A signal processing and determination method based on capillary electrophoresis nucleic acid fragment analysis

Abstract

The present invention discloses a signal processing and determination method based on capillary electrophoresis nucleic acid fragment analysis, which relates to the field of bioinformatics, and includes the steps of: obtaining the original waveform sampling signal after capillary electrophoresis processing after multiple polymerase chain reactions; obtaining the peak type feature point search after peak type optimization by preprocessing the original waveform sampling signal; performing preliminary screening of signal peak target points under comparison and judgment of the peak type feature points based on a preset position interval and a peak height threshold; performing proportional combination analysis based on the feature points obtained by the search, and marking the target target points under secondary judgment of the signal peak based on the analysis results. The present invention sets a specific algorithm for secondary identification according to the characteristics of the signal peak, such as the shape and height, which greatly improves the accuracy of target target point determination, and the combination of the three algorithms allows users to choose to start one or more algorithms according to actual conditions, providing higher flexibility.

Description

Technical Field

[0001] The present invention relates to the field of bioinformatics, and in particular to a signal processing and determination method based on capillary electrophoresis nucleic acid fragment analysis. Background Art

[0002] Fragment analysis (FA) is a molecular biology technique that performs nucleic acid analysis by measuring fluorescently labeled nucleic acid fragments of different sizes. It is widely used in genotyping, DNA fingerprint analysis, mutation detection, etc. in the fields of medicine, environment, and agricultural research.

[0003] The principle is to design primers for different target sites, and the products amplified from each target site differ by at least 1 bp. The primers are synthesized by a third party and labeled with fluorescence at the end. Different size fragments are labeled with the same or different color fluorescence, and multiple polymerase chain reactions (PCR) are performed so that all nucleic acid fragments of the target sites are fluorescently labeled. After product pretreatment, capillary electrophoresis is performed, and the products of different fragment sizes and different color fluorescence labels are distinguished and identified using a fluorescence detector.

[0004] In order to achieve fully automatic analysis, foreign countries started research on nucleic acid fragment analysis and the development of corresponding software earlier. Currently, the more famous commercial software is the software attached to the gene analyzer developed by ABI in the United States. This software can analyze and display the collected raw data, identify the position and peak height of each peak in the electrophoresis spectrum, and thus determine the length of the nucleic acid fragment (peak position) and the corresponding content (peak height). By comparing the obtained peak position and peak height with the preset position interval and peak height threshold, the target point can be identified.

[0005] During the fragment analysis process, due to problems such as nucleic acid concentration, instrument status fluctuations, or reagent and consumables quality, baseline disturbances occur, and interference peaks appear in the preset position range. False positives are prone to occur when only a single peak height threshold is used for judgment, which greatly troubles the interpretation of the target. Summary of the invention

[0006] In order to more accurately identify the target and avoid misjudgment, the present invention proposes a signal processing and determination method based on capillary electrophoresis nucleic acid fragment analysis, comprising the steps of:

[0007] S1: obtain the original waveform sampling signal after multiple polymerase chain reaction and capillary electrophoresis processing;

[0008] S2: Preprocessing the original waveform sampling signal to obtain the peak feature point search after the peak shape is optimized;

[0009] S3: Preliminary screening of signal peak target points based on comparison and judgment of peak feature points based on preset position intervals and peak height thresholds;

[0010] S4: Perform proportional combination analysis based on the feature points obtained by the search, and mark the target points under secondary judgment of the signal peak based on the analysis results.

[0011] Furthermore, in the step S2, the preprocessing includes baseline subtraction and peak height lossless filtering.

[0012] Furthermore, the baseline subtraction adopts a window sliding minimum value extraction method to remove abnormal interference baseline signals.

[0013] Furthermore, the peak height lossless filtering uses median filtering and polynomial smoothing to remove the cluttered peaks.

[0014] Furthermore, in step S2, the peak characteristic point is the inflection point of the first-order derivative of the sampling signal.

[0015] Furthermore, in the step S4, the ratio combination analysis includes peak height ratio analysis, peak area / peak height ratio analysis and mirror image ratio analysis.

[0016] Furthermore, the peak height ratio analysis is specifically as follows:

[0017] The peak height of the target mark signal peak is extracted and compared with the peak height of the wave within the preset distance range to obtain the signal-to-noise ratio data. If the signal-to-noise ratio data contains a signal-to-noise ratio less than the first set threshold, the signal peak is determined to be an interference signal and the target mark is cancelled.

[0018] Furthermore, the peak area / peak height ratio analysis is specifically as follows:

[0019] The peak height and peak area of ​​the target mark signal peak are extracted. If the ratio of the peak area to the peak height is greater than a second set threshold, the signal peak is determined to be an interference signal and the target mark is cancelled.

[0020] Furthermore, the mirror ratio analysis is specifically as follows:

[0021] The peak width of the target marker signal peak is used as the extraction range, and the peak position of the target marker signal peak is used as the extraction starting point. The waveform signal of the peak within the extraction range around the extraction starting point is obtained, and the correlation coefficient is obtained after mirror processing. If the correlation coefficient is less than the third set threshold, the signal peak is determined to be an interference signal and the target target mark is cancelled.

[0022] Compared with the prior art, the present invention has at least the following beneficial effects:

[0023] (1) The signal processing and determination method based on capillary electrophoresis nucleic acid fragment analysis described in the present invention sets a specific algorithm for secondary identification according to the shape and height of the signal peak, which greatly improves the accuracy of target point determination, assists experimenters in result determination, and significantly improves work efficiency;

[0024] (2) Three different algorithms are provided for signal peak identification. Users can choose to start one or more algorithms according to actual conditions, providing high flexibility;

[0025] (3) Linear compensation of the peak interval amplitude of the filtered signal is used to eliminate signal loss during the filtering process. This ensures the accuracy of the signal peak height and prevents changes in the actual signal strength caused by filtering. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 A step diagram of a signal processing and determination method based on capillary electrophoresis nucleic acid fragment analysis;

[0027] Figure 2 is the original waveform sampling signal;

[0028] Figure 3 It is a schematic diagram of the signal before and after baseline subtraction correction;

[0029] Figure 4 Schematic diagram of the signal before and after peak height lossless filtering;

[0030] Figure 5 Search for peak locations and mark peak graphs for peak feature points;

[0031] Figure 6 Open the before and after identification result graph for peak height ratio analysis;

[0032] Figure 7 It is the distribution diagram of the s / h ratio of normal peak and interference peak;

[0033] Figure 8 Open the before and after identification result graph for peak area / peak height ratio analysis;

[0034] Fig. 9 Open the before and after recognition result graph for mirror ratio analysis;

[0035] Description of the drawings: In the drawings, the horizontal axis of the coordinate system is the fragment size, and the vertical axis is the peak height. DETAILED DESCRIPTION

[0036] The following are specific embodiments of the present invention and the accompanying drawings to further describe the technical solution of the present invention, but the present invention is not limited to these embodiments.

[0037] In biological science research, capillary electrophoresis is often used to separate and analyze biological macromolecules such as proteins, amino acids, peptides, and nucleotides. Especially in genetic analysis, it can be used for genotyping, DNA fingerprint analysis, and mutation detection. By combining fluorescent labeling technology, the detection sensitivity can be enhanced, so that specific sequences in trace samples can be clearly distinguished. In order to solve the problem of interference in target interpretation caused by nucleic acid concentration, instrument status fluctuations, etc. during fragment analysis, such as Figure 1 As shown, the present invention proposes a signal processing and determination method based on capillary electrophoresis nucleic acid fragment analysis, comprising the steps of:

[0038] S1: obtain the original waveform sampling signal after multiple polymerase chain reaction and capillary electrophoresis processing;

[0039] S2: Preprocessing the original waveform sampling signal to obtain the peak feature point search after the peak shape is optimized;

[0040] S3: Preliminary screening of signal peak target points based on comparison and judgment of peak feature points based on preset position intervals and peak height thresholds;

[0041] S4: Perform proportional combination analysis based on the feature points obtained by the search, and mark the target points under secondary judgment of the signal peak based on the analysis results.

[0042] In the present invention, the original waveform sampling signal (such as Figure 2 As shown in the figure, considering the differences in signal baselines between different channels of the original signal, it is necessary to ensure the consistency of channel space signal values ​​through baseline subtraction. To this end, we use the window sliding minimum value extraction method. The window width generally ensures that there must be a baseline signal point in the window in most cases. Therefore, take a length of 10 consecutive half-peak widths, and the window sliding step is 1 half-peak width. Each time a window is slid, the window minimum value is output. Continuous sliding can eventually obtain a set of window minimum value sequences. The sequence is subjected to median filtering and low-pass filtering to remove abnormal interfering baseline signals. As shown in the figure Figure 3 Shown is a schematic diagram of the signal before and after baseline subtraction correction.

[0043] Then, we perform median filtering (filter window width can be set) + SG filtering (polynomial smoothing, polynomial order can be set, filter window width can be set) on the signal after deducting the baseline, and eliminate the noise peaks caused by baseline jumps through filtering to smooth the signal. After filtering, the peak signal is gain-amplified to ensure that the peak height value remains unchanged. Figure 4 As shown, it is a schematic diagram of the signal before and after the peak height lossless filtering.

[0044] After the signal is preprocessed, in order to facilitate the subsequent identification of interference signals, we also need to search for the characteristic points of the peak type. Specifically, Figure 5 As shown, according to the signal after lossless filtering, the peak position search is performed on the signal above the dynamic threshold, corresponding to the first-order derivative crossing the zero point, and the first-order derivative near the zero point is monotonically decreasing. The peak foot point is preliminarily determined by the second-order derivative maximum point. When the first-order derivative corresponding to the extreme point of the second-order derivative of the left peak foot is less than or equal to 0, the extreme point is used as the left peak foot point, otherwise the search continues to the left until the first-order derivative is less than or equal to 0, and the point is used as the left peak foot point; when the first-order derivative corresponding to the extreme point of the second-order derivative of the right peak foot is greater than or equal to 0, the extreme point is used as the right peak foot point, otherwise the search continues to the right until the first-order derivative is greater than or equal to 0, and the point is used as the right peak foot point. Then, according to the characteristic inflection point obtained by the search, the peak position and peak height of each signal peak can be determined, and they are compared with the preset position interval and peak height threshold, and the signal peaks in the position interval and whose peak height exceeds the threshold are screened out for subsequent secondary judgment processing.

[0045] Specifically, the present invention proposes a secondary determination method that uses multiple analysis methods to perform independent or multiple combined determinations, including peak height ratio analysis, peak area / peak height ratio analysis, and mirror ratio analysis. Among them:

[0046] Peak height ratio analysis: In the preset position interval, the peak signal P with a peak height exceeding the threshold has a peak position of x and a peak height of h1. If there is a distance d, there are other peaks in the range [xd, x+d] with peak heights of h2, h3, ..., h n , assuming that P is the true target peak and the others are interference peaks, calculate the signal-to-noise ratio snr1=h1 / h2, snr2=h1 / h3, …, snr n-1 =h1 / h n .

[0047] Then, there exists snr_min=min{snr1,snr2,…snr n-1}≤n, n is the set threshold value, the value is 3-5 (specific adjustment is made according to the operating status of the instrument and reagents), then the peak P is an interference signal. Figure 6 Middle a shows the result of not enabling peak height ratio analysis, and the interfering signal is identified as the target (16 yellow). Figure 6 Middle b shows the result after turning on the peak height ratio analysis, identifying the interfering signal and canceling the mark of the target.

[0048] Peak area / peak height ratio analysis: For the peak signal peak P within the preset position interval and whose peak height exceeds the threshold, its peak height is h1 and its peak area is s1. If s1 / h1≥m, where m is the set threshold, the peak is an interference signal.

[0049] The m value is determined by the distribution characteristics of the s / h ratio of normal peaks and interference peaks in the data, such as Figure 7As shown, take the corresponding optimal value (6.1 here).

[0050] Figure 8 In a, the result without the peak area / peak height ratio analysis is shown, and the interference signal is recognized as the target target (68 in yellow). Figure 8 In b, the result after the peak area / peak height ratio analysis is shown, and the interference signal is recognized and the label of the target target is cancelled.

[0051] Mirror ratio analysis: For the peak signal peak P within the preset position interval and with the peak height exceeding the threshold, the point where the peak position is located is i, the left peak foot inflection point position is iL, and the right peak foot inflection point position is iR. d = min{i - iL, iR - i}, extract the left peak signal yL from i - d to i, extract the right peak signal yR from i to i + d, mirror-process the yR signal at the point i + d to get yR', calculate the correlation coefficient r based on yL and yR', and when r < rh, then this peak is an interference signal. Through data verification, the optimal value of rh is 95%.

[0052] Fig. 9 In a, the result without the mirror ratio analysis is shown, and the interference signal is recognized as the target target (82 in yellow). Fig. 9 In b, the result after the mirror ratio analysis is shown, and the interference signal is recognized and the label of the target target is cancelled.

[0053] On the basis of signal preprocessing, the peak signals initially marked as target targets are secondarily filtered through the above analysis method, thus avoiding the false positive problem caused by the single peak height threshold determination.

[0054] In summary, a signal processing and determination method based on capillary electrophoresis nucleic acid fragment analysis described in the present invention sets a specific algorithm for secondary recognition according to the characteristics such as the shape and height of the signal peak, which greatly improves the accuracy of target target determination, assists the experimenter in result determination, and significantly improves the work efficiency. Three different algorithms are provided for signal peak recognition, and the user can select to start one algorithm or multiple algorithms according to the actual situation, providing high flexibility.

[0055] And the linear compensation for the amplitude value of the filtered signal peak type interval is used to eliminate the signal loss in the filtering process, which can ensure the accuracy of the signal peak height and prevent the change of the true signal intensity caused by filtering.

[0056] It should be noted that all directional indications (such as up, down, left, right, front, back...) in the embodiments of the present invention are only used to explain the relative position relationship and movement conditions between components in a certain specific posture (as shown in the drawings). If this specific posture changes, then the directional indications will also change accordingly.

[0057] In addition, in the present invention, descriptions such as "first", "second", "one", etc. are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0058] In the present invention, unless otherwise clearly specified and limited, the terms "connection", "fixation", etc. should be understood in a broad sense. For example, "fixation" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements, unless otherwise clearly defined. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0059] In addition, the technical solutions between the various embodiments of the present invention can be combined with each other, but it must be based on the fact that ordinary technicians in the field can implement it. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

Claims

1. A signal processing and determination method based on capillary electrophoresis nucleic acid fragment analysis, characterized in that: Includes steps: S1: obtain the original waveform sampling signal after multiple polymerase chain reaction and capillary electrophoresis processing; S2: Preprocessing the original waveform sampling signal to obtain the peak feature point search after the peak shape is optimized; S3: Preliminary screening of signal peak target points based on comparison and judgment of peak feature points based on preset position intervals and peak height thresholds; S4: Perform proportional combination analysis based on the feature points obtained by the search, and mark the target points under secondary determination of the signal peak based on the analysis results; In the step S4, the ratio combination analysis includes peak height ratio analysis, peak area / peak height ratio analysis and mirror ratio analysis; The peak height ratio analysis is specifically as follows: Extract the peak height of the target mark signal peak, and compare it with the peak height of the wave within the preset distance range to obtain signal-to-noise ratio data. If the signal-to-noise ratio data contains a signal-to-noise ratio less than a first set threshold, the signal peak is determined to be an interference signal and the target mark is cancelled; The peak area / peak height ratio analysis is specifically as follows: Extracting the peak height and peak area of ​​the target mark signal peak, if the ratio of the peak area to the peak height is greater than a second set threshold, determining the signal peak as an interference signal and canceling the target mark; The mirror ratio analysis is specifically as follows: The peak width of the target marker signal peak is used as the extraction range, and the peak position of the target marker signal peak is used as the extraction starting point. The waveform signal of the peak within the extraction range around the extraction starting point is obtained, and the correlation coefficient is obtained after mirror processing. If the correlation coefficient is less than the third set threshold, the signal peak is determined to be an interference signal and the target target mark is cancelled.

2. A signal processing and determination method based on capillary electrophoresis nucleic acid fragment analysis as claimed in claim 1, characterized in that: In the step S2, the preprocessing includes baseline subtraction and peak height lossless filtering.

3. A signal processing and determination method based on capillary electrophoresis nucleic acid fragment analysis as claimed in claim 2, characterized in that: The baseline subtraction adopts a window sliding minimum value extraction method to remove abnormal interference baseline signals.

4. A signal processing and determination method based on capillary electrophoresis nucleic acid fragment analysis as claimed in claim 2, characterized in that: The peak height lossless filtering uses median filtering and polynomial smoothing to remove the cluttered peaks.

5. The signal processing and determination method based on capillary electrophoresis nucleic acid fragment analysis according to claim 1, characterized in that: In the step S2, the peak characteristic point is the inflection point of the first-order derivative of the sampling signal.

Images