of the correlation analysis server.

[0061] Based on the above, the present invention provides a method for evaluating and screening the antidepressant activity of salamander lipids. This method can be used in a salamander lipid antidepressant activity evaluation and screening system comprised of a lipid extraction module, a lipidomics detection platform, an activity evaluation module, a correlation analysis server, and a quality biomarker database. The number of lipid extraction modules can include one or more; for example, three modules can be used, specifically lipid extraction modules 110a, 110b, and 110c. It is understood that the system may also include other equipment, such as a sample pretreatment workstation, a data storage server, and a result visualization terminal.

[0062] The data transmission method used in this system can be taken as Ethernet access technology. The correlation analysis server mainly refers to the analysis software deployed in the laboratory or cloud computing platform, which is used to provide correlation analysis services between lipidomics data and activity evaluation data. Through lipidomics detection, activity evaluation, correlation analysis and key lipid screening and other technical means, the system can achieve accurate evaluation of the antidepressant activity of salamander lipids and construction of quality markers.

[0063] Understandably, each lipid extraction module in this system can be used to communicate with the giant salamander raw material storage system, which mainly refers to a low-temperature freezer or sample library (e.g., ultra-low temperature freezer, liquid nitrogen tank, etc.) for storing giant salamander samples.

[0064] Taking the lipid extraction module 110a as an example, it can be used to communicate with a giant salamander raw material storage system. For instance, this raw material storage system can store giant salamander tissue samples, lipid extraction reagents, and standards. Furthermore, the lipidomics detection platform in this embodiment of the invention can establish three different types of logical links, specifically including logical link L1, logical link L2, and logical link L3, as detailed below: Logical Link L1: This refers to the lipid extraction module 110a connecting to the core switching network via wired or wireless access to achieve communication with the lipidomics detection platform.

[0065] Logical Link L2: This refers to the data link established between the lipid extraction module 110a and the activity evaluation module. It is mainly used for acquiring activity evaluation samples (e.g., lipid extract sample distribution, concentration gradient preparation, etc.). For example, the lipid extraction module 110a can provide a sample interface for the activity evaluation module to access. Correspondingly, the activity evaluation module can connect to the lipid extraction module 110a through a sample receiving protocol, thereby realizing a communication connection with the lipid extraction module 110a.

[0066] Logical Link L3 refers to the feedback link established between the lipid extraction module 110a and the correlation analysis server. This link can be used to acquire analytical result data or to ensure the dynamic updating of lipid data during the lipid screening process. It should be understood that lipid data updates here can include: adding new lipid types, adjusting lipid content, and correcting identification results.

[0067] In other words, the aforementioned lipidomics detection platform can not only be used to establish a communication connection with the lipid extraction module 110a through logical link L1, but also to achieve data interaction with the activity evaluation module or the correlation analysis server through logical link L2 or logical link L3.

[0068] In one possible implementation, please see Figure 2 , Figure 2 This is a simplified illustration of a method for evaluating and screening the antidepressant activity of lipids in giant salamanders, provided in an embodiment of the present invention. Figure 2 .like Figure 2 As shown, this method can be applied to a system for evaluating and screening the antidepressant activity of salamander lipids, consisting of a lipid database management server, a mass spectrometry data processing cluster, a machine learning computing engine, and a result visualization engine. Specifically, the number of lipid extraction modules can be exemplified by three modules: lipid extraction module 210a, lipid extraction module 210b, and lipid extraction module 210c. It is understood that... Figure 2 This is just an illustration; the system may also include other devices, such as an index server and a caching system. Figure 2 It is not shown in the middle.

[0069] Compared to the system shown, the system also deploys an additional lipid database management server. This lipid database management server can perform structured analysis and quality verification of lipid data in one stop. That is, the lipid database management server can be used to perform peak detection, noise removal, matching identification and other processing on lipid mass spectrometry data. The lipid database management server can be a high-performance computing node to provide data management, index update and other services for administrators.

[0070] like Figure 2 As shown, the lipid database management server communicates with the lipid extraction module via logical link L1, and can also synchronize data with the results visualization engine via logical link L3. For a detailed description of the lipid database management server's communication, please refer to the description of the lipidomics detection platform shown above; it will not be repeated here.

[0071] Understandably, since there is a lipid database management server in the system, the lipid samples extracted by these different lipid extraction modules (e.g., lipid extraction module 210a, lipid extraction module 210b, and lipid extraction module 210c) can be processed by the lipid database management server and then subjected to peak detection and qualitative identification by the mass spectrometry data processing cluster, thereby realizing intelligent analysis and multidimensional evaluation of lipid data.

[0072] It should be understood that in this invention, the management device (e.g., a lipid database management server or a correlation analysis server) can perform some or all of the steps in the various embodiments. These steps or operations are merely examples, and the embodiments of this invention can also perform other operations or variations thereof. Furthermore, the steps can be performed in different orders as presented in the various embodiments, and it is not necessary to perform all the operations in the embodiments of this invention. For ease of understanding, the management device in this invention can be taken as an example of a lipidomics detection platform to illustrate specific implementation methods for performing lipidomics detection.

[0073] In one possible implementation, during system operation, the lipidomics detection platform can synchronize the lipid index tables held by both parties with the lipid database management server at preset intervals (e.g., 30 minutes) to facilitate subsequent matching and identification of lipid molecules. The index table can include lipid codes and mass spectrometry feature identifiers. The index used to map lipid molecule information can be dynamically updated based on lipid identification. For example, if the mass spectrometry feature dimension used subsequently is 1024, then the mass spectrometry feature corresponding to each index is a high-dimensional vector. In other words, after a synchronization, the mapping relationship held by the lipidomics detection platform can include unique lipid codes, mass spectrometry feature identifiers, and lipid metadata.

[0074] For example, please refer to Table 1, which is a schematic table of lipid indexes held by the lipidomics detection platform provided in this embodiment of the invention under different lipid categories. As shown in Table 1: Table 1

[0075] Triglycerides, phospholipids, sphingolipids, and free fatty acids can all be used to represent different lipid types. As shown in Table 1 above, the lipid codes held by the lipidomics detection platform under the triglyceride category and the lipid codes held under the phospholipid category belong to different lipid types.

[0076] For ease of explanation, in this embodiment of the invention, the currently activated lipid index can be referred to as the first index, and the historically synchronized index can be referred to as the second index. The historical synchronization occurred earlier than the current activation time.

[0077] As shown in Table 1 above, the lipid index codes differ for different lipid categories. This means that the lipid abstraction rules are not fixed during system operation but can be dynamically adjusted based on lipid type and content, which enhances the flexibility of lipidomics detection. Furthermore, in the case of parallel analysis of multiple lipid categories, the last synchronized index table will participate in the subsequent construction of the lipid feature network, thereby accurately identifying the association between lipids and activity.

[0078] For a better understanding of the lipidomics detection process described above, please refer to the example provided. Figure 3 , Figure 3 This is a schematic flowchart illustrating a method for evaluating and screening the antidepressant activity of salamander lipids provided in an embodiment of the present invention. Figure 3 As shown, this method can be executed jointly by a lipid extraction module and related equipment in the system (e.g., a lipid extraction module, a lipidomics detection platform, an activity evaluation module, and a correlation analysis server). The method may include at least steps S301-S306: In step S301, upon receiving a giant salamander lipid extract sample, the lipid extraction module sends a lipidomics detection request to the lipidomics detection platform.

[0079] Correspondingly, the lipidomics detection platform can receive lipidomics detection request information from the lipid extraction module.

[0080] The lipidomics detection request information here is used to request the execution of lipidomics detection. This request information may include at least one of the following: lipid extract sample identifier, sample concentration, and extraction batch number. The lipid extract sample here contains lipid components from giant salamanders, such as triglycerides, phospholipids, sphingolipids, and free fatty acids.

[0081] Understandably, this request information is obtained by the lipid extraction module in response to a detection operation for the lipid extract, which may be a sample loading operation performed on the sample autosampler.

[0082] Step S302: The lipidomics detection platform performs lipidomics detection on the lipid extract sample.

[0083] In one possible implementation, after receiving a lipidomics detection request, the lipidomics detection platform can directly perform lipidomics detection according to a preset detection method, determine the mass spectrometry characteristics of lipid molecules, and then generate standardized lipidomics data based on a lipid index table.

[0084] For example, a lipidomics detection platform can map detected lipid molecules to mass spectrometry feature vector sequences and lipid categories to classification label sequences. In other words, the lipidomics data here can include lipid molecule feature vectors (i.e., a set of mass spectrometry feature codes) as well as lipid abundance information.

[0085] In one possible implementation, to ensure data accuracy, the lipidomics detection platform needs to verify the legitimacy of the detection results after generating the lipidomics data. If the detection results are determined to be within the range of the giant salamander lipid database, the lipidomics detection platform can send the lipidomics data to the correlation analysis server. If the detection results are determined to be outside the range, an error notification is sent directly to the lipid extraction module.

[0086] For example, after system initialization, the lipidomics detection platform can acquire lipid database information from giant salamanders and establish a mapping relationship between lipid coding and mass spectrometry features. For ease of understanding, please refer to Table 2, which is a lipid index mapping table stored in a system according to an embodiment of the present invention. Specifically, as shown in Table 2: Table 2

[0087] For the same lipid code, it can correspond to multiple mass spectrometry feature dimensions. This can be understood as the lipid information corresponding to the lipid code being represented by mass spectrometry from multiple feature perspectives.

[0088] In step S303, the activity evaluation module receives the lipid extract sample and performs an antidepressant activity evaluation. Accordingly, the activity evaluation module can receive the aforementioned lipid extract sample. The activity evaluation result is obtained by the activity evaluation module through in vitro cell model or in vivo animal model experiments on the lipid extract sample. In one possible implementation, the activity evaluation module can construct a multidimensional activity feature matrix A. When a certain lipid sample l exhibits activity f at time t, elements are set... Otherwise set By performing activity type classification and evaluation operations on the activity feature matrix, multi-dimensional activity evaluation results are generated.

[0089] Step S304: The correlation analysis server constructs a lipid-activity correlation model and performs correlation analysis operations. In one possible implementation, the correlation analysis server can construct a lipid-activity correlation map G, which includes a node set V and an edge set E. The node set V represents lipid molecules, activity indicators, biomarkers, etc., and the edge set E represents the correlation and regulatory relationships between lipids and activity indicators.

[0090] In step S305, the correlation analysis server generates the key active lipid screening results. Correspondingly, the quality biomarker database can receive these key active lipid screening results. These screening results may include a list of key active lipids, quality biomarker profiles, and activity contribution scores.

[0091] Step S306: The quality biomarker database is constructed based on the screening results. This quality biomarker database may include lipid molecule names, lipid molecule structural formulas, mass spectrometry characteristic ions, retention times, concentration ranges, activity contribution scores, and batch-to-batch coefficients of variation. For example, the quality biomarker database is determined by the following formula: ; in, This refers to the name of a lipid molecule. The structure of a lipid molecule is shown. These are characteristic ions for mass spectrometry. To retain the time, For concentration range, Score the contribution to activity. This represents the batch-to-batch variation coefficient.

[0092] The following will elaborate on different scenarios for evaluating and screening the antidepressant activity of salamander lipids: Implementation Method 1: For an example, please refer to [link / reference]. Figure 4 , Figure 4 This is an interactive illustration of a lipidomics detection method provided in an embodiment of the present invention. Figure 1 .like Figure 4 As shown, this method can be executed jointly by the lipid extraction module and related equipment in the system. The method may include at least steps S401-S407: Step S401: The lipidomics detection platform acquires the lipidomics detection method parameters. For example, the lipidomics detection platform downloads the latest ultra-high performance liquid chromatography-high resolution mass spectrometry (UHPLC-MS / MS) method parameters from the method parameter library.

[0093] In step S402, the lipid extraction module receives a sample of giant salamander lipid extract. This sample contains lipid components from the giant salamander.

[0094] In step S403, the lipid extraction module sends a detection request to the lipidomics detection platform.

[0095] Step S404: The lipidomics detection platform performs lipidomics detection. This lipidomics detection is performed using ultra-high performance liquid chromatography-high resolution mass spectrometry (UHPLC-MS / MS), with data acquired in both positive and negative ion modes.

[0096] In step S405, the lipidomics detection platform performs peak detection and noise reduction. For example, the lipidomics detection platform uses a baseline correction algorithm and a noise threshold filtering method to process the raw mass spectrometry data.

[0097] Step S406: The lipidomics detection platform performs lipid molecule identification. For example, the platform identifies lipid molecules based on a giant salamander lipid database through precise mass matching and secondary mass spectrometry fragmentation comparison. The overall matching degree is determined by the following formula: ; in, To achieve accurate quality matching scores, The score for matching fragments in the second-order mass spectrometry. , For weighting.

[0098] In step S407, the lipidomics detection platform outputs a lipidomics data report, simultaneously generating lipid molecular composition information, relative abundance information, and lipid characteristic spectrum data.

[0099] In this embodiment of the invention, full-spectrum lipid detection performed by ultra-high performance liquid chromatography-high resolution mass spectrometry (UHPLC-HDMS) enables precise identification and quantitative analysis of lipid molecules in salamander lipid extracts. This process replaces traditional single-lipid type detection, outputting multi-dimensional lipidomics data reports and improving the comprehensiveness and accuracy of lipidomics detection.

[0100] Implementation Method Two: For example, please see Figure 5 , Figure 5 This is an interactive illustration of multi-level activity evaluation provided by an embodiment of the present invention. Figure 2 The method may include at least steps S501 to S509: Step S501: The activity evaluation module obtains lipid extract samples.

[0101] In step S502, the activity evaluation module performs in vitro cell model evaluation. For example, the activity evaluation module uses a PC12 cell injury model, an SH-SY5Y cell neuroinflammation model, and a microglia activation model to detect cell viability, neurotransmitter content, inflammatory factor levels, and oxidative stress indicators.

[0102] In step S503, the activity evaluation module performs a zebrafish behavioral model evaluation. For example, the activity evaluation module uses a zebrafish depression model induced by chronic unpredictable stress to detect behavioral indicators in the open domain test, suspended log test, and sucrose preference test.

[0103] In step S504, the activity evaluation module performs a mouse CUMS model evaluation. For example, the activity evaluation module uses a mouse chronic unpredictable stress depression model to detect behavioral indicators in the sucrose preference test, tail suspension test, forced swimming test, and open field test.

[0104] Step S505: The activity evaluation module constructs multi-level activity evaluation results.

[0105] In step S506, the activity evaluation module performs activity data standardization processing. For example, the activity evaluation module uses the Z-score standardization method to process activity data from different models, eliminating dimensional differences.

[0106] Step S507: The activity evaluation module calculates the comprehensive activity score. For example, the comprehensive activity score is determined by the following formula: ; in, For the score of the j-th activity index, The weight of the j-th activity index, This represents the total number of activity indicators.

[0107] Step S508: The activity evaluation module generates an activity evaluation report.

[0108] Step S509: The activity evaluation module outputs activity evaluation data to the correlation analysis server.

[0109] In this embodiment of the invention, the system achieves a comprehensive evaluation of the antidepressant activity of lipids in giant salamanders through a multi-level evaluation system combining in vitro cell models and in vivo animal models. In vitro models are used for high-throughput initial screening, while in vivo models are used for confirmatory verification; the combination of both improves the accuracy and reliability of activity evaluation.

[0110] Implementation method three: For example, please see Figure 6 , Figure 6 This is an interactive illustration of a key active lipid screening method provided in an embodiment of the present invention. Figure 3 The method may include at least steps S601-S607: Step S601: Obtain lipidomics data and activity evaluation data from the correlation analysis server.

[0111] In step S602, the association analysis server performs Pearson correlation coefficient analysis. For example, the association analysis server calculates the Pearson correlation coefficient between lipid abundance and activity indicators, and screens candidate lipids with a correlation coefficient threshold ≥ 0.6 or ≤ -0.6.

[0112] Step S603: The association analysis server performs PCA and PLS-DA dimensionality reduction analysis. For example, the association analysis server uses PCA to retain dimensionality reduction results with ≥3 principal components and ≥70% cumulative explained variance, and uses PLS-DA to screen significant variables with VIP values ​​≥1.0.

[0113] In step S604, the association analysis server performs machine learning association modeling. For example, the association analysis server uses the random forest algorithm (500 decision trees), the SVM algorithm (RBF kernel function), and the BP neural network algorithm (network structure input layer-hidden layer-output layer) to construct the association analysis model.

[0114] Step S605: The association analysis server performs key active lipid screening. For example, the association analysis server uses a multi-index weighted comprehensive scoring algorithm to calculate contribution weight, difference fold, stability index, and repeatability validation results, and screens key active lipids with a comprehensive score ≥ 0.7.

[0115] Step S606: The correlation analysis server generates a quality biomarker profile. For example, the quality biomarker profile includes the lipid molecule name, structural formula, mass spectrometry characteristic ions, retention time, concentration range, activity contribution score, and batch-to-batch coefficient of variation.

[0116] Step S607: The association analysis server outputs a list of key active lipids to the quality biomarker database.

[0117] In this embodiment of the invention, the system overcomes the limitations of traditional single statistical methods in analyzing complex lipid-activity relationships by using the correlation analysis capability of multi-algorithm fusion, and achieves accurate identification and screening of key active lipids.

[0118] Implementation Method 4: For example, please see Figure 7 , Figure 7 This is a schematic diagram of a closed-loop management interaction based on quality markers provided in an embodiment of the present invention. Figure 4 The method may include at least steps S701-S707: Step S701: The activity evaluation module acquires the sample to be tested.

[0119] Step S702: The activity evaluation module constructs an activity prediction model based on quality markers.

[0120] For example, the activity prediction model uses a multiple linear regression model, with input parameters being the type, abundance, and proportion of quality markers, and output parameters being the predicted antidepressant activity score. Model validation employs leave-one-out cross-validation or 5-fold cross-validation.

[0121] Step S703: The activity evaluation module performs rapid prediction of sample activity.

[0122] Step S704: The association analysis server performs a quality control assessment.

[0123] For example, assess the completeness of key active lipid species (target ≥90%), content (determined by concentration range), ratio (ratio deviation ≤ ±20%), and batch-to-batch coefficient of variation (CV ≤ 25%).

[0124] Step S705: When the quality control assessment result does not meet the preset standard, the system triggers the model optimization process.

[0125] Step S706: The association analysis server re-executes the association analysis and key lipid screening.

[0126] Step S707: The quality marker database is updated based on the results of the reanalysis.

[0127] In this embodiment of the invention, the system links quality markers with an activity prediction model, enabling comprehensive monitoring and dynamic optimization of the quality of giant salamander lipid products. The entire process achieves a closed-loop operation of "model construction - quality control - parameter optimization - model iteration," ensuring that the high-throughput screening method always meets the actual needs of giant salamander lipid activity evaluation.

[0128] It is understandable that the lipidomics detection methods mentioned above can be applied not only to lipid screening but also to lipid quality control, such as monitoring changes in the content of key active lipids during lipid extraction. In other words, with lipidomics detection, the system can directly perform dynamic monitoring of lipid content without the need for a complete activity evaluation experiment.

[0129] For example, regardless of the source of the lipid extract, the lipid extraction module can send a detection request containing the lipid extract sample identifier to the lipidomics detection platform. The lipidomics detection platform generates mass spectrometry data for the lipid extract. When the lipidomics detection platform determines that the content of key active lipids is below a threshold, the mass biomarker database outputs an early warning message.

[0130] For example, the quality control index is determined by the following formula: ; in, Let be the measured concentration of the i-th key active lipid. This is the reference concentration for the i-th key active lipid. This represents the total number of key active lipids.

[0131] In the various embodiments of the present invention, unless otherwise specified or in case of logical conflict, the terminology and / or descriptions of different embodiments are consistent and can be referenced by each other. Technical features in different embodiments can be combined to form new embodiments according to their inherent logical relationships.

[0132] It is understood that the various numerical designations used in the embodiments of the present invention are merely for descriptive convenience and are not intended to limit the scope of the embodiments of the present invention. The order of the process numbers does not imply the order of execution; the execution order of each process should be determined by its function and internal logic.

[0133] The above are merely preferred embodiments of the technical solutions of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0134] To enable those skilled in the art to fully understand and implement this invention, the following further supplements the specific implementation principle of this invention in conjunction with the specific application scenario of evaluating and screening the antidepressant activity of salamander lipids.

[0135] Step 1: During lipid extraction and lipidomics detection, the experimenter logs into the lipid sample preparation system through the lipid extraction module and inputs the raw information of the sample containing salamander lipid extract. After acquiring the raw sample, the lipid extraction module 110a transmits the sample data to the lipidomics detection platform via logical link L2. The lipidomics detection platform uses a preset ultra-high performance liquid chromatography-high resolution mass spectrometry (UHPLC-MS / MS) technique to comprehensively detect triglycerides, phospholipids, sphingolipids, free fatty acids, and their derivatives in the salamander lipid extract. It obtains the mass spectrometric characteristics, relative abundance information, and lipid characteristic spectral data of lipid molecules, generating a standardized lipidomics data report.

[0136] Step 2: During the antidepressant activity evaluation, the lipid extraction module 110a transfers aliquots of lipid extract samples to the activity evaluation module. The activity evaluation module first performs high-throughput screening using in vitro cell models (PC12 cell damage model, SH-SY5Y cell neuroinflammation model, and microglia activation model), detecting cell viability, neurotransmitter secretion levels (5-HT, DA, NE), inflammatory factor levels (IL-1β, TNF-α), and oxidative stress indicators (SOD activity, MDA content). Then, it uses a zebrafish depression model and a mouse CUMS model for in vivo confirmation and validation, detecting antidepressant phenotypic data such as motor behavior, exploratory behavior, and sucrose preference behavior. The activity evaluation module integrates the in vitro and in vivo evaluation data to generate a multi-level activity evaluation report.

[0137] Step 3: During the association analysis and key lipid screening, the association analysis server receives lipidomics data from the lipidomics detection platform and activity evaluation data from the activity evaluation module via logical link L1. The association analysis server first uses Pearson correlation coefficient analysis for initial screening, identifying candidate lipids significantly correlated with activity indicators. Then, it uses PCA and PLS-DA for dimensionality reduction analysis to identify key principal components. Next, it constructs an association analysis model using random forest, SVM, and neural network algorithms. Finally, it uses a multi-indicator weighted comprehensive scoring algorithm (contribution weight 0.4, difference fold 0.25, stability index 0.2, repeatability validation result 0.15) to screen key active lipids with a comprehensive score ≥0.7, generating quality biomarker profiles and storing them in the quality biomarker database.

[0138] Step four, during quality control and closed-loop optimization, involves the activity evaluation module constructing an activity prediction model (multiple linear regression model or machine learning regression model) based on a quality biomarker library to rapidly predict the antidepressant activity of unknown samples. The correlation analysis server dynamically monitors key active lipids in raw materials, extraction processes, intermediate products, and finished products, assessing species completeness (target ≥90%), content (determined by concentration range), ratio (ratio deviation ≤ ±20%), and batch-to-batch coefficient of variation (CV ≤ 25%). When the quality control evaluation results do not meet preset standards, the system triggers a model optimization process, iteratively adjusting the algorithm parameters, screening thresholds, and prediction model coefficients of the correlation analysis server, forming a closed-loop mechanism of "model construction - quality control - parameter optimization - model iteration" to ensure that the high-throughput screening method always meets the actual needs of salamander lipid activity evaluation.

[0139] All content not described in detail in this specification is existing technology known to those skilled in the art. The hardware equipment involved in this technical solution, such as the lipidomics detection platform, correlation analysis server, and activity evaluation module, all adopt standard laboratory equipment, and their internal logic operations are executed by the processor through pre-stored software programs.

[0140] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program goods. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program goods embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0141] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program goods according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0142] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0143] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0144] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for evaluating and high-throughput screening the antidepressant activity of lipids in giant salamanders, characterized in that, include: Obtain lipid extract samples from giant salamanders; Lipidomics detection was performed on lipid extract samples to generate lipid molecular composition information, relative abundance information, and lipid characteristic spectral data; The antidepressant activity of lipid extract samples was evaluated, and activity evaluation data were obtained. We conducted a coupled analysis of lipid characterization data and activity evaluation data to construct a correlation model between lipid characteristics and antidepressant activity. A library of quality biomarkers was constructed by screening key bioactive lipids based on an association model.

2. The method for evaluating and high-throughput screening the antidepressant activity of lipids in giant salamanders according to claim 1, characterized in that, The lipid extract sample includes at least one of the following: triglyceride lipids, phospholipid lipids, sphingolipid lipids, free fatty acid lipids, and fat-soluble vitamin derivatives.

3. The method for evaluating and high-throughput screening the antidepressant activity of lipids in giant salamanders according to claim 1, characterized in that, When performing lipidomics assays on lipid extract samples, the following are included: A full-spectrum lipid detection method was established based on liquid chromatography-high-resolution mass spectrometry. Data acquisition is based on positive and negative ion modes.

4. The method for evaluating and high-throughput screening the antidepressant activity of lipids in giant salamanders according to claim 3, characterized in that, When establishing a full-spectrum lipid detection method based on liquid chromatography-high-resolution mass spectrometry, it includes: An ACQUITY UPLC HSS T3 column was used, with the column temperature set to 45℃. Mobile phase A was acetonitrile / water containing 0.1% formic acid and 10 mM ammonium formate, and mobile phase B was acetonitrile / isopropanol containing 0.1% formic acid and 10 mM ammonium formate. The following parameters were used: The flow rate was set to 0.35 mL / min, and the gradient elution program was as follows: 0-2 min to maintain 5% B, 2-15 min to linearly increase from 5% B to 80% B, 15-20 min to maintain 80% B, 20-22 min to decrease from 80% B to 5% B, and 22-25 min to maintain 5% B.

5. The method for evaluating and high-throughput screening the antidepressant activity of lipids in giant salamanders according to claim 3, characterized in that, When performing data-dependent acquisition based on positive and negative ion modes, it includes: An electrospray ionization source is used, with ionization modes including positive ion mode and negative ion mode, wherein: The capillary voltage is set to 3.0kV or 2.8kV, the cone voltage is set to 30V, the collision energy is set to 15-40eV, and the mass scan range is set to m / z 200-1800.

6. The method for evaluating and high-throughput screening the antidepressant activity of lipids in giant salamanders according to claim 1, characterized in that, When evaluating the antidepressant activity of lipid extract samples, the following should be included: A depression-related in vitro cell damage model was constructed, and lipid extract samples were applied to the in vitro cell damage model. The cell viability after treatment with lipid extract was detected to evaluate the protective effect of lipid extract against cell damage. The neurotransmitter secretion levels after treatment with lipid extracts were measured to evaluate the regulatory effect of lipid extracts on nerve function. The expression levels of inflammatory factors after treatment with lipid extracts were detected to evaluate the inhibitory effect of lipid extracts on inflammatory responses. Oxidative stress indicators were detected after treatment with lipid extracts to evaluate the ameliorative effect of lipid extracts on oxidative damage. The antidepressant activity of the lipid extract was comprehensively evaluated based on the cell survival rate, neurotransmitter secretion level, inflammatory factor expression level, and oxidative stress indicators.

7. The method for evaluating and high-throughput screening the antidepressant activity of lipids in giant salamanders according to claim 6, characterized in that, Constructing an in vitro cell damage model related to depression, and applying a lipid extract sample to the in vitro cell damage model, includes: PC12 cell injury model, SH-SY5Y cell neuroinflammation model and microglia activation model were constructed to simulate neuronal injury, neuroinflammation and neuroimmune imbalance, respectively. Inducing stimuli were used to treat various cells to establish an in vitro cell damage model related to depression. The lipid extract samples were prepared into concentration gradients of 12.5 μg / mL, 25 μg / mL, 50 μg / mL, 100 μg / mL and 200 μg / mL, respectively. Different concentrations of lipid extract samples were applied to the PC12 cell injury model, SH-SY5Y cell neuroinflammation model, and microglia activation model, respectively. Based on the cellular response results after treatment with lipid extract samples of various concentrations, the dose-response relationship of lipid extracts was established, and lipid extract samples with antidepressant activity were screened.

8. The method for evaluating and high-throughput screening the antidepressant activity of lipids in giant salamanders according to claim 1, characterized in that, When performing coupled analysis of lipid characterization data and activity evaluation data, the following steps are taken: obtaining lipid characteristic variables from the lipid characterization data and antidepressant activity indicators from the activity evaluation data. Pearson correlation coefficient analysis was used to calculate the correlation coefficient between each lipid characteristic variable and each antidepressant activity index, and candidate active lipid characteristics were screened based on the correlation coefficient. Principal component analysis was performed on the screened candidate active lipids to extract principal components characterizing the main changes in lipid data. Based on the principal components, partial least squares discriminant analysis was used to establish the mapping relationship between lipid characteristics and activity evaluation results, and key lipid characteristic variables were screened. Using key lipid characteristic variables as input parameters and activity evaluation results as output parameters, a correlation analysis model between lipid characteristic spectrum data and activity evaluation data is constructed using random forest algorithm, support vector machine algorithm or neural network algorithm; The contribution of each lipid characteristic variable to the antidepressant activity was calculated based on the association analysis model, and the target lipid components associated with the antidepressant activity were identified.

9. The method for evaluating and high-throughput screening the antidepressant activity of lipids in giant salamanders according to claim 8, characterized in that, Also includes: Obtain the candidate active lipid molecules screened by the correlation analysis model; The contribution weight, fold change, stability index, and replication results of each candidate active lipid molecule were extracted. The contribution weights, difference factors, stability indicators, and repeated validation results are standardized to obtain the corresponding standardized evaluation indicators. The standardized evaluation indicators are weighted and fused based on preset weights to obtain the comprehensive score of each candidate active lipid molecule. Candidate active lipid molecules are screened based on the comprehensive score, and lipid molecules with a comprehensive score greater than or equal to 0.7 are identified as key active lipids. The key active lipids were incorporated into a quality biomarker library to establish a lipid quality biomarker system for evaluating antidepressant activity.

10. The method for evaluating and high-throughput screening the antidepressant activity of lipids in giant salamanders according to claim 1, characterized in that, Also includes: Obtain validation opinions on the screening results of key active lipids; Record the model performance indicators of the correlation analysis model under different parameter configurations, and establish a model performance evaluation database; Based on the verification opinions and model performance indicators, the accuracy of the prediction results of the correlation analysis model is evaluated. Based on the evaluation results, analyze the impact of algorithm parameters, screening thresholds, and prediction model coefficients on model performance indicators, and determine the parameters to be optimized. The algorithm parameters, active lipid screening thresholds, and prediction model coefficients in the association analysis server were iteratively adjusted, and the association analysis model was reconstructed based on the adjusted parameters. The key active lipid screening was re-executed using the updated association analysis model, and new validation opinions were obtained. Based on the new validation feedback, continue the model evaluation, parameter optimization, and model reconstruction processes to form a closed-loop optimization mechanism encompassing model building, validation feedback, parameter optimization, and model iteration.

Images