Electronic microarray chip data processing and signal extraction method and system, storage medium and electronic equipment

Abstract

The invention belongs to the technical field of biological detection signal processing, and particularly relates to an electronic microarray chip data processing and signal extraction method and system. The method comprises the steps that stable dot matrix inner and outer well data are obtained through integral screening, the voltage difference is calculated, and preprocessing is carried out through double smoothing filtering, standardization and PCA dimension reduction; first-order and second-order derivatives of the preprocessed data are calculated and smoothed; searching and positioning a signal starting point st and an end point ed at the front section of the smoothed first-order derivative signal through a sliding window, and repairing a missing point based on second-order derivative characteristics and linear fitting; and finally, performing baseline fitting according to st and ed, and calculating an area difference between a signal curve and a baseline as an effective semaphore. The method effectively solves the problems of weak anti-interference capability, inaccurate start and end point positioning, large semaphore calculation deviation and the like in the prior art, and has the advantages of high data processing precision, high adaptability and high automation degree.

Description

Technical Field

[0001] This invention relates to the field of signal processing technology for electronic microarray chips, and in particular to a method, system, storage medium, and electronic device for data processing and signal extraction of electronic microarray chips. Background Technology

[0002] In the field of electronic microarray chip detection, data processing and signal extraction are the core links to achieve accurate detection. The core requirement is to detect and identify whether the target sequence is combined from the original data containing the effective signal inside the array and the background outside the array, so as to determine the positive or negative of the sample.

[0003] The main defects of existing technologies include: (1) Weak anti-interference ability: The original data contains interference such as baseline drift and random noise. Traditional smoothing filtering has limited effect on suppressing complex noise, resulting in a large error in the identification of the start and end points of subsequent signals.

[0004] (2) The signal generation stage is not precise: The existing signal calculation takes into account all frames in each reaction stage without combining the signal gradient change characteristics and without considering the start and end of the location signal generation stage. Especially in scenarios with weak signals or gradual signal changes, it is easy to miss or falsely detect.

[0005] (3) Poor data adaptability: The signal characteristics of different detection batches and different chip sites are different. Existing methods lack an adaptive adjustment mechanism and are difficult to adapt to diverse data scenarios.

[0006] (4) Insufficient accuracy in signal quantity calculation: The fitting deviation between the signal curve and the baseline is not considered. The signal quantity is calculated by simple integration or difference, which makes it difficult to quantitatively distinguish the response intensity, resulting in a large deviation between the result and the actual effective signal quantity. Summary of the Invention

[0007] The technical problem to be solved by this invention is to design a method, system, storage medium and electronic device for data processing and signal extraction of electronic microarray chips, so as to realize the differentiation and processing of effective data within the array and background data outside the array. Through multi-step data preprocessing, gradient analysis, extreme point location and signal quantity calculation, the start and end points of the signal generation stage are accurately extracted and the accurate signal quantity is obtained, thereby solving the existing technical problems.

[0008] To address the aforementioned technical problems, this invention provides a method for data processing and signal extraction of an electronic microarray chip, comprising the following steps: Step S1: Data Preprocessing: Read the raw data obtained from the electronic microarray chip detection. The raw data includes data inside the array and data outside the array. Stable external wells and effective internal wells are selected by differential denoising. The difference between the mean inside the array and the mean outside the array is calculated to obtain the voltage difference data. The voltage difference data is then subjected to double smoothing filtering and standardization. Finally, PCA dimensionality reduction and data reconstruction are used to obtain the preprocessed voltage difference data.

[0009] Step S2: Gradient calculation and smoothing: Baseline correction is performed on the preprocessed voltage difference data, the first and second derivatives of the corrected data are calculated, and then double smoothing filtering is performed on the first and second derivatives to obtain smoothed gradient data.

[0010] Step S3: Local Minimum Location: Search for local minima in the first half of the smoothed first derivative data (default first 40 frames) using a sliding window, satisfying the condition that the center of the window is less than or equal to the value in the left window and less than the value in the right window. Based on the search results, filter to obtain the initial candidate points for the signal start point st and the signal end point ed.

[0011] Step S4: Semaphore Calculation: Based on st and ed obtained from positioning, calculate the effective semaphore and mean value during the signal generation stage to improve the reliability of signal recognition.

[0012] In a further step of this invention, the differential denoising screening operation in step S1 includes: calculating the integral of each data curve in each reaction stage; firstly, selecting stable wells outside the lattice in the first 8 reaction stages; based on the integral of each well curve, sorting them from smallest to largest in the 8 reaction stages, wells with an integral between 25% and 75% in each stage are considered stable wells; then, in the high-mer reaction, selecting wells within the stable lattice as effective wells within the lattice; based on the integral of each well curve in the high-mer reaction stage, wells that satisfy the condition of being higher than the average value outside the effective lattice within at least half of the high-mer reaction stages.

[0013] In a further step of this invention, in step S1, the preprocessing of the voltage difference data performs the following operations: Dual smoothing filtering: Median filtering (window size 7) and uniform filtering (window size 7) are applied sequentially to the voltage difference data to further suppress random noise and baseline drift, thereby improving data smoothness.

[0014] Standardization: The smoothed data is standardized so that the mean is 0 and the standard deviation is 1, thus eliminating dimensional differences.

[0015] PCA dimensionality reduction and data reconstruction: The first principal component is extracted through principal component analysis (PCA), and the inverse transformation reconstruction of the data is completed based on the first principal component. This achieves dimensionality reduction and redundancy removal of the data, retains the core features, and obtains the preprocessed voltage difference data.

[0016] In a further step of this invention, during baseline correction, the voltage difference data reconstructed by PCA is subtracted from the value of the initial frame to obtain the voltage difference data after baseline correction; during dual smoothing filtering, median filtering (window size 7) and uniform filtering (window size 7) are performed on the first derivative and the second derivative respectively to suppress the noise introduced during gradient calculation and obtain smoothed first and second derivative data.

[0017] Furthermore, in step S3 of this invention, the minimum point selection rule is as follows: If at least two local minima are found, the first two are taken as the initial candidate points for st and ed.

[0018] If only one local minimum point is found, then that point is taken as st, and missing minimum points are repaired to obtain candidate points ed.

[0019] If no local minimum point is found, the minimum value of the previous n frames (n=10) of the signal is taken as st, and candidate points of ed are determined by repairing missing minimum points or filtering by the minimum absolute value point.

[0020] Furthermore, in this invention, missing minimum points are repaired based on gradient features and signal change trends, specifically including the following steps: Step S31: Analyze signal features: For samples with missing ed candidate points, extract the voltage difference data segment after st, calculate the second derivative values ​​of the three points of the segment's start, mid, and end, and extract the points where the overall second derivative sign changes (reflecting the position of the signal change direction conversion).

[0021] Step S32: Determine the repair area: Based on the sign of the second derivative values ​​of the three points, and combined with the point where the sign of the second derivative changes, determine the repair search area of ​​ed. When the second derivative of the three points is positive, ed searches before the first point where the sign of the second derivative changes after st; otherwise, it searches between the first and second points where the sign changes after st.

[0022] Step S33: Linear Fitting and Crossover Point Location: Perform linear fitting on the signal within the repair search area, calculate the difference sequence between the fitted line and the signal curve, search for the positive and negative crossover points of the sequence, and use the crossover point positions as candidate points for the repaired ed.

[0023] Furthermore, in this invention, step S4 specifically includes the following steps: Step S41: Based on the signal values ​​corresponding to st and ed, the baseline of the signal generation stage is obtained by using a linear fitting method.

[0024] Step S42: Calculate the area difference between the first derivative signal curve and the baseline and the average signal quantity per unit length, as the core signal quantity index.

[0025] This invention also provides an electronic microarray chip data processing and signal extraction system, which performs the aforementioned electronic microarray chip data processing and signal extraction method, and includes the following modules: The data preprocessing module is used to perform differential denoising to screen stable outer wells and effective inner wells of the lattice and calculate the difference between the mean data curves of the two to obtain the initial voltage difference data. The module then performs dual smoothing filtering, standardization, PCA dimensionality reduction and data reconstruction on the initial voltage difference data, and finally outputs the preprocessed voltage difference data.

[0026] The gradient calculation and smoothing module is used to perform baseline correction on the preprocessed voltage difference data, calculate the first and second derivatives of the corrected data, perform median filtering and uniform filtering on the first and second derivatives, and output smoothed gradient data.

[0027] The dual local minimum point localization module is used to search for local minimum points in the first half of the smoothed first derivative data through a sliding window, and to obtain initial candidate points for the signal start point st and the signal end point ed based on the search results.

[0028] The missing local minimum point repair module is used to extract the voltage difference data segment after st when only one local minimum point is found or no local minimum point is found, resulting in the missing signal termination point ed. Based on the second derivative values ​​and the change points of the second derivative sign of the segment's start, mid, and end points, the repair search area of ​​ed is determined, and ed is located by linear fitting and the intersection point of the difference sequence.

[0029] The semaphore calculation module is used to perform baseline fitting based on the signal start point st and signal end point ed obtained from the positioning. Based on the fitted baseline, it calculates the area difference between the first derivative signal curve and the baseline and the average signal quantity per unit length, generating core semaphore indices.

[0030] The present invention also provides an electronic device, comprising: At least one processor; and At least one memory communicatively connected to the processor; The memory stores instructions that can be executed by a processor, which are then executed by the processor to cause the electronic device to perform the aforementioned electronic microarray chip data processing and signal extraction method.

[0031] The present invention also provides a computer-readable storage medium storing computer-executable instructions thereon, which, when executed by a processor, implement the aforementioned method for data processing and signal extraction of an electronic microarray chip.

[0032] The electronic microarray chip data processing and signal extraction method and system of this invention are based on a combined preprocessing scheme of "dual smoothing + PCA dimensionality reduction and reconstruction". It enhances noise suppression through the synergistic effect of median filtering and uniform filtering, and combines PCA dimensionality reduction to accurately preserve core signal features, effectively solving the problem of incomplete separation of data inside and outside the array. This invention constructs a dual minimum point localization model of "local minimum search + gradient features + cross-point screening", integrating the amplitude and trend features of the signal to improve the accuracy of start and end point localization. Simultaneously, it designs an adaptive missing point repair mechanism to solve the problem of missed minimum point detection in complex scenarios. This invention proposes a signal quantity calculation method based on baseline fitting, combining linear baseline fitting and integral operations to achieve accurate calculation of effective signal quantity, providing multi-dimensional basis for signal recognition and improving the reliability of detection results. Compared with existing technologies, this invention has the following beneficial effects: (1) Strong anti-interference capability: By combining the dual smoothing mechanism of "median filtering + uniform filtering" with PCA dimensionality reduction and redundancy removal, random noise and baseline drift are effectively suppressed, and the effective signal inside the dot matrix and the interference data outside the dot matrix are accurately separated, thereby improving data quality.

[0033] (2) Accurate positioning of start and end points: By combining local minimum search, gradient feature analysis and cross-point screening, the start and end points of the signal are accurately located. An adaptive repair mechanism is designed for missing points to adapt to complex scenarios such as weak signals and smooth signals.

[0034] (3) Wide data adaptability: Through standardized processing, adaptive window setting and gradient feature adaptive analysis, it can adapt to signal data of different detection batches and different chip sites, without the need for manual parameter adjustment, thus reducing the complexity of operation.

[0035] (4) Accurate signal quantity calculation: The integral calculation method based on baseline fitting fully considers the deviation between the signal and the baseline, and the obtained signal quantity is closer to the actual effective signal strength. Attached Figure Description

[0036] The specific embodiments of the present invention will be further explained below with reference to the accompanying drawings.

[0037] Figure 1 This is a flowchart of the electronic microarray chip data processing and signal extraction method of the present invention.

[0038] Figure 2(a) shows the original well data in the lattice before screening.

[0039] Figure 2(b) shows the original well data within the filtered lattice.

[0040] Figure 3(a) shows the voltage difference before pretreatment.

[0041] Figure 3(b) shows the voltage difference after pretreatment.

[0042] Figure 4(a) is a schematic diagram of calculating the second derivative through the three points of the starting point, midpoint and ending point of the negative phase voltage difference curve segment.

[0043] Figure 4(b) is a schematic diagram of calculating the second derivative through the three points of the starting point, midpoint and ending point of the voltage difference curve segment in the positive phase.

[0044] Figure 5(a) is a schematic diagram of determining the termination point ed by linear fitting of the first derivative curve of the negative phase and the location of the intersection point.

[0045] Figure 5(b) is a schematic diagram of determining the termination point ed by linear fitting of the first derivative curve of the positive phase and the location of the intersection point.

[0046] Figure 6(a) is a schematic diagram of the difference between the solution signal curve and the baseline during the negative phase.

[0047] Figure 6(b) is a schematic diagram of the difference between the positive phase solution signal curve and the baseline.

[0048] Figure 7 This is a system block diagram of the electronic microarray chip data processing and signal extraction system of the present invention. Detailed Implementation Example 1

[0049] Combination Figure 1 As shown, the electronic microarray chip data processing and signal extraction method of this embodiment specifically includes the following steps: Step S1: Data Preprocessing: Read the raw data obtained from the electronic microarray chip detection. The raw data includes data inside the array and data outside the array. Stable external wells and effective internal wells are selected by differential denoising. The difference between the mean inside the array and the mean outside the array is calculated to obtain the voltage difference data. The voltage difference data is then subjected to double smoothing filtering and standardization. Finally, PCA dimensionality reduction and data reconstruction are used to obtain the preprocessed voltage difference data.

[0050] In this embodiment, preferably, in step S1, the differential denoising screening operation includes: calculating the integral of each data curve in each reaction stage; firstly, selecting stable wells outside the lattice in the first 8 reaction stages; based on the integral of each well curve, sorting them from smallest to largest in the 8 reaction stages, wells with an integral between 25% and 75% in each stage are considered stable wells; then, in the high-mer reaction, selecting wells within the stable lattice as effective wells within the lattice; based on the integral of each well curve in the high-mer reaction stage, wells that satisfy the condition of being higher than the average value outside the effective lattice in at least half of the high-mer reaction stages.

[0051] In this embodiment, preferably, in step S1, the preprocessing of the voltage difference data performs the following operations: Dual smoothing filtering: Median filtering and uniform filtering are performed sequentially on the voltage difference data. In this embodiment, the window size of the median filter is 7, and the window size of the uniform filter is 7, which further suppresses random noise and baseline drift and improves data smoothness.

[0052] Standardization: The smoothed data is standardized so that the mean is 0 and the standard deviation is 1, thus eliminating dimensional differences.

[0053] PCA dimensionality reduction and data reconstruction: The first principal component is extracted through principal component analysis (PCA), and the inverse transformation reconstruction of the data is completed based on the first principal component. This achieves dimensionality reduction and redundancy removal of the data, retains the core features, and obtains the preprocessed voltage difference data.

[0054] Step S1 completes the noise suppression and standardization of the original data, achieving the initial separation and optimization of effective data within the dot matrix and data outside the dot matrix.

[0055] Step S2: Gradient calculation and smoothing: Baseline correction is performed on the preprocessed voltage difference data, the first and second derivatives of the corrected voltage difference data are calculated, and then double smoothing filtering is performed on the first and second derivatives to obtain the smoothed gradient data.

[0056] In this embodiment, preferably, in step S2, when performing baseline correction, the voltage difference data after PCA reconstruction is subtracted from the value of the initial frame to obtain the voltage difference data after baseline correction; when performing dual smoothing filtering, median filtering and uniform filtering are performed on the first derivative and the second derivative respectively. In this embodiment, the window size of median filtering is 7, and the window size of uniform filtering is 7, to suppress the noise introduced during gradient calculation and obtain smoothed first and second derivative data.

[0057] Step S2 calculates the first and second derivatives of the preprocessed voltage difference data, using the first derivative data as a signal to provide characteristic basis for subsequent localization of minimum points.

[0058] Step S3: Local Minimum Location: Search for local minima in the first half of the smoothed first derivative data (default first 40 frames) using a sliding window. In this embodiment, the size of the left window is set to 3 and the size of the right window is set to 6. The center of the window is less than or equal to the value in the left window and less than the value in the right window. Based on the search results, the initial candidate points of the signal start point st and the signal end point ed are obtained.

[0059] In this embodiment, preferably, the minimum point selection rule in step S3 is as follows: If at least two local minima are found, the first two are taken as the initial candidate points for st and ed.

[0060] If only one local minimum point is found, then that point is taken as st, and missing minimum points are repaired to obtain candidate points ed.

[0061] If no local minimum point is found, the minimum value of the previous n frames (n=10) of the signal is taken as st, and candidate points of ed are determined by repairing missing minimum points or filtering by the minimum absolute value point.

[0062] In this embodiment, the missing minimum points are preferably repaired based on gradient features and signal change trends, specifically including the following steps: Step S31: Analyze signal features: For samples with missing ed candidate points, extract the voltage difference data segment after st, calculate the second derivative values ​​of the three points of the segment's start, mid, and end, and extract the points where the overall second derivative sign changes (reflecting the position of the signal change direction conversion).

[0063] Step S32: Determine the repair area: Based on the sign of the second derivative values ​​of the three points, and combined with the point where the sign of the second derivative changes, determine the repair search area of ​​ed. When the second derivative of the three points is positive, ed searches before the first point where the sign of the second derivative changes after st; otherwise, it searches between the first and second points where the sign changes after st.

[0064] Step S33: Linear Fitting and Crossover Point Location: Perform linear fitting on the signal within the repair search area, calculate the difference sequence between the fitted line and the signal curve, search for the positive and negative crossover points of the sequence, and use the crossover point positions as candidate points for the repaired ed.

[0065] Step S3 locates two key local minima during the first derivative signal generation stage, namely the signal start point st and the signal termination point ed. These two points correspond to the baseline positions before and after the generation stage and are the core basis for dividing the signal generation stage.

[0066] Step S4: Semaphore Calculation: Based on st and ed obtained from positioning, calculate the effective semaphore and mean value during the signal generation stage to improve the reliability of signal recognition.

[0067] In this embodiment, step S4 preferably includes the following steps: Step S41: Based on the signal values ​​corresponding to st and ed, the baseline of the signal generation stage is obtained by using a linear fitting method.

[0068] Step S42: Calculate the area difference between the first derivative signal curve and the baseline and the average signal quantity per unit length, as the core signal quantity index.

[0069] Step S4 calculates the effective signal quantity and mean value during the signal generation stage based on the st and ed obtained from the positioning, thereby improving the reliability of signal recognition.

[0070] Figures 2(a), 2(b), 3(a), 3(b), 4(a), 4(b), 5(a), 5(b), and 6(a) and 6(b) illustrate the data processing procedure, where: Figure 2(a) and Figure 2(b) show the filtering of the original dot matrix signal and the extraction of stable data. Figure 2(a) shows the original well data in the dot matrix before filtering and Figure 2(b) shows the original well data in the dot matrix after filtering.

[0071] Figures 3(a) and 3(b) illustrate the denoising process for voltage difference data. Figure 3(a) shows the voltage difference before preprocessing, and Figure 3(b) shows the voltage difference after preprocessing.

[0072] Figures 4(a) and 4(b) show the curve characteristics of the voltage difference in the two reaction stages. Figure 4(a) shows the calculation of the second derivative using the three points of the negative stage voltage difference curve segment: the start point, the midpoint, and the end point. Figure 4(b) shows the calculation of the second derivative using the three points of the positive stage voltage difference curve segment: the start point, the midpoint, and the end point. In the figures, green dots indicate that the second derivative is positive, blue dots indicate that it is negative, and red dots indicate the start point, the midpoint, and the end point of the segment. When searching for the first derivative using extreme points in the first 40 frames, if only one minimum value can be found, the second derivative calculated using these three points on the corresponding voltage difference curve is used to determine which segment of the first derivative should be searched for based on its sign. When the second derivative of the three points is positive, the search for ed is performed before the first second derivative sign change point after st; otherwise, the search is performed between the first and second sign change points after st.

[0073] Figures 5(a) and 5(b) show the curve characteristics of the first derivative of the two reaction stages. Figure 5(a) shows the termination point ed established by linear fitting of the first derivative curve of the negative stage and intersection point location. Figure 5(b) shows the termination point ed established by linear fitting of the first derivative curve of the positive stage and intersection point location. The red line in the figure is the straight line obtained by linear fitting the signal in the corresponding search area. The intersection point of the fitted straight line and the signal curve is recorded as the found termination point ed. The orange triangle marks the finally found st and ed.

[0074] Figures 6(a) and 6(b) illustrate the calculation of the signal quantity. Figure 6(a) is a schematic diagram of the difference between the calculated signal curve and the baseline in the negative phase, and Figure 6(b) is a schematic diagram of the difference between the calculated signal curve and the baseline in the positive phase. The baseline is established by st and ed, and the difference between the signal curve and the baseline is calculated to obtain the signal quantity. Example 2

[0075] Combination Figure 7 As shown, the electronic microarray chip data processing and signal extraction system of this embodiment executes the electronic microarray chip data processing and signal extraction method in Embodiment 1, and includes the following modules: The data preprocessing module is used to perform differential denoising to screen stable outer wells and effective inner wells of the lattice and calculate the difference between the mean data curves of the two to obtain the initial voltage difference data. The module then performs dual smoothing filtering, standardization, PCA dimensionality reduction and data reconstruction on the initial voltage difference data, and finally outputs the preprocessed voltage difference data.

[0076] In this embodiment, the data preprocessing module specifically includes a differential denoising unit, a dual smoothing filter unit, a standardization unit, and a PCA dimensionality reduction and reconstruction unit. The differential denoising unit is used to screen out stable lattice external wells whose integrals in each stage are in the range of 25% to 75% based on the integrals of each well curve in the first 8 reaction stages. In the high-MER reaction stage, at least half of the effective lattice internal wells whose integrals in the high-MER reaction stage are higher than the mean of the effective lattice external wells are screened out based on the integrals of each well curve. The difference between the mean data curves of the effective lattice internal wells and the stable lattice external wells is calculated to obtain the voltage difference data. The dual smoothing filter unit is used to perform dual smoothing filtering on the voltage difference data. The standardization unit is used to standardize the smoothed voltage difference data. The PCA dimensionality reduction and reconstruction unit is used to extract the first principal component through principal component analysis and complete the inverse transformation and reconstruction of the data based on the first principal component.

[0077] The gradient calculation and smoothing module is used to perform baseline correction on the preprocessed voltage difference data, calculate the first and second derivatives of the corrected data, perform median filtering and uniform filtering on the first and second derivatives, and output smoothed gradient data.

[0078] In this embodiment, the gradient calculation and smoothing module specifically includes a baseline correction unit, a gradient calculation unit, and a gradient smoothing unit. The baseline correction unit is used to subtract the value of the initial frame from the voltage difference data after PCA reconstruction to complete the baseline correction. The gradient calculation unit is used to calculate the first and second derivatives of the corrected voltage difference data. The gradient smoothing unit is used to perform median filtering with a window size of 7 and uniform filtering with a window size of 7 on the first and second derivatives respectively.

[0079] The dual local minimum point localization module is used to search for local minimum points in the first half of the smoothed first derivative data through a sliding window, and to obtain initial candidate points for the signal start point st and the signal end point ed based on the search results.

[0080] The missing local minimum point repair module is used to extract the voltage difference data segment after st when only one local minimum point is found or no local minimum point is found, resulting in the missing signal termination point ed. Based on the second derivative values ​​and the change points of the second derivative sign of the segment's start, mid, and end points, the repair search area of ​​ed is determined, and ed is located by linear fitting and the intersection point of the difference sequence.

[0081] Specifically, in this embodiment, the missing minimum point repair module includes a signal feature analysis unit, a repair region determination unit, and an intersection point location unit. The signal feature analysis unit is used to calculate the second derivative values ​​of the start point, midpoint, and end point of the voltage difference data segment, and to extract the points of change in the sign of the second derivative of the entire data segment. The repair region determination unit is used to determine the repair search region of ed based on the positive and negative values ​​of the second derivative values ​​and the points of change in the sign of the second derivative. The intersection point location unit is used to perform linear fitting on the signal within the repair search region, calculate the difference sequence between the fitted line and the signal curve, and search for the positive and negative intersection points of the difference sequence as the repaired ed.

[0082] The semaphore calculation module is used to perform baseline fitting based on the signal start point st and signal end point ed obtained from the positioning. Based on the fitted baseline, it calculates the area difference between the first derivative signal curve and the baseline and the average signal quantity per unit length, generating core semaphore indices. Example 3

[0083] The electronic device in this embodiment includes: At least one processor; and At least one memory communicatively connected to the processor; The memory stores instructions that can be executed by a processor, which are then executed by the processor to cause the electronic device to perform the electronic microarray chip data processing and signal extraction method in Embodiment 1. Example 4

[0084] The computer-readable storage medium of this embodiment stores computer-executable instructions. When the processor executes the computer-executable instructions, it implements the electronic microarray chip data processing and signal extraction method as described in Embodiment 1.

[0085] Many specific details have been set forth in the foregoing description to provide a thorough understanding of the present invention. However, the above description is merely a preferred embodiment of the present invention, and the present invention can be implemented in many other ways different from those described herein. Therefore, the present invention is not limited to the specific embodiments disclosed above. Furthermore, any person skilled in the art can make many possible variations and modifications to the technical solutions of the present invention, or modify them into equivalent embodiments, using the methods and techniques disclosed above, without departing from the scope of the present invention. Any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention, without departing from the content of the present invention, shall still fall within the protection scope of the present invention.

Claims

1. A method for data processing and signal extraction of an electronic microarray chip, characterized in that: Includes the following steps: Step S1: Data preprocessing: Read the raw data obtained by the electronic microarray chip detection. The raw data includes data inside the array and data outside the array. Stable external wells and effective internal wells are selected by differential denoising. The difference between the mean inside the array and the mean outside the array is calculated to obtain the voltage difference data. The voltage difference data is then subjected to double smoothing filtering and standardization. Finally, PCA dimensionality reduction and data reconstruction are used to obtain the preprocessed voltage difference data. Step S2: Gradient calculation and smoothing: Baseline correction is performed on the preprocessed voltage difference data, the first and second derivatives of the corrected data are calculated, and then double smoothing filtering is performed on the first and second derivatives to obtain the smoothed gradient data. Step S3: Local Minimum Location: Search for local minima in the first half of the smoothed first derivative data using a sliding window, satisfying that the center of the window is less than or equal to the value in the left window and less than the value in the right window. Based on the search results, the initial candidate points for the signal start point st and the signal end point ed are obtained. Step S4: Semaphore Calculation: Based on st and ed obtained from positioning, calculate the effective semaphore and mean value during the signal generation phase.

2. The method for data processing and signal extraction of electronic microarray chips according to claim 1, characterized in that: In step S1, the differential denoising screening operation includes: calculating the integral of each data curve in each reaction stage; firstly, selecting stable wells outside the lattice in the first 8 reaction stages; based on the integral of each well curve, sorting them from smallest to largest in the 8 reaction stages, wells with an integral between 25% and 75% in each stage are considered stable wells; then, in the high-mer reaction, selecting wells within the stable lattice as effective wells within the lattice; based on the integral of each well curve in the high-mer reaction stages, wells that satisfy the condition of being higher than the mean of effective outside the lattice in at least half of the high-mer reaction stages.

3. The method for data processing and signal extraction of electronic microarray chips according to claim 1, characterized in that: In step S1, the preprocessing of the voltage difference data is performed as follows: Dual smoothing filtering: Median filtering and uniform filtering are applied sequentially to the voltage difference data; Standardization: The smoothed data is standardized so that the mean is 0 and the standard deviation is 1, thus eliminating dimensional differences. PCA dimensionality reduction and data reconstruction: The first principal component is extracted through principal component analysis, and the inverse transformation and reconstruction of the data is completed based on the first principal component. This achieves dimensionality reduction and redundancy removal of the data, retains the core features, and obtains the preprocessed voltage difference data.

4. The method for data processing and signal extraction of electronic microarray chips according to claim 1, characterized in that: In step S2, when performing baseline correction, the voltage difference data after PCA reconstruction is subtracted from the value of the initial frame to obtain the voltage difference data after baseline correction; when performing dual smoothing filtering, median filtering and uniform filtering are performed on the first derivative and the second derivative respectively to suppress the noise introduced during gradient calculation and obtain the smoothed first and second derivative data.

5. The method for data processing and signal extraction of electronic microarray chips according to claim 1, characterized in that: In step S3, the minimum point selection rule is as follows: If at least two local minima are found, the first two are taken as the initial candidate points for st and ed. If only one local minimum point is found, then that point is taken as st, and missing minimum points are repaired to obtain candidate points ed; If no local minimum point is found, the minimum value of the signal in the previous n frames is taken as st, and candidate points for ed are determined by repairing missing minimum points or filtering by the minimum absolute value point.

6. The method for data processing and signal extraction of electronic microarray chips according to claim 5, characterized in that: Missing local points are repaired based on gradient features and signal change trends, specifically including the following steps: Step S31: Analyze signal features: For samples with missing ed candidate points, extract the voltage difference data segment after st, calculate the second derivative values ​​at the start, middle and end points of the segment, and extract the points where the sign of the second derivative changes as a whole in the segment; Step S32: Determine the repair area: Based on the sign of the second derivative values ​​of the three points, and combined with the point where the sign of the second derivative changes, determine the repair search area of ​​ed. When the second derivative of the three points is positive, ed searches before the first point where the sign of the second derivative changes after st; otherwise, it searches between the first and second points where the sign changes after st. Step S33: Linear Fitting and Crossover Point Location: Perform linear fitting on the signal within the repair search area, calculate the difference sequence between the fitted line and the signal curve, search for the positive and negative crossover points of the sequence, and use the crossover point positions as candidate points for the repaired ed.

7. The method for data processing and signal extraction of electronic microarray chips according to claim 1, characterized in that: Step S4 specifically includes the following steps: Step S41: Based on the signal values ​​corresponding to st and ed, obtain the baseline of the signal generation stage using a linear fitting method; Step S42: Calculate the area difference between the first derivative signal curve and the baseline and the average signal quantity per unit length, as the core signal quantity index.

8. A data processing and signal extraction system for an electronic microarray chip, characterized in that: The method for data processing and signal extraction of an electronic microarray chip as described in any one of claims 1-7 includes the following modules: The data preprocessing module is used to perform differential denoising to screen stable outer wells and effective inner wells of the lattice and calculate the difference between the mean data curves of the two to obtain the initial voltage difference data. The module then performs dual smoothing filtering, standardization, PCA dimensionality reduction and data reconstruction on the initial voltage difference data, and finally outputs the preprocessed voltage difference data. The gradient calculation and smoothing module is used to perform baseline correction on the preprocessed voltage difference data, calculate the first and second derivatives of the corrected data, perform median filtering and uniform filtering on the first and second derivatives, and output smoothed gradient data. The dual local minimum point localization module is used to search for local minimum points in the first half of the smoothed first derivative data through a sliding window, and to obtain initial candidate points for the signal start point st and the signal end point ed based on the search results. The missing local minimum point repair module is used to extract the voltage difference data segment after st when only one local minimum point is found or no local minimum point is found, resulting in the missing signal termination point ed. Based on the second derivative values ​​of the segment's start point, mid point, and end point, as well as the points where the second derivative sign changes, the repair search area of ​​ed is determined, and ed is located by linear fitting and the intersection point of the difference sequence. The semaphore calculation module is used to perform baseline fitting based on the signal start point st and signal end point ed obtained from the positioning. Based on the fitted baseline, it calculates the area difference between the first derivative signal curve and the baseline and the average signal quantity per unit length, generating core semaphore indices.

9. An electronic device, characterized in that: include: At least one processor; as well as At least one memory communicatively connected to the processor; The memory stores instructions that can be executed by a processor, which are executed by the processor to cause the electronic device to perform the electronic microarray chip data processing and signal extraction method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: It stores computer execution instructions, and when the processor executes the computer execution instructions, it implements the electronic microarray chip data processing and signal extraction method as described in any one of claims 1-7.

Images