Screening method and system for synthetic biological probiotics

Abstract

The invention belongs to the technical field of synthetic biological probiotic screening, and particularly relates to a synthetic biological probiotic screening method and system.The method comprises the steps that a target area in a DNA sequence is searched, K-mer frequencies in the target area are counted, the frequencies of K-mers with the same basic groups at the first k positions are sequentially put into a set according to the letter sequence of the residual basic groups of the K-mers, and the K-mers are screened; splicing all the sets into a frequency characteristic matrix of the target area in rows according to the alphabetic sequence of the first k bases, forming a multi-channel frequency characteristic matrix by the frequency characteristic matrixes of all the target areas, and inputting the multi-channel frequency characteristic matrix into a neural network model with a channel attention module for training; and obtaining a multi-channel frequency characteristic matrix of the to-be-screened synthetic biological probiotics from the DNA sequence of the to-be-screened synthetic biological probiotics, and inputting the multi-channel frequency characteristic matrix into the trained neural network model to obtain scores of the synthetic biological probiotics. The primary screening device can improve the primary screening efficiency of the synthetic biological probiotics.

Description

Technical Field

[0001] The present invention belongs to the technical field of screening of synthetic biological probiotics, and specifically relates to a screening method and system for synthetic biological probiotics. Background Art

[0002] Synthetic biological probiotics are based on traditional probiotics and are genetically modified to enhance their functions, such as improving their therapeutic effects on specific diseases, enhancing their adaptability to specific environments, and giving them new functions, such as producing specific bioactive substances, degrading environmental pollutants, etc. Synthetic biological probiotics have broad application prospects in the fields of medicine, food, agriculture, and environmental protection. Genetic screening of synthetic probiotics mainly relies on traditional molecular biology methods, including genome sequencing, gene editing technology, high-throughput screening, and functional verification. Although these methods can meet research and application needs to a certain extent, there are still some shortcomings. For example, traditional methods require a lot of time and experimental resources, and it is difficult to fully cover the complexity and diversity of gene combinations.

[0003] The rapid development of artificial intelligence has provided new solutions for the screening of synthetic biological probiotics. By using artificial intelligence technologies such as machine learning and deep learning, potential functional gene combinations can be mined from massive amounts of genetic data, gene editing strategies can be predicted and optimized, and screening efficiency and accuracy can be greatly improved. However, how to efficiently and accurately screen out probiotics that are beneficial to the human body is still a difficult problem that needs to be solved. Summary of the invention

[0004] The purpose of the present invention is to provide a method for screening synthetic biological probiotics to solve the problems raised in the above background technology.

[0005] A method for screening synthetic biological probiotics comprises the following steps: Obtain the DNA sequences of beneficial bacteria and non-beneficial bacteria in the training samples, search for regions with the same position and the same DNA fragments in all DNA sequences, and use the remaining regions in the DNA sequences as target regions; wherein the length of the DNA fragment is not less than a preset value; Count the K-mer frequencies in the target region, put the frequencies of K-mers with the same first k bases into a set in alphabetical order of the remaining bases of the K-mer, concatenate all the sets into a frequency feature matrix of the target region in alphabetical order of the first k bases, construct a multi-channel frequency feature matrix from the frequency feature matrices of all the target regions, and input the multi-channel frequency feature matrix into a neural network model with a channel attention module for training; A multi-channel frequency feature matrix of the synthetic biological probiotics to be screened is obtained from the DNA sequence of the synthetic biological probiotics to be screened, and the multi-channel frequency feature matrix is ​​input into the trained neural network model to obtain the score of the synthetic biological probiotics.

[0006] Preferably, the frequencies of K-mers with the same first k bases are put into a set in alphabetical order of the bases in the remaining part of the K-mer, specifically: Obtain the frequencies corresponding to all K-mers with the same first k bases, sort the frequencies in the order of the remaining base letters, and put the sorted frequencies into the set of first k base identifiers.

[0007] Preferably, the channel attention module specifically includes: A channel weight calculation unit is used to count the functional elements of the target area, set different weights for different functional elements, calculate the weights of all functional elements in the target area, normalize the weights of all target areas to obtain the normalized channel weights, and thus obtain the channel weight vector; A channel attention calculation unit is used to calculate a channel attention vector through a channel attention mechanism; The fusion unit is used to fuse the channel weight vector and the channel attention vector, and use the fusion result as the output of the channel attention module.

[0008] Preferably, the channel attention module specifically includes: A channel weight unit is used to obtain, for each target region, the frequency of occurrence of each base at the same position in the target region in all beneficial bacteria of the training sample, and obtain the frequency of occurrence of each base at the same position in the target region in all non-beneficial bacteria in the training sample, obtain the contribution of the target region to distinguishing beneficial bacteria from non-beneficial bacteria based on the frequency, normalize the contribution of all target regions to obtain the normalized channel weight, and thus obtain the channel weight vector; Channel attention unit, used to calculate the channel attention vector through the channel attention mechanism; The fusion unit is used to fuse the channel weight vector and the channel attention vector, and use the fusion result as the output of the channel attention module.

[0009] Preferably, the fusion of the channel weight vector and the channel attention vector is specifically as follows: The channel weight vector and the channel attention vector are input into the fully connected layer and then passed through an activation function to obtain a fusion result.

[0010] On the other hand, the present invention provides a screening system for synthetic biological probiotics, comprising the following modules: A deduplication module is used to obtain the DNA sequences of beneficial bacteria and non-beneficial bacteria in the training samples, search for regions with the same position and the same DNA fragments in all DNA sequences, and use the remaining regions in the DNA sequences as target regions; wherein the length of the DNA fragment is not less than a preset value; The model training module is used to count the K-mer frequencies in the target region, put the frequencies of K-mers with the same first k bases into a set in alphabetical order of the remaining bases of the K-mer, concatenate all the sets into a frequency feature matrix of the target region in row order according to the alphabetical order of the first k bases, form a multi-channel frequency feature matrix from the frequency feature matrices of all the target regions, and input the multi-channel frequency feature matrix into a neural network model with a channel attention module for training; The screening module is used to obtain a multi-channel frequency feature matrix of the synthetic biological probiotics to be screened from the DNA sequence of the synthetic biological probiotics to be screened, and input the multi-channel frequency feature matrix into the trained neural network model to obtain the score of the synthetic biological probiotics.

[0011] Preferably, the frequencies of K-mers with the same first k bases are put into a set in alphabetical order of the bases in the remaining part of the K-mer, specifically: Obtain the frequencies corresponding to all K-mers with the same first k bases, sort the frequencies in the order of the remaining base letters, and put the sorted frequencies into the set of first k base identifiers.

[0012] Preferably, the channel attention module specifically includes: A channel weight calculation unit is used to count the functional elements of the target area, set different weights for different functional elements, calculate the weights of all functional elements in the target area, normalize the weights of all target areas to obtain the normalized channel weights, and thus obtain the channel weight vector; A channel attention calculation unit is used to calculate a channel attention vector through a channel attention mechanism; The fusion unit is used to fuse the channel weight vector and the channel attention vector, and use the fusion result as the output of the channel attention module.

[0013] Preferably, the channel attention module specifically includes: A channel weight unit is used to obtain, for each target region, the frequency of occurrence of each base at the same position in the target region in all beneficial bacteria of the training sample, and obtain the frequency of occurrence of each base at the same position in the target region in all non-beneficial bacteria in the training sample, obtain the contribution of the target region to distinguishing beneficial bacteria from non-beneficial bacteria based on the frequency, normalize the contribution of all target regions to obtain the normalized channel weight, and thus obtain the channel weight vector; Channel attention unit, used to calculate the channel attention vector through the channel attention mechanism; The fusion unit is used to fuse the channel weight vector and the channel attention vector, and use the fusion result as the output of the channel attention module.

[0014] Preferably, the fusion of the channel weight vector and the channel attention vector is specifically as follows: The channel weight vector and the channel attention vector are input into the fully connected layer and then passed through an activation function to obtain a fusion result.

[0015] In addition, the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the method described above is implemented.

[0016] Finally, the present invention also proposes a computer device, which includes at least a processor and a readable storage medium, wherein a computer program is stored on the readable storage medium, and the computer program implements the above method when executed by the processor.

[0017] Compared with the prior art, the present invention has the following beneficial effects: In order to screen out synthetic biological probiotics, the present invention searches out target regions of DNA sequences, constructs a K-mer frequency feature matrix for each target region, and multiple target regions constitute a multi-channel frequency feature matrix. The multi-channel frequency feature matrix is ​​used to identify beneficial bacteria, and channel attention is applied to the identification of the multi-channel frequency feature matrix. In addition, when calculating the channel attention, it not only relies on the K-mer frequency, but also calculates according to the characteristics of each target region. At the same time, the functions or similarities of the K-mer frequency characteristics and the base sequences of the target regions are utilized, thereby achieving more accurate identification and screening of synthetic biological probiotics. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a flow chart of Embodiment 1; Figure 2 A schematic diagram of the frequency characteristic matrix of a target area; Figure 3 It is a schematic diagram of multi-channel frequency characteristic matrix; Figure 4 This is a structural diagram of the second embodiment. DETAILED DESCRIPTION

[0019] In this article, relational terms such as first and second, etc. are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprise", "include" 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. In the absence of further restrictions, the elements defined by the statement "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0020] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] In a first embodiment of the present invention, Figure 1 As shown, the present invention provides a method for screening synthetic biological probiotics, comprising the following steps: S1, obtaining the DNA sequences of beneficial bacteria and non-beneficial bacteria in the training sample, searching for regions with the same position and the same DNA fragment in all DNA sequences, and taking the remaining regions in the DNA sequence as the target region; wherein the length of the DNA fragment is not less than a preset value; Bacteria are divided into beneficial bacteria and non-beneficial bacteria, such as lactobacillus, bifidobacterium, etc., and non-beneficial bacteria are divided into harmful bacteria and harmless bacteria, such as Escherichia coli and Staphylococcus aureus, which are harmful bacteria, and Staphylococcus epidermidis and Providencia belong to harmless bacteria or neutral bacteria, which are generally harmless to humans. Synthetic biological probiotics are modified probiotics, which can be used to treat diseases, detect harmful substances, decompose harmful substances, etc. After designing and constructing and modifying probiotics through carriers, it is generally necessary to undergo preliminary screening and culture experiments to determine which are synthetic biological probiotics. When a large number of synthetic biological probiotics are present, the workload of screening out beneficial bacteria is relatively large. The present invention uses artificial intelligence to preliminarily screen out beneficial bacteria, and then conducts culture experiments, which can reduce a lot of workload.

[0022] The DNA sequences in the training samples are obtained from the database. The training samples include beneficial bacteria and non-beneficial bacteria. The DNA sequences in the database are from some public databases and some self-tested bacterial DNA sequences. Preferably, the training samples are different strains of the same species. For example, if the goal is to synthesize new Escherichia coli, the DNA sequences of Escherichia coli of different strains are used as training samples. Species with close genetic relationships can also be used to ensure that the length of the DNA sequence is the same. Since the bacterial DNA of the close genetic relationship or different strains is also relatively similar, in order to reduce the amount of data for subsequent analysis, the regions with the same position and the same DNA fragment in all DNA sequences in the training samples are searched. These regions are in the same position in the DNA sequence and the DNA fragments are the same, which is not helpful for distinguishing beneficial bacteria from non-beneficial bacteria. The remaining regions in the DNA sequence are used as target regions. There will be multiple target regions, and the continuous DNA sequence between every two identical regions is a target region. Among them, if the DNA fragment is one or two bases, the DNA sequence will be too messy. In one embodiment, the length of the DNA fragment is not less than the preset value.

[0023] S2, counting the K-mer frequencies in the target region, putting the frequencies of K-mers with the same first k bases into a set in alphabetical order of the remaining bases of the K-mer, concatenating all the sets into a frequency feature matrix of the target region in row order according to the alphabetical order of the first k bases, forming a multi-channel frequency feature matrix from all the frequency feature matrices of the target region, and inputting the multi-channel frequency feature matrix into a neural network model with a channel attention module for training; For each target region of each DNA sequence in the training sample, the K-mer frequency is counted, where K-mer is a DNA sequence substring of length K, for example, K=2, then the substrings are: AA, AT, AC, AG, TA, TT, TC, TG, CA, CT, CC, CG, GA, GT, GC, GG, where A is adenine, T is thymine, C is cytosine, and G is guanine. In the present invention, K-mer represents a specific DNA substring of length K, and K-mers represent all DNA substrings of length K, for example, AA is a K-mer, and the above sixteen substrings are K-mers. Wherein, K is a positive integer greater than 2. The K-mer frequency is the frequency of a substring of length K appearing in a DNA sequence. In a more specific embodiment, a K-mer is slid in the target region with a step size of 1. If the target region is the same as the K-mer, the K-mer appears once, otherwise it does not appear, and the K-mer frequency is calculated in this way.

[0024] After calculating the frequencies of all K-mers, the frequencies of K-mers with the same first k bases are put into a set in the alphabetical order of the remaining bases of the K-mer. Specifically, the frequencies corresponding to all K-mers with the same first k bases are obtained, and the frequencies are sorted in the order of the remaining base letters. The sorted frequencies are put into the set of the first k base identifiers, where the remaining bases are the bases from the Kkth to the Kth bases of the K-mer. Assuming K=3, k=2, the first two bases of the four 3-mers AAA, AAT, AAC, and AAG are the same. If their corresponding frequencies are p1, p2, p3, and p4 respectively, and the alphabetical order of the remaining bases of the 3-mer is A, C, G, and T, they are put into a set in this order as {p1, p3, p4, and p2}. Similarly, the corresponding sets of ACA, ACT, ACC, and ACG can be obtained. In a preferred embodiment, k is also a positive integer, and k <K。 Figure 2 The frequency feature matrix of a target area with k=1 and K=3 is shown.

[0025] After obtaining the set corresponding to all K-mers, the set is spliced ​​into a frequency feature matrix by row according to the alphabetical order of the first k bases. Taking the above example, when k=2, there are 16 combinations of the first two bases. According to the alphabetical order of the first k bases, AA is in the first row, AC is in the second row, and so on. Finally, the frequency feature matrix of the target area is obtained. When there are multiple target areas, each target area corresponds to a frequency feature matrix. Multiple frequency feature matrices of multiple target areas constitute a multi-channel frequency feature matrix. Each channel is a frequency feature matrix of a target area. The multi-channel frequency feature matrix is ​​arranged in the order in which the target areas appear in the DNA sequence, such as Figure 3 shown.

[0026] Different channels in the multi-channel frequency feature matrix represent the characteristics of different target areas. Different target areas have different positions and functions. The present invention further uses channel attention to pay attention to different channels, i.e., target areas, dynamically adjust the importance of each channel, and enhance the impact of different target areas on the results. In a more specific embodiment, the neural network model with a channel attention module is a SENet model or a CBAM model, etc., which is of course not limited to the above models, and can also be other classification models with an attention module added.

[0027] The K-mer frequency matrix can reflect some characteristics of the target region, but the functional information of the target region is lost. In a specific embodiment, Channel attention module, specifically including: A channel weight calculation unit is used to count the functional elements of the target area, set different weights for different functional elements, calculate the weights of all functional elements in the target area, normalize the weights of all target areas to obtain the normalized channel weights, and thus obtain the channel weight vector; A channel attention calculation unit is used to calculate a channel attention vector through a channel attention mechanism; The fusion unit is used to fuse the channel weight vector and the channel attention vector, and use the fusion result as the output of the channel attention module.

[0028] Among them, the functional element is a subsequence or substring with a specific function in the DNA sequence, including but not limited to codons, regulatory elements, enhancers, promoters, etc. Due to the different functions of different functional elements, different weights are set for different functional elements, and the weights of all functional elements in the target area are calculated to obtain the weight of the target area. The weights of all target areas are normalized to obtain the normalized channel weights. Since one target area corresponds to a frequency feature matrix, the channel attention and channel weight of the frequency feature matrix are combined to recalculate the channel attention, that is, the channel weight and the channel attention vector are fused. Assuming there are 3 target areas, there will be 3 frequency feature matrices, and there will be three normalized channel weights. The attention of each channel is obtained through the frequency feature matrix of 3 channels, that is, the channel attention vector containing 3 elements. The 3 channel weights and the channel attention vector containing 3 elements are fused, and the fusion result is used as the output of the attention module. Among them, there are many fusion methods, including but not limited to bitwise multiplication, fusion using MLP, etc.

[0029] Different target areas have different differences between beneficial bacteria and non-beneficial bacteria. Some target areas have large differences. These areas are likely to be functional areas of beneficial bacteria and non-beneficial bacteria. In one embodiment, the channel attention module specifically includes: A channel weight unit is used to obtain the frequency of occurrence of each base at the same position in the target region of all beneficial bacteria in the training sample, and obtain the frequency of occurrence of each base at the same position in the target region of all non-beneficial bacteria in the training sample, and obtain the contribution of the target region to distinguishing beneficial bacteria from non-beneficial bacteria based on the frequency, and normalize the contribution of all target regions to obtain the normalized channel weight, thereby obtaining a channel weight vector; Channel attention unit, used to calculate the channel attention vector through the channel attention mechanism; A fusion unit is used to fuse the channel weight vector and the channel attention vector, and use the fusion result as the output of the channel attention module; For each target region, each position is a base such as one of A, T, G, and C. For the same position in the same target region, the probability of each base appearing in all beneficial bacteria is counted. For example, in the second target region, the probability of A appearing at the 18th position is 0.8, the probability of T appearing is 0.1, the probability of G appearing is 0.1, and the probability of C appearing is 0; similarly, the probability of each base appearing in all non-beneficial bacteria is counted. Taking the above position as an example, the probability of A appearing is 0.2, the probability of T appearing is 0.5, the probability of G appearing is 0.1, and the probability of C appearing is 0.2.

[0030] Based on the above two frequencies, the contribution of the beneficial bacteria and non-beneficial bacteria in the target area is obtained. The contribution refers to the contribution of distinguishing beneficial bacteria or non-beneficial bacteria. In one embodiment, it is specifically as follows: the absolute difference of the frequency corresponding to the same base in the same position is calculated, so as to obtain the absolute difference sum of the frequencies of the four bases in the position, and the average value of the absolute difference sum of the frequencies of all positions in the target area is calculated, and the average value is used as the contribution. Still taking the above as an example, all are the 18th, and the absolute difference of the frequency corresponding to the base ATGC is 0.6, 0.4, 0, 0.2, and the absolute difference sum of the frequency is 1.2, and the absolute difference of the frequency is the absolute value of the frequency difference. Assuming that there are 20 bases in the target area 2, the average value of the absolute difference sum of the frequencies corresponding to the 20 bases is calculated, and this average value is used as the contribution. The greater the difference of the target area in beneficial bacteria and non-beneficial bacteria, the greater the contribution to distinguishing beneficial bacteria and non-beneficial bacteria.

[0031] In yet another embodiment, the channel attention module only includes a channel weight unit, and the channel weight vector output by the channel weight unit is used as the weight value of each channel.

[0032] Among them, each target region corresponds to a weight, and the channel weight vector is obtained according to the order in which the target regions appear in the DNA sequence. For example, if there are 10 target regions, the channel weight vector has 10 elements, and the channel attention vector also has 10 elements. The elements in the channel attention vector are calculated by performing a channel attention mechanism on the multi-channel frequency feature matrix. The channel attention mechanism adopts the attention mechanism including but not limited to SENet and STN.

[0033] After obtaining the fusion result, the fusion result and the multi-channel frequency feature matrix are multiplied to achieve attention to different channels.

[0034] S3, obtaining a multi-channel frequency feature matrix of the synthetic biological probiotics to be screened from the DNA sequence of the synthetic biological probiotics to be screened, and inputting the multi-channel frequency feature matrix into the trained neural network model to obtain the score of the synthetic biological probiotics.

[0035] After the neural network model is trained, for the DNA sequence of the synthetic biological probiotics to be screened, the multi-channel frequency matrix is ​​obtained in the same manner as in S1 and S2. Since the neural network model includes a channel attention module, the channel attention of the synthetic biological probiotics to be identified is also calculated. The process is the same as above and will not be repeated here. The multi-channel frequency matrix is ​​input into the trained neural network model to obtain the score of the synthetic biological probiotics. The higher the score, the greater the possibility that it is a probiotic.

[0036] In a second embodiment of the present invention, Figure 4 As shown, the present invention proposes a screening system for synthetic biological probiotics, comprising the following modules: The deduplication module is used to obtain the DNA sequences of beneficial bacteria and non-beneficial bacteria in the training samples, search for regions with the same position and the same DNA fragments in all DNA sequences, and use the remaining regions in the DNA sequences as target regions; wherein the length of the DNA fragment is not less than a preset value; The model training module is used to count the K-mer frequencies in the target region, put the frequencies of K-mers with the same first k bases into a set in alphabetical order of the remaining bases of the K-mer, concatenate all the sets into a frequency feature matrix of the target region in row order according to the alphabetical order of the first k bases, form a multi-channel frequency feature matrix from the frequency feature matrices of all the target regions, and input the multi-channel frequency feature matrix into a neural network model with a channel attention module for training; The screening module is used to obtain a multi-channel frequency feature matrix of the synthetic biological probiotics to be screened from the DNA sequence of the synthetic biological probiotics to be screened, and input the multi-channel frequency feature matrix into the trained neural network model to obtain the score of the synthetic biological probiotics.

[0037] Preferably, the frequencies of K-mers with the same first k bases are put into a set in alphabetical order of the remaining bases of the K-mer, specifically: Get the frequencies corresponding to all K-mers with the same first k bases, sort the frequencies in the order of the remaining base letters, and put the sorted frequencies into the set of first k base identifiers.

[0038] Preferably, the channel attention module specifically includes: A channel weight calculation unit is used to count the functional elements of the target area, set different weights for different functional elements, calculate the weights of all functional elements in the target area, normalize the weights of all target areas to obtain the normalized channel weights, and thus obtain the channel weight vector; A channel attention calculation unit is used to calculate a channel attention vector through a channel attention mechanism; The fusion unit is used to fuse the channel weight vector and the channel attention vector, and use the fusion result as the output of the channel attention module.

[0039] Preferably, the channel attention module specifically includes: A channel weight unit is used to obtain the frequency of occurrence of each base at the same position in the target region of all beneficial bacteria in the training sample, and obtain the frequency of occurrence of each base at the same position in the target region of all non-beneficial bacteria in the training sample, and obtain the contribution of the target region to distinguishing beneficial bacteria from non-beneficial bacteria based on the frequency, and normalize the contribution of all target regions to obtain the normalized channel weight, thereby obtaining a channel weight vector; Channel attention unit, used to calculate the channel attention vector through the channel attention mechanism; The fusion unit is used to fuse the channel weight vector and the channel attention vector, and use the fusion result as the output of the channel attention module.

[0040] Preferably, the channel weight vector and the channel attention vector are fused, specifically: The channel weight vector and channel attention vector are input into the fully connected layer and then passed through the activation function to obtain the fusion result.

[0041] In a third embodiment of the present invention, the present invention further proposes a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the method of the first embodiment is implemented.

[0042] In a fourth embodiment of the present invention, the present invention further proposes a computer device, which includes at least a memory and a processor. A computer program is stored in the memory, and when the computer program is executed by the processor, the method of the first embodiment is implemented.

[0043] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by adding a necessary general hardware platform, and of course can also be implemented by combining hardware and software. Based on such an understanding, the above technical solution can essentially or in other words be embodied in the form of a computer product, and the present invention can be implemented in the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes. Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it, and other embodiments may also be used. Although the present invention has been described in detail with reference to the aforementioned embodiments, ordinary technicians in this field should understand that they can still modify the technical solutions recorded in the aforementioned embodiments, or replace some of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for screening synthetic biological probiotics, characterized in that: The following steps are involved: Obtain the DNA sequences of beneficial bacteria and non-beneficial bacteria in the training samples, search for regions with the same position and the same DNA fragments in all DNA sequences, and use the remaining regions in the DNA sequences as target regions; wherein the length of the DNA fragment is not less than a preset value; Count the K-mer frequencies in the target region, put the frequencies of K-mers with the same first k bases into a set in alphabetical order of the remaining bases of the K-mer, concatenate all the sets into a frequency feature matrix of the target region in alphabetical order of the first k bases, construct a multi-channel frequency feature matrix from the frequency feature matrices of all the target regions, and input the multi-channel frequency feature matrix into a neural network model with a channel attention module for training; A multi-channel frequency feature matrix of the synthetic biological probiotics to be screened is obtained from the DNA sequence of the synthetic biological probiotics to be screened, and the multi-channel frequency feature matrix is ​​input into the trained neural network model to obtain the score of the synthetic biological probiotics.

2. The method for screening synthetic biological probiotics according to claim 1, characterized in that: The frequencies of K-mers with the same first k bases are put into a set in alphabetical order of the remaining bases of the K-mer, specifically: Obtain the frequencies corresponding to all K-mers with the same first k bases, sort the frequencies in the order of the remaining base letters, and put the sorted frequencies into the set of first k base identifiers.

3. The method for screening synthetic biological probiotics according to claim 1, characterized in that: The channel attention module specifically includes: A channel weight calculation unit is used to count the functional elements of the target area, set different weights for different functional elements, calculate the weights of all functional elements in the target area, normalize the weights of all target areas to obtain the normalized channel weights, and thus obtain the channel weight vector; A channel attention calculation unit is used to calculate a channel attention vector through a channel attention mechanism; The fusion unit is used to fuse the channel weight vector and the channel attention vector, and use the fusion result as the output of the channel attention module.

4. The method for screening synthetic biological probiotics according to claim 1, characterized in that: The channel attention module specifically includes: A channel weight unit is used to obtain, for each target region, the frequency of occurrence of each base at the same position in the target region in all beneficial bacteria of the training sample, and obtain the frequency of occurrence of each base at the same position in the target region in all non-beneficial bacteria in the training sample, obtain the contribution of the target region to distinguishing beneficial bacteria from non-beneficial bacteria based on the frequency, normalize the contribution of all target regions to obtain the normalized channel weight, and thus obtain the channel weight vector; Channel attention unit, used to calculate the channel attention vector through the channel attention mechanism; The fusion unit is used to fuse the channel weight vector and the channel attention vector, and use the fusion result as the output of the channel attention module.

5. The method for screening synthetic biological probiotics according to claim 3 or 4, characterized in that: The fusion of the channel weight vector and the channel attention vector is specifically as follows: The channel weight vector and the channel attention vector are input into the fully connected layer and then passed through an activation function to obtain a fusion result.

6. A screening system for synthetic biological probiotics, characterized in that: Includes the following modules: A deduplication module is used to obtain the DNA sequences of beneficial bacteria and non-beneficial bacteria in the training samples, search for regions with the same position and the same DNA fragments in all DNA sequences, and use the remaining regions in the DNA sequences as target regions; wherein the length of the DNA fragment is not less than a preset value; The model training module is used to count the K-mer frequencies in the target region, put the frequencies of K-mers with the same first k bases into a set in alphabetical order of the remaining bases of the K-mer, concatenate all the sets into a frequency feature matrix of the target region in row order according to the alphabetical order of the first k bases, form a multi-channel frequency feature matrix from the frequency feature matrices of all the target regions, and input the multi-channel frequency feature matrix into a neural network model with a channel attention module for training; The screening module is used to obtain a multi-channel frequency feature matrix of the synthetic biological probiotics to be screened from the DNA sequence of the synthetic biological probiotics to be screened, and input the multi-channel frequency feature matrix into the trained neural network model to obtain the score of the synthetic biological probiotics.

7. The screening system for synthetic biological probiotics according to claim 6, characterized in that: The frequencies of K-mers with the same first k bases are put into a set in alphabetical order of the remaining bases of the K-mer, specifically: Obtain the frequencies corresponding to all K-mers with the same first k bases, sort the frequencies in the order of the remaining base letters, and put the sorted frequencies into the set of first k base identifiers.

8. The screening system for synthetic biological probiotics according to claim 6, characterized in that: The channel attention module specifically includes: A channel weight calculation unit is used to count the functional elements of the target area, set different weights for different functional elements, calculate the weights of all functional elements in the target area, normalize the weights of all target areas to obtain the normalized channel weights, and thus obtain the channel weight vector; A channel attention calculation unit is used to calculate a channel attention vector through a channel attention mechanism; The fusion unit is used to fuse the channel weight vector and the channel attention vector, and use the fusion result as the output of the channel attention module.

9. The screening system for synthetic biological probiotics according to claim 6, characterized in that: The channel attention module specifically includes: A channel weight unit is used to obtain, for each target region, the frequency of occurrence of each base at the same position in the target region in all beneficial bacteria of the training sample, and obtain the frequency of occurrence of each base at the same position in the target region in all non-beneficial bacteria in the training sample, obtain the contribution of the target region to distinguishing beneficial bacteria from non-beneficial bacteria based on the frequency, normalize the contribution of all target regions to obtain the normalized channel weight, and thus obtain the channel weight vector; Channel attention unit, used to calculate the channel attention vector through the channel attention mechanism; The fusion unit is used to fuse the channel weight vector and the channel attention vector, and use the fusion result as the output of the channel attention module.

10. A computer storage device, wherein a computer program is stored on the storage device, characterized in that: When the computer program is executed by a processor, the computer program implements the method according to any one of claims 1 to 5.

Images