Insect bionic odor detection method based on ligand recognition mechanism and aversion training

Abstract

This invention discloses an insect biomimetic odor detection method based on a ligand recognition mechanism and aversion training, which relates to the field of biological detection technology. The method includes matching insects and target volatile organic compounds (VOCs) based on the ligand recognition mechanism; performing aversion olfactory training on the matched insects and target VOCs; using a CNN hybrid model to discriminate the identification and spatial distribution of various insects and target VOCs after the aversion olfactory training, and obtaining the relationship between the spatial distribution of insects and odor sources; based on this relationship between the spatial distribution of insects and odor sources, using multi-point recognition to create a VOC distribution map of the insect target area, thus completing insect biomimetic odor detection based on the ligand recognition mechanism and aversion training. This invention solves the problems of existing methods that lack superior detection sensitivity, repeatability, and stability.

Description

Technical Field

[0001] The present invention relates to the field of biological detection technology, and in particular to an insect bionic odor detection method based on ligand recognition mechanism and aversion training. Background Art

[0002] Insects offer incomparable advantages over dogs. However, their small size and abundance also create practical challenges. Unlike dogs, they cannot be restrained or tracked. Once released, their small size, high speed, and even mimicry in certain environments make them difficult to identify and track. Exploiting this biological sense of smell often relies on training based on preference, such as by linking bees' nectar preferences and police dogs' meat preferences with targets. While these naturally occurring associations undoubtedly offer advantages, they also come with numerous limitations, particularly when it comes to exploiting insect olfaction. Training typically leverages insects' hunger and mating needs. However, hunger-based preferences are easily affected by the insect's state. Honeybees' nectar preferences are a special case. Because bees are social animals, and flower-foraging bees' natural duty is to forage for nectar, they tend to exhibit nectar-seeking behaviors regardless of their state. However, commonly used model organisms, such as the fruit fly, lack this characteristic. The reproductive and mating needs based on pheromones are more susceptible to interference. For example, these preferences are often affected by factors such as insect age, mating status, sex, light intensity, and temperature. Even the same insect may exhibit different behaviors at different stages. Therefore, biased preference training is clearly ineffective in maintaining a stable insect sensitivity.

[0003] Compared with preference training, punishment-based aversion training is theoretically more suitable for our goals. However, most of the conditional training conducted by researchers today is irrational training. According to the principle mechanism of biological olfaction, different species have different sensitivities to the same odor. This is because the ability of their olfactory receptors to bind to odor molecules, as well as the subsequent neural responses, are different. Therefore, when we select subjects for training, we usually do not know whether the selected insects can smell the target odor molecules or whether they are sensitive to the target odor molecules. In other words, most of the olfactory training we conduct is blind, and these will affect the behavior that the insects ultimately exhibit. Importantly, this irrational olfactory training means that the training results it brings are also unstable. Summary of the Invention

[0004] In response to the above-mentioned deficiencies in the prior art, the present invention provides an insect bionic odor detection method based on ligand recognition mechanism and aversion training, which solves the problem that the existing methods have insufficient detection sensitivity, repeatability and stability.

[0005] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is: an insect bionic odor detection method based on ligand recognition mechanism and aversion training, comprising the following steps:

[0006] S1: Matching insects and target VOCs based on ligand recognition mechanism;

[0007] S2: aversive olfactory training with matching insects and target VOCs;

[0008] S3: For the insects and target VOCs trained with aversive olfactory perception, the CNN hybrid model is used to identify various insects and their spatial distribution, and the relationship between the spatial distribution of insects and odor sources is obtained;

[0009] S4: Based on the relationship between insect spatial distribution and odor sources, multi-point recognition is used to depict the VOC distribution map of the insect target area, completing insect bionic odor detection based on ligand recognition mechanism and aversion training.

[0010] Furthermore, the S1 includes the following sub-steps:

[0011] S11: Search and compare the target VOC with the molecules that the insects can smell in the iORbase database;

[0012] S12: Based on the search and comparison results, determine whether the target VOC already exists in the iORbase database. If so, complete the matching of the insect and the target VOC. If not, use molecular simulation technology to replace the training insect to complete the matching of the insect and the target VOC.

[0013] Furthermore, the step S2 includes the following sub-steps:

[0014] S21: A set number of insects were placed in a training tube lined with a copper mesh and were made to smell the target VOC and then the ambient gas;

[0015] S22: stimulation is performed when the insect smells the target VOC, constituting a training cycle;

[0016] S23: Use a gradient voltage increase and perform three training cycles for each voltage to obtain an appropriate training voltage to complete the aversive olfactory training for the matched insects and target VOCs.

[0017] Furthermore, the step S3 includes the following sub-steps:

[0018] S31: placing the insects that have completed the aversion olfactory training in a transparent white-bottomed octagonal cage, placing an odor source containing the target VOC outside the transparent white-bottomed octagonal cage, and observing the spatial distribution data, density distribution data, and time required to form a stable distribution of the insects in the transparent white-bottomed octagonal cage;

[0019] S32: Preprocessing the spatial distribution data, density distribution data, and time data required to form a stable distribution;

[0020] S33: Using the preprocessed spatial distribution data, density distribution data, and the time data required to form a stable distribution, the CNN hybrid model is trained and evaluated to obtain the final CNN hybrid model;

[0021] S34: Deploy the final CNN hybrid model to actual application scenarios to identify various insects and their spatial distribution, and obtain the relationship between the spatial distribution of insects and odor sources.

[0022] Furthermore, the data preprocessing in S32 includes image data normalization and data enhancement, wherein the image data normalization includes scaling the image pixel values ​​to be within the range of 0-1, and the data enhancement includes rotation, scaling and cropping.

[0023] Furthermore, the CNN hybrid model in S33 includes a CNN feature extraction layer, a feature processing and fusion layer, and other model layers connected in sequence;

[0024] The CNN feature extraction layer includes an input layer, a convolution layer, an activation layer, and a pooling layer connected in sequence, wherein the input layer is used to receive image data, the convolution layer uses multiple convolution kernels to extract local features of the image, the activation layer applies a ReLU activation function, and the pooling layer is used to reduce feature dimensions while retaining important information;

[0025] The feature processing and fusion layer includes a fully connected layer, which flattens the features extracted by the CNN feature extraction layer into a one-dimensional vector;

[0026] The other model layers include a recurrent layer and a classification and regression layer connected in sequence. The recurrent layer uses a long short-term memory network LSTM to process sequence data. The classification and regression layer uses a fully connected layer to perform classification tasks, regression tasks, or access a support vector machine SVM for prediction tasks.

[0027] The beneficial effects of the present invention are:

[0028] It is proposed to use mathematical models based on algorithms to automatically identify insect individuals and distributions, realize insect olfactory bionics, and solve the bottleneck that existing technologies cannot track and monitor changes in insect behavior.

[0029] The insect olfactory structure and function database iORbase and molecular simulation technology are used to achieve rational selection of insect olfactory training molecules, avoiding interference of insensitive odor molecules in the experiment.

[0030] The proposed insect-inspired olfactory recognition system will be highly portable and provide comprehensive odor source identification information. For example, it will be adaptable to aerospace, space, sea, and ground platforms, allowing it to be used with various mobile devices and simultaneously determine the location and concentration of odor sources. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 This is a flow chart of an insect bionic odor detection method based on ligand recognition mechanism and aversion training.

[0032] Figure 2 Schematic diagram of a T-shaped maze.

[0033] Figure 3 Schematic diagram of the reverse tracking algorithm principle.

[0034] Figure 4 This is the overall schematic diagram of the odor source detection and prediction system.

[0035] Figure 5 Schematic diagram of the test results with 9 repetitions.

[0036] Figure 6 This is the distribution diagram of fruit flies' responses to acetaldehyde and air as a control when they are not trained.

[0037] Figure 7 This is the distribution diagram of fruit flies in response to acetaldehyde and air during training.

[0038] Figure 8 Schematic diagram of the octagonal cage test results. DETAILED DESCRIPTION

[0039] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0040] like Figure 1 As shown, an insect bionic odor detection method based on ligand recognition mechanism and aversion training includes the following steps:

[0041] S1: Matching insects and target VOCs based on ligand recognition mechanism;

[0042] S2: aversive olfactory training with matching insects and target VOCs;

[0043] S3: For the insects and target VOCs trained with aversive olfactory perception, the CNN hybrid model is used to identify various insects and their spatial distribution, and the relationship between the spatial distribution of insects and odor sources is obtained;

[0044] S4: Based on the relationship between insect spatial distribution and odor sources, multi-point recognition is used to depict the VOC distribution map of the insect target area, completing insect bionic odor detection based on ligand recognition mechanism and aversion training.

[0045] The S1 includes the following steps:

[0046] S11: Search and compare the target VOC with the molecules that the insects can smell in the iORbase database;

[0047] S12: Based on the search and comparison results, determine whether the target VOC already exists in the iORbase database. If so, complete the matching of the insect and the target VOC. If not, use molecular simulation technology to replace the training insect (or use MD to analyze its structure and compare the molecule with the closest structure) to complete the matching of the insect and the target VOC.

[0048] The S2 includes the following steps:

[0049] S21: A set number of insects were placed in a training tube lined with a copper mesh and were made to smell the target VOC and then the ambient gas;

[0050] S22: stimulation is performed when the insect smells the target VOC, constituting a training cycle;

[0051] S23: Use a gradient voltage increase and perform three training cycles for each voltage to obtain an appropriate training voltage to complete the aversive olfactory training for the matched insects and target VOCs.

[0052] In one embodiment of the present invention, approximately 100 fruit flies are placed in a training tube lined with a copper mesh and are exposed to two odors (target VOC (CS+) and ambient gas (CS-)) for 1 minute each, followed by 45 seconds of fresh air. When and only when the fruit flies smell CS+, they are stimulated by an electric shock stimulator. This electric shock stimulation, together with the CS+, lasts for 1 minute and consists of 12 1.5-second AC pulses. This process constitutes a training cycle. By using a gradient voltage increase and performing three training cycles at each voltage, the most suitable training voltage can be obtained.

[0053] During the test, the trained fruit flies were placed in a T-maze, such as Figure 2Figure 1 shows the training tube (A) with an electric shock stimulator, the maze corridor (B), and the sorting chambers (C and D). Blue arrows indicate the direction of insect movement, and red arrows indicate the direction of odorant introduction. The two arms of the maze were filled with the two odors the fruit flies had smelled during training: CS+, which accompanied an electric shock, and CS-, which did not. The fruit flies had 2 minutes to choose between the two sorting chambers. After making their choice, they were separated and counted. They then chose the CS+ and compared it with a novel odorant to observe their avoidance effect on the CS+.

[0054] The S3 includes the following sub-steps:

[0055] S31: placing the insects that have completed the aversion olfactory training in a transparent white-bottomed octagonal cage, placing an odor source containing the target VOC outside the transparent white-bottomed octagonal cage, and observing the spatial distribution data, density distribution data, and time required to form a stable distribution of the insects in the transparent white-bottomed octagonal cage;

[0056] In theory, insects should be concentrated in the direction of the straight line connecting the cage and the odor source, and the closer to the odor source, the greater the distribution density. Points with higher concentrations require less time to establish a steady state.

[0057] S32: Preprocessing the spatial distribution data, density distribution data, and time data required to form a stable distribution;

[0058] S33: Using the preprocessed spatial distribution data, density distribution data, and the time data required to form a stable distribution, the CNN hybrid model is trained and evaluated to obtain the final CNN hybrid model;

[0059] S34: Deploy the final CNN hybrid model to actual application scenarios to identify various insects and their spatial distribution, and obtain the relationship between the spatial distribution of insects and odor sources.

[0060] The data preprocessing in S32 includes image data normalization and data enhancement. The image data normalization includes scaling the image pixel values ​​to a range of 0-1, and the data enhancement includes rotation, scaling, and cropping.

[0061] The CNN hybrid model in S33 includes a CNN feature extraction layer, a feature processing and fusion layer, and other model layers connected in sequence;

[0062] The CNN feature extraction layer includes an input layer, a convolution layer, an activation layer, and a pooling layer connected in sequence, wherein the input layer is used to receive image data, the convolution layer uses multiple convolution kernels to extract local features of the image, the activation layer applies a ReLU activation function, and the pooling layer is used to reduce feature dimensions while retaining important information;

[0063] The feature processing and fusion layer includes a fully connected layer, which flattens the features extracted by the CNN feature extraction layer into a one-dimensional vector;

[0064] The other model layers include a recurrent layer and a classification and regression layer connected in sequence. The recurrent layer uses a long short-term memory network LSTM to process sequence data. The classification and regression layer uses a fully connected layer to perform classification tasks, regression tasks, or access a support vector machine SVM for prediction tasks.

[0065] In this embodiment, the CNN hybrid model is mainly composed of a mixture of recurrent neural networks (RNN), long short-term memory networks (LSTM), support vector machines (SVM), decision trees, etc., and is used to process sequence data, enhance the non-spatial characteristics of features, or perform classification and regression tasks. According to the algorithm principle, in the actual detection process, each search point corresponds to the two-dimensional distribution state of an insect. However, it is also possible that the distribution within the device should be the same when it is very far away from the target and when it is located on the target. Therefore, the entire search process can be recorded using video, and the distribution of each point is obtained by extracting video frames. Although in general, it is sufficient to obtain planar data relative to the lens, in this case, an LSTM-based RNN must be used to analyze the search process to assist in determining whether it is close to the target and predict the insect's final landing point on the z-axis. After determining that it is close to the target, the upgraded data can be classified and judged using SVM, and the entire process can be connected in series using a decision tree model.

[0066] The model training process includes:

[0067] Initialization: Randomly initialize weights for CNN and other model layers;

[0068] Forward propagation: extract features through CNN and pass them to other model layers;

[0069] Loss function calculation: Select the loss function based on the task type, such as cross entropy loss for classification tasks;

[0070] Backpropagation: calculate gradients and update network weights;

[0071] Optimization: Use optimization algorithms such as Adam and SGD to adjust weights.

[0072] The model evaluation process includes:

[0073] Use the validation set to evaluate model performance;

[0074] Adjust hyperparameters to optimize the model.

[0075] Finally, the trained model is deployed to actual applications, and necessary model compression and acceleration can be performed to run efficiently on mobile devices or servers.

[0076] In this embodiment, based on the Fruit Fly Optimization Algorithm (FOA), this solution sets the following algorithm, which is called the reverse tracking algorithm:

[0077] Different from FOA, this solution fixes the origin of the coordinate system at the center of the device, that is, the coordinate system is established with itself as the reference, and the rectangular coordinate system is replaced by the polar coordinate system, such as Figure 3 As shown; for any fruit fly individual X i , and their spatial positions in the device are expressed as (ρ i ,θ i ) to describe; when the fruit fly group in the device becomes stable, a certain distribution is formed. At this time, we i Set a threshold for ρ and divide the coordinate system θ ([0, 2π]) into eight equal intervals. Classify and merge qualified individuals based on the value of θ, and select the interval with the largest number of individuals. Because aversive learning is used, iterate in the opposite direction of the median of the interval selected in the above step and repeat the above steps.

[0078] like Figure 4 As shown in the figure, A is the object being measured (odor source), B is the insect behavior chamber, C is the video capture device (camera), D is the neural network processor (computer or single-chip microcomputer), E is the result output device (display), and F is the insect being used. As the insect sniffs the odor source in the behavior chamber, its distribution eventually stabilizes. At this point, the video capture device located directly above the behavior chamber can capture and record the distribution, transmit the results to the neural network processor for processing and analysis, and finally display them on the result output device.

[0079] In this embodiment, the fruit flies selected through training can sniff the target odor molecules, and then the insects are placed in a device, and the device is divided into eight compartments by partitions, so theoretically there are different concentrations of odor molecules in different compartments. After the odor source is placed and a location is selected, the trained insects will perceive the odor molecules at that point, forming a certain distribution within the device. The camera located on the device takes a picture, and the image is analyzed by the trained mathematical model based on the convolutional neural network to obtain the spatial relationship between the point and the odor source, achieving an effect similar to that of a compass. Repeating this process and identifying and analyzing multiple points, you can get a distribution map of odor molecules in the area, and realize the search and detection of the odor source, such as Figure 5 shown.

[0080] Again, acetaldehyde was used as the target VOC, and the insects to be trained were fruit flies. The data showed that fruit flies did not have a clear preference for acetaldehyde, so without training, there was no difference in the distribution of fruit flies to acetaldehyde and the control air, such as Figure 6 Then, the punishment training was carried out by using electrical stimulation. The training conditions were AC pulse wave, 30% duty cycle, 20V voltage, 1 minute of electrical stimulation, 1 minute of control, 45 seconds of air purification, 8 cycles, 25.6℃ temperature, 52% humidity, and the training effect was tested using a Y-shaped maze. The test duration was 2 minutes. Figure 7 As shown, A is the beginning of the test, acetaldehyde is introduced to the left side and air is introduced to the right side. The chi-square test shows no difference in distribution; B is 2 minutes after the test, the fruit flies have formed obvious avoidance behavior towards acetaldehyde. The chi-square test shows that the distribution of the results is significantly different, indicating that the training is successful. Finally, the octagonal cage test is carried out. Figure 8 As shown, A is the actual octagonal cage test system. The square box is in a sealed state except for the air inlet and outlet, with the air inlet on the left and the air outlet on the right. B is the test result of acetaldehyde molecules in the octagonal cage. The red arrow is the vector synthesis direction of the system's airflow field, so the distinguishing point for the chi-square test is the red line segment on the right side of B. After acetaldehyde gas was introduced and the test lasted 2 minutes, more fruit flies chose the right side of the distinguishing point, that is, the fruit flies formed an avoidance behavior towards acetaldehyde. The chi-square test showed that the distribution of the results was significantly different, and the training was successful.

[0081] This invention implements rational aversion olfactory training based on ligand recognition mechanisms. Specifically, it uses molecular simulation and docking to predict OR-VOC matching relationships, which serve as a prerequisite for aversion training. This allows for rational experimental guidance and overcomes the obstacles and limitations of preference training that are susceptible to interference. Furthermore, it achieves "insect olfactory biomimetic" based on theories such as deep learning. Through computational biology, it reflects insect swarm behavior toward odor sources and mimics the decision-making process of the insect brain during sniffing, improving the speed and reliability of target judgment and decision-making.

[0082] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and it should be understood that the scope of protection of the present invention is not limited to such specific descriptions and embodiments. Those skilled in the art can make various other specific variations and combinations based on the technical teachings disclosed in the present invention without departing from the essence of the present invention, and such variations and combinations are still within the scope of protection of the invention.

Claims

1. An insect bionic odor detection method based on ligand recognition mechanism and aversion training, characterized in that: The following steps are involved: S1: Matching insects and target VOCs based on ligand recognition mechanism; The S1 includes the following steps: S11: Search and compare the target VOC with the molecules that the insects can smell in the iORbase database; S12: Based on the search and comparison results, determine whether the target VOC is already in the iORbase database. If so, complete the matching of the insect and the target VOC. If not, use molecular simulation technology to replace the training insect to complete the matching of the insect and the target VOC. S2: aversive olfactory training with matching insects and target VOCs; S3: For the insects and target VOCs trained with aversive olfactory perception, the CNN hybrid model is used to identify various insects and their spatial distribution, and the relationship between the spatial distribution of insects and odor sources is obtained; The S3 includes the following sub-steps: S31: placing the insects that have completed the aversion olfactory training in a transparent white-bottomed octagonal cage, placing an odor source containing the target VOC outside the transparent white-bottomed octagonal cage, and observing the spatial distribution data, density distribution data, and time required to form a stable distribution of the insects in the transparent white-bottomed octagonal cage; S32: Preprocessing the spatial distribution data, density distribution data, and time data required to form a stable distribution; S33: Using the preprocessed spatial distribution data, density distribution data, and the time data required to form a stable distribution, the CNN hybrid model is trained and evaluated to obtain the final CNN hybrid model; S34: Deploy the final CNN hybrid model to actual application scenarios to identify various insects and their spatial distribution, and obtain the relationship between insect spatial distribution and odor sources; The CNN hybrid model in S33 includes a CNN feature extraction layer, a feature processing and fusion layer, and other model layers connected in sequence; The CNN feature extraction layer includes an input layer, a convolution layer, an activation layer, and a pooling layer connected in sequence, wherein the input layer is used to receive image data, the convolution layer uses multiple convolution kernels to extract local features of the image, the activation layer applies a ReLU activation function, and the pooling layer is used to reduce feature dimensions while retaining important information; The feature processing and fusion layer includes a fully connected layer, which flattens the features extracted by the CNN feature extraction layer into a one-dimensional vector; The other model layers include a loop layer and a classification and regression layer connected in sequence, wherein the loop layer uses a long short-term memory network (LSTM) to process sequence data, and the classification and regression layer uses a fully connected layer to perform classification tasks, regression tasks, or access a support vector machine (SVM) for prediction tasks; S4: Based on the relationship between insect spatial distribution and odor sources, multi-point recognition is used to depict the VOC distribution map of the insect target area, completing insect bionic odor detection based on ligand recognition mechanism and aversion training.

2. The insect bionic odor detection method based on ligand recognition mechanism and aversion training according to claim 1 is characterized in that: The S2 includes the following steps: S21: A set number of insects were placed in a training tube lined with a copper mesh and were made to smell the target VOC and then the ambient gas; S22: stimulation is performed when the insect smells the target VOC, constituting a training cycle; S23: Use a gradient voltage increase and perform three training cycles for each voltage to obtain an appropriate training voltage to complete the aversive olfactory training for the matched insects and target VOCs.

3. The insect bionic odor detection method based on ligand recognition mechanism and aversion training according to claim 1 is characterized in that: The data preprocessing in S32 includes image data normalization and data enhancement. The image data normalization includes scaling the image pixel values ​​to a range of 0-1, and the data enhancement includes rotation, scaling, and cropping.

Images