Artificial intelligence-assisted method for screening for antioxidant peptides in oils

Abstract

The present invention relates to the technical field of bioinformatics, and disclosed is an artificial intelligence-assisted method for screening for antioxidant peptides in oils. In the present invention, by combining deep learning and virtual enzymolysis, a novel method for rapidly and efficiently screening for antioxidant peptides prepared by protein enzymolysis is researched and developed. In the present invention, taking acquisition of antioxidant peptides as an objective, an antioxidant peptide dataset and a test set are constructed, and thus, on the basis of the dataset, a Bi-LSTM high-accuracy deep learning strategy is proposed, and is trained to recognize the antioxidant peptides. Then, a Bi-LSTM is used to predict the antioxidant activity of polypeptides obtained by virtual enzymolysis. Taking a black bean protein as an example, it is found using the strategy of the present invention that hydrolysis of the black bean protein by trypsin can more effectively yield antioxidant peptides. After ABTS- and DPPH-based activity validation, the antioxidant peptides are used for lipophilicity measurement, and the peptides having lipophilicity can exert better antioxidant effects in oils, providing a research basis for the safety control of oil thermal processing processes.

Description

A method for screening antioxidant peptides in oils based on artificial intelligence Technical Field

[0001] This invention relates to a method for screening antioxidant peptides in oils and fats based on artificial intelligence, belonging to the field of bioinformatics technology. Background Technology

[0002] During production and storage, vegetable oils are prone to oxidative rancidity under the influence of external environmental factors such as light, heat, moisture, oxygen, metal ions, and enzymes. This results in the production of various small molecule substances such as aldehydes, ketones, and acids, which not only reduce the quality of the oils but also pose certain toxic hazards to the human body when ingested.

[0003] In order to inhibit the oxidation of plant oil products, current research mainly uses antioxidants, including synthetic antioxidants and natural antioxidants. However, to avoid the potential harm caused by synthetic antioxidants such as BHA, BHT, and TBHQ, their use is gradually being replaced by green and safe natural antioxidants.

[0004] Natural antioxidant peptides are a class of small molecule peptide chains with antioxidant activity. They can inhibit lipid oxidation through various mechanisms, such as quenching free radicals, inhibiting oxidation reactions, and capturing oxidation products. However, there are thousands of antioxidant peptide structures, and their biological activity varies depending on their structure. Current experimental methods for screening antioxidant peptides are time-consuming and labor-intensive.

[0005] Currently, artificial intelligence is widely used for screening bioactive peptides, offering advantages such as speed, efficiency, and high accuracy. Through virtual enzymatic digestion and deep learning, bioactive peptides can be rapidly and accurately extracted from proteins. Virtual enzymatic digestion, a highly efficient protein hydrolysis sequence prediction technique based on protein amino acid sequences and protease cleavage sites, is considered an effective alternative to traditional enzymatic digestion methods. However, it can only provide peptide sequences and cannot predict activity. Deep learning, as an important branch of artificial intelligence, automatically learns algorithms and models from data, extracting patterns, rules, and knowledge from large datasets and utilizing this knowledge for tasks such as prediction, classification, and identification. In the field of bioactive peptide screening, deep learning technology can further explore the potential relationship between peptide sequences and their bioactivity, thereby achieving efficient screening and prediction of bioactive peptides.

[0006] Despite significant advancements in artificial intelligence (AI) technology for screening bioactive peptides, limitations remain in the screening and prediction of antioxidant peptides. In particular, while antioxidant peptides from various sources have emerged in recent years, few have been systematically screened and predicted using AI methods. Furthermore, the effectiveness of the same antioxidant peptide varies significantly across different environments, making the screening and application of antioxidant peptides more complex and challenging.

[0007] In summary, there is an urgent need for an artificial intelligence-assisted method for screening antioxidant peptides in oils. By combining the advantages of artificial intelligence technology, a large amount of information on antioxidant peptides can be quickly obtained, and antioxidant peptides that can play a role in oils can be efficiently screened. This will play an important role in promoting the widespread application and in-depth research of antioxidant peptides in oils. Summary of the Invention

[0008] To address the aforementioned issues, this invention provides a method for screening antioxidant peptides in lipids based on artificial intelligence. This method utilizes deep learning and virtual enzymatic hydrolysis to obtain antioxidant peptides, ensuring high accuracy. Combined with lipid antioxidant experiments, it uncovers novel lipid antioxidant peptides, providing a research foundation for lipid antioxidant studies.

[0009] In a first aspect, the present invention provides a method for screening antioxidant peptides in oils based on artificial intelligence-assisted screening, the method comprising:

[0010] Step S1: Collect antioxidant peptides from multiple databases, delete duplicate data and remove samples with inconsistent experimental results to obtain a positive dataset containing antioxidant peptides, randomly extract an equal amount of peptide sequences matching the length of the positive dataset as a negative control, and divide the positive and negative sample datasets into training and test sets.

[0011] Step S2: Input the dataset from step S1 into a bidirectional long short-term memory network, use the training set to learn and summarize the features of antioxidant peptides, and use the validation set data to validate the model to obtain an antioxidant peptide prediction model.

[0012] Step S3: Obtain the amino acid sequence of the raw protein from the NCBI website, and use different enzymes in the Novo Pro Labs virtual enzymatic digestion program to perform virtual enzymatic digestion on the raw protein to obtain peptides after virtual enzymatic digestion of the raw protein with different enzymes.

[0013] Step S4: Use the antioxidant peptide prediction model obtained in step S2 to predict the antioxidant activity of the peptides obtained in step S3, and calculate the percentage of peptides with antioxidant activity. Use the enzyme with the highest percentage for the actual enzymatic hydrolysis of the raw material protein to obtain antioxidant peptides.

[0014] Step S5: Randomly select the antioxidant peptides obtained in step S4 to determine the ABTS and DPPH free radical scavenging activities to verify the antioxidant activity;

[0015] Step S6: Import the antioxidant peptide sequence obtained in step S5 into ACD / Labs software to calculate the logP of the compound;

[0016] Step S7: Use the enzymes selected in step S3 that can yield more antioxidant peptides to perform actual enzymatic digestion of the protein, and identify, separate and purify the antioxidant peptides.

[0017] The antioxidant peptides prepared in steps S8 and S7 were added to vegetable oil and subjected to heating and frying simulations. The inhibition rates of the antioxidant peptides on the formation of primary and secondary oxidation products after heat processing of vegetable oil and their inhibitory effects on the formation of harmful volatile aldehydes were determined.

[0018] In one embodiment of the present invention, in step S1, each peptide contains 2-20 amino acids, and the positive and negative sample datasets are randomly divided into a training set and a test set in a ratio of 8:2.

[0019] In one embodiment of the present invention, in step S2, the characteristics of peptides in the training set are learned and summarized, and the peptides in the validation set are validated, and "0" and "1" are output, where "0" indicates that the peptide does not have antioxidant activity, and "1" indicates that the peptide has antioxidant activity.

[0020] In one embodiment of the present invention, in step S3, the length of the polypeptide sequence after virtual enzymatic hydrolysis of the raw material protein by different enzymes is less than 20.

[0021] In one embodiment of the present invention, in step S4, the peptide is judged to have antioxidant activity based on the output of the antioxidant peptide prediction model, where "0" indicates that the peptide does not have antioxidant activity and "1" indicates that the peptide has antioxidant activity.

[0022] In one embodiment of the present invention, in step S6, logP greater than 0 indicates fat solubility, and the higher the value, the greater the fat solubility.

[0023] In one embodiment of the present invention, in step S8, the types of vegetable oil include, but are not limited to, one or more of the following: peanut oil, sunflower seed oil, palm oil, soybean oil, olive oil, rapeseed oil, cottonseed oil, perilla seed oil, rice bran oil, flaxseed oil, safflower seed oil, tea seed oil, palm fruit oil, coconut oil, cocoa bean oil, almond oil, tung oil seed oil, corn germ oil, wheat germ oil, sesame seed oil, pumpkin seed oil, walnut oil, grape seed oil, flaxseed oil, pumpkin seed oil, cocoa butter, and algae oil.

[0024] In one embodiment of the present invention, in step S2, the bidirectional long short-term memory network merges the outputs of two directional RNNs, one processing the sequence forward and the other processing the sequence backward.

[0025] In a second aspect, the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method for screening antioxidant peptides in oils based on artificial intelligence.

[0026] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the method for screening oil antioxidant peptides based on artificial intelligence.

[0027] The present invention has the following beneficial effects:

[0028] This invention develops a novel, rapid, and efficient method for screening proteins and preparing antioxidant peptides through enzymatic hydrolysis by combining deep learning and virtual enzymatic hydrolysis. Aiming to obtain antioxidant peptides, this invention constructs an antioxidant peptide dataset and a test set. Based on this dataset, a high-precision Bi-LSTM deep learning strategy is proposed and trained to identify antioxidant peptides. Then, Bi-LSTM is used to predict the antioxidant activity of peptides obtained through virtual enzymatic hydrolysis.

[0029] Taking black soybean protein as an example, the strategy of this invention revealed that trypsin hydrolysis of black soybean protein yields better antioxidant peptides. After ABTS and DPPH activity verification, the antioxidant peptides were used for lipophilicity determination. Peptides with lipophilicity can exert better antioxidant effects on oils, providing a research basis for the safe control of oil thermal processing. Attached Figure Description

[0030] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0031] Figure 1 is a flowchart of the Bi-LSTM deep learning model for antioxidant peptides provided by the present invention.

[0032] Figure 2 shows the 5x cross-training and validation results of the antioxidant peptide Bi-LSTM deep learning model provided by this invention.

[0033] Figure 3 is a graph verifying the ABTS and DPPH free radical scavenging activity of the antioxidant peptides of the present invention.

[0034] Figure 4 shows the antioxidant activity of the primary and secondary products of the antioxidant peptide lipids of this invention. Detailed Implementation

[0035] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0036] Unless otherwise specified, all experimental materials, reagents and instruments used in the embodiments of the present invention are commercially available. Unless otherwise specified, all technical means used in the embodiments are conventional means well known to those skilled in the art.

[0037] Example 1

[0038] This invention provides a method for screening antioxidant peptides in oils and fats based on artificial intelligence, the method comprising:

[0039] Step S1: Collect antioxidant peptides from multiple databases, delete duplicate data and remove samples with inconsistent experimental results to obtain a positive dataset containing antioxidant peptides, randomly extract an equal amount of peptide sequences matching the length of the positive dataset as a negative control, and divide the positive and negative sample datasets into training and test sets.

[0040] Step S2: Input the dataset from step S1 into a bidirectional long short-term memory network, use the training set to learn and summarize the features of antioxidant peptides, and use the validation set data to validate the model to obtain an antioxidant peptide prediction model.

[0041] Step S3: Obtain the amino acid sequence of the raw protein from the NCBI website, and use different enzymes in the Novo Pro Labs virtual enzymatic digestion program to perform virtual enzymatic digestion on the raw protein to obtain peptides after virtual enzymatic digestion of the raw protein with different enzymes.

[0042] Step S4: Use the antioxidant peptide prediction model obtained in step S2 to predict the antioxidant activity of the peptides obtained in step S3, and calculate the percentage of peptides with antioxidant activity. Use the enzyme with the highest percentage for the actual enzymatic hydrolysis of the raw material protein to obtain antioxidant peptides.

[0043] Step S5: Randomly select the antioxidant peptides obtained in step S4 to determine the ABTS and DPPH free radical scavenging activities to verify the antioxidant activity;

[0044] Step S6: Import the antioxidant peptide sequence obtained in step S5 into ACD / Labs software to calculate the logP of the compound;

[0045] Step S7: Use the enzymes selected in step S3 that can yield more antioxidant peptides to perform actual enzymatic digestion of the protein, and identify, separate and purify the antioxidant peptides.

[0046] The antioxidant peptides prepared in steps S8 and S7 were added to vegetable oil and subjected to heating and frying simulations. The inhibition rates of the antioxidant peptides on the formation of primary and secondary oxidation products after heat processing of vegetable oil and their inhibitory effects on the formation of harmful volatile aldehydes were determined.

[0047] Optionally, in step S1, each peptide contains 2-20 amino acids, and the positive and negative sample datasets are randomly divided into training and test sets in a ratio of 8:2.

[0048] Optionally, in step S2, the characteristics of peptides in the training set are learned and summarized, and the peptides in the validation set are validated, outputting "0" and "1", where "0" indicates that the peptide does not have antioxidant activity, and "1" indicates that the peptide has antioxidant activity.

[0049] Optionally, in step S3, the length of the polypeptide sequence after virtual enzymatic hydrolysis of the raw material protein by different enzymes is less than 20.

[0050] Optionally, in step S4, the peptide is judged to have antioxidant activity based on the output of the antioxidant peptide prediction model, where "0" indicates that the peptide does not have antioxidant activity and "1" indicates that the peptide has antioxidant activity.

[0051] Optionally, in step S6, logP greater than 0 indicates fat solubility, and the higher the value, the greater the fat solubility.

[0052] Optionally, in step S8, the types of vegetable oil include, but are not limited to, one or more of the following: peanut oil, sunflower seed oil, palm oil, soybean oil, olive oil, rapeseed oil, cottonseed oil, perilla seed oil, rice bran oil, flaxseed oil, safflower seed oil, tea seed oil, palm fruit oil, coconut oil, cocoa soybean oil, almond oil, tung oil seed oil, corn germ oil, wheat germ oil, sesame seed oil, pumpkin seed oil, walnut oil, grape seed oil, flaxseed oil, pumpkin seed oil, cocoa butter, and algae oil.

[0053] Optionally, in step S2, the bidirectional long short-term memory network merges the outputs of two directional RNNs, one processing the sequence forward and the other processing the sequence backward.

[0054] Example 2

[0055] As shown in Figures 1-4, this invention provides a method for screening antioxidant peptides in lipids, the method comprising:

[0056] Step 1: Input "Antioxidative peptide" into multiple databases including DFBP (http: / / www.cqudfbp.net / ), BIOPEP-UWM (https: / / biochemia.uwm.edu.pl / biopep / start_biopep.php), Antimicrobial Peptide Database (https: / / aps.unmc.edu / ), PlantPepDB (http: / / 14.139.61.8 / PlantPepDB / index.php), and FermFooDb (https: / / webs.iiitd.edu.in / raghava / fermfoodb / ) to collect peptides with antioxidant activity. Duplicate data and samples with inconsistent experimental results were removed, resulting in a positive dataset containing 1,511 antioxidant peptides, each containing 2-20 amino acids. An equal amount of peptide sequence matching the length of the positive dataset was randomly extracted as a negative control. The positive and negative sample datasets were randomly divided into training and test sets in an 8:2 ratio. A 5-fold cross-validation was then used to split the training and test sets. The model dataset is structured as shown in Table 1.

[0057] Table 1. Composition of the model dataset

[0058] Step 2: The constructed Bidirectional Long Short-Term Memory (Bi-LSTM) network is a type of Recurrent Neural Network (RNN). It combines the outputs of two directional RNNs, one processing the sequence forward and the other processing the sequence backward, effectively capturing long-term dependencies in sequence data. The main process is shown in Figure 1. The amino acid sequence of the peptide is input into the Bi-LSTM neural network. By learning and summarizing the features of peptides in the training set and validating the peptides in the validation set, the output is "0" and "1". "0" indicates that the peptide does not have antioxidant activity, and "1" indicates that the peptide has antioxidant activity, thus obtaining an antioxidant peptide prediction model. The Bi-LSTM deep learning model achieves an accuracy of 0.8969 on a 5x cross-validation set. The model's accuracy (0.9043), precision (0.9767), sensitivity (0.8284), specificity (0.9802), MCC (0.8181), and AUC (0.9043) on the independent test dataset all demonstrate its excellent predictive ability. The accuracy evaluation of the Bi-LSTM deep learning model is shown in Table 2.

[0059] Table 2 Accuracy of Bi-LSTM Model

[0060] Step 3: Search for and obtain the amino acid sequence of black bean protein from the National Center for Biotechnology Information (NCBI) website (https: / / www.ncbi.nlm.nih.gov). Use different enzymes in the Novo Pro Labs / (https: / / www.novoprolabs.com / ) virtual enzymatic hydrolysis program to virtually hydrolyze the raw material protein, including alkaline protease, trypsin, Arg-C protease, chymotrypsin, Asp-N endopeptidase, and pepsin (pH>1.3), to obtain peptides after virtual enzymatic hydrolysis of the raw material protein with different enzymes.

[0061] Step 4: Use the antioxidant peptide prediction model obtained in Step 2 to predict the antioxidant activity of the peptides obtained in Step 3, and calculate the percentage of peptides with antioxidant activity. Use the enzyme with the highest percentage for the actual enzymatic hydrolysis of the raw protein to obtain antioxidant peptides. The results are shown in Table 3.

[0062] Table 3. Number of antioxidant peptides obtained from virtual enzymatic digestion of black soybean protein

[0063] Step 5: To verify the accuracy of the deep learning model, a subset of the predicted peptides with antioxidant activity were selected for ABTS and DPPH free radical scavenging experiments to verify their antioxidant activity. The results are shown in Figure 4.

[0064] Step 6: Import the antioxidant peptide sequence obtained in Step 5 into ACD / Labs software to calculate the logP of the compound. Generally, logP greater than 0 indicates lipophilicity, and the higher the value, the greater the lipophilicity. Table 4 shows the calculated lipophilicity values ​​of some antioxidant peptides.

[0065] Table 4. Calculation of lipophilicity of antioxidant peptides

[0066] Step 7: Use the trypsin selected in Step 3, which yields more antioxidant peptides, to perform actual enzymatic hydrolysis of black soybean protein, and then identify, separate, and purify the antioxidant peptides. The preparation of antioxidant peptides is carried out in accordance with a method for preparing amphiphilic antioxidant peptides and the method in its application disclosed in patent document CN117567551A.

[0067] Step 8: Randomly select 4 antioxidant peptides from Step 7 to conduct an experiment on the inhibitory effect of lipid oxidation. Before use, add the pre-prepared natural antioxidant peptides to the vegetable oil at an addition amount of 0.01% of the total amount of oil, stir evenly, and after fully dissolving, heat the vegetable oil to about 180℃ and continue heating for 2-12 hours. Determine the inhibition rate of the antioxidant peptides on the formation of primary and secondary oxidation products of vegetable oil and their inhibitory effect on harmful volatile aldehydes.

[0068] Specifically, step 7 includes:

[0069] Step 7-1: Determination of the inhibition rate of primary oxidation products;

[0070] The peptide was dissolved in oil, with BHT synthesized antioxidant as a positive control. The solution was heated to 180℃ for 6 hours. After heating, 10 μL was placed in a 96-well plate, and 200 μL of chloroform-methanol solution (7:3), 20 μL of 3% ammonium thiocyanate solution, and 20 μL of 2 mM ferrous chloride solution were added sequentially. The plate was incubated at room temperature for 3 minutes, and the absorbance was measured at 490 nm using a microplate reader. This absorbance was recorded as the peroxide value (A). The inhibition rate (%) of the sample against primary oxidation products of oil was calculated using the following formula:

[0071] Where: I is the inhibition rate; A0 is the absorbance value of the oil; A is the absorbance value of the oil after the addition of the amphiphilic peptide;

[0072] Step 7-2: Determination of the inhibition rate of secondary oxidation products;

[0073] Weigh 37.5 mg of thiobarbituric acid and 1.5 g of trichloroacetic acid, dissolve them in deionized water, add 100 μL of hydrochloric acid, and bring the volume to 10 mL with deionized water. Take 50 μL of the heated oil and add 100 μL of thiobarbituric acid reagent, mix well, and place in a boiling water bath for 15 min. Then place in ice water for 10 min to cool to room temperature. Centrifuge at 2000 r / min for 10 min, collect the supernatant, and measure the absorbance at 532 nm using a microplate reader. The inhibition rate (%) of the sample against secondary oxidation products of the oil is calculated using the following formula:

[0074] Where: I is the inhibition rate; A0 is the absorbance value of the oil; A is the absorbance value of the oil after the addition of the amphiphilic antioxidant peptide;

[0075] Step 7-3: Determination of the content of harmful volatile aldehydes in frying oil:

[0076] Blank control: Take 100g of blank antibiotic-free rapeseed oil sample, place it in a 500ml beaker, place it on a heatable magnetic stirrer, heat it to 180±5℃ within 10min, divide it into 4 groups of heating for 2h, 4h, 6h and 8h, and take oil samples at the end of each frying to determine the content of oxidized α,β unsaturated aldehydes in the oil sample.

[0077] Peptide-added frying test: Before heating, 100 ppm of the prepared antioxidant peptide was added to 100 g of sample oil to prepare the oil sample for the frying test, with BHT synthetic antioxidant as a positive control. The frying test was conducted according to the frying test method described in the blank control.

[0078] Accurately weigh 2g of oxidized oil sample into a 20mL headspace vial, add a stir bar and mix thoroughly; place the sample in a constant temperature device at 50℃ for 15min, then insert the solid phase microextraction head into the sample vial and extract for 60min. After extraction, insert it into the GC inlet and desorb for 10min.

[0079] Gas phase conditions: Carrier gas was He (>99.999%), flow rate was 0.8 mL / min. Column oven temperature program: initial temperature 40℃, hold for 6 min, increase to 80℃ at 4℃ / min; then increase to 86℃ at 2℃ / min, hold for 3 min; then increase to 120℃ at 4℃ / min, hold for 2 min; finally increase to 250℃ at 12℃ / min, hold for 2 min. Injector: 230℃, split ratio 5:1. Column: HP-5MS conventional capillary column (30 m × 0.25 mm × 0.25 μm).

[0080] Mass spectrometry conditions: transfer line 280℃, quadrupole 150℃, ion source 230℃; electron impact (EI) ion source; electron quantization 70 eV; solvent delay 3 min; full scan mode; mass scan range m / z 35~500.

[0081] The total amount of harmful volatile aldehydes after frying is shown in Table 5, and the inhibition of the formation of harmful volatile aldehydes by antioxidant peptides is shown in Table 6.

[0082] Table 5 Total amount of harmful volatile aldehydes after frying oils

[0083] Table 6. Inhibition of harmful volatile aldehyde formation in oils by antioxidant peptides.

[0084] As shown above, the four oil-based antioxidant peptides not only exhibited better inhibitory effects on primary and secondary oxidation products of vegetable oils than the synthetic antioxidant BHT, but also reduced the total oxidized α,β-unsaturated aldehydes to varying degrees under the same frying time. Among them, YAP showed relatively better antioxidant effects at most time points, and its total oxidized α,β-unsaturated aldehyde content was lower than that of the control group CON and BHT in most cases. The antioxidant effects of IFS and BHT were similar; under different frying times, their total oxidized α,β-unsaturated aldehyde content was reduced to some extent relative to CON, but not as much as YAP. The antioxidant effects of IFQ and YPL were less stable, but overall they demonstrated a certain antioxidant effect.

[0085] In summary, trypsin hydrolysis of black soybean protein is the optimal strategy for producing and screening antioxidant peptides in oils. The validated peptide sequences are highly consistent with the theoretical antioxidant peptide effects. The oil antioxidant properties of the peptides obtained through enzymatic hydrolysis have also been well demonstrated. Therefore, the method for obtaining antioxidant peptides in oils based on deep learning and virtual enzymatic hydrolysis provided by this invention not only provides guidance for the selection of hydrolytic proteases but also obtains peptide sequences with antioxidant activity, enabling the screening of high-performance oil-based antioxidant peptides.

[0086] Furthermore, the present invention provides a computer device, which may include a processor, a memory, a network interface, and a database connected via a system bus. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The network interface of the computer device is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it causes the processor to perform the steps of a method for screening antioxidant peptides in oils based on artificial intelligence, as described in any of the above embodiments.

[0087] The working process, working details and technical effects of the computer device provided in this embodiment can be found in the embodiment above regarding a method for screening antioxidant peptides in oils based on artificial intelligence, and will not be repeated here.

[0088] Furthermore, the present invention also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of a method for screening antioxidant peptides in oils based on artificial intelligence as described in any of the above embodiments. The computer-readable storage medium refers to a data storage carrier, which may include, but is not limited to, floppy disks, optical disks, hard disks, flash memory, USB flash drives, and / or memory sticks, etc. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.

[0089] The working process, working details, and technical effects of the computer-readable storage medium provided in this embodiment can be found in the above embodiment regarding a method for screening antioxidant peptides in oils based on artificial intelligence, and will not be repeated here.

[0090] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM).

[0091] Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for screening antioxidant peptides in oils based on artificial intelligence-assisted screening, characterized in that, The method includes: Step S1: Collect antioxidant peptides from multiple databases, delete duplicate data and remove samples with inconsistent experimental results to obtain a positive dataset containing antioxidant peptides, randomly extract an equal amount of peptide sequences matching the length of the positive dataset as a negative control, and divide the positive and negative sample datasets into training and test sets. Step S2: Input the dataset from step S1 into a bidirectional long short-term memory network, use the training set to learn and summarize the features of antioxidant peptides, and use the validation set data to validate the model to obtain an antioxidant peptide prediction model. Step S3: Obtain the amino acid sequence of the raw protein from the NCBI website, and use different enzymes in the Novo Pro Labs virtual enzymatic digestion program to perform virtual enzymatic digestion on the raw protein to obtain peptides after virtual enzymatic digestion of the raw protein with different enzymes. Step S4: Use the antioxidant peptide prediction model obtained in step S2 to predict the antioxidant activity of the peptides obtained in step S3, and calculate the percentage of peptides with antioxidant activity. Use the enzyme with the highest percentage for the actual enzymatic hydrolysis of the raw material protein to obtain antioxidant peptides. Step S5: Randomly select the antioxidant peptides obtained in step S4 to determine the ABTS and DPPH free radical scavenging activities to verify the antioxidant activity; Step S6: Import the antioxidant peptide sequence obtained in step S5 into ACD / Labs software to calculate the logP of the compound; Step S7: Use the enzymes selected in step S3 that can yield more antioxidant peptides to perform actual enzymatic digestion of the protein, and identify, separate and purify the antioxidant peptides. The antioxidant peptides prepared in steps S8 and S7 were added to vegetable oil and subjected to heating and frying simulations. The inhibition rates of the antioxidant peptides on the formation of primary and secondary oxidation products after heat processing of vegetable oil and their inhibitory effects on the formation of harmful volatile aldehydes were determined.

2. The method for screening antioxidant peptides in oils based on artificial intelligence assisted screening according to claim 1, characterized in that, In step S1, each peptide contains 2-20 amino acids, and the positive and negative sample datasets are randomly divided into training and test sets in a ratio of 8:

2.

3. The method for screening antioxidant peptides in oils based on artificial intelligence assisted screening according to claim 1, characterized in that, In step S2, the characteristics of peptides in the training set are learned and summarized, and the peptides in the validation set are validated. The output is "0" and "1", where "0" indicates that the peptide does not have antioxidant activity and "1" indicates that the peptide has antioxidant activity.

4. The method for screening antioxidant peptides in oils based on artificial intelligence assisted screening according to claim 1, characterized in that, In step S3, the length of the polypeptide sequence after virtual enzymatic hydrolysis of the raw material protein by different enzymes is less than 20.

5. The method for screening antioxidant peptides in oils based on artificial intelligence assisted screening according to claim 1, characterized in that, In step S4, the peptide is judged to have antioxidant activity based on the output of the antioxidant peptide prediction model. "0" indicates that the peptide does not have antioxidant activity, and "1" indicates that the peptide has antioxidant activity.

6. The method for screening antioxidant peptides in oils based on artificial intelligence assisted screening according to claim 1, characterized in that, In step S6, logP greater than 0 indicates fat solubility, and the higher the value, the greater the fat solubility.

7. The method for screening antioxidant peptides in oils based on artificial intelligence assisted screening according to claim 1, characterized in that, In step S8, the types of vegetable oil include, but are not limited to, one or more of the following: peanut oil, sunflower seed oil, palm oil, soybean oil, olive oil, rapeseed oil, cottonseed oil, perilla seed oil, rice bran oil, flaxseed oil, safflower seed oil, tea seed oil, palm fruit oil, coconut oil, cocoa soybean oil, almond oil, tung oil seed oil, corn germ oil, wheat germ oil, sesame seed oil, pumpkin seed oil, walnut oil, grape seed oil, flaxseed oil, pumpkin seed oil, cocoa butter, and algae oil.

8. The method for screening antioxidant peptides in oils based on artificial intelligence assisted screening according to claim 1, characterized in that, In step S2, the bidirectional long short-term memory network merges the outputs of two directed RNNs, one processing the sequence forward and the other processing the sequence backward.

9. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method for screening oil antioxidant peptides based on artificial intelligence as described in any one of claims 1-8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the method for screening oil antioxidant peptides based on artificial intelligence as described in any one of claims 1-8.

Images