A milk quality detection system and method based on gene sequence
Through a gene sequence-based detection system, using technologies such as gene sequencing and deep learning, the problem of time-consuming, labor-intensive and accurate detection methods of existing milk quality detection methods is solved, and fast and accurate milk quality evaluation and optimization control are achieved, which improves the Detection efficiency and stability.
Patent Information
- Application Number
- CN202311849502.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-12-29
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2043-12-29
AI Technical Summary
The existing milk quality detection methods have problems such as time-consuming and labor-intensive, environmental pollution, complex operation and difficulty in obtaining accurate detection results, especially chemical analysis and biological analysis methods, and spectral analysis requires pretreatment of milk to affect the original state.
Adopt a detection system based on gene sequences, including gene sequencing module, gene database module, intelligent matching module, quality evaluation module, control optimization module and control execution module. Through technologies such as gene sequencing, deep learning, linear regression and genetic algorithms, rapid , accurate milk quality detection and optimization control.
It realizes rapid and accurate evaluation and optimization control of milk quality, improves detection efficiency, accuracy and comparability, and can dynamically monitor and intelligently adjust milk quality to ensure its nutritional value, safety and taste stability.
Smart Images

Figure CN117805326B_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of food quality detection, and in particular relates to a milk quality detection system and method based on gene sequences. Background Art
[0002] Milk is one of the important food sources for human beings. Its quality directly affects human health and nutrition. The quality parameters of milk include protein content, fat content, lactic acid bacteria content and antibiotic residues, etc. These parameters reflect the nutritional value, freshness, safety and quality of milk. Therefore, accurate, fast and convenient testing and evaluation of milk quality is of great significance to ensure the production, processing, circulation and consumption of milk.
[0003] At present, the quality detection of milk mainly adopts chemical analysis, biological analysis, spectral analysis and other methods. Although these methods can provide relatively reliable test results, they also have some disadvantages, such as:
[0004] Chemical analysis methods require a large amount of reagents, instruments and manpower, which is time-consuming and labor-intensive, and may cause certain environmental pollution;
[0005] Biological analysis methods require the cultivation and identification of microorganisms, which takes a long time, is complex to operate, and is affected by environmental conditions;
[0006] The spectral analysis method requires pretreatment of milk, such as dilution, filtration, homogenization, etc., which affects the original state of the milk and is interfered by the complex composition and structure of milk, making it difficult to obtain accurate spectral signals.
[0007] Therefore, it is necessary to develop a new milk quality detection system that can overcome the above limitations and achieve rapid determination and optimal control of milk quality parameters. Summary of the invention
[0008] In view of the above problems, the present invention provides a milk quality detection system based on gene sequences, which can quickly, accurately and comprehensively evaluate the quality of milk, and perform corresponding control and optimization according to the evaluation results to improve the nutritional value, safety and taste of milk. At the same time, it can also trace and analyze the causes and links of milk quality differences to find out the source and influencing factors of quality problems, and provide a basis for quality improvement and prevention. The system includes the following modules:
[0009] A gene sequencing module, used for determining the gene sequence in the milk sample to obtain first gene sequence data;
[0010] A gene database module, used for storing known second gene sequence data related to milk quality and its corresponding standard quality parameters;
[0011] An intelligent matching module, used to match the first gene sequence data obtained by the gene sequencing module with the second gene sequence data in the gene database module to obtain a matching result;
[0012] The quality assessment module is used to calculate the quality parameters of the milk, including protein content, fat content, lactic acid bacteria content and antibiotic residue, based on the matching results, and to assess the quality of the milk to obtain a quality assessment result;
[0013] The control optimization module is used to generate corresponding control optimization strategies according to the quality assessment results, including adjusting the temperature, acidity and oxidation degree of milk, and outputting control signals;
[0014] The control execution module is used to receive the control signal output by the control optimization module and perform corresponding control operations on the milk.
[0015] The beneficial effects of this implementation scheme are that, through the gene sequencing module, the gene sequence data in the milk sample can be obtained quickly and accurately without complex chemical analysis or microbial culture; through the gene database module, the existing gene knowledge base can be used to scientifically evaluate and standardize the milk quality, thereby improving the comparability and traceability of the milk quality; through the intelligent matching module, the deep learning method can be used to efficiently encode, extract features and calculate similarity of the gene sequence data, thereby improving the matching accuracy and sensitivity of the milk quality; through the quality assessment module, the linear regression and fuzzy logic methods can be used to calculate and evaluate the quality parameters of the milk, thereby improving the quantification and objectivity of the milk quality; through the control optimization module, the genetic algorithm and reinforcement learning methods can be used to optimize the quality parameters of the milk and generate strategies, thereby improving the controllability and adaptability of the milk quality; through the control execution module, the milk can be subjected to corresponding control operations according to the control signal, thereby improving the stability and consistency of the milk quality.
[0016] In a preferred embodiment, the system also includes a gene editing module for performing targeted editing on the gene sequence in the milk according to the quality assessment results, and transmitting the results of the targeted editing to the control optimization module.
[0017] Under preferred implementation conditions, the beneficial effect of this embodiment is that the gene sequence in milk is modified in a targeted manner using gene editing technology, thereby improving the quality parameters of milk, such as increasing protein content, reducing fat content, increasing lactic acid bacteria content or reducing antibiotic residues; the results of targeted editing can also be fed back to the control optimization module, thereby adjusting the control optimization strategy to make it more suitable for the actual situation of the milk and improve the control effect and efficiency.
[0018] In a preferred implementation, the gene sequencing module of this system specifically includes:
[0019] The sample processing submodule is used to purify, lyse and amplify the milk sample to obtain DNA fragments in the milk;
[0020] A sequencing chip submodule is used to fix the DNA fragments on the sequencing chip and perform high-throughput sequencing to obtain the first gene sequence data in the milk;
[0021] The data transmission submodule is used to transmit the first gene sequence data to the intelligent matching module via a wireless network or a wired network.
[0022] Under the best implementation conditions, the beneficial effects of this embodiment are as follows: the sample processing submodule is used to effectively pre-process the milk sample, thereby improving the quality and quantity of DNA fragments and increasing the accuracy and sensitivity of gene sequencing; the sequencing chip submodule is used to perform high-throughput sequencing on the DNA fragments, thereby improving the speed and coverage of gene sequencing and increasing the amount of information and representativeness of gene sequencing; the data transmission submodule is used to quickly and securely transmit the first gene sequence data, thereby improving the real-time and reliability of gene sequencing and reducing the risk of data loss and damage.
[0023] In a preferred implementation, the intelligent matching module of the system specifically includes:
[0024] An encoding submodule, used to encode the first gene sequence data and the second gene sequence data respectively to obtain an encoded first gene sequence vector and a second gene sequence vector;
[0025] A feature extraction submodule, used to input the encoded first gene sequence vector and the second gene sequence vector into a pre-trained deep neural network respectively, to obtain a feature vector of the first gene sequence and a feature vector of the second gene sequence;
[0026] The similarity calculation submodule is used to calculate the cosine similarity between the first gene sequence data and the second gene sequence data according to the feature vector of the first gene sequence and the feature vector of the second gene sequence to obtain a similarity matrix. Specifically, the cosine similarity between the first gene sequence data and the second gene sequence data is calculated using the following formula:
[0027]
[0028] Where a and b represent the feature vector of the first gene sequence and the feature vector of the second gene sequence, respectively. n is the dimension of the feature vector. cos(θ) is the cosine similarity between the two feature vectors, indicating the similarity between the two gene sequences. θ is the angle between the two feature vectors, reflecting the difference between the two gene sequences. ||a|| and ||b|| are the moduli of the two feature vectors, indicating the size of the two feature vectors, respectively. a i and b i are the eigenvalues of the eigenvector of the first gene sequence and the eigenvector of the second gene sequence in the i-th dimension respectively;
[0029] The matching judgment submodule is used to perform threshold judgment on the similarity matrix, and judge whether the first gene sequence data matches the second gene sequence data according to a preset threshold, so as to obtain a matching result.
[0030] Under the best implementation conditions, the beneficial effects of this embodiment are that the encoding submodule is used to effectively encode the gene sequence data, thereby reducing the dimension and complexity of the data and increasing the processability and comparability of the data; the feature extraction submodule is used to perform deep learning on the gene sequence data, thereby extracting the key features of the gene sequence and increasing the amount of information and discrimination of the data; the similarity calculation submodule is used to perform accurate similarity calculation on the gene sequence data, thereby reflecting the similarity and difference of the gene sequences and increasing the interpretability and evaluability of the data; the matching judgment submodule is used to perform intelligent matching judgment on the gene sequence data, thereby quickly judging whether the gene sequence matches according to the preset threshold, and increasing the decisionability and operability of the data.
[0031] In a preferred implementation, the quality assessment module of the system specifically includes:
[0032] The parameter calculation submodule is used to calculate the quality parameters of milk by linear regression according to the matching results to obtain the parameter calculation results;
[0033] A parameter comparison submodule is used to compare the parameter calculation results with the standard quality parameters stored in the gene database module to obtain parameter comparison results;
[0034] The evaluation and judgment submodule is used to evaluate the quality of milk based on the parameter comparison results and use fuzzy logic to obtain the quality evaluation result.
[0035] Under the best implementation conditions, the beneficial effects of this embodiment are that the parameter calculation submodule is used to effectively calculate the quality parameters of milk, so that according to the matching results, the parameters such as the protein content, fat content, lactic acid bacteria content and antibiotic residue of the milk are obtained, thereby increasing the quality indicators and quantifiability of the milk; the parameter comparison submodule is used to effectively compare the quality parameters of milk, so that the parameter calculation results are compared with the standard quality parameters stored in the gene database module to obtain parameter comparison results, thereby increasing the quality reference and evaluability of the milk; the evaluation and judgment submodule is used to effectively evaluate the quality of milk, so that according to the parameter comparison results, fuzzy logic is used to make a comprehensive judgment on the quality of milk, and a quality evaluation result is obtained, thereby increasing the quality grade and optimizability of the milk.
[0036] In a preferred implementation, the control optimization module of the system specifically includes:
[0037] The parameter optimization submodule is used to optimize the quality parameters of milk using a genetic algorithm according to the quality evaluation results to obtain parameter optimization results;
[0038] The strategy generation submodule is used to generate corresponding control optimization strategies according to the parameter optimization results, including adjusting the temperature, acidity and oxidation degree of milk;
[0039] The strategy evaluation submodule uses reinforcement learning to evaluate the control optimization strategy and obtain the strategy evaluation result;
[0040] The strategy update submodule is used to update the control optimization strategy according to the strategy evaluation results to improve the control effect, and output the control signal to the control execution module.
[0041] Under the best implementation conditions, the beneficial effects of this implementation scheme are as follows: the quality parameters of milk are effectively optimized by using the parameter optimization submodule, and then the quality parameters of milk are adaptively adjusted by using the genetic algorithm according to the quality evaluation results, so as to obtain the parameter optimization results, thereby increasing the quality improvement and improveability of milk; the control optimization strategy of milk is effectively generated by using the strategy generation submodule, and then a reasonable control optimization strategy is generated according to the parameter optimization results, including adjusting the temperature, acidity and oxidation degree of milk, and increasing the control scheme and selectivity of milk; the control optimization strategy is effectively evaluated by using the strategy evaluation submodule, and then the control optimization strategy is fed back and learned by using reinforcement learning, and then the strategy evaluation results are obtained, thereby increasing the control effect and optimizability of milk; the control optimization strategy is effectively updated by using the strategy update submodule, and then the control optimization strategy is dynamically adjusted according to the strategy evaluation results to improve the control effect, and the control signal is output to the control execution module, thereby increasing the control implementation and executability of milk.
[0042] In the best implementation, the system also includes a gene tracing module, which is used to trace and analyze the causes and links of differences in milk quality, so as to find out the source and influencing factors of quality problems and provide a basis for quality improvement and prevention.
[0043] Under the best implementation conditions, the beneficial effect of this embodiment is that the quality differences of milk can be effectively traced by using the gene tracing module so as to carry out quality improvement and preventive measures; the system can improve the traceability and improveability of milk quality and reduce the probability of quality problems and the degree of loss.
[0044] In a preferred implementation, the gene tracing module of this system specifically includes:
[0045] A gene comparison submodule is used to compare the first gene sequence data obtained by the gene sequencing module with the second gene sequence data in the gene database module to obtain a comparison result, including the similarity, difference and variation of the gene sequences;
[0046] A gene difference submodule, for identifying differences between the first gene sequence data and the second gene sequence data, including insertion, deletion, substitution or mutation, based on the comparison results obtained by the gene comparison submodule, and analyzing the effects of the differences on milk quality, including changes in protein content, fat content, lactic acid bacteria content or antibiotic residues;
[0047] A gene tracing submodule, for tracing the source information of the first gene sequence data in the milk based on the difference points obtained by the gene difference submodule and using the second gene sequence data stored in the gene database module, and analyzing the influence of the source information on the quality of the milk, including the influence on the nutritional value, safety and taste of the milk;
[0048] The gene reporting submodule is used to generate a gene tracing report based on the source information obtained by the gene tracing submodule, including the comparison results of the gene sequences, differences, source information, influencing factors and improvement suggestions, and send the gene tracing report to the control optimization module and the control execution module for quality improvement and preventive measures.
[0049] Under the best implementation conditions, the beneficial effect of this embodiment is that the gene tracing module is used to effectively trace the quality differences of milk, so as to identify the differences between the gene sequence in the milk and the standard gene sequence in the gene database according to the quality assessment results, and analyze the impact of the differences on the quality of milk, including changes in protein content, fat content, lactic acid bacteria content or antibiotic residues; the source information of the gene sequence in the milk is traced by using the standard gene sequence in the gene database, and the impact of the source information on the quality of milk is analyzed, including the impact on the nutritional value, safety and taste of the milk; the system can also generate a gene tracing report, including the comparison results, differences, source information, influencing factors and improvement suggestions of the gene sequence, and send the gene tracing report to the control optimization module and the control execution module for quality improvement and preventive measures; the system can improve the traceability and improveability of milk quality, and reduce the probability of quality problems and the degree of loss.
[0050] This embodiment also provides a method for detecting milk quality based on gene sequences, comprising the following steps:
[0051] Step 1, measuring the gene sequence in the milk sample to obtain first gene sequence data;
[0052] Step 2, matching the first gene sequence data with the second gene sequence data related to milk quality stored in the gene database to obtain a matching result;
[0053] Step 3, according to the matching results, the quality parameters of the milk are calculated, including protein content, fat content, lactic acid bacteria content and antibiotic residue, and the quality of the milk is evaluated to obtain a quality evaluation result;
[0054] Step 4: Generate a corresponding control optimization strategy based on the quality assessment results, including adjusting the temperature, acidity and oxidation degree of the milk to obtain a control signal;
[0055] Step 5: Perform corresponding control operations on the milk according to the control signal.
[0056] The beneficial effects of this implementation scheme are: using gene sequences as indicators of milk quality, thereby achieving rapid, accurate and non-destructive detection of milk quality, and improving detection efficiency and accuracy; using gene matching, quality assessment, control optimization and other steps, thereby achieving dynamic monitoring and intelligent adjustment of milk quality, and improving milk quality and stability; using control execution modules, thereby achieving automated and precise control and optimization of milk quality, and improving the efficiency and quality of milk production and processing. BRIEF DESCRIPTION OF THE DRAWINGS
[0057] Figure 1This is a schematic diagram of the system structure in Embodiment 1 of the present invention;
[0058] Figure 2 Schematic diagram of the system structure in Embodiment 2 of the present invention;
[0059] Figure 3 is a schematic diagram of the system structure in Example 7 of the present invention;
[0060] Figure 4 This is a flow chart of the method in Example 8 of the present invention. DETAILED DESCRIPTION
[0061] In order to enable those skilled in the art to better understand the technical solution, the technical solution is described in detail below in conjunction with the embodiments. The description in this section is only exemplary and explanatory and should not have any limiting effect on the scope of protection of the present invention.
[0062] Embodiment 1:
[0063] This embodiment provides a milk quality detection system based on gene sequence, and its structural diagram is shown in FIG. Figure 1 As shown, it includes the following modules:
[0064] The gene sequencing module is used to measure the gene sequence in the milk sample and obtain the first gene sequence data. The function of this module is to extract DNA from the milk sample and use high-throughput sequencing technology to obtain the gene sequence information in the milk. This information is an important indicator reflecting the source and characteristics of the milk, and is also an important basis for evaluating and controlling the quality of the milk.
[0065] The gene database module is used to store known second gene sequence data related to milk quality and its corresponding standard quality parameters. The function of this module is to establish a gene database, collect and organize various gene sequence data related to milk quality, including gene sequences of different breeds of dairy cows, gene sequences of different types of milk, gene sequences of milk with different processing technologies, etc., as well as the standard quality parameters corresponding to these gene sequence data, including protein content, fat content, lactic acid bacteria content, and antibiotic residues, etc., to provide data support for gene matching and quality assessment.
[0066] The intelligent matching module is used to match the first gene sequence data obtained by the gene sequencing module with the second gene sequence data in the gene database module to obtain a matching result. The function of this module is to use artificial intelligence technology to intelligently match the first gene sequence data obtained by the gene sequencing module with the second gene sequence data in the gene database module to find the most similar gene sequence and the degree of similarity, thereby determining the source and type of milk and providing a basis for quality assessment.
[0067] The quality assessment module is used to calculate the quality parameters of milk, including protein content, fat content, lactic acid bacteria content and antibiotic residue, according to the matching results, and evaluate the quality of milk to obtain quality assessment results. The function of this module is to calculate the quality parameters of milk, including protein content, fat content, lactic acid bacteria content and antibiotic residue, according to the matching results obtained by the intelligent matching module, using mathematical models. These parameters are important indicators reflecting the nutrition, safety and taste of milk, and are also important bases for controlling and optimizing the quality of milk. The module also uses fuzzy logic to evaluate the quality of milk and obtain quality assessment results, including excellent, good, medium and poor grades, to provide a reference for control optimization.
[0068] The control optimization module is used to generate corresponding control optimization strategies according to the quality assessment results, including adjusting the temperature, acidity and oxidation degree of milk, and outputting control signals. The function of this module is to design control optimization strategies according to the quality assessment results obtained by the quality assessment module to adjust the temperature, acidity and oxidation degree of milk. These parameters are important factors affecting the shelf life and flavor of milk, and are also important means to control and optimize the quality of milk. According to the deviation between the preset target value and the actual value, the module can use feedback control, adaptive control or other suitable control strategies to calculate the appropriate control signal and send it to the control execution module, so that the milk can run according to the expected target and achieve the best quality effect, while also avoiding or reducing the deterioration and contamination of milk.
[0069] The control execution module is used to receive the control signal output by the control optimization module and perform corresponding control operations on the milk. The function of this module is to receive the control signal output by the control optimization module, control the temperature, acidity and oxidation degree of the milk, and perform corresponding control operations on the milk by heating, cooling, acidification, anti-oxidation, etc., so that the quality of the milk reaches the best state and meets the needs and preferences of consumers.
[0070] This embodiment adopts a milk quality detection system based on gene sequence, which can realize a rapid, accurate and intelligent detection and control process of milk quality, improve the quality and benefit of milk, and reduce the risk and loss of milk.
[0071] Embodiment 2:
[0072] like Figure 2 As described above, this embodiment provides a milk quality detection system based on gene sequences, which, compared with Example 1, adds a gene editing module to achieve targeted editing of gene sequences in milk and improve the quality and characteristics of milk.
[0073] The gene editing module is used to perform targeted editing on the gene sequence in milk according to the quality assessment results, and transmit the results of the targeted editing to the control optimization module. The function of this module is to use gene editing technology to perform targeted editing on the gene sequence in milk according to the quality assessment results obtained by the quality assessment module, thereby changing the expression or function of certain genes in milk to improve the quality and characteristics of milk, such as increasing the protein content in milk, reducing the fat content in milk, increasing the lactic acid bacteria content in milk, or reducing the antibiotic residue in milk.
[0074] Specifically, the gene editing module includes:
[0075] The editing target submodule is used to determine the gene editing targets in milk based on the quality assessment results, that is, the gene sequences that need to be edited and their corresponding editing methods, such as insertion, deletion, substitution or mutation.
[0076] The editing execution submodule is used to edit the gene sequence in milk using gene editing technology according to the gene editing target determined by the editing target submodule to obtain the edited gene sequence. This submodule can use different gene editing technologies, such as CRISPR-Cas9, TALEN, ZFN, etc., and select appropriate editing tools and parameters according to different editing targets to achieve accurate and efficient editing of the gene sequence in milk.
[0077] The editing result submodule is used to analyze and evaluate the editing results based on the edited gene sequence obtained by the editing execution submodule, determine the effect and impact of the editing, and transmit the editing results to the control optimization module. This submodule can use technologies such as gene sequencing, gene chips, and gene expression analysis to verify and detect the edited gene sequence, evaluate the accuracy and stability of the editing, and the impact on the quality and characteristics of milk.
[0078] This embodiment provides a milk quality detection system based on gene sequences. The system uses gene editing technology to perform targeted editing of gene sequences in milk according to quality assessment results, thereby improving the quality and characteristics of milk and meeting the needs and preferences of different consumers.
[0079] Embodiment 3:
[0080] This embodiment provides a milk quality detection system based on gene sequence. Based on Embodiment 1, this embodiment specifically implements the gene sequencing module.
[0081] The gene sequencing module specifically includes:
[0082] The sample processing submodule is used to purify, lyse and amplify milk samples to obtain DNA fragments in milk. The function of this submodule is to extract DNA from milk samples, purify, lyse and amplify DNA using purification reagents, lysate and PCR reagents to obtain DNA fragments of sufficient quantity and quality. These fragments are an important source of genetic information in milk and an important basis for gene sequencing.
[0083] The sample processing submodule may use a magnetic bead-based DNA extraction method, which includes the following steps:
[0084] (1) adding a certain amount of magnetic beads to a milk sample, wherein the surface of the magnetic beads is coated with probes that can specifically bind to DNA in the milk to form a magnetic bead-DNA complex;
[0085] (2) using a magnetic device to separate the magnetic bead-DNA complex from the milk sample and remove impurities such as protein and fat in the milk;
[0086] (3) adding a certain amount of lysis solution to the magnetic bead-DNA complex, wherein the lysis solution contains an enzyme that can destroy the magnetic bead-DNA complex and release DNA fragments;
[0087] (4) A thermal cycler is used to amplify the DNA fragments. The amplification solution contains polymerase, primers and dNTPs that can replicate the DNA fragments. Through a certain temperature cycle, the number of DNA fragments is increased to obtain amplified DNA fragments.
[0088] The sequencing chip submodule is used to fix the DNA fragments on the sequencing chip and perform high-throughput sequencing to obtain the first gene sequence data in milk. The function of this submodule is to fix the DNA fragments on the sequencing chip through microfluidics technology, and use fluorescent sequencing technology, such as the Illumina platform, to perform high-throughput sequencing on the DNA fragments to obtain a large number of short fragment sequences, namely the first gene sequence data. These data are important references for reflecting the gene information in milk and are also an important basis for gene comparison and tracing.
[0089] The sequencing chip submodule can adopt a sequencing method based on fluorescence signals, which includes the following steps:
[0090] (1) annealing the amplified DNA fragment with a sequencing primer that is complementary to one end of the DNA fragment to form a single-stranded DNA;
[0091] (2) Fixing the single-stranded DNA on a sequencing chip. The sequencing chip has many micropores, and only one single-stranded DNA is fixed in each micropore to form a sequencing point;
[0092] (3) Add a fluorescently labeled dNTPs to the sequencing chip. The fluorescently labeled dNTPs complement the other end of the single-stranded DNA to form double-stranded DNA and emit a specific fluorescent signal. The color of the fluorescent signal corresponds to the base type of the dNTPs.
[0093] (4) using a fluorescence detector to record the fluorescence signal of each sequencing point, and determining the base type of each sequencing point according to the color of the fluorescence signal, thereby obtaining the gene sequence of each sequencing point;
[0094] (5) The gene sequences of all sequencing points are spliced together to form a complete gene sequence, which is output as the first gene sequence data in the milk.
[0095] The data transmission submodule is used to transmit the first gene sequence data to the intelligent matching module via a wireless network or a wired network.
[0096] The present embodiment provides a milk quality detection system based on gene sequence. The system can extract DNA fragments from milk samples quickly, sensitively and effectively by utilizing the magnetic bead method, thereby avoiding DNA degradation and contamination and improving the performance and quality of the sample processing submodule. By utilizing the fluorescence method, the DNA fragments can be sequenced with high throughput and high precision to obtain gene sequence data in milk, thereby improving the efficiency and accuracy of the sequencing chip submodule. By utilizing the cloud computing method, the gene sequence data can be safely and quickly transmitted to the intelligent matching module, thereby improving the security and speed of the data transmission submodule.
[0097] Embodiment 4:
[0098] This embodiment provides a milk quality detection system based on gene sequence. Based on Embodiment 1, this embodiment specifically implements the intelligent matching module.
[0099] The intelligent matching module specifically includes:
[0100] The encoding submodule is used to encode the first gene sequence data and the second gene sequence data respectively to obtain the encoded first gene sequence vector and the second gene sequence vector. The function of this submodule is to convert the first gene sequence data and the second gene sequence data into a numerical vector for subsequent feature extraction and similarity calculation.
[0101] The encoding submodule may adopt an encoding algorithm based on a bidirectional long short-term memory network (Bi-LSTM), which includes the following steps:
[0102] (1) converting the first gene sequence data and the second gene sequence data into strings consisting of four letters (A, T, C, G) respectively, representing the nucleotide composition of the gene sequence;
[0103] (2) Represent each letter in the string with a four-dimensional one-hot vector, that is, A is [1,0,0,0], T is [0,1,0,0], C is [0,0,1,0], and G is [0,0,0,1], to obtain the one-hot encoding of the gene sequence;
[0104] (3) The one-hot encoding is input into a pre-trained Bi-LSTM network, which consists of two LSTM layers, encoding the gene sequence from the forward and backward directions respectively to obtain the forward and backward hidden states of the gene sequence;
[0105] (4) The forward and backward hidden states are concatenated to obtain a bidirectional hidden state of the gene sequence as the encoding vector of the gene sequence, and the encoded first gene sequence vector and second gene sequence vector are output.
[0106] The feature extraction submodule is used to input the encoded first gene sequence vector and the second gene sequence vector into a pre-trained deep neural network to obtain the feature vector of the first gene sequence and the feature vector of the second gene sequence. The function of this submodule is to extract meaningful features from the encoded gene sequence vector for subsequent similarity calculation.
[0107] The feature extraction submodule can adopt a feature extraction algorithm based on a multi-scale convolution kernel, which includes the following steps:
[0108] (1) The encoded gene sequence vector is regarded as a two-dimensional matrix, in which each row represents a position of a gene sequence and each column represents a dimension of a gene sequence;
[0109] (2) Use convolution kernels of different sizes to perform convolution operations on the gene sequence matrix to obtain feature maps of different scales, each of which represents a local feature of the gene sequence;
[0110] (3) Perform a maximum pooling operation on each feature map to obtain a scalar that represents a global feature of the gene sequence;
[0111] (4) All global features are concatenated to obtain feature vectors of gene sequences, and the feature vectors of the first gene sequence and the second gene sequence are output.
[0112] The similarity calculation submodule is used to calculate the cosine similarity between the first gene sequence data and the second gene sequence data according to the feature vector of the first gene sequence and the feature vector of the second gene sequence to obtain a similarity matrix. Specifically, the cosine similarity between the first gene sequence data and the second gene sequence data is calculated using the following formula:
[0113]
[0114] Where a and b represent the feature vector of the first gene sequence and the feature vector of the second gene sequence, respectively. n is the dimension of the feature vector. cos(θ) is the cosine similarity between the two feature vectors, indicating the similarity between the two gene sequences. θ is the angle between the two feature vectors, reflecting the difference between the two gene sequences. ||a|| and ||b|| are the moduli of the two feature vectors, indicating the size of the two feature vectors, respectively. a i and b i are the eigenvalues of the eigenvector of the first gene sequence and the eigenvector of the second gene sequence in the i-th dimension respectively.
[0115] The matching judgment submodule is used to perform threshold judgment on the similarity matrix, and judge whether the first gene sequence data matches the second gene sequence data according to the preset threshold, and obtain a matching result. The function of this submodule is to determine the matching relationship between the first gene sequence data and the second gene sequence data according to the similarity matrix, so as to perform subsequent gene tracing and editing.
[0116] The matching judgment submodule may adopt a matching judgment algorithm based on logistic regression, which includes the following steps:
[0117] (1) The similarity matrix is used as input into a pre-trained logistic regression model, which consists of a linear layer and a sigmoid activation function, and outputs a probability value indicating the likelihood of whether the two gene sequences match;
[0118] (2) Compare the probability value with a preset threshold. If the probability value is greater than or equal to the threshold, the two gene sequences are judged to match, and the matching result is output as 1; if the probability value is less than the threshold, the two gene sequences are judged to be mismatched, and the matching result is output as 0.
[0119] This embodiment provides a milk quality detection system based on gene sequences. The system can quickly and accurately determine whether the gene sequence in milk matches the standard gene sequence by utilizing encoding, feature extraction, similarity calculation and matching judgment technology, thereby improving the efficiency and accuracy of the intelligent matching module.
[0120] Embodiment 5:
[0121] This embodiment provides a milk quality detection system based on gene sequences. Based on Embodiment 1, this embodiment specifically implements the quality assessment module.
[0122] The quality assessment modules include:
[0123] The parameter calculation submodule is used to calculate the quality parameters of milk by linear regression according to the matching results to obtain the parameter calculation results. The function of this submodule is to establish a linear regression model according to the matching results. This model is a statistical technique that can calculate the quality parameters of milk, such as protein content, fat content, lactose content, etc., by fitting the linear relationship in the data, providing a basis for milk quality assessment.
[0124] The parameter calculation submodule may adopt a parameter calculation algorithm based on the least square method, which includes the following steps:
[0125] (1) Perform data preprocessing on the matching results obtained by the data matching module, including removing outliers, missing values, and noise, to obtain a set of valid matching data;
[0126] (2) performing parameter calculation on the valid matching data, using a linear regression model, such as multiple linear regression, to establish the relationship between the quality parameters of milk and the gene sequence, and solving the parameters of the model, which can reflect the degree of influence of the gene sequence on the quality of milk;
[0127] (3) Based on the results of parameter calculation, the quality parameters of each milk sample are output as the basis for milk quality assessment.
[0128] The parameter comparison submodule is used to compare the parameter calculation results with the standard quality parameters stored in the gene database module to obtain the parameter comparison results. The function of this submodule is to compare the parameter calculation results with the standard quality parameters stored in the gene database module to calculate the quality deviation of milk, such as protein content deviation, fat content deviation, lactose content deviation, etc., to provide a basis for milk quality assessment.
[0129] The parameter comparison submodule may adopt a parameter comparison algorithm based on relative error, which includes the following steps:
[0130] (1) Performing data normalization on the parameter calculation results obtained by the parameter calculation submodule, including standardization, regularization and other operations, to obtain a set of normalized parameter calculation results;
[0131] (2) performing parameter comparison on the normalized parameter calculation results, and using a relative error formula, such as a percentage relative error, calculating the difference between each parameter calculation result and the standard quality parameter stored in the gene database module, wherein the difference can reflect the degree of deviation of the milk quality;
[0132] (3) Based on the results of the parameter comparison, the parameter comparison results of each milk sample are output as the basis for milk quality assessment.
[0133] The evaluation and judgment submodule is used to evaluate the quality of milk based on the parameter comparison results and use fuzzy logic to obtain the quality evaluation results. The function of this submodule is to establish a fuzzy logic model based on the parameter comparison results. This model is an artificial intelligence technology that can comprehensively evaluate the quality of milk by simulating the human way of thinking and obtain quality evaluation results such as excellent, good, medium, poor, etc., to provide a basis for milk quality management.
[0134] The evaluation and judgment submodule can adopt an evaluation and judgment algorithm based on fuzzy reasoning, which includes the following steps:
[0135] (1) The parameter comparison results obtained by the parameter comparison submodule are fuzzified, including defining fuzzy sets, fuzzy variables, and membership functions, to obtain a set of fuzzified parameter comparison results;
[0136] (2) Evaluate and judge the fuzzy parameter comparison results, and use a fuzzy reasoning mechanism, such as Mamdani reasoning, to infer the quality evaluation results of each milk sample according to pre-set fuzzy rules. The evaluation results can reflect the level of milk quality, such as excellent, good, medium, poor, etc.
[0137] (3) According to the evaluation and judgment results, the data is defuzzified, including defining the defuzzification method, calculating the defuzzification value and other operations, to obtain a set of defuzzified evaluation and judgment results, which are output as the quality evaluation results of the milk.
[0138] This embodiment provides a milk quality detection system based on gene sequences. The system can accurately evaluate the quality parameters of milk and improve the reliability and sensitivity of the quality assessment module by utilizing technologies such as linear regression, relative error and fuzzy logic.
[0139] Embodiment 6:
[0140] This embodiment provides a milk quality detection system based on gene sequence. Based on Embodiment 1, this embodiment specifically implements the control optimization module.
[0141] The control optimization module specifically includes:
[0142] The parameter optimization submodule is used to optimize the quality parameters of milk using genetic algorithms according to the quality assessment results to obtain parameter optimization results. The function of this submodule is to establish a genetic algorithm model based on the quality assessment results. This model is a heuristic search technology that can find the optimal solution for the quality parameters of milk by simulating the process of natural selection and genetic variation, providing a basis for the control optimization of milk.
[0143] The strategy generation submodule is used to generate corresponding control optimization strategies according to the parameter optimization results, including adjusting the temperature, acidity and oxidation degree of milk. The function of this submodule is to design control optimization strategies according to the parameter optimization results to improve the quality of milk and provide a basis for milk control execution.
[0144] The strategy generation submodule may adopt a rule-based strategy generation algorithm, which includes the following steps:
[0145] (1) Data decoding is performed on the parameter optimization results obtained by the parameter optimization submodule, that is, each binary string is converted into a corresponding quality parameter value to obtain a set of decoded parameter optimization results;
[0146] (2) generating strategies for the decoded parameter optimization results, that is, using a rule base, such as a fuzzy rule base, to generate corresponding control optimization strategies according to predefined rules. The control optimization strategies can reflect the adjustment direction and amplitude of the temperature, acidity and oxidation degree of the milk. For example, if the quality parameter value is lower than the standard value, the temperature of the milk is increased and the acidity and oxidation degree of the milk are reduced.
[0147] (3) Based on the results of strategy generation, the control optimization strategy corresponding to each quality parameter is output as the control optimization strategy for milk.
[0148] The strategy evaluation submodule uses reinforcement learning to evaluate the control optimization strategy and obtain the strategy evaluation result. The function of this submodule is to establish a reinforcement learning model based on the control optimization strategy. This model is an artificial intelligence technology that can learn and optimize the effect of the control optimization strategy through interaction with the environment, and provide feedback for the control optimization of milk.
[0149] The strategy evaluation submodule can adopt a reinforcement learning algorithm based on state, action, reward and value, which includes the following steps:
[0150] (1) According to the control optimization strategy, an action is selected, such as raising the temperature of milk by 10°C, and the action is executed to obtain feedback from the environment, such as changes in the quality parameters of milk;
[0151] (2) Based on the feedback from the environment, a reward is calculated, such as the negative value of the sum of squares of the deviations between the quality parameters of milk and the standard quality parameters, as the reward value of the action, and the reward result is obtained;
[0152] (3) Update a value based on the reward result, such as using a value update algorithm, such as Q-learning, to update the value of the current state and action based on the reward value of the action and the maximum value of the next state to obtain a value update result;
[0153] (4) Evaluate the effect of the control optimization strategy based on the value update result, such as using an evaluation algorithm, such as average value, to calculate the average value of the control optimization strategy as the strategy evaluation result, and output it to the strategy update submodule.
[0154] The strategy update submodule is used to update the control optimization strategy according to the strategy evaluation results to improve the control effect and output the control signal to the control execution module. The function of this submodule is to adjust the parameters of the control optimization strategy, such as the change range of temperature, acidity and oxidation degree, according to the strategy evaluation results to improve the effect of the control optimization strategy and provide guidance for the control execution of milk.
[0155] The policy update submodule can adopt a policy update algorithm based on gradient descent, which includes the following steps:
[0156] (1) Calculate a gradient based on the strategy evaluation result, such as using a gradient calculation algorithm, such as the chain rule, to calculate the partial derivative of the parameters of the control optimization strategy with respect to the strategy evaluation result as the gradient value, and obtain the gradient result;
[0157] (2) Update a parameter based on the gradient result, such as using a parameter update algorithm, such as the gradient descent method, to adjust the parameters of the control optimization strategy according to the gradient value and the learning rate, so that it moves in the opposite direction of the gradient to obtain the parameter update result;
[0158] (3) Generate a control signal based on the parameter update result, such as using a control signal generation algorithm, such as encoding, to convert the parameter update result into a control signal in a certain format, such as a digital or analog signal, obtain a control signal result, and output it to the control execution module.
[0159] This embodiment provides a milk quality detection system based on gene sequences. The system can effectively optimize and control the quality parameters of milk by utilizing genetic algorithms, rule bases, Q learning and other technologies, thereby improving the intelligence and adaptability of the control optimization module.
[0160] Embodiment 7:
[0161] like Figure 3 As shown, this embodiment provides a milk quality detection system based on gene sequence. Based on Example 1, this embodiment adds a gene tracing module and implements the gene tracing module in detail.
[0162] The gene traceability module is used to trace and analyze the causes and links of differences in milk quality, in order to find out the source and influencing factors of quality problems and provide a basis for quality improvement and prevention. The module uses gene databases and analysis software to determine the source, variety, production environment, processing and other information of milk based on the gene sequence and quality parameters of milk, so as to analyze the possible causes of differences in milk quality, such as genetic variation, feed composition, production hygiene, storage conditions, etc., and provide scientific basis and suggestions for improvement and prevention of milk quality.
[0163] The gene tracing module specifically includes:
[0164] The gene comparison submodule is used to compare the first gene sequence data obtained by the gene sequencing module with the second gene sequence data in the gene database module to obtain the comparison results, including the similarity, difference and variation of the gene sequences. The function of this submodule is to calculate the sequence matching degree between the gene sequence in milk and the reference sequence in the gene database using a gene comparison algorithm, such as BLAST, so as to evaluate the quality and characteristics of the gene sequence in milk and provide a basis for milk quality analysis.
[0165] The gene alignment submodule may use a BLAST-based gene alignment algorithm, which includes the following steps:
[0166] (1) performing data preprocessing on the first gene sequence data obtained by the gene sequencing module, including removing low-quality sequences, removing duplicate sequences, removing non-coding sequences, etc., to obtain a set of high-quality first gene sequence data;
[0167] (2) performing data indexing on the second gene sequence data in the gene database module, that is, using an indexing algorithm, such as a BWT algorithm, to compress and sort the second gene sequence data and construct a gene index table to facilitate rapid comparison;
[0168] (3) performing data comparison on the high-quality first gene sequence data, that is, using a comparison algorithm, such as the BLAST algorithm, to compare each first gene sequence data with the second gene sequence data according to the gene index table, find the best matching region, and calculate the matching score, which can reflect the similarity of the gene sequences;
[0169] (4) Performing data filtering on the results of data comparison, that is, using a filtering algorithm, such as the E-value algorithm, to filter out low-quality matches based on the matching scores and retain high-quality matches. High-quality matches can reflect the difference and variation of gene sequences;
[0170] (5) According to the result of data filtering, the comparison result of each first gene sequence data and the second gene sequence data is output, including similarity, difference and variation, as the gene comparison result.
[0171] The gene difference submodule is used to identify the difference points between the first gene sequence data and the second gene sequence data, including insertion, deletion, substitution or mutation, based on the comparison results obtained by the gene comparison submodule, and analyze the impact of the difference points on the quality of milk, including changes in protein content, fat content, lactic acid bacteria content or antibiotic residues. The role of this submodule is to use a gene difference analysis algorithm, such as MAFFT, based on the gene comparison results to find the difference position and difference type between the two gene sequences, so as to analyze the reasons for the difference between the gene sequence in milk and the reference sequence, as well as the impact of these differences on milk quality, and provide a reference for milk quality improvement and prevention.
[0172] The gene difference submodule may adopt a gene difference algorithm based on the comparison result, and the algorithm comprises the following steps:
[0173] (1) Performing data analysis on the alignment results obtained by the gene alignment submodule, that is, using an analysis algorithm, such as the SAMtools algorithm, to convert and sort the alignment results to obtain a set of standardized alignment results. The standardized alignment results can reflect the matching and variation of the gene sequences;
[0174] (2) performing difference identification on the standardized comparison results, that is, using a difference identification algorithm, such as the GATK algorithm, to identify the difference points between the first gene sequence data and the second gene sequence data according to the comparison results. The difference points can be divided into insertions, deletions, substitutions or mutations. Insertions and deletions can reflect changes in the length of the gene sequence, and substitutions and mutations can reflect changes in the content of the gene sequence;
[0175] (3) performing difference analysis on the results of difference identification, i.e., using a difference analysis algorithm, such as ANNOVAR algorithm, to analyze the influence of the difference points on the quality of milk according to the difference points, and the influence can be divided into changes in protein content, fat content, lactic acid bacteria content or antibiotic residue. The change in protein content can reflect the nutritional value of milk, the change in fat content can reflect the taste of milk, the change in lactic acid bacteria content can reflect the safety of milk, and the change in antibiotic residue can reflect the sanitary condition of milk;
[0176] (4) Based on the results of the differential analysis, the type and impact of each differential point are output as the differential result of the gene.
[0177] The gene tracing submodule is used to trace the source information of the first gene sequence data in the milk based on the difference points obtained by the gene difference submodule and the second gene sequence data stored in the gene database module, and analyze the impact of the source information on the quality of the milk, including the impact on the nutritional value, safety and taste of the milk. The function of this submodule is to construct the evolutionary relationship between the gene sequence in the milk and the reference sequence based on the gene difference points using a gene tracing algorithm, such as PhyloTree, so as to determine the source of the gene sequence in the milk, such as buffalo, sheep, donkey, etc., and the impact of these sources on the quality of the milk, and provide a basis for the quality control and optimization of the milk.
[0178] The gene tracing submodule may adopt a gene tracing algorithm based on difference points, which includes the following steps:
[0179] (1) Performing data screening on the difference points obtained by the gene difference submodule, that is, using a screening algorithm, such as frequency screening, to screen out representative difference points according to the frequency of the difference points. The representative difference points can reflect the characteristics and differences of the gene sequence;
[0180] (2) performing data matching on the screened difference points, i.e., using a matching algorithm, such as SNP matching, to match each difference point with the second gene sequence data stored in the gene database module to find the best matching result. The matching result can reflect the source information of the gene sequence, such as the breed, origin, and age of the cattle;
[0181] (3) performing data analysis on the matching results, i.e., using a data analysis algorithm, such as cluster analysis, to analyze the impact of the source information on the quality of the milk based on the matching results. The impact can be divided into the impact on the nutritional value, safety and taste of the milk. The nutritional value can reflect the breed and origin of the cattle, the safety can reflect the age and environment of the cattle, and the taste can reflect the feeding and management of the cattle;
[0182] (4) Based on the results of the data analysis, the impact of each source of information is output as the gene tracing result.
[0183] The gene report submodule is used to generate a gene traceability report based on the source information obtained by the gene traceability submodule, including the comparison results, differences, source information, influencing factors and improvement suggestions of the gene sequence, and send the gene traceability report to the control optimization module and the control execution module for quality improvement and preventive measures. The role of this submodule is to display the gene traceability results in the form of charts, maps, timelines, etc. based on the gene traceability results using a gene report generation algorithm, such as ReportMaker, so as to clearly show the quality and characteristics of the gene sequence in milk, as well as the relationship between these qualities and characteristics and the source of milk, and provide a basis and suggestions for quality improvement and prevention of milk.
[0184] This embodiment provides a milk quality detection system based on gene sequences. The system can comprehensively, accurately and quickly detect and analyze gene sequences in milk by utilizing modules such as gene comparison, gene difference, gene tracing and gene reporting, thereby improving the controllability and traceability of milk quality and providing a scientific basis and suggestions for the improvement and prevention of milk quality.
[0185] Embodiment 8:
[0186] This embodiment provides a method for detecting milk quality based on gene sequences, which can quickly and accurately detect gene sequences in milk, thereby evaluating the quality and safety of milk.
[0187] like Figure 4 As shown, the method specifically comprises the following steps:
[0188] Step 1, determining the gene sequence in the milk sample to obtain the first gene sequence data. By using DNA extraction and PCR amplification, the gene sequence related to milk quality is extracted and amplified from the milk sample, thereby obtaining the first gene sequence data. DNA extraction adopts any method suitable for extracting DNA from milk, such as thermal lysis method, phenol chloroform method, silica gel column method, etc., and DNA extraction is performed by a kit or other appropriate equipment to process and separate the milk sample. PCR amplification adopts any method suitable for amplifying DNA in milk, such as conventional PCR, real-time PCR, multiplex PCR, etc.; PCR amplification is performed by a PCR instrument or other appropriate equipment to input and amplify the extracted DNA; PCR amplification can use specific primers to amplify specific gene sequences in milk, such as milk protein genes, lactic acid bacteria genes, antibiotic genes, etc. related to milk quality; the first gene sequence data is obtained by the output signal of the PCR instrument or electrophoresis analysis.
[0189] Step 2, matching the first gene sequence data with the second gene sequence data related to milk quality stored in the gene database to obtain a matching result. The gene database can be any database storing gene sequence data related to milk quality, such as NCBI database, KEGG database, UniProt database, etc.; the gene database is connected to a computer or other appropriate device via the Internet or other appropriate means to input and query the first gene sequence data. Matching can be performed using any method suitable for comparing gene sequence data, such as BLAST algorithm, Smith-Waterman algorithm, Needleman-Wunsch algorithm, etc.; the matching result is obtained through the output signal of a computer or other appropriate device, and the matching result can represent the similarity or difference between the first gene sequence data and the second gene sequence data.
[0190] Step 3, according to the matching results, calculate the quality parameters of the milk, including protein content, fat content, lactic acid bacteria content and antibiotic residue, and evaluate the quality of the milk to obtain a quality evaluation result. The calculation adopts any method suitable for calculating the quality parameters of milk, such as mathematical formulas, statistical analysis, machine learning, etc.; the calculation is performed by a computer or other appropriate equipment to input and process the matching results; the quality parameters are obtained through the output signal of the computer or other appropriate equipment, and the quality parameters represent the content or proportion of various components in the milk; the evaluation adopts any method suitable for evaluating the quality of milk, such as the standard value method, the scoring method, the grade method, etc.; the evaluation is performed by a computer or other appropriate equipment to input and judge the quality parameters; the quality evaluation result is obtained through the output signal of the computer or other appropriate equipment, and the quality evaluation result represents the quality grade or score of the milk.
[0191] Step 4: Generate a corresponding control optimization strategy based on the quality assessment results, including adjusting the temperature, acidity and oxidation degree of the milk to obtain a control signal. The generation adopts any method suitable for generating a control optimization strategy, such as a rule base method, an expert system method, a neural network method, etc. The generation is performed by a computer or other appropriate device to input and analyze the quality assessment results; the control optimization strategy is obtained through the output signal of the computer or other appropriate device, and the control optimization strategy represents suggestions or guidance for various processing conditions of the milk; the control signal is performed by a controller or other appropriate device to set and adjust the temperature, acidity and oxidation degree of the milk.
[0192] Step 5, performing corresponding control operations on the milk according to the control signal. The control operation adopts any method suitable for controlling the milk, such as heating method, cooling method, acidification method, reduction method, etc. The control operation is performed through a heater, a cooler, an acidifier, a reducing agent, etc., so as to process and improve the milk.
[0193] This embodiment provides a method for detecting milk quality based on gene sequences, which can quickly and accurately detect gene sequences in milk, thereby evaluating the quality and safety of milk.
[0194] The beneficial effects of this embodiment are:
[0195] By utilizing DNA extraction and PCR amplification, gene sequences associated with milk quality can be extracted and amplified from milk samples, thereby obtaining first gene sequence data;
[0196] By using a gene database and a matching algorithm, the first gene sequence data can be matched with the second gene sequence data related to milk quality, thereby obtaining a matching result;
[0197] By utilizing the calculation and evaluation methods, the quality parameters of the milk can be calculated according to the matching results, and the quality of the milk can be evaluated, thereby obtaining a quality evaluation result;
[0198] By utilizing the generation and control method, the corresponding control optimization strategy can be generated according to the quality evaluation results, and the corresponding control operation can be performed on the milk according to the control signal, thereby improving the quality and safety of the milk.
[0199] The method of this embodiment can be widely used in the rapid extraction and detection of genomic DNA of various dairy products, for example, for detecting milk, milk powder, cheese, yogurt, etc.; the method of this embodiment can improve the quality and safety of dairy products, thereby improving the application value and market competitiveness of dairy products.
[0200] It should be noted that, in this article, the terms "include", "comprises" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. Specific examples are used in this article to illustrate the principle and implementation of the technical solution of the present invention. The above examples are only used to help understand the method of the present invention and its core idea. The above is only a preferred implementation of the present invention. It should be pointed out that due to the limitations of textual expression and the objective existence of infinite specific structures, for ordinary technicians in this technical field, without departing from the principle of the present invention, several improvements, modifications or changes can be made, and the above technical features can also be combined in an appropriate manner; these improvements, modifications, changes or combinations, or the direct application of the concept and technical solution of the invention to other occasions without improvement, should be regarded as the protection scope of the present invention.
Claims
1. A milk quality detection system based on gene sequence, characterized in that: include: A gene sequencing module, used for determining the gene sequence in the milk sample to obtain first gene sequence data; A gene database module, used for storing known second gene sequence data related to milk quality and its corresponding standard quality parameters; An intelligent matching module, used to match the first gene sequence data obtained by the gene sequencing module with the second gene sequence data in the gene database module to obtain a matching result; The quality assessment module is used to calculate the quality parameters of the milk, including protein content, fat content, lactic acid bacteria content and antibiotic residue, based on the matching results, and to assess the quality of the milk to obtain a quality assessment result; The control optimization module is used to generate corresponding control optimization strategies according to the quality assessment results, including adjusting the temperature, acidity and oxidation degree of milk, and outputting control signals; A control execution module is used to receive the control signal output by the control optimization module and perform corresponding control operations on the milk; The intelligent matching module specifically includes: An encoding submodule, used to encode the first gene sequence data and the second gene sequence data respectively to obtain an encoded first gene sequence vector and a second gene sequence vector; A feature extraction submodule, used to input the encoded first gene sequence vector and the second gene sequence vector into a pre-trained deep neural network respectively, to obtain a feature vector of the first gene sequence and a feature vector of the second gene sequence; The similarity calculation submodule is used to calculate the cosine similarity between the first gene sequence data and the second gene sequence data according to the feature vector of the first gene sequence and the feature vector of the second gene sequence to obtain a similarity matrix. Specifically, the cosine similarity between the first gene sequence data and the second gene sequence data is calculated using the following formula: in, and Represent the feature vector of the first gene sequence and the feature vector of the second gene sequence respectively, n is the dimension of the feature vector, is the cosine similarity between two feature vectors, indicating the similarity between two gene sequences. is the angle between the two eigenvectors, reflecting the degree of difference between the two gene sequences. and is the modulus of the two eigenvectors, which respectively represent the size of the two eigenvectors. and The feature vectors of the first gene sequence and the second gene sequence are respectively The eigenvalues in the dimensions; The matching judgment submodule is used to perform threshold judgment on the similarity matrix, and judge whether the first gene sequence data matches the second gene sequence data according to a preset threshold, so as to obtain a matching result; The system also includes a gene tracing module for tracing and analyzing the causes and links of milk quality differences to find out the source and influencing factors of quality problems and provide a basis for quality improvement and prevention; The gene tracing module specifically includes: A gene comparison submodule is used to compare the first gene sequence data obtained by the gene sequencing module with the second gene sequence data in the gene database module to obtain a comparison result, including the similarity, difference and variation of the gene sequences; A gene difference submodule, for identifying differences between the first gene sequence data and the second gene sequence data, including insertion, deletion, substitution or mutation, based on the comparison results obtained by the gene comparison submodule, and analyzing the effects of the differences on milk quality, including changes in protein content, fat content, lactic acid bacteria content or antibiotic residues; A gene tracing submodule, for tracing the source information of the first gene sequence data in the milk based on the difference points obtained by the gene difference submodule and using the second gene sequence data stored in the gene database module, and analyzing the influence of the source information on the quality of the milk, including the influence on the nutritional value, safety and taste of the milk; The gene reporting submodule is used to generate a gene tracing report based on the source information obtained by the gene tracing submodule, including the comparison results of the gene sequences, differences, source information, influencing factors and improvement suggestions, and send the gene tracing report to the control optimization module and the control execution module for quality improvement and preventive measures.
2. The milk quality detection system based on gene sequence according to claim 1, characterized in that: The system also includes a gene editing module, which is used to perform targeted editing on the gene sequence in the milk according to the quality assessment result, and transmit the result of the targeted editing to the control optimization module.
3. The milk quality detection system based on gene sequence according to claim 1, characterized in that: The gene sequencing module specifically includes: The sample processing submodule is used to purify, lyse and amplify the milk sample to obtain DNA fragments in the milk; A sequencing chip submodule is used to fix the DNA fragments on the sequencing chip and perform high-throughput sequencing to obtain the first gene sequence data in the milk; The data transmission submodule is used to transmit the first gene sequence data to the intelligent matching module via a wireless network or a wired network.
4. The milk quality detection system based on gene sequence according to claim 1, characterized in that: The quality assessment module specifically includes: The parameter calculation submodule is used to calculate the quality parameters of milk by linear regression according to the matching results to obtain the parameter calculation results; A parameter comparison submodule is used to compare the parameter calculation results with the standard quality parameters stored in the gene database module to obtain parameter comparison results; The evaluation and judgment submodule is used to evaluate the quality of milk based on the parameter comparison results and use fuzzy logic to obtain the quality evaluation result.
5. The milk quality detection system based on gene sequence according to claim 1, characterized in that: The control optimization module specifically includes: The parameter optimization submodule is used to optimize the quality parameters of milk using a genetic algorithm according to the quality evaluation results to obtain parameter optimization results; The strategy generation submodule is used to generate corresponding control optimization strategies according to the parameter optimization results, including adjusting the temperature, acidity and oxidation degree of milk; The strategy evaluation submodule uses reinforcement learning to evaluate the control optimization strategy and obtain the strategy evaluation result; The strategy update submodule is used to update the control optimization strategy according to the strategy evaluation results to improve the control effect, and output the control signal to the control execution module.
6. A gene sequence-based milk quality detection method using the milk quality detection system of any one of claims 1 to 5, characterized in that: The following steps are involved: Step 1, measuring the gene sequence in the milk sample to obtain first gene sequence data; Step 2, matching the first gene sequence data with the second gene sequence data related to milk quality stored in the gene database to obtain a matching result; Step 3, according to the matching results, the quality parameters of the milk are calculated, including protein content, fat content, lactic acid bacteria content and antibiotic residue, and the quality of the milk is evaluated to obtain a quality evaluation result; Step 4: Generate a corresponding control optimization strategy based on the quality assessment results, including adjusting the temperature, acidity and oxidation degree of the milk to obtain a control signal; Step 5: Perform corresponding control operations on the milk according to the control signal.
