A method and device for predicting the difficulty of DNA fragment synthesis

Abstract

The present application discloses a method and device for predicting the difficulty of DNA fragment synthesis, the method comprising: obtaining a difficulty feature group from a DNA fragment; inputting the difficulty feature group into a synthesis difficulty evaluation model, evaluating the synthesis difficulty of the DNA fragment by the synthesis difficulty evaluation model, and obtaining a difficulty score; predicting the synthesis difficulty of the DNA fragment according to the difficulty score by the synthesis difficulty evaluation model. The method provided by the present application can realize quantitative evaluation of the synthesis difficulty of a DNA fragment.

Description

Technical Field

[0001] The present application relates to the field of DNA synthesis, and in particular to a method and device for predicting the difficulty of DNA fragment synthesis. Background Art

[0002] During the synthesis of large DNA fragments, some difficult-to-synthesize fragments are often encountered, which seriously affects the overall progress. Repeated experiments and method modifications for these difficult-to-synthesize fragments require a lot of manpower, funds and time. If the difficult-to-synthesize sequences can be identified in advance, the synthesis methods and resource investment can be adjusted in advance, or the difficulty of sequence synthesis can be reduced by codon optimization, replacement of related genes, etc., thereby reducing the cost of large DNA fragment synthesis and assembly and improving the success rate.

[0003] The current method for analyzing the difficulty of DNA fragment synthesis is to predict the difficulty of sequence synthesis by analyzing the length of the repeat sequence, the number of secondary structures, and the change curve of the GC content of base pairs. However, this method is based on the experience of relevant technicians and lacks deductive reasoning based on data and analysis of the complex correlation between features. Summary of the invention

[0004] In view of this, the present application provides a method and device for predicting the difficulty of DNA fragment synthesis, aiming to achieve quantitative evaluation of the difficulty of DNA fragment synthesis.

[0005] In order to achieve the above objectives, this application provides the following technical solutions:

[0006] One aspect of the present application provides a method for predicting the difficulty of DNA fragment synthesis, the method comprising:

[0007] Obtaining difficulty feature groups from DNA fragments;

[0008] Inputting the difficulty feature group into a synthesis difficulty evaluation model, and performing synthesis difficulty evaluation on the DNA fragment using the synthesis difficulty evaluation model to obtain a difficulty score;

[0009] The synthesis difficulty of the DNA fragment is predicted according to the difficulty score by the synthesis difficulty evaluation model.

[0010] Optionally, each difficulty feature in the difficulty feature group includes: a repeat count of 12 base pairs bp, a repeat count of 6bp, a maximum value of the base pair GC content in a 20bp fragment, a minimum value of the information entropy in a 100bp fragment, an occurrence frequency of the trinucleotide GGC, and an overall free energy of the sequence.

[0011] Optionally, obtaining a difficulty feature group from the DNA fragment includes:

[0012] Extracting a first sequence feature group from the DNA fragment according to the perspectives of repeat sequence, hairpin structure, poly structure, GC content, information entropy, free energy, base frequency variance, and trinucleotide frequency;

[0013] Calculating importance scores of a plurality of sequence features included in the first sequence feature group;

[0014] Screening the sequence features included in the first sequence feature group according to the partial least squares dimensionality reduction method, removing the sequence features whose importance scores are lower than a preset first threshold, and obtaining a second sequence feature group;

[0015] The sequence features included in the second sequence feature group are screened according to the XGBoost algorithm, and sequence features with importance scores lower than a preset second threshold are removed to obtain the difficulty feature group.

[0016] Optionally, the method further includes:

[0017] Calculating the contribution value of each difficulty feature in the difficulty feature group to the difficulty score according to the additive interpretation SHAP method through the synthetic difficulty evaluation model;

[0018] The maximum value of the contribution value is determined according to the synthetic difficulty evaluation model, and the difficulty feature corresponding to the maximum value of the contribution value is determined.

[0019] Optionally, predicting the synthesis difficulty of the DNA fragment according to the difficulty score by using the synthesis difficulty evaluation model includes:

[0020] If the difficulty score is greater than or equal to a preset score threshold, the synthesis difficulty evaluation model is used to determine that the DNA fragment belongs to a difficult-to-synthesize category;

[0021] If the difficulty score is less than a preset score threshold, the synthesis difficulty evaluation model is used to determine that the DNA fragment belongs to the easy-to-synthesize category.

[0022] Another aspect of the present application provides a device for predicting the difficulty of DNA fragment synthesis, the device comprising: an acquisition module, a processing module and a prediction module;

[0023] The acquisition module is used to acquire a difficulty feature group from the DNA fragment;

[0024] The processing module is used to input the difficulty feature group into a synthesis difficulty evaluation model, and perform synthesis difficulty evaluation on the DNA fragment through the synthesis difficulty evaluation model to obtain a difficulty score;

[0025] The prediction module is used to predict the synthesis difficulty of the DNA fragment according to the difficulty score through the synthesis difficulty evaluation model.

[0026] Optionally, each difficulty feature in the difficulty feature group includes: a repeat count of 12 base pairs bp, a repeat count of 6bp, a maximum value of the base pair GC content in a 20bp fragment, a minimum value of the information entropy in a 100bp fragment, an occurrence frequency of the trinucleotide GGC, and an overall free energy of the sequence.

[0027] Optionally, the device further comprises: an extraction module;

[0028] The extraction module is used to extract the first sequence feature group from the DNA fragment according to the angles of repetitive sequence, hairpin structure, poly structure, GC content, information entropy, free energy, base frequency variance, and trinucleotide frequency;

[0029] The processing module is further used to calculate the importance scores of the plurality of sequence features included in the first sequence feature group;

[0030] The processing module is further used to screen the sequence features included in the first sequence feature group according to the partial least squares dimensionality reduction method, remove the sequence features whose importance scores are lower than a preset first threshold, and obtain a second sequence feature group;

[0031] The processing module is further used to screen the sequence features included in the second sequence feature group according to the XGBoost algorithm, remove the sequence features whose importance scores are lower than a preset second threshold, and obtain the difficulty feature group.

[0032] Optionally, the device is also used for:

[0033] The processing module is further used to calculate the contribution value of each difficulty feature in the difficulty feature group to the difficulty score by using the additive interpretation SHAP method;

[0034] The processing module is further used to determine the maximum value of the contribution value according to the synthetic difficulty evaluation model, and determine the difficulty feature corresponding to the maximum value of the contribution value.

[0035] Optionally, the prediction module is further used for:

[0036] If the difficulty score is greater than or equal to a preset score threshold, the prediction module is further used to determine that the DNA fragment belongs to a difficult-to-synthesize category through the synthesis difficulty evaluation model;

[0037] If the difficulty score is less than a preset score threshold, the prediction module is further configured to determine that the DNA fragment belongs to the easy-to-synthesize category through the synthesis difficulty evaluation model.

[0038] The present application discloses a method and device for predicting the difficulty of synthesizing a DNA fragment, the method comprising: obtaining a difficulty feature group from a DNA fragment; inputting the difficulty feature group into a synthesis difficulty evaluation model, evaluating the synthesis difficulty of the DNA fragment by the synthesis difficulty evaluation model, and obtaining a difficulty score; and predicting the synthesis difficulty of the DNA fragment according to the difficulty score by the synthesis difficulty evaluation model. The method provided by the present application can realize quantitative evaluation of the synthesis difficulty of a DNA fragment. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0040] Figure 1 A schematic diagram of a method for predicting the difficulty of DNA fragment synthesis provided in an embodiment of the present application;

[0041] Figure 2 A schematic diagram of a process for screening a difficulty feature group provided in an embodiment of the present application;

[0042] Figure 3 A schematic diagram of another method for predicting the difficulty of DNA fragment synthesis provided in an embodiment of the present application;

[0043] Figure 4 An example diagram showing the proportion of difficult-to-synthesize short fragments in a long DNA fragment provided in an embodiment of the present application;

[0044] Figure 5 An example diagram showing the difficulty scores of DNA long fragments and medium-length fragments provided in an embodiment of the present application;

[0045] Figure 6 An example diagram of a contribution value of a difficulty feature calculated by a SHAP algorithm provided in an embodiment of the present application;

[0046] Figure 7 A schematic diagram of the structure of a device for predicting the difficulty of DNA fragment synthesis provided in an embodiment of the present application;

[0047] Figure 8 This is an example diagram of a DNA fragment synthesis difficulty score provided in an embodiment of the present application. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0049] Combine the following Figure 1 , introduces a method for predicting the difficulty of DNA fragment synthesis provided by an embodiment of the present application. A method for predicting the difficulty of DNA fragment synthesis can be implemented through the following steps S11-S13.

[0050] S11: Obtaining difficulty feature groups from DNA fragments.

[0051] Specifically, a difficulty feature group is obtained from the DNA fragment, and the difficulty feature is used to analyze the synthesis difficulty of the DNA fragment.

[0052] Specifically, the difficulty feature group includes six important difficulty features: 12bp repeat count, 6bp repeat count, maximum value of base pair GC content in 20bp fragment, minimum value of information entropy in 100bp fragment, occurrence frequency of trinucleotide GGC and overall free energy of sequence.

[0053] Specifically, the screening method of the difficulty feature group is as follows: Figure 2 As shown in Figure 1, in this screening method, the area under the receiver operator characteristic curve (AUC) is used to measure the performance score of the model.

[0054] The following describes how the difficulty feature group in S11 is screened in conjunction with S111-S114.

[0055] S111: Extract the first sequence feature group from the DNA fragments according to the perspectives of repeat sequence, hairpin structure, poly structure, GC content, information entropy, free energy, base frequency variance, and trinucleotide frequency.

[0056] Specifically, sequence features related to synthesis difficulty can be extracted from DNA fragments according to the angles of repeat sequence, hairpin structure, poly structure, GC content, information entropy, free energy, base frequency variance, and trinucleotide frequency to form a first sequence feature group. The sequence feature group includes 108 dimensions of sequence features related to synthesis difficulty.

[0057] S112: Calculate the importance score of each sequence feature included in the first sequence feature group.

[0058] Specifically, each sequence feature included in the first sequence feature group is projected into a vector space, and the importance score of each sequence feature in the vector space is calculated by using the Partial Least Squares (PLS) dimensionality reduction method.

[0059] S113: Screening the sequence features included in the first sequence feature group according to the partial least squares dimensionality reduction method, removing the sequence features whose importance scores are lower than a preset first threshold, and obtaining a second sequence feature group.

[0060] Specifically, the sequence features included in the first sequence feature group are screened according to the partial least squares dimensionality reduction method, a PLS model is established, the sixth-to-last value of the importance score is used as the first threshold, and the five sequence features with importance scores less than the first threshold are removed to obtain the second sequence feature group, and the AUC is used to calculate whether the performance score of the current model reaches the preset third threshold.

[0061] If the performance score of the current model reaches the preset third threshold, the process proceeds to step S114.

[0062] If the performance score of the current model does not reach the preset third threshold, step S113 is performed again.

[0063] S114: The sequence features included in the second sequence feature group are screened according to the XGBoost algorithm, and sequence features whose importance scores are lower than a preset second threshold are removed to obtain a difficulty feature group.

[0064] Specifically, an XGBoost model is constructed according to a machine learning algorithm based on a gradient boosting tree (eXtreme GradientBoosting, XGBoost). The model uses a default gain method to calculate the importance scores of the sequence features included in the second sequence feature group during model establishment. The second to last value of the importance score is used as the second threshold. The sequence features with importance scores less than the second threshold are removed to obtain the difficulty feature group, and the AUC is used to calculate whether the performance score of the current model reaches the preset fourth threshold.

[0065] If the performance score of the current model reaches the preset fourth threshold, then the difficulty feature group is the desired difficulty feature group.

[0066] If the performance score of the current model does not reach the preset fourth threshold, step S114 is performed again.

[0067] Through a screening method provided by steps S111 to S114, six sequence features with the highest importance scores can be screened out from the sequence features contained in the DNA fragments to obtain a difficulty feature group.

[0068] It can be understood that the feature selection methods include but are not limited to PLS, regression model, ridge regression, minimum redundancy maximum correlation, etc.

[0069] S12: Input the difficulty feature group into the synthesis difficulty evaluation model, and evaluate the synthesis difficulty of the DNA fragment through the synthesis difficulty evaluation model to obtain a difficulty score.

[0070] Specifically, the difficulty feature group is input into the XGBoost algorithm to establish a model to obtain a synthesis difficulty evaluation model. The synthesis difficulty evaluation model evaluates the synthesis difficulty of the input DNA fragment to obtain a difficulty score.

[0071] It is understandable that the algorithms for building the model include but are not limited to XGBoost algorithm, support vector machine, random forest, logistic regression, etc.

[0072] S13: Predict the synthesis difficulty of the DNA fragment according to the difficulty score through the synthesis difficulty evaluation model.

[0073] Specifically, if the difficulty score is higher than a preset value, the synthesis difficulty evaluation model is used to determine that the DNA fragment belongs to the difficult-to-synthesize category.

[0074] If the difficulty score is lower than the preset value, the synthesis difficulty evaluation model is used to determine that the DNA fragment belongs to the easy synthesis category.

[0075] It is understandable that the probability of difficult synthesis predicted by the synthesis difficulty evaluation model is a difficulty score, which is a score ranging from 0 to 1, and the larger the value, the higher the difficulty. The preset value is 0.5 as an example. If the difficulty score is higher than 0.5, the DNA fragment belongs to the difficult synthesis category.

[0076] The method provided in the embodiment of the present application can quantitatively evaluate the synthesis difficulty of a DNA sequence. And through the machine learning algorithm, the complex interaction between multiple difficulty features such as repetition, GC content, information entropy, etc. can be learned in a data-driven manner, and the quantitative evaluation results are more accurate.

[0077] Combine the following Figure 3 , a method for predicting the difficulty of DNA fragment synthesis provided in an embodiment of the present application is introduced, which can be implemented through the following steps S31-S33.

[0078] In the embodiment of the present application, the difficulty score is represented by the difficult synthesis probability S-index.

[0079] S31: Divide the long DNA fragment into multiple short fragments of equal length.

[0080] Specifically, for a long DNA fragment to be synthesized, the long DNA fragment can be divided into a plurality of short fragments of equal length.

[0081] S32: Evaluate multiple short and long segments of equal length using a synthetic difficulty evaluation model to obtain difficulty scores and key difficulty features of each short and long segment of equal length.

[0082] Specifically, a plurality of short and long segments of equal length are evaluated by a synthetic difficulty evaluation model to obtain the S-index and key difficulty features of the short and long segments of equal length.

[0083] Specifically, the synthetic difficulty evaluation model calculates the contribution value of each difficulty feature to S-inde through an additive interpretation method for machine learning models (SHapleyAdditive exPlanations, SHAP).

[0084] like Figure 6 As shown, in these fragments of equal length, the 12-bp repeat count ( Figure 6 The value of Repeat ratio of 12-bp is small, indicating that the content of repeat sequences in such long and short fragments is low. Since the content of repeat sequences is low, it means that such long and short fragments are easy to synthesize, and its contribution to S-index is negative. Figure 6 In gray. Figure 6 The maximum GC content of base pairs in the 20 bp fragment ( Figure 6 MAX GC within 20-bp), minimum information entropy in a 100-bp fragment ( Figure 6 Min IE within100-bp) and the trinucleotide GGC ( Figure 6 The frequency of occurrence of GGC in the sequence is represented by a black box, indicating that the sequence containing the above-mentioned difficulty features has a high content of such fragments, making it difficult to synthesize such fragments. The black fragment with the minimum information entropy in the 100bp fragment is Figure 6 The longest one among them has the largest contribution to the S-index. Therefore, the key difficulty feature of these long and short fragments is the minimum value of information entropy in the 100bp fragment.

[0085] S33: According to the difficulty scores and key difficulty features of short and long fragments of equal length, the synthesis difficulty and key difficulty features of the DNA long fragment are obtained.

[0086] Specifically, a schematic diagram of the difficulty of synthesizing long DNA fragments is drawn based on the difficulty scores of short and long fragments of equal length. Figure 4 Taking the DNA fragment in as an example, the vertical axis distribution ratio (Distribution) is used to indicate the proportion of short fragments of equal length with various difficulty scores in the long DNA fragments.

[0087] according to Figure 4It is known that human chromosome 22 ( Figure 4 In H. sapiens chr22) has a greater proportion of difficult-to-synthesize fragments than easy-to-synthesize fragments, it can be judged that most of the DNA fragments of human chromosome 22 are difficult-to-synthesize DNA fragments, and the chromosome 2 of nematodes ( Figure 4 In C. elegans chrII), of which 37.6% of the short fragments of equal length were predicted to be difficult to synthesize fragments, chromosome 5 of Saccharomyces cerevisiae ( Figure 4 In S. cerevisiae synV, synV) among which 6.34% of the short fragments of equal length were predicted to be difficult to synthesize. It can be seen that the overall synthesis difficulty of synV is lower than that of chromosome 2 of nematodes and chromosome 22 of humans.

[0088] Specifically, Figure 8 For example, chromosome 2 of nematodes ( Figure 8 In C. elegans chrII) consists of 9 short and long fragments with scores of 0-1, among which the proportions of different difficulty score ranges are represented by different shapes. For difficult-to-synthesize fragments, the short DNA fragments with the highest difficulty scores can be analyzed based on the proportions, thereby obtaining key difficulty features.

[0089] Specifically, Figure 5 For example, the vertical axis is used to represent the difficulty score S-index, and the horizontal axis is used to represent the position on the chromosome (Chromosome Locatiao). Figure 5 In the sequence, each fragment of equal length is 2 kilobases (kb).

[0090] according to Figure 5 It can be seen that among the 50 2kb short fragments in the figure, 29 have an S-index higher than 0.5, which means that the overall synthesis difficulty of this long DNA fragment is higher than 0.5.

[0091] Specifically, according to the key difficulty characteristics of short and long fragments of equal length, the one with the highest repetition ratio is the key difficulty characteristic of the long fragment DNA.

[0092] Through the method provided in the embodiment of the present application, it is possible to calculate the difficulty score of a single sample and the corresponding key difficulty features for the single sample, and for long fragments, it is possible to calculate the overall synthetic difficulty and key difficulty features by splicing.

[0093] Combine the following Figure 7 , introducing a device for predicting the difficulty of DNA fragment synthesis provided by the present application, the device includes: an acquisition module 701, a processing module 702 and a prediction module 703.

[0094] The acquisition module 701 is used to acquire a difficulty feature group from a DNA fragment.

[0095] The processing module 702 is used to input the difficulty feature group into the synthesis difficulty evaluation model, and perform synthesis difficulty evaluation on the DNA fragment through the synthesis difficulty evaluation model to obtain a difficulty score.

[0096] The prediction module 703 is used to predict the synthesis difficulty of the DNA fragment according to the difficulty score through the synthesis difficulty evaluation model.

[0097] Specifically, the difficulty feature group includes: 12bp repeat count, 6bp repeat count, maximum value of base pair GC content in 20bp fragment, minimum value of information entropy in 100bp fragment, occurrence frequency of trinucleotide GGC and overall free energy of sequence.

[0098] Specifically, the device further includes: an extraction module 704 .

[0099] The extraction module 704 is used to extract the first sequence feature group from the DNA fragment according to the angles of repetitive sequence, hairpin structure, poly structure, GC content, information entropy, free energy, base frequency variance, and trinucleotide frequency.

[0100] The processing module 702 is further configured to calculate the importance score of each sequence feature included in the first sequence feature group.

[0101] The processing module 702 is further used to screen the sequence features included in the first sequence feature group according to the partial least squares dimensionality reduction method, remove the sequence features whose importance scores are lower than a preset first threshold, and obtain a second sequence feature group.

[0102] The processing module 702 is further used to screen the sequence features included in the second sequence feature group according to the XGBoost algorithm, remove the sequence features whose importance scores are lower than a preset second threshold, and obtain a difficulty feature group.

[0103] Specifically, the processing module 702 is further configured to calculate the contribution value of each difficulty feature in the difficulty feature group to the difficulty score by using the additive interpretation SHAP method.

[0104] The processing module 702 is further configured to obtain a difficulty feature corresponding to the maximum value according to the maximum value of the contribution value.

[0105] Specifically, the prediction module 703 is also used for:

[0106] If the difficulty score is higher than a preset value, the prediction module 703 is further used to determine whether the DNA fragment belongs to the difficult-to-synthesize category through a synthesis difficulty evaluation model.

[0107] If the difficulty score is lower than a preset value, the prediction module 703 is further used to determine whether the DNA fragment belongs to the easy-to-synthesize category through a synthesis difficulty evaluation model.

[0108] The device provided in the embodiment of the present application can quantitatively evaluate the synthesis difficulty of a DNA sequence. And through the machine learning algorithm, the complex interaction between multiple difficulty features such as repetition, GC content, information entropy, etc. can be learned in a data-driven manner, and the quantitative evaluation results are more accurate.

[0109] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0110] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0111] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical business division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0112] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0113] In addition, each business unit in each embodiment of the present application can be integrated into a processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software business units.

[0114] If the integrated unit is implemented in the form of a software business unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0115] Those skilled in the art will appreciate that in one or more of the above examples, the services described in the present invention may be implemented using hardware, software, firmware, or any combination thereof. When implemented using software, the services may be stored in a computer-readable medium or transmitted as one or more instructions or codes on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein communication media include any media that facilitates the transmission of computer programs from one place to another. Storage media may be any available media that can be accessed by a general-purpose or special-purpose computer.

[0116] The above specific implementation modes further describe the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above are only specific implementation modes of the present invention.

[0117] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person of ordinary skill in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for predicting the difficulty of DNA fragment synthesis, characterized in that: The method comprises: Obtaining difficulty feature groups from DNA fragments; Inputting the difficulty feature group into a synthesis difficulty evaluation model, and performing synthesis difficulty evaluation on the DNA fragment using the synthesis difficulty evaluation model to obtain a difficulty score; Predicting the synthesis difficulty of the DNA fragment according to the difficulty score by using the synthesis difficulty evaluation model; Each difficulty feature in the difficulty feature group includes: 12bp repeat count, 6bp repeat count, maximum value of base pair GC content in 20bp fragment, minimum value of information entropy in 100bp fragment, occurrence frequency of trinucleotide GGC and overall free energy of sequence; The method of obtaining a difficulty feature group from a DNA fragment includes: extracting a first sequence feature group from the DNA fragment according to the angles of a repeating sequence, a hairpin structure, a poly structure, a GC content, information entropy, free energy, base frequency variance, and a trinucleotide frequency; calculating importance scores of a plurality of sequence features included in the first sequence feature group; screening the sequence features included in the first sequence feature group according to a partial least squares dimensionality reduction method, removing sequence features with importance scores lower than a preset first threshold, and obtaining a second sequence feature group; screening the sequence features included in the second sequence feature group according to an XGBoost algorithm, removing sequence features with importance scores lower than a preset second threshold, and obtaining the difficulty feature group.

2. The method according to claim 1, characterized in that The method further comprises: Calculating the contribution value of each difficulty feature in the difficulty feature group to the difficulty score according to the additive interpretation SHAP method through the synthetic difficulty evaluation model; The maximum value of the contribution value is determined according to the synthetic difficulty evaluation model, and the difficulty feature corresponding to the maximum value of the contribution value is determined.

3. The method according to claim 1, characterized in that The step of predicting the synthesis difficulty of the DNA fragment according to the difficulty score by using the synthesis difficulty evaluation model includes: If the difficulty score is greater than or equal to a preset score threshold, the synthesis difficulty evaluation model is used to determine that the DNA fragment belongs to a difficult-to-synthesize category; If the difficulty score is less than a preset score threshold, the synthesis difficulty evaluation model is used to determine that the DNA fragment belongs to the easy-to-synthesize category.

4. A device for predicting the difficulty of DNA fragment synthesis, characterized in that: The device comprises: an acquisition module, a processing module and a prediction module; The acquisition module is used to acquire a difficulty feature group from the DNA fragment; The processing module is used to input the difficulty feature group into a synthesis difficulty evaluation model, and perform synthesis difficulty evaluation on the DNA fragment through the synthesis difficulty evaluation model to obtain a difficulty score; The prediction module is used to predict the synthesis difficulty of the DNA fragment according to the difficulty score through the synthesis difficulty evaluation model; Each difficulty feature in the difficulty feature group includes: 12bp repeat count, 6bp repeat count, maximum value of base pair GC content in 20bp fragment, minimum value of information entropy in 100bp fragment, occurrence frequency of trinucleotide GGC and overall free energy of sequence; The device also includes: an extraction module; The extraction module is used to extract the first sequence feature group from the DNA fragment according to the angles of repetitive sequence, hairpin structure, poly structure, GC content, information entropy, free energy, base frequency variance, and trinucleotide frequency; The processing module is further used to calculate the importance scores of the plurality of sequence features included in the first sequence feature group; The processing module is further used to screen the sequence features included in the first sequence feature group according to the partial least squares dimensionality reduction method, remove the sequence features whose importance scores are lower than a preset first threshold, and obtain a second sequence feature group; The processing module is further used to screen the sequence features included in the second sequence feature group according to the XGBoost algorithm, remove the sequence features whose importance scores are lower than a preset second threshold, and obtain the difficulty feature group.

5. The device according to claim 4, characterized in that The device is also used for: The processing module is further used to calculate the contribution value of each difficulty feature in the difficulty feature group to the difficulty score by using the additive interpretation SHAP method; The processing module is further used to determine the maximum value of the contribution value according to the synthetic difficulty evaluation model, and determine the difficulty feature corresponding to the maximum value of the contribution value.

6. The device according to claim 4, characterized in that The prediction module is also used to: If the difficulty score is greater than or equal to a preset score threshold, the prediction module is further used to determine that the DNA fragment belongs to a difficult-to-synthesize category through the synthesis difficulty evaluation model; If the difficulty score is less than a preset score threshold, the prediction module is further configured to determine that the DNA fragment belongs to the easy-to-synthesize category through the synthesis difficulty evaluation model.

Images