Sequencing result automatic interpretation method, system and equipment based on machine learning

Abstract

A sequencing result automatic interpretation method based on machine learning comprises the following steps: collecting different types of sequencing results as data sources in a data source database to obtain different types of sequencing results in the data source database, and removing repeated, irrelevant or noise data from original data; reading related information of a sequencing result of the. Ab1 file, and extracting features; the features and the corresponding labels are merged, and a structured data table is generated; constructing a corresponding machine learning model according to the task selection; training the model by using the training set, and adjusting hyper-parameters to optimize the performance of the model; calling an evaluation standard to evaluate the performance of the model; and packaging the calling model, and automatically interpreting a sequencing result.

Description

Technical Field

[0001] The present invention relates to the field of computers, and in particular to a method, system and device for automatically interpreting sequencing results based on machine learning. Background Art

[0002] In modern biotechnology, especially in the fields of genomics, transcriptomics and personalized medicine, the demand for analysis of sequencing data is increasing. Although the continuous development of bioinformatics tools and software has provided many solutions for data analysis, the traditional manual interpretation method still generally relies on manual experience judgment of data.

[0003] These manual analysis methods are not only time-consuming and labor-intensive, but also easily affected by subjective factors, which may lead to inconsistency and deviation in results. Therefore, how to reduce the reliance on manual experience and ensure the consistency of analysis results are urgent issues to be solved. Summary of the invention

[0004] One of the purposes of the present invention is to provide a method, system and device for automatic interpretation of sequencing results based on machine learning, which can realize efficient and automatic interpretation of sequencing results by constructing a machine learning model, thereby greatly improving the speed and accuracy of sequencing result interpretation.

[0005] One of the purposes of the present invention is to provide a method, system and device for automatic interpretation of sequencing results based on machine learning, which can provide an automatic interpretation model for sequencing results, automatically learn from sequencing data, obtain and analyze potential rules and relationships, and provide analysis for deciphering complex biological data.

[0006] One of the purposes of the present invention is to provide a method, system and device for automatically interpreting sequencing results based on machine learning, which can use machine learning algorithms to automatically learn the patterns and rules of data, thereby quickly processing and analyzing massive amounts of sequencing data, thereby improving the efficiency of data analysis.

[0007] One of the purposes of the present invention is to provide a method, system and device for automatically interpreting sequencing results based on machine learning, which can automatically process through algorithms, avoid errors and biases that may be caused by human factors, and improve the accuracy and reliability of the results.

[0008] One of the purposes of the present invention is to provide a method, system and device for automatic interpretation of sequencing results based on machine learning, which can provide an automatic interpretation model for sequencing results, reduce dependence on manual labor, and reduce labor costs.

[0009] In order to achieve at least one of the inventive purposes of the present invention, the present invention provides a method for automatically interpreting sequencing results based on machine learning, comprising the following steps:

[0010] Collect different types of sequencing results as data sources in the data source database

[0011] Obtain different types of sequencing results from the data source database and remove duplicate, irrelevant or noisy data from the original data;

[0012] Read the sequencing result-related information of the .ab1 file and extract features;

[0013] Merge the features with the corresponding labels to generate a structured data table;

[0014] Build the corresponding machine learning model according to the task selection; use the training set to train the model and adjust the hyperparameters to optimize the model performance;

[0015] Retrieving the evaluation criteria to assess the performance of the model; and

[0016] Encapsulate the calling model and automatically interpret the sequencing results.

[0017] In some embodiments, the method for automatic interpretation of sequencing results based on machine learning also includes the steps of: processing the extracted feature data and merging them with the corresponding labels into a structured DataFrame data table, and exporting the constructed data set to CSV format to facilitate learning and processing of the corresponding machine learning model.

[0018] In some embodiments, the method for automatic interpretation of sequencing results based on machine learning also includes the steps of: calling the predict method to generate predictions on the test set, and calculating the accuracy, precision, recall, and F1 score as evaluation indicators.

[0019] In some embodiments, the method for automatic interpretation of sequencing results based on machine learning further includes the step of using a grid search algorithm to find the optimal parameter value.

[0020] In some embodiments, the method for automatic interpretation of sequencing results based on machine learning further includes the step of adjusting parameters to continue training the model until the model evaluation index meets the requirements.

[0021] In some embodiments, the method for automatic interpretation of sequencing results based on machine learning also includes the steps of: calling the dump() function in the joblib library to save the trained model; encapsulating the saved model into a callable interface, and the sequencing results can be obtained by externally requesting the interface.

[0022] According to another aspect of the present invention, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the automatic interpretation method of sequencing results based on machine learning are performed.

[0023] According to another aspect of the present invention, there is also provided a device for automatically interpreting sequencing results based on machine learning, comprising:

[0024] Memory, for storing software applications,

[0025] A processor is used to execute the software application, and each program of the software application correspondingly executes each step of the automatic interpretation method of sequencing results based on machine learning.

[0026] According to another aspect of the present invention, there is also provided a system for automatic interpretation of sequencing results based on machine learning, comprising a data collection unit, a data preprocessing unit, a feature extraction unit, a data set construction unit, a model training unit, a model evaluation unit and an interpretation result unit; the data collection unit collects sequencing result files as a data source; the data preprocessing unit preprocesses the data; the feature extraction unit reads the sequencing result related information of the .ab1 file and extracts features; the data set construction unit merges the features and corresponding labels into a structured data table; the model training unit constructs a corresponding machine learning model according to the task, trains the model using a training set, and adjusts hyperparameters to optimize model performance; the model evaluation unit applies evaluation criteria to evaluate the performance of the model; the interpretation result unit encapsulates and calls the model to automatically interpret the sequencing results.

[0027] In some embodiments, the result interpretation unit includes a trained model saving module, a model encapsulation module and a sequencing result automatic interpretation module; the trained model saving module is configured to execute: calling the dump() function in the joblib library to save the trained model; the model encapsulation module encapsulates the saved model into a callable interface; the sequencing result automatic interpretation module automatically interprets the sequencing results in response to the interpretation request of the external request interface. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a flowchart of the steps of a method for automatically interpreting sequencing results based on machine learning according to an embodiment of the present invention. DETAILED DESCRIPTION

[0029] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art can think of other obvious variations. The basic principles of the present invention defined in the following description can be applied to other embodiments, variations, improvements, equivalents, and other technical solutions that do not deviate from the spirit and scope of the present invention.

[0030] It is to be understood that the term "one" should be understood as "at least one" or "one or more", that is, in one embodiment, the number of an element may be one, while in another embodiment, the number of the element may be multiple, and the term "one" should not be understood as a limitation on the quantity.

[0031] The present invention is an invention related to a computer program. Figure 1 The flowchart of the automatic interpretation method of sequencing results based on machine learning of the present invention is shown, which explains the solution for solving the problem proposed by the present invention, based on the computer program processing flow, and controlling or processing the external object or internal object of the computer by executing the computer program compiled according to the above process by the computer. Through the automatic interpretation method of sequencing results based on machine learning of the present invention, it is possible to use a computer system to construct a machine learning model to realize efficient automatic interpretation of sequencing results, greatly improve the speed and accuracy of sequencing result interpretation; it is possible to provide an automatic interpretation model for sequencing results, automatically learn from sequencing data, obtain and analyze potential laws and relationships, and provide a method for cracking complex biological data analysis; it is possible to use a machine learning algorithm to automatically learn the patterns and laws of data, so as to quickly process and analyze massive sequencing data, and improve the efficiency of data analysis; it is possible to automatically process through an algorithm, avoid errors and biases that may be caused by human factors, and improve the accuracy and reliability of the results; it is possible to provide an automatic interpretation model for sequencing results, reduce dependence on manual labor, and reduce labor costs.

[0032] It is understandable that the "computer" referred to in the present invention does not only refer to desktop computers, laptops, tablets and other devices, but also includes other intelligent electronic devices that can run according to programs and process data.

[0033] like Figure 1 More specifically, the automatic interpretation method of sequencing results based on machine learning includes the following steps:

[0034] S100: Collect sequencing result files as data sources;

[0035] S200: preprocessing the data;

[0036] S300: Read the sequencing result related information of the .ab1 file and extract features;

[0037] S400: Merging the features with the corresponding labels to generate a structured data table;

[0038] S500: Build the corresponding machine learning model according to the task selection; train the model using the training set and adjust the hyperparameters to optimize the model performance;

[0039] S600: Retrieving the evaluation criteria to evaluate the performance of the model;

[0040] S700: Encapsulates and calls the model to automatically interpret sequencing results.

[0041] Preferably, in step S300, the sequencing result-related information of the .ab1 file is read by writing a Python script.

[0042] It is worth mentioning that, through the automatic interpretation method of sequencing results based on machine learning of the present invention, the corresponding machine learning model can automatically learn the patterns and rules of the data according to the task selection, so as to quickly process and analyze massive sequencing data. Compared with the traditional manual interpretation method, this greatly improves the efficiency of data analysis. Through algorithm automation processing, errors and biases that may be caused by human factors are avoided, and the accuracy and reliability of the results are improved. The machine learning interpretation method reduces the dependence on manual labor and reduces labor costs.

[0043] More specifically, in a specific embodiment, the step S100 further includes the following steps:

[0044] Different types of sequencing results are collected as data sources in the data source database.

[0045] The step S200 further includes the following steps:

[0046] Obtain different types of sequencing results from the data source database and remove duplicate, irrelevant or noisy data from the original data.

[0047] The step S300 further includes the following steps:

[0048] The .ab1 file is read using the read() function of the Bio.SeqIO module in the BioPython library; the function returns a SeqRecord object that stores the biological sequence and its related annotation information; another dictionary under abif_raw is obtained by accessing the "abif_raw" key of the annotations attribute of the SeqRecord object; the dictionary contains some basic information about the sequencing data and more detailed raw data items, and the quality score, sequence, peak spacing, signal intensity, noise intensity, raw signal data for each channel that has not been further processed or analyzed, and signal data for each channel that has been processed or analyzed are obtained from the dictionary as feature data.

[0049] In a specific preferred embodiment, for example:

[0050] from Bio import SeqIO

[0051] record=SeqIO.read(ab1file,'abi')

[0052] In a specific preferred embodiment, for example:

[0053] allAnnotations=record.annotations['abif_raw']

[0054] The step S400 further includes the following steps:

[0055] The extracted feature data is processed and merged with the corresponding labels into a structured DataFrame data table, and the constructed data set is exported to CSV format for subsequent machine learning processing.

[0056] The step S500 further includes the following steps:

[0057] Call the train_test_split function in the Scikit-Learn library to split the data set into two parts: training set and test set in a ratio of 8:2;

[0058] In a specific preferred embodiment, for example:

[0059] # Split the data set, test_size = 0.2 means the test set accounts for 20%

[0060] X_train,X_test,y_train,y_test=train_test_split(X,y,test_size=0.2,random_state=42)

[0061] Furthermore, the step S500 further includes the following steps:

[0062] Import the CatBoost library and create a CatBoost Classifier instance, setting the learning rate, number of trees, depth, class weight parameters, and L2 regularization term;

[0063] In a specific preferred embodiment, for example:

[0064] from catboost import CatBoostClassifier

[0065] #Create a CatBoostClassifier instance, set the learning rate (learning_rate), the number of trees (iterations), the maximum depth of the tree (depth), the class weights (class_weights), and the L2 regularization term (l2_leaf_reg)

[0066] model = CatBoostClassifier(

[0067] learning_rate=0.1,#learning rate

[0068] iterations=500,#number of trees

[0069] depth=6,# maximum depth of the tree

[0070] class_weights=None, #class weight, None means no weight adjustment

[0071] l2_leaf_reg=3, #L2 regularization term, used to control overfitting )

[0073] Furthermore, the step S500 further includes the following steps:

[0074] Call the fit method to train the model on the training set.

[0075] In a specific preferred embodiment, for example:

[0076] model.fit(X_train,y_train,verbose=100)

[0077] The step S600 further includes the following steps:

[0078] Call the predict method to generate predictions on the test set;

[0079] In a specific preferred embodiment, for example:

[0080] y_pred=model.predict(X_test)

[0081] Furthermore, the step S600 further includes the following steps:

[0082] Calculate accuracy, precision, recall, and F1 score as evaluation indicators;

[0083] In a specific preferred embodiment, for example:

[0084] Accuracy is defined as:

[0085]

[0086] Among them, TP is a true positive example, TN is a true negative example, FP is a false positive example, and FN is a false negative example; the precision rate is defined as Precision, and the formula is:

[0087]

[0088] The recall rate is defined as Recall, and the formula is:

[0089]

[0090] The F1 score is defined as F1 Score, and the formula is:

[0091]

[0092] Furthermore, the step S600 further includes the following steps:

[0093] Calculate the aforementioned evaluation indicators by calling the scikit-learn library;

[0094] In a specific preferred embodiment, for example:

[0095] # Calculate evaluation metrics

[0096] accuracy=accuracy_score(y_test,y_pred)

[0097] precision=precision_score(y_test,y_pred,average='macro')

[0098] recall=recall_score(y_test,y_pred,average='macro')

[0099] f1=f1_score(y_test,y_pred,average='macro'

[0100] Furthermore, the step S600 further includes the following steps:

[0101] Use a grid search algorithm to find the optimal parameter values;

[0102] In a specific preferred embodiment, for example:

[0103]

[0104]

[0105] Furthermore, the step S600 further includes the following steps:

[0106] Adjust the parameters and continue training the model until the model evaluation indicators meet the requirements.

[0107] The step S700 further includes the following steps:

[0108] Call the dump() function in the joblib library to save the trained model; encapsulate the saved model into a callable interface, and the sequencing results can be obtained by external request interface.

[0109] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided in the form of methods, systems or computer program products. Therefore, the present invention may take the form of an all-hardware embodiment, an all-software embodiment, or an embodiment combining software and hardware.

[0110] The present invention can be embedded in a computer program product, which includes all the features that enable the automatic interpretation method of sequencing results based on machine learning described herein to be implemented. The computer program product is contained in one or more computer-readable storage media, and the computer-readable storage medium has a computer-readable program code contained therein. According to another aspect of the present invention, a computer-readable storage medium is also provided, on which a computer program is stored, and the computer program can perform the steps of the method of the present invention when executed by a processor. A computer storage medium is a medium for storing a certain discontinuous physical quantity in a computer memory. Computer storage media include, but are not limited to, semiconductors, disk storage, magnetic cores, magnetic drums, magnetic tapes, laser disks, etc. It can be understood by those skilled in the art that computer storage media are not limited to the aforementioned examples, and the aforementioned examples are merely examples and are not limited to the present invention.

[0111] According to another aspect of the present invention, there is also provided an automatic interpretation device for sequencing results based on machine learning, the automatic interpretation device for sequencing results based on machine learning comprising: a software application, a memory for storing the software application, and a processor for executing the software application. Each program of the software application can correspondingly execute the steps in the automatic interpretation method for sequencing results based on machine learning of the present invention.

[0112] A typical combination of hardware and software may be a general-purpose computer system with a computer program, which, when loaded and executed, controls the computer system so as to perform the method for automatic interpretation of sequencing results based on machine learning disclosed in the present invention.

[0113] It will be understood by those skilled in the art that the automatic interpretation device for sequencing results based on machine learning can be embodied as a desktop computer, a notebook, a mobile smart device, etc., but the foregoing is only used as an example and also includes other intelligent analysis devices equipped with the software application of the present invention.

[0114] It will be appreciated by those skilled in the art that the automatic interpretation method of sequencing results based on machine learning of the present invention can be implemented by hardware, software, or a combination of software and hardware. The present invention can be implemented in a centralized manner in at least one computer system, or in a decentralized manner by different parts distributed in several interconnected computer systems. Any computer system or other device that can implement the method is applicable. The combination of common hardware and software can be a general-purpose computer system with a computer program installed, and the computer system is controlled by installing and executing the program to operate according to the automatic interpretation method of sequencing results based on machine learning.

[0115] Corresponding to the embodiment of the method of the present invention, according to another aspect of the present invention, there is also provided a system for automatically interpreting sequencing results based on machine learning, which is the application of the method for automatically interpreting sequencing results based on machine learning of the present invention in the improvement of computer programs. The system for automatically interpreting sequencing results based on machine learning can utilize a computer system to construct a machine learning model to realize efficient and automatic interpretation of sequencing results, greatly improving the speed and accuracy of sequencing result interpretation; it can provide an automatic interpretation model for sequencing results, automatically learn from sequencing data, obtain and analyze potential laws and relationships, and provide a method for cracking complex biological data analysis; it can use machine learning algorithms to automatically learn the patterns and laws of data, thereby quickly processing and analyzing massive amounts of sequencing data, improving the efficiency of data analysis; it can be processed automatically by algorithms, avoiding errors and biases that may be caused by human factors, and improving the accuracy and reliability of the results; it can provide an automatic interpretation model for sequencing results, reduce dependence on manual labor, and reduce labor costs.

[0116] Specifically, the automatic interpretation system of sequencing results based on machine learning includes a data collection unit, a data preprocessing unit, a feature extraction unit, a data set construction unit, a model training unit, a model evaluation unit and an interpretation result unit. The data collection unit collects sequencing result files as a data source; the data preprocessing unit preprocesses the data; the feature extraction unit reads the sequencing result related information of the .ab1 file and extracts features; the data set construction unit merges the features and corresponding labels into a structured data table; the model training unit constructs the corresponding machine learning model according to the task, trains the model using the training set, and adjusts the hyperparameters to optimize the model performance; the model evaluation unit applies the evaluation criteria to evaluate the performance of the model; the interpretation result unit encapsulates the call model and automatically interprets the sequencing results.

[0117] Through the various units of the automatic interpretation system of sequencing results based on machine learning, the machine learning algorithm can automatically learn the patterns and rules of the data, so as to quickly process and analyze massive sequencing data. Compared with the traditional manual interpretation method, this greatly improves the efficiency of data analysis; through the automatic processing of the algorithm, the errors and biases that may be caused by human factors are avoided, and the accuracy and reliability of the results are improved; through the machine learning interpretation method, the dependence on manual labor is reduced, and the labor cost is reduced.

[0118] More specifically, in a preferred embodiment of the present invention, the data collection unit includes a sequencing result collection module and a data source database module. The sequencing result collection module is used to collect different types of sequencing results, and the data source database module is used to obtain the collected different types of sequencing results and store them as data sources in the data source database for subsequent calls.

[0119] The data preprocessing unit includes a sequencing result acquisition module and a raw data preprocessing module. The sequencing result acquisition module acquires different types of sequencing results in the data source database in the data source database module, and the raw data preprocessing module removes duplicate, irrelevant or noise data from the raw data.

[0120] The feature extraction unit includes an annotation information acquisition module, a raw data item dictionary module, and a feature data acquisition module. The annotation information acquisition module is configured to execute: read the .ab1 file using the read() function of the Bio.SeqIO module in the BioPython library, and the function returns a SeqRecord object, which stores the biological sequence and its related annotation information. The raw data item dictionary module is configured to execute: access the "abif_raw" key of the annotations attribute of the SeqRecord object to obtain another dictionary under abif_raw, which contains some basic information of the sequencing data and more detailed raw data items. The feature data acquisition module is configured to execute: obtain the quality score, sequence, peak spacing, signal intensity, noise intensity, raw signal data under each channel that has not been further processed or analyzed, and signal data under each channel that has been processed or analyzed as feature data from the dictionary obtained from the raw data item dictionary module.

[0121] The data set construction unit includes a data set splitting module and a learning model construction module. The data set splitting module calls the train_test_split function in the Scikit-Learn library to split the data set into two parts, a training set and a test set, in a ratio of 8:2; the learning model construction module is configured to execute: import the CatBoost library, create a CatBoost Classifier instance, set the learning rate, the number of trees, the depth, the category weight parameter, and the L2 regularization term.

[0122] The model training unit includes a training set acquisition module and a training module. The training set acquisition module acquires the training set split by the data set split module. The training module is configured to execute: calling a fit method to train the model on the training set of the training set acquisition module.

[0123] The model evaluation unit includes a prediction generation module and an evaluation index calculation module. The prediction generation module calls the predict method to generate predictions on the test set, and the evaluation index calculation module calculates accuracy, precision, recall, and F1 score as evaluation indicators.

[0124] More specifically, in a specific preferred embodiment, the evaluation index calculation module also includes an accuracy calculation module, a precision calculation module, a recall calculation module and an F1 score calculation module.

[0125] The accuracy calculation module is configured to perform the calculation of the accuracy evaluation index, and defines the accuracy as Accuracy, and the formula is:

[0126]

[0127] Among them, TP is a true positive example, TN is a true negative example, FP is a false positive example, and FN is a false negative example.

[0128] The precision calculation module is configured to perform calculation of the precision evaluation index, and defines the precision as Precision, and the formula is:

[0129]

[0130] The recall rate calculation module is configured to perform the calculation of the recall rate evaluation index, and the recall rate is defined as Recall, and the formula is:

[0131]

[0132] The F1 score calculation module is configured to perform calculation of the F1 score evaluation index, and defines the F1 score as F1Score, and the formula is:

[0133]

[0134] The evaluation index calculation module is further configured to execute: calling the scikit-learn library to calculate the evaluation index.

[0135] The model evaluation unit also includes an optimal parameter value module and a parameter adjustment module, wherein the optimal parameter value module is configured to execute: calling a grid search algorithm to find an optimal parameter value. The parameter adjustment module is used to adjust parameters to optimize model performance, and the training module of the model training unit continues to train the model based on the parameters adjusted by the parameter adjustment module until the model evaluation index meets the requirements in the model evaluation unit.

[0136] The reading result unit includes a trained model saving module, a model encapsulation module and a sequencing result automatic reading module. The trained model saving module is configured to execute: calling the dump() function in the joblib library to save the trained model. The model encapsulation module encapsulates the saved model into a callable interface. The sequencing result automatic reading module automatically reads the sequencing result in response to the reading request of the external request interface.

[0137] It will be appreciated by those skilled in the art that the present invention has been described with reference to the flowcharts and / or block diagrams of the methods, systems and computer program products according to the present invention. Each box in the flowchart and / or block diagram, and the combination of boxes in the flowchart and / or block diagram, can obviously be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions (the instructions are passed through the processor of the computer or other programmable data processing device) generate a device for implementing the functions specified in one or more boxes of the flowchart and / or block diagram.

[0138] It should be understood by those skilled in the art that the embodiments of the present invention described above and shown in the accompanying drawings are only examples and do not limit the present invention. The purpose of the present invention has been fully and effectively achieved. The functional and structural principles of the present invention have been demonstrated and explained in the embodiments, and the embodiments of the present invention may be deformed or modified in any way without departing from the principles.

Claims

1. A method for automatically interpreting sequencing results based on machine learning, characterized in that: The automatic interpretation method of sequencing results based on machine learning comprises the following steps: Collect different types of sequencing results as data sources in the data source database Obtain different types of sequencing results from the data source database and remove duplicate, irrelevant or noisy data from the original data; Read the sequencing result-related information of the .ab1 file and extract features; Merge the features with the corresponding labels to generate a structured data table; Build the corresponding machine learning model according to the task selection; use the training set to train the model and adjust the hyperparameters to optimize the model performance; Retrieving the evaluation criteria to assess the performance of the model; and Encapsulate the calling model and automatically interpret the sequencing results.

2. The automatic interpretation method of sequencing results based on machine learning as described in claim 1, wherein the automatic interpretation method of sequencing results based on machine learning further comprises the steps of: processing the extracted feature data and merging them with the corresponding labels into a structured DataFrame data table, and exporting the constructed data set to CSV format to facilitate the learning and processing of the corresponding machine learning model constructed.

3. The method for automatic interpretation of sequencing results based on machine learning as described in claim 1, wherein the method for automatic interpretation of sequencing results based on machine learning also includes the steps of: calling the predict method to generate predictions on the test set, and calculating the accuracy, precision, recall, and F1 score as evaluation indicators.

4. The method for automatically interpreting sequencing results based on machine learning as described in claim 3, wherein the method for automatically interpreting sequencing results based on machine learning further comprises the step of: using a grid search algorithm to find the optimal parameter value.

5. The automatic interpretation method of sequencing results based on machine learning as described in claim 4, wherein the automatic interpretation method of sequencing results based on machine learning further comprises the step of adjusting parameters to continue training the model until the model evaluation index meets the requirements.

6. The automatic interpretation method of sequencing results based on machine learning as described in claim 5, wherein the automatic interpretation method of sequencing results based on machine learning further comprises the steps of: calling the dump() function in the joblib library to save the trained model; encapsulating the saved model into a callable interface, and the sequencing results can be obtained by an external request interface.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for automatically interpreting sequencing results based on machine learning described in any one of claims 1 to 6 are performed.

8. A device for automatically interpreting sequencing results based on machine learning, characterized in that: include: Memory, for storing software applications, A processor is used to execute the software application, and each program of the software application correspondingly executes each step of the automatic interpretation method of sequencing results based on machine learning as described in any one of claims 1 to 6.

9. A system for automatically interpreting sequencing results based on machine learning, characterized in that: The automatic interpretation system for sequencing results based on machine learning includes a data collection unit, a data preprocessing unit, a feature extraction unit, a data set construction unit, a model training unit, a model evaluation unit and an interpretation result unit; the data collection unit collects sequencing result files as a data source; the data preprocessing unit preprocesses the data; the feature extraction unit reads the sequencing result related information of the .ab1 file and extracts features; The data set construction unit merges the features and corresponding labels into a structured data table; the model training unit constructs the corresponding machine learning model according to the task, trains the model using the training set, and adjusts the hyperparameters to optimize the model performance; the model evaluation unit applies the evaluation criteria to evaluate the performance of the model; the result interpretation unit encapsulates the call model and automatically interprets the sequencing results.

10. The automatic interpretation system for sequencing results based on machine learning as described in claim 9, wherein the interpretation result unit includes a trained model saving module, a model encapsulation module and a sequencing result automatic interpretation module; the trained model saving module is configured to execute: calling the dump() function in the joblib library to save the trained model; the model encapsulation module encapsulates the saved model into a callable interface; the sequencing result automatic interpretation module responds to the interpretation request of the external request interface to automatically interpret the sequencing results.

Images