Method and device for identifying nanopore state signal type based on deep learning, electronic equipment and storage medium

Abstract

The invention provides a nanopore state signal type identification method based on deep learning. The method comprises the following steps: acquiring multiple pieces of state signal data from nanopore sequencing signal data; calculating the fragment number of each piece of state signal data in the plurality of pieces of state signal data according to a preset threshold value; judging whether each piece of state signal data is abnormal state signal data or not according to the fragment number of each piece of state signal data; deleting the abnormal state signal data from the plurality of state signal data to obtain normal state signal data; and identifying the type of the normal state signal data by using the trained deep learning model.

Description

Technical Field

[0001] The present invention relates to the field of bioinformatics, and more specifically to a method, device, electronic device and storage medium for identifying nanopore sequencing signal types based on deep learning. Background Art

[0002] Nanopore sequencing signals refer to the changes in the perforation current recorded when a DNA molecule passes through a nanopore. This change is mainly caused by the different currents of different bases in the nanopore. When analyzing nanopore sequencing signals, an important step is to perform state detection on the nanopore sequencing signals. The core task of state detection is to identify which state the nanopore sequencing signal is in, so as to provide a basis for the next analysis strategy.

[0003] Nanopore sequencing signals are one-dimensional time series signals. There are two existing technical solutions for detecting the state of such time series signals. One is the traditional time series data analysis method, which converts the time series data from the time domain to the frequency domain, and performs similarity calculation and threshold filtering on the frequency domain signal by manually extracting special sequence features to identify the state of the signal.

[0004] However, traditional time series data analysis methods can only identify some target signals with obvious features and easy to identify in some cases, and have poor robustness in other cases. For different types of target signals, they need to design manual feature extraction separately, which is inefficient. Summary of the invention

[0005] In order to solve the technical problems existing in the existing problems, in a first aspect of the present disclosure, a method for identifying the type of nanopore sequencing signal based on deep learning is provided, comprising the following steps:

[0006] Acquire a plurality of state signal data from the nanopore sequencing signal data;

[0007] Calculate the number of fragments of each state signal data in the plurality of state signal data according to a preset threshold;

[0008] Determine whether each state signal data is abnormal state signal data according to the number of fragments of each state signal data;

[0009] Deleting abnormal state signal data from the plurality of state signal data to obtain normal state signal data;

[0010] Use the trained deep learning model to identify the type of normal state signal data.

[0011] According to an embodiment of the present invention, judging whether each state signal data is abnormal state signal data according to the number of fragments of each state signal data includes:

[0012] If the number of segments of the status signal data is within a preset range, the status signal data is determined to be abnormal status signal data.

[0013] According to an embodiment of the present invention, judging whether each state signal data is abnormal state signal data according to the number of fragments of each state signal data includes:

[0014] Calculate the first median and standard deviation of the state signal data according to the state signal data;

[0015] If the number of segments of the status signal data is less than a preset number, the first median of the status signal data is greater than a preset median, and the standard deviation of the status signal data is less than a preset standard deviation, the status signal data is determined to be abnormal status signal data.

[0016] According to an embodiment of the present invention, judging whether each state signal data is abnormal state signal data according to the number of fragments of each state signal data includes:

[0017] Calculate the first median and standard deviation of the state signal data according to the state signal data;

[0018] Calculate the absolute difference of the state signal data according to the second median of the first preset number of signal points in the state signal data and the third median of the second preset number of signal points in the state signal data;

[0019] If the number of fragments of the state signal data is less than a preset number, the first median of the state signal data is less than a preset median, the standard deviation of the state signal data is greater than a preset standard deviation, and the absolute difference of the state signal data is less than a preset absolute difference, the state signal data is judged to be abnormal state signal data.

[0020] According to an embodiment of the present invention, the deep learning model includes a convolutional neural network model.

[0021] According to an embodiment of the present invention, the convolutional neural network model includes a normalization layer, a first convolutional layer, a first pooling layer, a second convolutional layer, a second pooling layer, and a fully connected layer connected in sequence.

[0022] According to an embodiment of the present invention, the deep learning model is trained using normal state signal data.

[0023] In a second aspect of the present disclosure, a device for identifying a type of nanopore sequencing signal based on deep learning is provided, comprising the following modules:

[0024] An acquisition module, used to acquire multiple state signal data from the nanopore sequencing signal data; a calculation module, used to calculate the number of fragments of each state signal data in the multiple state signal data according to a preset threshold;

[0025] A judging module, used for judging whether each state signal data is abnormal state signal data according to the number of fragments of each state signal data;

[0026] The deletion module is used to delete abnormal status signal data from multiple status signal data.

[0027] Get normal state signal data;

[0028] The recognition module is used to recognize the type of normal state signal data using the trained deep learning model.

[0029] In a third aspect of the present disclosure, an electronic device is provided, comprising:

[0030] at least one processor; and

[0031] a memory communicatively connected to at least one processor; wherein,

[0032] The memory stores instructions that can be executed by at least one processor. The instructions are executed by the at least one processor so that the at least one processor can perform any one of the above methods.

[0033] In a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to cause a computer to execute any one of the above methods.

[0034] Based on the various aspects provided by the present disclosure, a method, device, electronic device and non-transitory computer-readable storage medium for identifying the type of nanopore sequencing signal based on deep learning are provided. These aspects can improve robustness while identifying nanopore sequencing signals with lower complexity, solving the problems of poor robustness of traditional time series data analysis methods and high labor costs of feature analysis of machine learning methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 4 is a flowchart of a method for training a deep learning model for identifying nanopore sequencing signal types according to an embodiment of the present disclosure.

[0036] Figure 2 is a schematic diagram of an exemplary image generated based on the collected nanopore sequencing signal data.

[0037] Figure 3 It is through state recognition Figure 2 Schematic diagram of three pictures obtained by classifying the exemplary picture shown.

[0038] Figure 4 4 is a flowchart of a method for identifying nanopore sequencing signal types based on deep learning according to an embodiment of the present disclosure.

[0039] Figure 5 Schematic diagram of cutting fragments according to an embodiment of the present disclosure.

[0040] Figure 6 An example of different states of a nanopore sequencing signal is shown.

[0041] Figure 7 Another example of different states of a nanopore sequencing signal is shown.

[0042] Figure 8 Yet another example of different states of a nanopore sequencing signal is shown.

[0043] Fig. 9 is a schematic diagram of a device for identifying nanopore sequencing signal types based on deep learning according to an embodiment of the present disclosure.

[0044] Fig.10 An example of a deep learning model is shown. DETAILED DESCRIPTION

[0045] The specific embodiments of the present invention will be described in detail below. It should be noted that the embodiments described herein are only for illustration and are not intended to limit the present invention. In the following description, a large number of specific details are set forth in order to provide a thorough understanding of the present invention. However, it is obvious to those of ordinary skill in the art that these specific details do not have to be used to implement the present invention. In other examples, in order to avoid confusing the present invention, known circuits, materials or methods are not specifically described.

[0046] Throughout the specification, references to "one embodiment," "an embodiment," "an example," or "an example" mean that a particular feature, structure, or characteristic described in conjunction with the embodiment or example is included in at least one embodiment of the present invention. Thus, the phrases "in one embodiment," "in an embodiment," "an example," or "an example" appearing in various places throughout the specification do not necessarily all refer to the same embodiment or example. Furthermore, particular features, structures, or characteristics may be combined in one or more embodiments or examples in any appropriate combinations and / or subcombinations.

[0047] It should be understood that when an element is said to be "coupled to" or "connected to" another element, it can be directly coupled or connected to the other element or there may be intermediate elements. On the contrary, when an element is said to be "directly coupled to" or "directly connected to" another element, there are no intermediate elements.

[0048] Furthermore, as used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0049] It will be understood that a noun corresponding to a term in the singular form may include one or more things unless the relevant context clearly indicates otherwise. As used herein, each of phrases such as "A or B," "at least one of A and B," "at least one of A or B," "A, B or C," "at least one of A, B and C," and "at least one of A, B or C" may include all possible combinations of items listed together with a corresponding one of the plurality of phrases. As used herein, terms such as "1st" and "2nd" or "first" and "second" may be used to simply distinguish a corresponding component from another component and do not limit the component in other respects (e.g., importance or order).

[0050] As used herein, the term "module" may include units implemented in hardware, software, or firmware, and may be used interchangeably with other terms (e.g., "logic," "logic block," "portion," or "circuit"). A module may be a single integrated component adapted to perform one or more functions or a minimum unit or portion of the single integrated component. For example, according to an embodiment, a module may be implemented in the form of an application specific integrated circuit (ASIC).

[0051] It should be understood that the various embodiments of the present disclosure and the terms used therein are not intended to limit the technical features set forth herein to specific embodiments, but include various changes, equivalent forms or replacement forms for the corresponding embodiments. Unless otherwise clearly defined herein, all terms will be given their broadest possible interpretation, including the meanings implied in the specification and the meanings understood by those skilled in the art and / or defined in dictionaries, papers, etc.

[0052] In addition, it should be understood by those skilled in the art that the drawings provided herein are for illustrative purposes and are not necessarily drawn to scale. For the description of the drawings, similar reference numerals may be used to refer to similar or related elements. The present disclosure will be described exemplarily with reference to the drawings below.

[0053] The following describes in detail with reference to the accompanying drawings a method, apparatus, electronic device, and non-transitory computer-readable storage medium for identifying nanopore sequencing signal types based on deep learning according to an embodiment of the present invention.

[0054] Figure 1 is a flowchart of a method for training a deep learning model for identifying nanopore sequencing signal types according to an embodiment of the present disclosure. Figure 1 A method 100 for training a deep learning model for identifying nanopore sequencing signal types according to an embodiment of the present disclosure is described.

[0055] like Figure 1As shown, the training method 100 of the deep learning model for identifying the type of nanopore sequencing signal according to the embodiment of the present disclosure includes step S110 and step S210. In step S110, the training data of the deep learning model is obtained. In step S120, the deep learning model is trained using the obtained training data, thereby obtaining a trained deep learning model for identifying the type of nanopore sequencing signal.

[0056] Specifically, step S110 further includes step S1101 and step S1102.

[0057] Wherein, in step S1101, nanopore sequencing signal data is collected.

[0058] For example, nanopore sequencing signal data can be collected by a nanopore sequencing device. When collecting signals, for example, a single-channel nanopore sequencing device and a computer can be used. The target sequencing signal data is collected by the nanopore sequencing device, and the collected sequencing signal data and other information are saved as an h5 file structure and stored in real time in a storage device of the computer.

[0059] The nanopore sequencing signal includes a state signal, which refers to a signal segment from the time when the pore current value decreases until it rises back to the pore current value, and the pore current value is known before the test data is collected.

[0060] The nanopore sequencing signal data may also be existing signal data collected in advance and stored in a computer system, rather than temporarily collected signal data, which is not specifically limited in the present disclosure.

[0061] In step S1102, training data is generated based on the collected nanopore sequencing signal data.

[0062] When making training data, first generate a picture based on the collected nanopore sequencing signal data. For example, with the current value as the vertical axis and the data collection number (the position in the data sequence) as the horizontal axis, you can generate a picture based on the collected nanopore sequencing signal data. Figure 2 The purpose of generating these images is to manually classify the state signal, and the images are not used as input to the model. Figure 2 In the exemplary graph shown, the current signal drops from the initial open hole current value at position 100 in the data sequence and rises back at position 500 (eg, Figure 2 ), then drops from the open hole current value again at position 600 and rises again at position 2000 (as shown in the reference numeral "1" in FIG. Figure 2 2), the current value at the opening drops for the third time at position 3000 and rises again at position 3500 (as shown in the figure mark "2" in FIG. Figure 2 As shown in the figure mark "3" in FIG. Figure 2 The exemplary picture shown includes a total of 3 status signals.

[0063] Next, the generated images are manually classified. Figure 2 The pictures shown in the figure are manually classified into three categories, named "0", "1" and "2". A single picture is generated for each category, such as Figure 3 As shown. The starting position and ending position of each category of state signal in the data sequence are recorded, and each state signal image is named "artificial classification name_starting position in the data sequence_ending position in the data sequence.png". In this way, according to the name of the image, the artificial classification name of the state signal image and the position of the state signal in the data sequence can be known, so as to intercept the corresponding data in the data sequence.

[0064] Next, according to the start position and end position of the data sequence of the picture name, the corresponding target data (status signal data) is intercepted in the data sequence.

[0065] In an optional embodiment, the target data (i.e., the state signal data) can be standardized, that is, for each data in the state signal data, the difference between the value of the data and the minimum value of the state signal data is divided by the difference between the maximum value of the state signal data and the minimum value of the state signal data, and the obtained value is used as the standardized value of the data.

[0066] The state signal data can be converted into a state signal vector, and each state signal data or state signal vector and the corresponding manual classification result form a training data. Figure 3 The state signal vector V1 is generated from the image 0_100_500.png in , and V1 and the corresponding manual classification result (the category named "0") constitute the first training data. Figure 3 The state signal vector V2 is generated from the image 1_600_2000.png in the image. V2 and the corresponding manual classification result (the category or type named "1") constitute the second training data. Figure 3 The state signal vector V3 is generated from the picture 2_3000_3500.png in the figure. V3 and the corresponding manual classification result (the category named "2") constitute the third training data.

[0067] In step S120, the deep learning model is trained using the training data obtained in step S110, and the model file is saved.

[0068] During training, the training data and the corresponding manual classification results need to be divided into training sets and test sets. In the training set production stage, you can use random extraction and self-set ratios to produce a data set for deep learning training. Then, set the model training parameters according to the application scenario, such as learning rate and number of training iterations, and train and test the model. If the test results are not ideal, you can try to increase the number of data sets, modify the model training parameters, and adjust the training. Repeat this process until the model training results meet the indicator requirements.

[0069] Figure 4 is a flow chart of a method for identifying nanopore sequencing signal types based on deep learning according to an embodiment of the present disclosure. Figure 4 A method 400 for identifying nanopore sequencing signal types based on deep learning according to an embodiment of the present disclosure is described.

[0070] In step S410, a plurality of state signal data are obtained from the nanopore sequencing signal data.

[0071] The acquired state signal data is nanopore sequencing signal data whose type is to be detected.

[0072] In order to obtain the state signal data, firstly, a single-channel nanopore sequencing device and a computer are used to collect the nanopore sequencing signal data, and then the signal segment data from the opening current value decreasing until the opening current value recovers is intercepted in the nanopore sequencing signal data, and the signal segment data is used as the obtained state signal data X1, X2, ..., X N , where N is a natural number greater than 1, and the starting position and ending position of the signal segment data in the entire nanopore sequencing signal data are recorded. The sequencing signal data may include one or more signal segment data, and the type identification is performed for each intercepted signal segment data.

[0073] In step S420, the number of segments of each state signal data in the plurality of state signal data is calculated according to a preset threshold.

[0074] The signal value of the opening current may be 200. The preset threshold may be 180. A state signal data may be cut at a position where the signal value is equal to 180 to obtain the number of segments of the state signal data. The preset threshold may be 160, 170, 180, etc. Figure 5 Schematic diagram of the cutting fragment according to the embodiment of the present disclosure. Figure 5 , cut at the position where the y-axis (signal value of the status signal data) is equal to 180 to obtain 3 fragments, and the value is 3.

[0075] In step S430, it is determined whether each state signal data is abnormal state signal data according to the number of segments of each state signal data.

[0076] In a specific embodiment, judging whether each state signal data is abnormal state signal data according to the number of segments of each state signal data includes: judging the state signal data is abnormal state signal data if the number of segments of the state signal data is within a preset range.

[0077] Specifically, the preset range may be 100 to 1500, 200 to 1500, etc. For example, if the preset range is 100 to 1500, and the number of segments of a state signal data is 3, the state signal data is not abnormal state signal data. For example, if the preset range is 100 to 1500, and the number of segments of a state signal data is 100, the state signal data is abnormal state signal data.

[0078] In a specific embodiment, judging whether each state signal data is abnormal state signal data according to the number of fragments of each state signal data includes: calculating a first median and a standard deviation of the state signal data according to the state signal data; if the number of fragments of the state signal data is less than a preset number, the first median of the state signal data is greater than the preset median, and the standard deviation of the state signal data is less than the preset standard deviation, judging that the state signal data is abnormal state signal data.

[0079] Specifically, the preset median may be 180, 181, etc., and the preset standard deviation may be 5, 10, 25, etc. For example, it may be determined that state signal data having a number of segments less than 3, a first median of state signal data greater than 180, and a standard deviation of the state signal data less than 5 is abnormal state signal data.

[0080] A state signal data may include multiple signal points, each signal point corresponds to a signal value, and the first median and standard deviation can be calculated through the signal values ​​of the multiple signal points.

[0081] In a specific embodiment, judging whether each state signal data is abnormal state signal data according to the number of fragments of each state signal data includes: calculating a first median and a standard deviation of the state signal data according to the state signal data; calculating an absolute difference of the state signal data according to a second median of a first preset number of signal points in the state signal data and a third median of a second preset number of signal points in the state signal data; if the number of fragments of the state signal data is less than a preset number, the first median of the state signal data is less than a preset median, the standard deviation of the state signal data is greater than a preset standard deviation, and the absolute difference of the state signal data is less than a preset absolute difference, the state signal data is judged to be abnormal state signal data.

[0082] Specifically, the first preset number may be 5% or 6% of all signal points in the state signal data. The second preset number may be 5% or 6% of all signal points in the state signal data. The preset absolute difference may be 10, 11, etc. The median of the first 5% of signal points in the state signal data may be obtained as the second median, and the median of the last 5% of signal points in the state signal data may be obtained as the third median; the absolute value of the difference between the second median and the third median is calculated as the absolute difference of the state signal data.

[0083] For example, it can be determined that state signal data having a fragment number less than 3, a first median of state signal data less than 180, a standard deviation of state signal data greater than 25, and an absolute difference of state signal data less than 10 is abnormal state signal data.

[0084] In step S440, the abnormal state signal data is deleted from the plurality of state signal data to obtain the normal state signal data.

[0085] The abnormal state signal data can be deleted from the multiple state signal data to obtain the normal state signal data, thereby avoiding the recognition of the abnormal state signal data.

[0086] In step S450, the type of normal state signal data is identified using the trained deep learning model. For example, the trained deep learning model can be Figure 1 The model shown is obtained by training according to the training method of the deep learning model of the embodiment of the present disclosure.

[0087] The deep learning model includes a convolutional neural network model. The convolutional neural network model includes a normalization layer, a first convolutional layer, a first pooling layer, a second convolutional layer, a second pooling layer, and a fully connected layer connected in sequence. The deep learning model can be trained using normal state signal data.

[0088] Specifically, the normal state signal data in step S440 is input into the trained deep learning model to obtain a prediction result. For example, the prediction result can be one of the three values ​​0, 1, and 2. In addition, a picture of each state signal identified can be output according to the prediction result. For example, the picture can be named "prediction result name_starting position in nanopore sequencing signal data_ending position in nanopore sequencing signal data.png". The names and formats of the pictures described here are only examples, and the present disclosure is not limited thereto.

[0089] Figures 6 to 8Examples of different states of nanopore sequencing signals are shown. It can be seen that some states contain a large amount of data, and it takes a lot of time and effort to manually determine the states of all signals, and the accuracy of the determination results is not high. The deep learning model trained using the training method disclosed in the present invention can identify the states of a large number of signals with high accuracy, thereby improving efficiency and saving time.

[0090] Fig. 9 Schematic diagram of a device for identifying nanopore sequencing signal types based on deep learning according to an embodiment of the present disclosure. Fig. 9 The device 800 for identifying the type of nanopore sequencing signal based on deep learning according to an embodiment of the present disclosure is described. The identification device 800 includes an acquisition module 810 , a calculation module 820 , a judgment module 830 , a deletion module 840 and an identification module 850 .

[0091] The acquisition module 810 is used to acquire a plurality of state signal data from the nanopore sequencing signal data.

[0092] The calculation module 820 is used to calculate the number of segments of each state signal data among the multiple state signal data according to a preset threshold.

[0093] The judging module 830 is used to judge whether each state signal data is abnormal state signal data according to the number of fragments of each state signal data;

[0094] The deletion module 840 is used to delete the abnormal state signal data from the multiple state signal data to obtain the normal state signal data;

[0095] The identification module 850 is used to identify the type of normal state signal data using the trained deep learning model.

[0096] The training device of the deep learning model for identifying the type of nanopore sequencing signal according to the embodiment of the present disclosure is described. The training device includes an acquisition module and a training module.

[0097] The acquisition module is used to acquire training data for the deep learning model, wherein the acquisition module acquires nanopore sequencing signal data, performs state recognition on the acquired nanopore sequencing signal data, and intercepts state signal data X1, X2, ..., X corresponding to the recognized type. N , for the intercepted state signal data X1, X2, ..., C N A state signal vector is generated, where N is a natural number greater than 1, and the training data includes the result of type recognition and the state signal data.

[0098] The training module is used to train the deep learning model using the acquired training data, thereby obtaining a trained deep learning model.

[0099] The deep learning models in the present disclosure include convolutional neural network models, etc.

[0100] A specific example of the deep learning model in the present disclosure can be a convolutional neural network model built with the open source TensorFlow framework. In a convolutional neural network, the input layer receives the original signal data (normal state signal data, that is, the vector form of normal state signal data). The convolution layer extracts features from the input data through a convolution operation to generate a feature group. The pooling layer downsamples the feature group to reduce the data dimension and improve the generalization ability of the model. The fully connected layer converts the output of the pooling layer into a one-dimensional vector and performs a nonlinear transformation through an activation function. The convolutional neural network model of the present invention is composed of a normalization layer and two groups of convolutional layers plus a pooling layer to form a convolutional neural network encoder part. The original signal is directly passed into the convolutional neural network as input, and after being encoded by the convolutional neural network, it is passed into a fully connected layer for the final classification judgment output.

[0101] Fig.10 An example of a deep learning model is shown.

[0102] The feature extraction method described in the present disclosure is not limited to the above-mentioned specific embodiments. Those skilled in the art can make various improvements, deformations, deletions and substitutions based on the abnormal state signal data judgment method disclosed in the present disclosure. For example, the screening conditions can be increased, reduced or replaced. As an example, the absolute difference restriction condition can be added to the following steps: If the number of fragments of the state signal data is within a preset range, the state signal data is judged to be abnormal state signal data.

[0103] The deep learning-based nanopore sequencing signal type identification method and deep learning model training method described in the present disclosure are not limited to state recognition of nanopore sequencing signals, but can also be extended to the field of classification and identification of other time domain and frequency domain signals.

[0104] Although multiple components are shown in the above block diagrams, those skilled in the art will appreciate that the embodiments of the present invention may be implemented without one or more components or by combining certain components.

[0105] Although the various steps are described above according to the order shown in the drawings, those skilled in the art should understand that the various steps may be performed in a different order, or the embodiments of the present invention may be implemented without one or more of the above steps.

[0106] It can be understood from the foregoing that the electronic components of one or more systems or devices may include, but are not limited to, at least one processing unit, a memory, and a communication bus or communication device that couples the various components including the memory to the processing unit. The system or device may include or may access various device-readable media. The system memory may include device-readable storage media in the form of volatile and / or non-volatile memory (e.g., read-only memory (ROM) and / or random access memory (RAM)). By way of example and not limitation, the system memory may also include an operating system, application programs, other program modules, and program data.

[0107] Embodiments may be implemented as systems, methods, or program products. Thus, embodiments may take the form of all-hardware embodiments or embodiments that include software (including firmware, resident software, microcode, etc.), which may be collectively referred to herein as "circuits," "modules," or "systems." In addition, embodiments may take the form of a program product embodied in at least one device-readable medium having device-readable program code embodied thereon.

[0108] Combinations of device-readable storage media may be used. In the context of this document, a device-readable storage medium ("storage medium") may be any tangible, non-signal medium that may contain or store a program consisting of program code configured to be used by or in conjunction with an instruction execution system, apparatus, or device. For purposes of this disclosure, a storage medium or device should be interpreted as being non-transitory, i.e., not including signal or propagation media.

[0109] The present disclosure is presented for purposes of illustration and description, but is not intended to be exhaustive or limiting. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to illustrate the principles and practical applications, and to enable those of ordinary skill in the art to understand the various embodiments of the present disclosure with various modifications suitable for the intended specific use.

Claims

1. A method for identifying the type of nanopore state signal based on deep learning, comprising the following steps: Acquire a plurality of state signal data from the nanopore sequencing signal data; Calculate the number of fragments of each state signal data in the plurality of state signal data according to a preset threshold; Determine whether each state signal data is abnormal state signal data according to the number of fragments of each state signal data; Deleting abnormal state signal data from the plurality of state signal data to obtain normal state signal data; The type of the normal state signal data is identified using a trained deep learning model.

2. The identification method according to claim 1, wherein: The step of judging whether each state signal data is abnormal state signal data according to the number of fragments of each state signal data comprises: If the number of segments of the status signal data is within a preset range, the status signal data is determined to be abnormal status signal data.

3. The identification method according to claim 1, wherein: The step of judging whether each state signal data is abnormal state signal data according to the number of fragments of each state signal data comprises: Calculate a first median and a standard deviation of the state signal data according to the state signal data; If the number of segments of the status signal data is less than a preset number, the first median of the status signal data is greater than a preset median, and the standard deviation of the status signal data is less than a preset standard deviation, the status signal data is determined to be abnormal status signal data.

4. The identification method according to claim 1, wherein: The step of judging whether each state signal data is abnormal state signal data according to the number of fragments of each state signal data comprises: Calculate a first median and a standard deviation of the state signal data according to the state signal data; Calculate the absolute difference of the state signal data according to the second median of the first preset number of signal points in the state signal data and the third median of the second preset number of signal points in the state signal data; If the number of fragments of the status signal data is less than a preset number, the first median of the status signal data is less than a preset median, the standard deviation of the status signal data is greater than a preset standard deviation, and the absolute difference of the status signal data is less than a preset absolute difference, the status signal data is judged to be abnormal status signal data.

5. The identification method according to claim 1, wherein: The deep learning model includes a convolutional neural network model.

6. The identification method according to claim 5, wherein: The convolutional neural network model includes a normalization layer, a first convolutional layer, a first pooling layer, a second convolutional layer, a second pooling layer, and a fully connected layer, which are connected in sequence.

7. The identification method according to claim 1, wherein: The deep learning model is trained using normal state signal data.

8. A device for identifying nanopore state signal types based on deep learning, comprising the following modules: An acquisition module, used for acquiring a plurality of state signal data from the nanopore sequencing signal data; A calculation module, used for calculating the number of fragments of each state signal data in the plurality of state signal data according to a preset threshold; A judging module, used for judging whether each state signal data is abnormal state signal data according to the number of fragments of each state signal data; A deletion module, used for deleting abnormal state signal data from the plurality of state signal data to obtain normal state signal data; An identification module is used to identify the type of the normal state signal data using a trained deep learning model.

9. An electronic device, comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1-7.

Images