Spectral peak identification method, equipment, medium and product based on iterative algorithm
By adopting an iterative algorithm-based method in mass spectroscopic peak recognition, combining signal-to-noise ratio and peak interval threshold parameters, the problem of inaccurate identification of traditional methods in noise environments is solved, and higher accuracy and better mass spectroscopic peak recognition are achieved. test results.
Patent Information
- Application Number
- CN202210488617.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-05-06
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2042-05-06
AI Technical Summary
Traditional mass spectrometry peak recognition methods are difficult to accurately identify mass spectrometry peaks in the presence of noise, resulting in inaccuracy of recognition results and the impact of instrument detection results.
The spectrum peak recognition method based on iterative algorithm is adopted to calculate the signal-to-noise ratio estimation value of each data point through the average iterative signal-to-noise ratio estimation algorithm. Combining the signal-to-noise ratio threshold and peak interval threshold parameters, iteratively find and adjust the spectrum peaks to improve the Accuracy of identification.
In the mass spectral data under the influence of noise, spectral peaks can be more accurately identified, the accuracy of noise estimation can be optimized, the accuracy of identification results can be improved, and the useless peaks can be effectively screened in the case of multiple interference peaks.
Smart Images

Figure CN115078519B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of spectrum peak recognition, and in particular to a spectrum peak recognition method, equipment, medium and product based on an iterative algorithm. Background Art
[0002] Matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF-MS) is used for the identification of microorganisms and the analysis of biomacromolecules including proteins and nucleic acids. It has the characteristics of high throughput, high sensitivity and high accuracy. The MALDI-TOF-MS method for detecting the new coronavirus is to amplify the target gene fragment through specific primers, then extend the target site probe specifically with a single base, and then detect the base at the extension site through mass spectrometry technology to determine the type of virus and the type of mutation. The application of this technology in many fields such as genetic disease screening, tumor mutation detection, methylation detection, medication guidance, pathogen detection and functional medicine health management is becoming increasingly in-depth, and it has become an indispensable molecular diagnostic technology for precision medicine.
[0003] The most important thing in the application of MALDI-TOF-MS technology is to identify the mass peaks of different types of ions, that is, to identify the mass peaks of different charge-to-mass ratios. The accuracy of peak identification is closely related to the correctness of the diagnosis result, so the accuracy of peak detection and identification is an important indicator affecting the detection capability of the instrument, and a high-accuracy mass peak identification method must be used.
[0004] The traditional peak identification method is to examine the changing trend of the data signal intensity within a certain range, and a peak value is considered to be a peak value if it meets the trend of first rising and then falling. The traditional method cannot accurately distinguish the mass spectrum peak with interference noise. If the spectrum has noise jumps that meet the set range requirements, the traditional peak identification will mistakenly think that the noise is a mass spectrum peak, which seriously affects the accuracy of identification and the detection results of the instrument. The spectrum of the mass spectrum peak is characterized by a small half-peak width and many peaks. The traditional peak identification algorithm does not set enough identification conditions for the mass spectrum, and is not suitable for the mass spectrum results of MALDI-TOF-MS. The number of identified peaks is large and most of them are peaks of no interest, which affects the accuracy of the identification results. Summary of the invention
[0005] In order to achieve the above-mentioned purpose and other advantages according to the present invention, a first object of the present invention is to provide a spectrum peak identification method based on an iterative algorithm, comprising the following steps:
[0006] S1, input the original spectrum data, and calculate the signal-to-noise ratio estimation value of each data point based on the mean iteration signal-to-noise ratio estimation algorithm;
[0007] S2, traverse all data points in the spectrum, find data points that meet the signal-to-noise ratio threshold setting and have a trend of first rising and then falling, and mark them as candidate spectrum peak points;
[0008] S3, judging whether the spectrum peak candidate point satisfies the peak interval threshold parameter;
[0009] S4, if yes, retain the candidate peak point and output it;
[0010] S5, otherwise remove the spectrum peak candidate point;
[0011] S6. Traverse all candidate spectral peak points, repeat steps S3 to S5, and obtain spectral peaks that meet the conditions in the original spectrum data.
[0012] Further, the SI step includes the following steps:
[0013] S11, dividing the original spectrum data into a plurality of data groups with the same number of points, each group having N data points, and performing calculations on each group of data;
[0014] S12. Calculate the expected value E(X) and standard deviation STDEV(X) of the data set, where X represents the vector constituting the data points, the maximum value max of the intensity of the data set, the threshold η of the screening data set, and the threshold formula of the screening data is:
[0015] S13, set the number of histogram bins of the original spectrogram data to N bin , the histogram is divided into N bin Segment, number of data points per segment IN size =η / N count , parameter N count The parameters are set, and all data points larger than η in the original spectrum data are discarded, and data points smaller than η are retained for histogram statistics, and IN size Divide the data into segments according to length and calculate the average intensity of each segment;
[0016] S14, performing histogram statistics on the data of each partition, calculating the mean and variance of the histogram of the data segment, the intensity corresponding to the histogram mean n is the noise estimation value, and the formula of the noise estimation value is n=mean(hist);
[0017] S15, repeating steps S13 and S14, iteratively calculating the histogram several times and determining a final noise estimation value, the signal-to-noise ratio estimation value of the center point of the window is the intensity of the point divided by the noise estimation value.
[0018] Furthermore, the step S2 includes the following steps:
[0019] Traverse all data points in the spectrum, and look for points that satisfy the signal-to-noise ratio estimate value greater than the set signal-to-noise ratio threshold. At the same time, the intensity of this point is greater than the intensity of several points on its left and several points on its right, and the signal-to-noise ratio estimates of the i-th point on its left and the i-th point on its right are greater than the threshold; in order from far to near, the intensity of several points on the left increases successively, and the intensity of several points on the right also increases successively. If the above conditions are met, it is considered to be a spectrum peak point, marked and recorded as a spectrum peak candidate point.
[0020] Furthermore, the threshold value η of the screening data is greater than or equal to 0.
[0021] Furthermore, in the step S2, all data points in the spectrum are traversed to sequentially search for points satisfying the condition that the signal-to-noise ratio estimate is greater than the set signal-to-noise ratio threshold, and at the same time, the intensity of the point is greater than the intensity of the first point on its left, the second point on its left, the first point on its right, and the second point on its right, the signal-to-noise ratio estimate of the first point on its left and the first point on its right is greater than the threshold, the intensity of the second point on the left is less than the first point on the left, the intensity of the second point on the right is less than the first point on the right, and the signal-to-noise ratio estimate of the second point on the left and the second point on the right is greater than the threshold. If the above conditions are met, it is considered to be a spectrum peak point and is marked and recorded as a spectrum peak candidate point.
[0022] Furthermore, the S3 step includes the following steps:
[0023] All spectral peak candidate points are sorted from small to large according to the intensity. The distance between the spectral peak candidate point and the first point on its right is less than the peak interval threshold multiplied by the minimum peak interval, the intensity of the first point on the right is greater than the spectral peak candidate point, the distance between the spectral peak candidate point and the first point on its left is less than the peak interval threshold multiplied by the minimum peak interval, and the intensity of the first point on the left is less than the candidate point. The candidate peak points that meet the above conditions are retained as results, and the candidate peak points that do not meet any point are removed; the minimum peak interval is the minimum value of the interval between adjacent peaks within the spectral peak candidate point, and the peak interval threshold is a multiple of the difference between peaks allowed to be retained when setting the screening peaks. The higher the peak interval threshold is set, the farther the allowed peak interval is, and the more peaks are retained.
[0024] Furthermore, the step S6 includes the following steps:
[0025] According to step S3, all spectral peak candidate points are screened, and the sorted spectral peak candidate points are used as seed centers. Iteration is performed in the sorting order. When all candidate points are traversed, the peak screening is terminated, and the mass spectrum peaks that meet the conditions in the original data are obtained according to each candidate point that meets the screening conditions.
[0026] The second object of the present invention is to provide an electronic device, comprising: a memory on which a program code is stored; a processor, which is connected to the memory and implements a spectrum peak identification method based on an iterative algorithm when the program code is executed by the processor.
[0027] A third object of the present invention is to provide a computer-readable storage medium having program instructions stored thereon, which implement a spectrum peak identification method based on an iterative algorithm when the program instructions are executed.
[0028] A fourth object of the present invention is to provide a computer program product, comprising a computer program / instruction, which, when executed by a processor, implements a spectrum peak identification method based on an iterative algorithm.
[0029] Compared with the prior art, the present invention has the following beneficial effects:
[0030] The present invention provides a spectral peak recognition method based on an iterative algorithm. The iterative method is used to search, recalibrate and adjust the identified spectral peaks. The effect of refining the peak center can be achieved when the peak boundaries overlap. The peak interval threshold parameter is used to screen out useless peaks. This method can better identify the spectral peaks of the mass spectrum.
[0031] The present invention utilizes the algorithm idea of iterative calculation and adopts the signal-to-noise ratio estimation method of mean iteration. In mass spectrometry data where noise cannot be completely removed, peak recognition can be performed based on the estimated noise, and the signal-to-noise ratio estimation value is used as a threshold to identify the spectrum peak in combination with the data intensity. The accuracy of noise estimation can be optimized on mass spectrometry data with noise, and a more accurate spectrum peak recognition result can be obtained.
[0032] Common spectral peak recognition methods do not screen the spectral peaks of interest after recognition, which is not suitable for the case of multiple interfering peaks. It will cause the peaks of interest to be submerged in a large number of peaks of no interest, and it is impossible to achieve a good recognition result. The spectral peak recognition method based on an iterative algorithm proposed in the present invention can set the parameter conditions of spectral peak recognition during the recognition process, delete useless peaks during the recognition process, and thus obtain a recognition result that meets the requirements better.
[0033] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention and implement it according to the contents of the specification, the following is a detailed description of the preferred embodiments of the present invention in conjunction with the accompanying drawings. The specific implementation of the present invention is given in detail by the following embodiments and their accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0034] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:
[0035] Figure 1 The process of the spectrum peak identification method based on the iterative algorithm of Example 1 is as follows Figure 1 ;
[0036] Figure 2The process of the spectrum peak identification method based on the iterative algorithm of Example 1 is as follows Figure 2 ;
[0037] Figure 3 This is a schematic diagram of an electronic device according to Embodiment 2. DETAILED DESCRIPTION
[0038] The present invention is further described below in conjunction with the accompanying drawings and specific implementation methods. It should be noted that, under the premise of no conflict, the various embodiments or technical features described below can be arbitrarily combined to form a new embodiment.
[0039] Example 1
[0040] Spectral peak identification methods based on iterative algorithms, such as Figure 1 , Figure 2 As shown, the following steps are included:
[0041] S1. Inputting the original spectrum data, and calculating the signal-to-noise ratio estimation value of each data point based on the mean iteration signal-to-noise ratio estimation algorithm; specifically, comprising the following steps:
[0042] S11, dividing the original spectrum data into a plurality of data groups with the same number of points, each group having N data points, and performing calculations on each group of data;
[0043] S12. Calculate the expected value E(X) and standard deviation STDEV(X) of the data set, where X represents the vector constituting the data points, the maximum value max of the intensity of the data set, the threshold η of the screening data set, and the threshold formula of the screening data is: Generally, the threshold η is greater than or equal to 0.
[0044] S13, set the number of histogram bins of the original spectrogram data to N bin , the histogram is divided into N bin Segment, number of data points per segment IN size =η / N count , parameter N count The parameters are set, and all data points larger than η in the original spectrum data are discarded, and data points smaller than η are retained for histogram statistics, and IN size Divide the data into segments according to length and calculate the average intensity of each segment;
[0045] S14, performing histogram statistics (hist) on the data of each partition, calculating the mean and variance of the histogram of the data segment, the intensity corresponding to the histogram mean n is the noise estimation value, and the formula of the noise estimation value is n=mean(hist);
[0046] S15, repeating steps S13 and S14, iteratively calculating the histogram several times, for example three times, and determining a final noise estimation value, wherein the signal-to-noise ratio estimation value of the center point of the window is the intensity of the point divided by the noise estimation value.
[0047] The present invention uses a signal-to-noise ratio estimation method based on mean iteration. There are multiple noise influences in the operation of the MALDI-TOF-MS instrument. Noise still exists in the data after preprocessing or filtering. The noise is distributed in the entire range of the spectrum, affecting the accuracy of peak identification. The signal-to-noise ratio estimation method based on mean iteration can accurately estimate the noise in the spectrum and improve the accuracy of the spectrum peak identification algorithm.
[0048] S2, traverse all data points in the spectrum, find data points that meet the signal-to-noise ratio threshold setting and have a trend of first rising and then falling, and mark them as candidate peak points. Specifically, the following steps are included:
[0049] Traverse all data points in the spectrum, and look for points that satisfy the signal-to-noise ratio estimate value greater than the set signal-to-noise ratio threshold. At the same time, the intensity of this point is greater than the intensity of several points on its left and several points on its right, that is, it satisfies the peak graph trend of first rising and then falling; the signal-to-noise ratio estimate value of the i-th point on its left and the i-th point on its right is greater than the threshold; in the order from far to near, the intensity of several points on the left increases successively, and the intensity of several points on the right also increases successively. If the above conditions are met, it is considered to be a spectrum peak point, marked and recorded as a spectrum peak candidate point.
[0050] In this embodiment, the intensity of the point is compared with the intensities of the two points on its left and the two points on its right. Specifically, all data points in the spectrum are traversed, and points are sequentially searched for which the signal-to-noise ratio estimation value is greater than the set signal-to-noise ratio threshold, and at the same time, the intensity of the point is greater than the intensity of the first point on its left, the second point on its left, the first point on its right, and the second point on its right, and the signal-to-noise ratio estimation values of the first point on its left and the first point on its right are greater than the threshold, the intensity of the second point on the left is less than the first point on the left, the intensity of the second point on the right is less than the first point on the right, and the signal-to-noise ratio estimation values of the second point on the left and the second point on the right are greater than the threshold. If the above conditions are met, it is considered to be a spectrum peak point, marked and recorded as a spectrum peak candidate point.
[0051] S3, determine whether the spectrum peak candidate point meets the peak interval threshold parameter; specifically, sort all the spectrum peak candidate points from small to large according to the intensity, the distance between the spectrum peak candidate point and the first point on its right is less than the peak interval threshold multiplied by the minimum peak spacing, the intensity of the first point on the right is greater than the spectrum peak candidate point, the distance between the spectrum peak candidate point and the first point on its left is less than the peak interval threshold multiplied by the minimum peak spacing, and the intensity of the first point on the left is less than the candidate point, the candidate peak points that meet the above conditions are retained as the result, that is, S4, if yes, the spectrum peak candidate points are retained and output; the candidate peak points that do not meet any point are removed, that is, S5, otherwise the spectrum peak candidate points are removed.
[0052] Among them, the minimum peak spacing is the minimum value of the interval between adjacent peaks within the spectrum peak candidate point, and the peak spacing threshold is the multiple of the difference between the peaks allowed to be retained when setting the screening peaks. When the peak spacing threshold is set higher, the allowed peak spacing is farther, and the more peaks are retained.
[0053] S6, traverse all the peak candidate points, repeat steps S3 to S5, and obtain the peaks that meet the conditions in the original spectrum data. Specifically, all the peak candidate points are screened according to step S3, and the sorted peak candidate points are used as seed centers, and iterate in the sorting order. When all candidate points have been traversed, the peak screening ends, and the mass spectrum peaks that meet the conditions in the original data are obtained according to each candidate point that meets the screening conditions.
[0054] The present invention proposes an iterative spectral peak recognition method based on spectral peak parameters, which is used for spectral peak recognition of multiple interference peaks. The recognition results of interest are obtained through the threshold set by the parameters, and the interference spectral peaks that do not meet the parameters are deleted. The accuracy of the peak recognition algorithm is improved by using iterative calculation and parameter setting.
[0055] Example 2
[0056] An electronic device 200, such as Figure 3 As shown, it includes but is not limited to: a memory 201 on which program codes are stored; a processor 202, which is connected to the memory, and when the program codes are executed by the processor, a spectrum peak recognition method based on an iterative algorithm is implemented. For a detailed description of the method, reference can be made to the corresponding description in the above method embodiment, which will not be repeated here.
[0057] Example 3
[0058] A computer-readable storage medium stores program instructions, and when the program instructions are executed, a spectrum peak identification method based on an iterative algorithm is implemented. For a detailed description of the method, reference may be made to the corresponding description in the above method embodiment, and no further description is given here.
[0059] Example 4
[0060] A computer program product includes a computer program / instruction, which implements a spectrum peak identification method based on an iterative algorithm when executed by a processor. For a detailed description of the method, reference may be made to the corresponding description in the above method embodiment, which will not be repeated here.
[0061] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.
[0062] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.
[0063] The above are only embodiments of this specification and are not intended to limit one or more embodiments of this specification. For those skilled in the art, one or more embodiments of this specification may have various changes and transformations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of one or more embodiments of this specification shall be included in the scope of the claims of one or more embodiments of this specification. One or more embodiments of this specification One or more embodiments of this specification One or more embodiments of this specification One or more embodiments of this specification.
Claims
1. A spectrum peak identification method based on an iterative algorithm, characterized in that: The following steps are involved: S1, input the original spectrum data, and calculate the signal-to-noise ratio estimation value of each data point based on the mean iteration signal-to-noise ratio estimation algorithm; S2, traverse all data points in the spectrum, find data points that meet the signal-to-noise ratio threshold setting and have a trend of first rising and then falling, and mark them as candidate spectrum peak points; S3, judging whether the spectrum peak candidate point satisfies the peak interval threshold parameter; S4, if yes, retain the candidate peak point and output it; S5, otherwise remove the spectrum peak candidate point; S6, traverse all the candidate spectral peak points, repeat steps S3 to S5, and obtain the spectral peaks that meet the conditions in the original spectrum data; The S1 step includes the following steps: S11, dividing the original spectrum data into a plurality of data groups with the same number of points, each group having data points, and calculate for each set of data; S12. Calculate the expected value of this group of data and standard deviation ,in, Represents the vector that constitutes the data point, the maximum intensity of this group of data , the threshold for filtering data in this group , the threshold formula for filtering data is ; S13, set the number of histogram bins of the original spectrogram data to , the histogram is divided into Segments, number of data points per segment ,parameter As the parameter to be set, the original spectrum data is larger than All data less than The data points are counted as histograms, and the Divide the data into segments according to length and calculate the average intensity of each segment; S14. Perform histogram statistics on the data of each partition, and calculate the mean and variance of the histogram of the data segment. The intensity corresponding to the histogram mean n is the noise estimation value. The formula for the noise estimation value is: ; S15, repeating steps S13 and S14, iteratively calculating the histogram several times and determining a final noise estimation value, the signal-to-noise ratio estimation value of the center point of the window is the intensity of the point divided by the noise estimation value; The S2 step includes the following steps: Traverse all data points in the spectrum, and search for points that satisfy the signal-to-noise ratio estimate value greater than the set signal-to-noise ratio threshold. At the same time, the intensity of the point is greater than the intensity of several points on its left and several points on its right, and the signal-to-noise ratio estimate value of the i-th point on its left and the i-th point on its right is greater than the threshold. In the order from far to near, the intensity of several points on the left increases successively, and the intensity of several points on the right also increases successively. If the above conditions are met, it is considered to be a spectrum peak point, marked and recorded as a spectrum peak candidate point; The S3 step includes the following steps: All spectrum peak candidate points are sorted from small to large according to the intensity. The distance between the spectrum peak candidate point and the first point on its right is less than the peak interval threshold multiplied by the minimum peak interval, the intensity of the first point on the right is greater than the spectrum peak candidate point, the distance between the spectrum peak candidate point and the first point on its left is less than the peak interval threshold multiplied by the minimum peak interval, and the intensity of the first point on the left is less than the candidate point. The candidate peak points that meet the above conditions are retained as the result, and the candidate peak points that do not meet any point are removed; the minimum peak interval is the minimum value of the interval between adjacent peaks in the spectrum peak candidate point, and the peak interval threshold is a multiple of the difference between peaks allowed to be retained when setting the screening peaks. When the peak interval threshold is set higher, the allowed peak interval is farther, and the more peaks are retained; The step S6 comprises the following steps: According to step S3, all spectral peak candidate points are screened, and the sorted spectral peak candidate points are used as seed centers. Iteration is performed in the sorting order. When all candidate points are traversed, the peak screening is terminated, and the mass spectrum peaks that meet the conditions in the original data are obtained according to each candidate point that meets the screening conditions.
2. The spectrum peak identification method based on iterative algorithm according to claim 1, characterized in that: The threshold value of the screening data Greater than or equal to 0.
3. The spectrum peak identification method based on iterative algorithm according to claim 1, characterized in that : In the S2 step, all data points in the spectrum are traversed to find points in turn whose signal-to-noise ratio estimation value is greater than the set signal-to-noise ratio threshold value, and at the same time, the intensity of the point is greater than the intensity of the first point on the left, the second point on the left, the first point on the right and the second point on the right, the signal-to-noise ratio estimation value of the first point on the left and the first point on the right is greater than the threshold, the intensity of the second point on the left is less than the first point on the left, the intensity of the second point on the right is less than the first point on the right, and the signal-to-noise ratio estimation value of the second point on the left and the second point on the right is greater than the threshold. If the above conditions are met, it is considered to be a spectrum peak point, marked and recorded as a spectrum peak candidate point.
4. An electronic device, characterized in that: include: A memory having program code stored therein; A processor is coupled to the memory and implements the method according to any one of claims 1 to 3 when the program code is executed by the processor.
5. A computer-readable storage medium, characterized in that: Program instructions are stored thereon, and when the program instructions are executed, the method according to any one of claims 1 to 3 is implemented.
6. A computer program product comprising computer instructions, characterized in that: When the computer instructions are executed by a processor, the method according to any one of claims 1 to 3 is implemented.
Citation Information
Patent Citations
Ion detection
CN103262205A
Post-processing method of Raman spectrum weak signal extraction
CN111089856A
