A method and system for predicting activity of lb2 cas12a protein mutants
By digitally encoding the amino acid sequence of the Lb2Cas12a protein and training the model, a random forest regression model was constructed, which solved the problem of the difficulty in efficiently screening Lb2Cas12a protein mutants in the existing technology, and realized rapid and accurate prediction and screening of mutant activity.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- WUHAN UNIV
- Filing Date
- 2022-12-07
- Publication Date
- 2026-05-08
AI Technical Summary
Existing technologies require extensive expert knowledge and experimentation to find efficient mutants of the Lb2Cas12a protein, and have limitations, making it difficult to accurately select important mutation sites.
The amino acid sequence was digitally encoded using the AAindex of KARS160105, and a random forest regression model was constructed. The model was trained and validated using training, validation, and test sets to predict the activity of single mutants of the Lb2Cas12a protein.
Within three minutes, all possible single mutants can be traversed, and efficient mutants can be quickly screened out, saving manpower and resources and improving the accuracy and efficiency of mutant activity prediction.
Smart Images
Figure CN116153403B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of biomedical technology, and in particular to a method and system for predicting the activity of Lb2Cas12a protein mutants. Background Technology
[0002] The CRISPR / Cas system, as the most widely used gene editing system, has attracted the attention of many research teams worldwide. Researchers hope to improve the CRISPR / Cas system by reducing off-target rates, increasing the cleavage efficiency of Cas proteins, and overcoming PAM limitations. CRISPR / Cas12a has a lower off-target rate than CRISPR / Cas9 and comparable editing efficiency. Lb2Cas12a is a protein in the Cas12a family. Compared to other proteins in the Cas12a family, such as AsCas12f, LbCas12a, and FnCas12a, Lb2Cas12a has lower off-target effects and comparable cleavage activity. Could we make the Lb2Cas12a protein have even higher editing efficiency? Based on this idea, we began searching for mutants that could increase the editing efficiency of the Lb2Cas12a protein. The traditional process of finding more efficient protein mutants requires a great deal of expert knowledge, including an understanding of the properties of amino acids and the interactions between amino acids and between amino acids and nucleotides. Only then can a number of target sites be selected for experimental verification. Moreover, the selected target sites are a very small part of the entire amino acid sequence of the protein, and some important mutation sites may be missed, thus having certain limitations.
[0003] Therefore, there is an urgent need to develop a method and system for predicting the activity of Lb2Cas12a protein mutants that is simple to operate and highly accurate. Summary of the Invention
[0004] The purpose of this invention is to provide a method and system for predicting the activity of Lb2Cas12a protein mutants. This model can be used for rapid initial screening when searching for highly efficient mutants of Lb2Cas12a protein, thereby saving a lot of manpower and resources.
[0005] To achieve the above objectives, the present invention adopts the following technical solution:
[0006] In a first aspect of the present invention, a method for predicting the activity of Lb2Cas12a protein mutants is provided, the method comprising:
[0007] Obtain the amino acid sequence of the Lb2Cas12a protein mutant and the corresponding raw data of enzyme activity;
[0008] The amino acid sequence was digitally encoded using the AAindex index numbered KARS160105 to obtain the digital amino acid sequence.
[0009] The digitized amino acid sequence is processed to obtain an original dataset consisting of a first feature value and a second feature value. The original dataset is then randomly divided into a training set, a validation set, and a test set. The mean of the digitized amino acid sequence is taken as the first feature value. The mutation position is obtained by comparing the amino acid sequence of the Lb2Cas12a protein mutant with the original sequence, and the number of the mutation position in the digitized amino acid sequence is taken as the second feature value.
[0010] Construct a preliminary data model containing hyperparameters, and repeatedly train and validate the preliminary dataset processing model using data from the training set and validation set to obtain an adjusted dataset processing model;
[0011] The adjusted dataset processing model was tested using data from the test set to obtain a prediction model for Lb2Cas12a protein mutant activity.
[0012] Based on the number of amino acids in the Lb2Cas12a protein and the fact that each amino acid position can be mutated into 20 amino acids, a single mutant dataset of the Lb2Cas12a protein was obtained.
[0013] The Lb2Cas12a protein mutant activity prediction model was used to process the single mutant dataset of the Lb2Cas12a protein one by one to obtain the prediction results of the Lb2Cas12a protein single mutant activity.
[0014] Furthermore, the step of using the AAindex numbered KARS160105 to digitally encode the amino acid sequence to obtain a digitized amino acid sequence specifically includes:
[0015] The amino acids A, L, R, K, N, M, D, F, C, P, Q, S, E, T, G, W, H, Y, I, and V correspond to the following numbers: 1.00, 5.00, 8.12, 7.00, 5.00, 5.40, 5.17, 7.00, 2.33, 4.00, 5.86, 1.67, 6.00, 3.25, 0.00, 11.10, 6.71, 8.88, 3.25, and 3.25.
[0016] In a second aspect of the invention, a system for predicting the activity of Lb2Cas12a protein mutants is provided, the system comprising:
[0017] The raw data acquisition module is used to obtain the amino acid sequence of the Lb2Cas12a protein mutant and the corresponding raw data of enzyme activity.
[0018] The digitization encoding module is used to digitize the amino acid sequence using the AAindex index numbered KARS160105 to obtain the digitized amino acid sequence.
[0019] The data processing module is used to process the digitized amino acid sequence to obtain an original dataset composed of a first feature value and a second feature value, and randomly divide the original dataset into a training set, a validation set, and a test set; wherein, the mean of the digitized amino acid sequence is taken as the first feature value; based on the comparison between the amino acid sequence of the Lb2Cas12a protein mutant and the original sequence, the mutation position is obtained, and the number of the mutation position in the digitized amino acid sequence is taken as the second feature value;
[0020] The adjusted dataset processing model building module is used to build a preliminary data model containing hyperparameters. The preliminary dataset processing model is repeatedly trained and validated using data in the training set and validation set to obtain the adjusted dataset processing model.
[0021] The adjusted dataset processing model building module is used to test the adjusted dataset processing model using data in the test set to obtain the adjusted dataset processing model.
[0022] The model testing module is used to test the adjusted dataset processing model using data in the test set to obtain the Lb2Cas12a protein mutant activity prediction model.
[0023] The module for obtaining single mutant datasets of Lb2Cas12a protein is used to obtain single mutant datasets of Lb2Cas12a protein based on the number of amino acids in Lb2Cas12a protein and the fact that each amino acid position may mutate into 20 amino acids.
[0024] The Lb2Cas12a protein mutant activity prediction module is used to process the single mutant dataset of the Lb2Cas12a protein one by one using the Lb2Cas12a protein mutant activity prediction model to obtain the prediction results of Lb2Cas12a protein mutant activity.
[0025] In a third aspect of the invention, a system for predicting the activity of Lb2Cas12a protein mutants is provided, the system comprising:
[0026] A processor and a memory, the memory being coupled to the processor, the memory storing instructions that, when executed by the processor, cause the Lb2Cas12a protein mutant activity prediction to perform the steps of the method.
[0027] In a fourth aspect of the invention, a computer-readable storage medium is provided having a computer program stored thereon, characterized in that the computer program, when executed by a processor, implements the steps of the method.
[0028] One or more technical solutions in the embodiments of the present invention have at least the following technical effects or advantages:
[0029] This invention provides a method for predicting the activity of Lb2Cas12a protein mutants. A predictive model for the cleavage efficiency of Lb2Cas12a mutant proteins in the CRISPR / Cas system was constructed. Using this model, all possible single mutants of the Lb2Cas12a protein were identified within three minutes, yielding the mutant sequences and their predicted enzyme activities. This model can be used for rapid initial screening when searching for highly efficient Lb2Cas12a protein mutants, thus saving significant manpower and resources. Since this technique is effective for Lb2Cas12a, it can be transferred to other proteases, and is likely to be effective in the initial screening of high-activity mutants of other proteases as well. Attached Figure Description
[0030] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0031] Figure 1 A flowchart illustrating a method for predicting the activity of Lb2Cas12a protein mutants provided in this embodiment of the invention;
[0032] Figure 2 The graph shows the cleavage efficiency of wild-type and mutant proteins at the CTLA4 target. The letters on the horizontal axis represent: WT for wild-type, and AJ for ten mutants: A: K686; B: K564R; C: K575R; D: F577R; E: K580R; F: F585R; G: K597R; H: K602R; I: K603R; J: H626R. The vertical axis represents the cleavage efficiency of the ten mutants relative to the wild-type, and the numbers at the top of the bars indicate the experimentally obtained cleavage efficiency.
[0033] Figure 3The graph shows the cleavage efficiency of wild-type and mutant proteins at the PRKCH target site. The letters on the horizontal axis represent: WT for wild-type, and AJ for ten mutants: A: K686; B: K564R; C: K575R; D: F577R; E: K580R; F: F585R; G: K597R; H: K602R; I: K603R; J: H626R. The vertical axis represents the cleavage efficiency of the ten mutants relative to the wild-type, and the numbers at the top of the bars indicate the experimentally obtained cleavage efficiency. Detailed Implementation
[0034] The present invention will be described in detail below with reference to specific embodiments and examples, thereby making the advantages and various effects of the present invention more clearly apparent. Those skilled in the art should understand that these specific embodiments and examples are for illustrative purposes only and are not intended to limit the present invention.
[0035] Throughout this specification, unless otherwise specified, the terminology used herein should be understood as having the meaning commonly used in the art. Therefore, unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. In the event of any conflict, this specification shall prevail.
[0036] Unless otherwise specified, all raw materials, reagents, instruments and equipment used in this invention can be obtained by purchasing them from the market or by existing methods.
[0037] The following will describe in detail a method for predicting the activity of Lb2Cas12a protein mutants according to the present invention, with reference to specific embodiments.
[0038] Example 1: A method for predicting the activity of Lb2Cas12a protein mutants
[0039] This invention provides a method for predicting the activity of Lb2Cas12a protein mutants, such as... Figure 1 As shown, the method includes:
[0040] S1. Obtain the amino acid sequence of the Lb2Cas12a protein mutant and the corresponding raw data of enzyme activity;
[0041] As one specific implementation method:
[0042] The original data is:
[0043] Table 1
[0044]
[0045]
[0046] S2. The amino acid sequence is digitized using the AAindex index numbered KARS160105 to obtain the digitized amino acid sequence; the AAindex index is shown in Table 2.
[0047] Table 2
[0048]
[0049] For example, if the amino acid sequence is ALRKNMDFCPQSETGWHYIV, then the digitized encoding is: 1.00, 5.00, 8.12, 7.00, 5.00, 5.40, 5.17, 7.00, 2.33, 4.00, 5.86, 1.67, 6.00, 3.25, 0.00, 11.10, 6.71, 8.88, 3.25, 3.25.
[0050] S3. The digitized amino acid sequence is processed to obtain an original dataset consisting of a first feature value and a second feature value. The original dataset is then randomly divided into a training set, a validation set, and a test set. The mean of the digitized amino acid sequence is taken as the first feature value. The mutation position is obtained by comparing the amino acid sequence of the Lb2Cas12a protein mutant with the original sequence, and the number of the mutation position in the digitized amino acid sequence is taken as the second feature value. The mean and position of the 34 samples are shown in Table 3.
[0051] Table 3
[0052] sample mean Location sample mean Location Sample 1 4.94213101 0. Sample 18 4.94072139 396. Sample 2 4.94289386 68. Sample 19 4.94370647 445. Sample 3 4.94026534 75. Sample 20 4.93873134 445. Sample 4 4.94358209 75. Sample 21 4.94273632 455. Sample 5 4.93891376 137. Sample 22 4.94213101 551. Sample 6 4.93746269 147. Sample 23 4.94307629 571. Sample 7 4.94130182 164. Sample 24 4.94120232 755. Sample 8 4.94710614 203. Sample 25 4.93809287 755. Sample 9 4.94482587 203. Sample 26 4.94441128 805. Sample 10 4.94157546 238. Sample 27 4.94213101 805. Sample 11 4.94344113 238. Sample 28 4.94576285 1003. Sample 12 4.93968491 338. Sample 29 4.94289386 1003. Sample 13 4.93678275 338. Sample 30 4.94756219 1003. Sample 14 4.9414262 379. Sample 31 4.94037313 1011. Sample 15 4.94087894 381. Sample 32 4.94025705 1011. Sample 16 4.94218905 388. Sample 33 4.94037313 1095. Sample 17 4.94083748 396. Sample 34 4.93746269 1114.
[0053] The original dataset is randomly divided into a training set, a validation set, and a test set, mainly using the train_test_split() function for random division; the ratio of the number of samples in the training set, validation set, and test set is 7:1.5:1.5.
[0054] S4. Construct a preliminary data model containing hyperparameters, and repeatedly train the preliminary dataset processing model using the data in the training set and validation set to obtain the adjusted dataset processing model.
[0055] The initial dataset processing model is a random forest regression model. The hyperparameters of the initial dataset processing model are random_state = 42, n_estimators = 21, max_depth = 7, min_samples_split = 3, min_samples_leaf = 1, and the remaining parameters are set to default values.
[0056] The specific method for repeatedly training the preliminary dataset processing model using data from the training set and validation set includes adjusting hyperparameters based on the performance of the validation set; the adjusted dataset processing model is a random forest regression model.
[0057] S5. Test the adjusted dataset processing model using the data in the test set to obtain the Lb2Cas12a protein mutant activity prediction model.
[0058] The Lb2Cas12a protein mutant activity prediction model is a random forest regression model. The hyperparameters of the Lb2Cas12a protein mutant activity prediction model are random_state=42, n_estimators=21, max_depth=7, min_samples_split=3, min_samples_leaf=1, and the remaining hyperparameters are set to default values.
[0059] The R2 and RMSE values on the training set are 0.8744749414671693 and 7.71304300703707, respectively.
[0060] The R² and RMSE values on the validation set are 0.679893692715831 and 7.644514860852709, respectively.
[0061] The R2 and RMSE on the test set are 0.8414481394244004 and 4.613458763625018, respectively.
[0062] S6. Based on the number of amino acids in the Lb2Cas12a protein and the fact that each amino acid position may mutate into 20 amino acids, obtain a dataset of single mutants of the Lb2Cas12a protein.
[0063] The Lb2Cas12a protein single mutant dataset is a dataset consisting of all possible single-point mutations of the unknown Lb2Cas12a protein.
[0064] Since the Lb2Cas12a protein has 1206 amino acids, and proteins are composed of twenty amino acids, meaning there are 1206 positions, and each position has twenty possibilities, the Lb2Cas12a protein single mutant dataset is a dataset consisting of 1206 multiplied by 20.
[0065] S7. The Lb2Cas12a protein mutant activity prediction model is used to process the single mutant dataset of the Lb2Cas12a protein one by one to obtain the prediction results of Lb2Cas12a protein mutant activity.
[0066] (1) The amino acids of the dataset consisting of all possible single-point mutations of the unknown Lb2Cas12a protein were digitized according to the AAindex index numbered KARS160105.
[0067] (2) Feature extraction was performed on the dataset consisting of all possible single-point mutations of the unknown Lb2Cas12a protein. The first feature was the average value of all amino acids after digitization, and the second feature was the location of the mutation.
[0068] (3) Use these two features as inputs to the Lb2Cas12a protein mutant activity prediction model.
[0069] (4) The model provides predictions.
[0070] Example 2: Validation of the model's good performance in predicting the activity of single mutants of the Lb2Cas12a protein.
[0071] I. Experimental verification steps:
[0072] 1. The DNA sequence encoding the Lb2Cas12a protein was mutated using a two-step PCR method;
[0073] The two-step PCR and validation steps are as follows:
[0074] ① Based on the location of the mutation and the information of the amino acid change, design an upstream primer containing the mutation site by looking up the codon table. Select a suitable downstream primer 500-1000 bp downstream of the upstream primer. Then, use the plasmid to be mutated as a template to perform the first PCR amplification.
[0075] ② Take a small amount of PCR amplification product for gel electrophoresis to confirm successful PCR amplification. After confirming successful PCR amplification, recover the PCR product and determine its concentration.
[0076] ③ Using the recovered product of the first PCR reaction as primers and the template from the first PCR reaction as templates, a second PCR reaction is performed to amplify the entire plasmid.
[0077] ④ Take a small amount of the second PCR product for gel electrophoresis. After confirming that the second PCR amplification was successful, recover the PCR product.
[0078] ⑤ The product of the second PCR was digested with DpnI enzyme in a 37°C water bath for 30 minutes to digest the original template containing methyl groups. The enzyme digestion reaction system is shown in Table 4:
[0079] Table 4
[0080]
[0081] ⑥ Add 10 μl of the enzyme digestion product to DH5α cells, incubate the mixture on ice for 30 minutes, heat shock at 42°C for 90 seconds, and then incubate on ice for 5 minutes to allow the mutant plasmid to transform into the cells. Then, expand the cells in LB liquid medium at 37°C for 45 minutes.
[0082] ⑦ Spread the expanded cell culture onto LB solid medium containing ampicillin and incubate at 37°C for 12 hours.
[0083] ⑧ Pick a single colony with a needle and place it in 500 μl of LB liquid medium. Incubate in a shaker at 37°C until turbid. Then, take 200 μl and send it for sequencing.
[0084] ⑨ After the sequencing results are obtained, the sequencing sequence is compared with the target gene sequence. If there is a mismatch at the mutation site but a good match at other sites, it means that the mutation was successful.
[0085] 2. Transfect plasmids containing mutant DNA into E. coli DH5α prokaryotic cells to express the protein;
[0086] 3. The expressed Lb2Cas12a protein was transduced into HEK29 human renal epithelial cells;
[0087] 4. Subsequently, a T7E1 restriction enzyme digestion experiment was performed to obtain the cleavage band of the mutant Lb2Cas12a protein.
[0088] 5. The results of grayscale analysis of the strips are as follows: Figure 2 and Figure 3 As shown.
[0089] II. Experimental Results
[0090] The cleavage efficiency of wild-type and mutant proteins at the CTLA4 target site is as follows: Figure 2 As shown, the cleavage efficiency of wild-type and mutant proteins at the PRKCH target site is as follows: Figure 3 As shown.
[0091] Figure 2 The cleavage efficiency of wild-type and mutant strains at the CTLA4 target was demonstrated. Figure 2We can observe that in the first replicate experiment targeting CTLA4, the cleavage efficiency of the wild-type was 0.434, while that of mutant F was 0.458, representing a 5.5% increase in cleavage efficiency compared to the wild-type. In the second replicate experiment, the cleavage efficiency of the wild-type was 0.423, while that of mutant F was 0.451, representing a 6.6% increase in cleavage efficiency compared to the wild-type. In the third replicate experiment, the cleavage efficiency of the wild-type was 0.421, while that of mutant F was 0.461, representing a 9.5% increase in cleavage efficiency compared to the wild-type. Overall, the average cleavage efficiency of mutant F relative to the wild-type across the three replicate experiments was 1.072, representing a 7.2% increase in cleavage efficiency compared to the wild-type. In the three replicate experiments, except for mutant F, whose cleavage efficiency was higher than that of wild type, the cleavage efficiency of the other mutants was lower than that of wild type. Among them, mutants B and J showed a more significant decrease in cleavage efficiency compared to wild type. This indicates that K at position 564 and H at position 626 of Lb2Cas12a protein are important for the cleavage efficiency of Lb2Cas12a protein at the CTLA4 target site. Mutations at these two sites are more damaging to Lb2Cas12a protein.
[0092] Figure 3 The data shows the cleavage efficiency of the Lb2Cas12a protein mutant and wild-type at the PRKCH target. In the first replicate, the wild-type cleavage efficiency was 0.29, mutant E's cleavage efficiency was 0.416 (a 43.4% increase compared to wild-type), and mutant J's cleavage efficiency was 0.412 (a 42% increase compared to wild-type). In the second replicate, the wild-type cleavage efficiency was 0.395, mutant E's cleavage efficiency was 0.433 (a 9.6% increase compared to wild-type), and mutant J's cleavage efficiency was 0.451 (a 14.2% increase compared to wild-type). In the third replicate, the wild-type cleavage efficiency was 0.358, mutant E's cleavage efficiency was 0.381 (a 6.4% increase compared to wild-type), and mutant J's cleavage efficiency was 0.391 (a 9.2% increase compared to wild-type). Mutants E and J showed higher cleavage efficiency at the target PRKCH site than the wild type in three replicate experiments. In the three replicate experiments, the average cleavage efficiency of mutant E relative to the wild type was 1.198, an increase of 19.8% on average, while the average cleavage efficiency of mutant J relative to the wild type was 1.218, an increase of 21.8% on average.
[0093] comprehensive Figure 2 and Figure 3Specifically, mutant F showed a 7% increase in cleavage efficiency at target CTLA4 compared to the wild type, while mutants E and J showed approximately a 20% increase in cleavage efficiency at target PRKCH compared to the wild type. These two figures, in addition to cleavage efficiency, also show information about the target specificity of the mutants. Mutant F showed a 7% increase in cleavage efficiency at target CTLA4 compared to the wild type, but its cleavage efficiency at target PRKCH was lower, averaging a 26.6% decrease across three replicate experiments. This indicates that in Lb2Cas12a, F at position 585 is significantly associated with the cleavage of Lb2Cas12a protein at target PRKCH. Mutants E and J showed approximately a 20% increase in cleavage efficiency at the target site PRKCH compared to the wild type, but their cleavage efficiency at the target site CTLA4 was lower. At the CTLA4 target site, mutant E's average cleavage efficiency decreased by 12.3% compared to the wild type, and mutant J's average cleavage efficiency decreased by 36.3% compared to the wild type. This indicates that the K at position 580 and the H at position 626 of the Lb2Cas12a protein are closely related to the cleavage of the Lb2Cas12a protein at the target site PRKCH. Mutations at these two sites reduced the cleavage efficiency of the Lb2Cas12a protein at the PRKCH target site.
[0094] III. The results of the activity prediction of the Lb2Cas12a protein single mutant using the model are as follows: The mutant with increased activity relative to the wild type was obtained by this method, indicating that this method is effective in discovering improved protein mutation sites.
[0095] Example 3: A system for predicting the activity of Lb2Cas12a protein mutants.
[0096] This invention provides a system for predicting the activity of Lb2Cas12a protein mutants, the system comprising:
[0097] The raw data acquisition module is used to obtain the amino acid sequence of the Lb2Cas12a protein mutant and the corresponding raw data of enzyme activity.
[0098] The digitization encoding module is used to digitize the amino acid sequence using the AAindex index numbered KARS160105 to obtain the digitized amino acid sequence.
[0099] The data processing module is used to process the digitized amino acid sequence to obtain an original dataset composed of a first feature value and a second feature value, and randomly divide the original dataset into a training set, a validation set, and a test set; wherein, the mean of the digitized amino acid sequence is taken as the first feature value; based on the comparison between the amino acid sequence of the Lb2Cas12a protein mutant and the original sequence, the mutation position is obtained, and the number of the mutation position in the digitized amino acid sequence is taken as the second feature value;
[0100] The adjusted dataset processing model building module is used to build a preliminary data model containing hyperparameters. The preliminary dataset processing model is repeatedly trained and validated using data in the training set and validation set to obtain the adjusted dataset processing model.
[0101] The adjusted dataset processing model building module is used to test the adjusted dataset processing model using data in the test set to obtain the adjusted dataset processing model.
[0102] The model testing module is used to test the adjusted dataset processing model using data in the test set to obtain the Lb2Cas12a protein mutant activity prediction model.
[0103] The module for obtaining single mutant datasets of Lb2Cas12a protein is used to obtain single mutant datasets of Lb2Cas12a protein based on the number of amino acids in Lb2Cas12a protein and the fact that each amino acid position may mutate into 20 amino acids.
[0104] The Lb2Cas12a protein mutant activity prediction module is used to process the single mutant dataset of Lb2Cas12a protein one by one using the Lb2Cas12a protein mutant activity prediction model to obtain the prediction results of Lb2Cas12a protein mutant activity.
[0105] Example 4: A system for predicting the activity of Lb2Cas12a protein mutants.
[0106] This invention provides a system for predicting the activity of Lb2Cas12a protein mutants, the system comprising:
[0107] A processor and a memory, the memory being coupled to the processor, the memory storing instructions that, when executed by the processor, cause the system for predicting the activity of the Lb2Cas12a protein mutant to perform the steps of the method described in Example 1.
[0108] Example 5: A computer-readable storage medium
[0109] This invention provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the method of embodiment 1.
[0110] Of course, the computer-executable instructions provided in the embodiments of the present invention are not limited to the method operations described above, but can also perform related operations in the methods provided in any embodiment of the present invention.
[0111] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause an electronic device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.
[0112] It is worth noting that the various units and modules included in the above embodiments are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy differentiation and are not used to limit the scope of protection of the present invention.
[0113] Finally, it should be noted that the terms “comprising,” “including,” or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.
[0114] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.
[0115] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.
Claims
1. A method for predicting the activity of Lb2Cas12a protein mutants, characterized in that, The method includes: Obtain the amino acid sequence of the Lb2Cas12a protein mutant and the corresponding raw data of enzyme activity; The amino acid sequence was digitally encoded using the AAindex index numbered KARS160105 to obtain the digital amino acid sequence. The digitized amino acid sequence is processed to obtain an original dataset consisting of a first feature value and a second feature value. The original dataset is then randomly divided into a training set, a validation set, and a test set. The mean of the digitized amino acid sequence is taken as the first feature value. The mutation position is obtained by comparing the amino acid sequence of the Lb2Cas12a protein mutant with the original sequence, and the number of the mutation position in the digitized amino acid sequence is taken as the second feature value. A preliminary data model with hyperparameters is constructed, and the preliminary data model is repeatedly trained and validated using data from the training set and validation set to obtain an adjusted dataset processing model. The adjusted dataset processing model is a random forest regression model. The hyperparameters of the adjusted dataset processing model are random_state=42, n_estimators=21, max_depth=7, min_samples_split=3, min_samples_leaf=1, and the remaining hyperparameters are set to default values. The adjusted dataset processing model was tested using data from the test set to obtain a prediction model for Lb2Cas12a protein mutant activity. Based on the number of amino acids in the Lb2Cas12a protein and the fact that each amino acid position can be mutated into 20 amino acids, a single mutant dataset of the Lb2Cas12a protein was obtained. The Lb2Cas12a protein mutant activity prediction model was used to process the single mutant dataset of the Lb2Cas12a protein one by one to obtain the prediction results of the Lb2Cas12a protein single mutant activity.
2. The method for predicting the activity of Lb2Cas12a protein mutants according to claim 1, characterized in that, The ratio of the number of samples in the training set, validation set, and test set is 7:1.5:1.
5.
3. The method for predicting the activity of Lb2Cas12a protein mutants according to claim 1, characterized in that, The preliminary data model is a random forest regression model. The hyperparameters of the preliminary data model are random_state=42, n_estimators=21, max_depth=7, min_samples_split=3, min_samples_leaf=1, and the remaining parameters are set to default values.
4. The method for predicting the activity of Lb2Cas12a protein mutants according to claim 1, characterized in that, The process of using the Lb2Cas12a protein mutant activity prediction model to process each single mutant dataset of the Lb2Cas12a protein to obtain the prediction results of the Lb2Cas12a protein single mutant activity specifically includes: The Lb2Cas12a protein single mutant dataset was digitized according to the AAindex index numbered KARS160105, and then feature extraction was performed. The first feature was the average value of all digitized amino acids, and the second feature was the mutation location. The two feature values were input into the Lb2Cas12a protein mutant activity prediction model to give the prediction results.
5. A system for predicting the activity of Lb2Cas12a protein mutants, characterized in that, The system includes: The raw data acquisition module is used to obtain the amino acid sequence of the Lb2Cas12a protein mutant and the corresponding raw data of enzyme activity. The digitization encoding module is used to digitize the amino acid sequence using the AAindex index numbered KARS160105 to obtain the digitized amino acid sequence. The data processing module is used to process the digitized amino acid sequence to obtain an original dataset composed of a first feature value and a second feature value, and randomly divide the original dataset into a training set, a validation set, and a test set; wherein, the mean of the digitized amino acid sequence is taken as the first feature value; based on the comparison between the amino acid sequence of the Lb2Cas12a protein mutant and the original sequence, the mutation position is obtained, and the number of the mutation position in the digitized amino acid sequence is taken as the second feature value; The adjusted dataset processing model building module is used to construct a preliminary data model containing hyperparameters. The preliminary data model is repeatedly trained and validated using data from the training and validation sets to obtain the adjusted dataset processing model. The adjusted dataset processing model is a random forest regression model. The hyperparameters of the adjusted dataset processing model are random_state=42, n_estimators=21, max_depth=7, min_samples_split=3, min_samples_leaf=1, and the remaining hyperparameters are set to default values. The model testing module is used to test the adjusted dataset processing model using data in the test set to obtain the Lb2Cas12a protein mutant activity prediction model. The module for obtaining single mutant datasets of Lb2Cas12a protein is used to obtain single mutant datasets of Lb2Cas12a protein based on the number of amino acids in Lb2Cas12a protein and the fact that each amino acid position may mutate into 20 amino acids. The Lb2Cas12a protein mutant activity prediction module is used to process the single mutant dataset of the Lb2Cas12a protein one by one using the Lb2Cas12a protein mutant activity prediction model to obtain the prediction results of Lb2Cas12a protein mutant activity.
6. A system for predicting the activity of Lb2Cas12a protein mutants, characterized in that, The system includes: A processor and a memory, the memory being coupled to the processor, the memory storing instructions that, when executed by the processor, cause the Lb2Cas12a protein mutant activity prediction to perform the steps of the method according to any one of claims 1-4.
7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1-4.