Rapid separation method for isotope aliasing peak of small ion trap mass spectrometer

Abstract

The invention discloses a rapid separation method for isotope aliasing peaks of a small ion trap mass spectrometer. The rapid separation method comprises the following steps: determining Gaussian parameters of a mass spectrum signal to be fitted by using progressive peak finding; setting dual judgment conditions according to the characteristics of the mass spectrum signal, and performing false peak removal; and optimizing the Gaussian parameters by using an LM algorithm with a weak peak protection mechanism to obtain spectrum peak information in the spectrogram. According to the method, high-precision separation of aliasing isotope peaks in a complex sample can still be kept under the condition of high-speed scanning, and the adaptability is high.

Description

Technical Field

[0001] The present invention belongs to the technical field of mass spectrometry analysis, and relates to a method for rapidly separating isotope overlapping peaks in a small ion trap mass spectrometer. Background Art

[0002] A mass spectrometer is an instrument used to analyze the structure of compounds and determine their components. Its basic principle is to ionize the molecular ions in a compound, separate them according to their mass-to-charge ratio (m / z), and measure their relative abundances. A mass spectrometer generally includes parts such as sample processing, ionization, mass analysis, and data processing. Due to its high sensitivity, good selectivity, high precision, and high resolution ability, the mass spectrometer has become the gold standard for chemical detection and identification and has been applied in various fields, such as environmental detection, biology, and medical detection.

[0003] Compared with other mass spectrometers, the linear ion trap has higher sensitivity and resolution, and can detect low-concentration compounds. Therefore, the linear ion trap is often used in small mass spectrometers. Small mass spectrometers have the characteristics of small size, light weight, and strong portability, and are suitable for on-site or field real-time analysis; however, the performance of small mass spectrometers is limited by power and volume, and their resolution, precision, detection limit, etc. are usually not as good as those of large mass spectrometers.

[0004] The realization of isotope separation in small mass spectrometers usually depends on high-resolution conditions, which often requires reducing the scanning speed to meet the resolution requirements. Mass spectrometry analysis requires rapid analysis and an increased scanning speed. However, increasing the scanning speed will lead to a decrease in the resolution of the mass spectrometry signal, resulting in mass spectrometry peak overlapping, making it difficult to effectively separate isotopes.

[0005] Therefore, there is an urgent need for a method for rapidly separating isotope overlapping peaks in a small ion trap mass spectrometer to effectively separate isotopes. Summary of the Invention

[0006] To solve the above technical problems, the purpose of the present invention is to provide a method for rapidly separating isotope overlapping peaks in a small ion trap mass spectrometer, which can solve the problem of peak overlapping in small mass spectrometers under rapid scanning conditions.

[0007] A method for rapidly separating isotope overlapping peaks in a small ion trap mass spectrometer according to the present invention includes:

[0008] Using progressive peak finding to determine the Gaussian parameters of the mass spectrometry signal to be fitted;

[0009] Setting double judgment conditions according to the characteristics of the mass spectrometry signal and performing pseudo-peak removal;

[0010] Using the LM algorithm with a weak peak protection mechanism to optimize the Gaussian parameters, obtaining the peak information in the spectrogram, and further separating the isotopes in the overlapping peaks.

[0011] Further, the mass spectrometry signals to be fitted are expressed by the following Gaussian mixture model:

[0012]

[0013] where f i (x) are the individual Gaussian components that make up the spectrogram; n is the total number of Gaussian components; A i , μ i and σ i represent the amplitude, expectation value, and standard deviation of the i-th Gaussian component respectively, which are Gaussian fitting parameters; Noise represents Gaussian background noise.

[0014] Further, the process of progressive peak finding is as follows:

[0015] Search for local maximum points of the amplitude from the left to the right of the original mass spectrogram, where the amplitude of the local maximum point needs to be greater than the set amplitude threshold, and assign the amplitude and position information of the local maximum points that meet the conditions to A i and μ i respectively;

[0016] Search to the left from this local maximum point until the position where the amplitude is A i / 2 is found, denoted as tg, and according to a set of Gaussian fitting parameters A i , μ i , σ i and the corresponding Gaussian component f i (x) are obtained;

[0017] Subtract the estimated Gaussian component from the original spectrogram, and continue to search to the right from the position μ i of the local maximum point, and repeat the above steps until no local maximum point that meets the conditions can be found, and multiple sets of Gaussian fitting parameters and corresponding Gaussian components are obtained.

[0018] Further, the double determination conditions are as follows:

[0019] Condition 1: If the distance between two estimated Gaussian components is less than the set threshold, it is considered that one of them is a false peak, and the Gaussian component with the larger amplitude is retained; the set threshold is the minimum value of the peak spacing of the mass spectrometry signals;

[0020] Condition 2: If the standard deviation of the estimated Gaussian component is greater than the sampling time corresponding to 1.2Th or less than the sampling time corresponding to 0.1Th, which does not conform to the actual situation, it is considered a false peak and removed.

[0021] Further, the optimization of the Gaussian fitting parameters using the LM algorithm with a weak peak protection mechanism is specifically as follows:

[0022] Divide the spectrogram into individual cluster peaks and process each cluster peak as follows:

[0023] Subtract the identified weak peak Gaussian components from each cluster peak and optimize the remaining non-weak peak Gaussian components using the LM algorithm;

[0024] After completing the optimization of the non-weak peak Gaussian parameters, subtract the fitted non-weak peak Gaussian components from the cluster peak and optimize and fit the estimated weak peak Gaussian components using the LM algorithm to obtain the spectral peak information in the spectrogram.

[0025] Further, when the distance between the central positions of two Gaussian components is less than 2Th equivalent sampling time lengths, these two Gaussian components will be classified into the same cluster peak.

[0026] Further, the definition of the weak peak is as follows: A i <0.05A MAX where A MAX represents the amplitude of the Gaussian component with the largest amplitude in the cluster peak.

[0027] Further, the spectral peak information is the Gaussian parameters in the Gaussian component, and isotopes in the overlapping peaks can be separated based on the spectral peak information.

[0028] A method for rapid separation of isotope overlapping peaks in a small ion trap mass spectrometer according to the present invention has the following beneficial effects:

[0029] (1) This method uses a progressive peak finding algorithm. Through gradual optimization and iterative search, it deeply explores the possible Gaussian components in the spectrogram. Compared with traditional methods, this algorithm can more accurately identify isotope peaks in overlapping peaks, especially performing well in complex overlapping signals. In addition, this algorithm supports automated processing without user intervention, significantly improving the analysis efficiency and reliability.

[0030] (2) This method fully considers the characteristics of the mass spectrometry signal and introduces a pseudo-peak removal strategy. By setting reasonable judgment conditions, it can effectively identify and remove pseudo-peaks, preventing them from causing incorrect iterations in the subsequent optimization process. Compared with existing technologies, the pseudo-peak removal strategy significantly improves the accuracy and stability of the fitting results.

[0031] (3) This method introduces a weak peak protection mechanism. This mechanism can prevent strong peak signals from masking weak peak signals during the optimization process, ensuring the complete retention of weak signals. Compared with traditional Gaussian fitting methods, the weak peak protection mechanism further guarantees the integrity of the data, making the analysis results more comprehensive and reliable.

[0032] (4) The present invention is specifically designed for small ion trap mass spectrometers and can achieve efficient isotope separation in on-site rapid analysis. Its progressive peak finding algorithm and automated processing capabilities significantly improve the analysis efficiency. For complex overlapping signals, the present invention can accurately separate overlapping isotope peaks, solving the limitations of traditional methods in dealing with complex signals. The present invention does not require modification of the instrument hardware, reducing costs and improving practicality. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 is a flowchart of a method for rapid separation of isotope overlapping peaks in a small ion trap mass spectrometer according to the present invention;

[0034] Figure 2 shows the algorithm separation effects of three typical compounds in Example of the present invention at different scanning rates. From top to bottom are clozapine, caffeine, and imatinib: (a) algorithm separation effect at a scanning speed of 500 Th / s; (b) algorithm separation effect at a scanning speed of 2000 Th / s; (c) algorithm separation effect at a scanning speed of 5000 Th / s;

[0035] Figure 3 shows the Gaussian separation effect of the full spectrum by the algorithm for complex samples in Example of the present invention at different scanning speeds: (a) Gaussian separation effect at a scanning speed of 500 Th / s; (b) Gaussian separation effect at a scanning speed of 2000 Th / s; (c) Gaussian separation effect diagram at a scanning speed of 5000 Th / s. DETAILED DESCRIPTION OF THE INVENTION

[0036] Gaussian decomposition technology is a signal processing technology. Its core lies in decomposing complex signals into a linear superposition of multiple Gaussian functions to better analyze signal characteristics. This technology has been widely used in fields such as radar echo decomposition. Its application in mass spectrometry analysis has not been fully explored. Mass spectrometry signals can be approximated by a series of Gaussian functions in the time domain, which provides a theoretical basis for the application of Gaussian decomposition technology in mass spectrometry signal processing.

[0037] Based on the above theoretical basis, the present invention provides a method for rapid separation of isotope overlapping peaks in a small ion trap mass spectrometer, including:

[0038] Step 1: Use progressive peak finding to determine the Gaussian parameters of the mass spectrometry signal to be fitted.

[0039] The mass spectrometry signal to be fitted is expressed by the following Gaussian mixture model:

[0040]

[0041] where f i(x) are the individual Gaussian components that make up the spectral graph; n is the total number of Gaussian components; A i , μ i and σ i respectively represent the amplitude, expected value, and standard deviation of the i-th Gaussian component, which are Gaussian fitting parameters; Noise represents the Gaussian background noise.

[0042] The process of progressive peak finding is as follows:

[0043] Search for local maximum points of the amplitude from the left to the right of the original mass spectrum. The amplitude of the local maximum points needs to be greater than the set amplitude threshold. Assign the amplitude and position information of the local maximum points that meet the conditions to A i and μ i respectively. The purpose of setting the amplitude threshold is to prevent background noise from being recognized as an implicit Gaussian component, and its specific value should refer to the noise amplitude in the spectral graph.

[0044] Search to the left from this local maximum point until the position where the amplitude is A i / 2 is found, denoted as tg. According to a set of Gaussian fitting parameters A i , μ i , σ i and the corresponding Gaussian component f i (x) are obtained.

[0045] Subtract the estimated Gaussian component from the original spectrum. Starting from the position μ of the local maximum point i continue to search to the right and repeat the above steps until no local maximum points that meet the conditions can be found, obtaining multiple sets of Gaussian fitting parameters and the corresponding Gaussian components.

[0046] Step 2: Set double determination conditions according to the mass spectrometry signal characteristics and perform pseudo-peak removal. The purpose of pseudo-peak removal is to avoid incorrect iteration of the subsequent Gaussian parameter optimization caused by pseudo-peaks generated during the Gaussian fitting process.

[0047] The double determination conditions are as follows:

[0048] Condition 1: If the distance between two estimated Gaussian components is less than the set threshold, then one of them is considered a pseudo-peak, and the Gaussian component with the larger amplitude is retained; the set threshold is the minimum value of the peak spacing of the mass spectrometry signal.

[0049] The mass spectrometry signal characteristics are related to the application scenario. Taking single-charge isotopes as an example, the peak spacing between two peaks will not be less than 0.8 Th. Combining the characteristics of the method of the present invention for separating single-charge isotope overlapping peaks, the set threshold is 0.8 Th equivalent sampling time length.

[0050] Condition 2: If the standard deviation of the estimated Gaussian component is greater than the sampling time corresponding to 1.2Th or less than the sampling time corresponding to 0.1Th, which does not conform to the actual situation, it is considered a spurious peak and removed.

[0051] Step 3: Use the LM algorithm with a weak peak protection mechanism to optimize the Gaussian parameters to obtain the spectral peak information in the spectrogram. The purpose of the weak peak protection mechanism is to avoid strong peak signals masking weak peak signals during the Gaussian optimization process, thereby improving the overall fitting accuracy. Specifically:

[0052] First, divide the spectrogram into individual cluster peaks, and then process each cluster peak as follows:

[0053] Subtract the identified weak peak Gaussian components from each cluster peak, and optimize the remaining non-weak peak Gaussian components using the LM algorithm.

[0054] After optimizing the Gaussian parameters of the non-weak peaks, subtract the fitted non-weak peak Gaussian components from the cluster peak, and optimize and fit the estimated weak peak Gaussian components using the LM algorithm to obtain the spectral peak information in the spectrogram.

[0055] In specific implementation, when the distance between the center positions of two Gaussian components is less than the equivalent sampling time length of 2Th, these two Gaussian components will be classified into the same cluster peak.

[0056] In specific implementation, the definition of a weak peak in the present invention is as follows: A i <0.05A MAX , where A MAX represents the amplitude of the Gaussian component with the largest amplitude in the cluster peak.

[0057] The spectral peak information obtained in Step 3 is the Gaussian parameters in the Gaussian component. According to the spectral peak information, isotopes in the overlapping peaks can be separated.

[0058] The following further describes specific application embodiments of the present invention.

[0059] Embodiment 1:

[0060] The purpose of this experimental example is to verify the actual application performance of the invented algorithm. In this embodiment, three typical compounds are selected for testing: 10 ug / mL clozapine (m / z 327), 10 ug / mL caffeine (m / z 194), and 1 ug / mL imatinib (m / z 494). The data is based on a small ion trap mass spectrometer, and mass spectrometry signals are collected at different scanning rates. All data has been processed by Gaussian smoothing denoising and amplitude normalization to eliminate intensity differences. Figure 2 The black curve in the figure shows the original mass spectrometry signals collected under different conditions, and the orange curve shows the Gaussian components separated by the algorithm.

[0061] Figure 2 (a) shows the spectra of each compound under a low-speed scan of 500 Th / s. The system exhibits excellent mass resolution and can clearly resolve the isotope peaks of each compound.

[0062] Figure 2 (b) shows the spectra of each compound under a medium-speed scan of 2000 Th / s. Aliasing phenomena start to appear in the spectra, and the algorithm of the present invention can accurately separate the isotope peaks.

[0063] Figure 2 (c) shows the spectra of each compound under a high-speed scan of 5000 Th / s. Due to the reduced resolution, the full width at half maximum is significantly broadened, the spectral peaks completely overlap, and the isotopes cannot be directly separated. However, the algorithm of the present invention can still accurately separate the isotopes.

[0064] Example 2:

[0065] The purpose of this experimental example is to verify the applicability of the algorithm of the invention to the full-spectrum analysis of mixed samples at different scan speeds, and to focus on evaluating the separation effect of Gaussian components of mixed samples at low, medium, and high scan speeds. The sample contains 10 ug / mL tetrabutylammonium bromide (m / z 242), 10 ug / mL amitriptyline (m / z 278), 10 ug / mL imatinib (m / z 494), and 30 ug / mL reserpine (m / z 609). The experimental results are as Figure 3 shown. The figure includes the full-spectrum diagram and its partial enlarged view. The blue solid line in the figure represents the original mass spectrometry signal, the black dashed line represents the Gaussian components separated by the algorithm, the red solid line represents the fitting result, the red numbers mark the peak center positions identified by the algorithm, and the purple cross marks the peak center positions identified by the professional peak-finding algorithm. The specific analysis is as follows:

[0066] Figure 3 (a) shows the full-spectrum diagram of the mixture at a scan speed of 500 Th / s, and all isotopes can be directly separated

[0067] Figure 3 (b) shows the full-spectrum diagram of the mixture at a scan speed of 2000 Th / s. The isotope peaks of amitriptyline (m / z 278) and imatinib (m / z 494) can still be resolved by the professional peak-finding function. The isotope peaks of tetrabutylammonium bromide (m / z 242) and reserpine (m / z 609) are difficult to directly separate using the peak-finding function. However, the algorithm of the present invention can achieve accurate separation

[0068] Figure 3 (c) shows the full-spectrum diagram of the mixture at a scan speed of 5000 Th / s. All isotopes cannot be resolved by the professional peak-finding function, and the algorithm of the present invention can still achieve accurate separation.

[0069] As can be seen from this embodiment, the isotope separation method proposed by the present invention can effectively separate Gaussian components under low-speed, medium-speed, and high-speed scanning conditions, and particularly shows significant advantages under high-speed scanning conditions, providing reliable technical support for rapid mass spectrometry analysis.

[0070] The foregoing are only preferred embodiments of the present invention and are not intended to limit the idea of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A rapid separation method for isotope overlapping peaks of a small ion trap mass spectrometer, characterized in that, Including: Determining the Gaussian parameters of the mass spectrometry signal to be fitted by progressive peak finding; Setting double determination conditions according to the characteristics of the mass spectrometry signal and performing pseudo-peak removal; Optimizing the Gaussian parameters using the LM algorithm with a weak peak protection mechanism to obtain the peak information in the spectrum, and further separating the isotopes in the overlapping peaks.

2. The rapid separation method for isotope overlapping peaks of a small ion trap mass spectrometer according to claim 1, characterized in that The mass spectrometry signal to be fitted is expressed by the following Gaussian mixture model: Among them, f i (x) is each Gaussian component that constitutes the spectrogram; n is the total number of Gaussian components; A i , μ i and σ i respectively represent the amplitude, expected value, and standard deviation of the i-th Gaussian component, which are Gaussian fitting parameters; Noise represents Gaussian background noise.

3. The rapid separation method for isotope overlapping peaks of a small ion trap mass spectrometer according to claim 1, wherein The process of the progressive peak finding is as follows: Search for local maximum points of amplitude from the left side to the right side of the original mass spectrum, where the amplitude of the local maximum points needs to be greater than the set amplitude threshold, and assign the amplitudes and position information of the local maximum points that meet the conditions to A i and μ i ; Search to the left from this local maximum point until a position with an amplitude of A i / 2 is found and denoted as tg. According to A set of Gaussian fitting parameters A i , μ i , σ i and the corresponding Gaussian component f i (x) are obtained; Subtract the estimated Gaussian components from the original spectrum, starting from the position μ of the local maximum points i Continue to search to the right and repeat the above steps until no local maximum points meeting the conditions can be found, obtaining multiple sets of Gaussian fitting parameters and the corresponding Gaussian components.

4. The rapid separation method for isotope overlapping peaks of a small ion trap mass spectrometer according to claim 1, characterized in that The double determination conditions are as follows: Condition 1: If the distance between two estimated Gaussian components is less than the set threshold, then one of them is considered a pseudo-peak, and the Gaussian component with a larger amplitude is retained; the set threshold is the minimum value of the peak spacing of the two peaks of the mass spectrometry signal; Condition 2: If the standard deviation of the estimated Gaussian component is greater than the sampling time corresponding to 1.2Th or less than the sampling time corresponding to 0.1Th, which does not conform to the actual situation, then it is considered a pseudo-peak and removed.

5. The rapid separation method for isotope overlapping peaks of a small ion trap mass spectrometer according to claim 1, wherein The specific optimization of the Gaussian fitting parameters using the LM algorithm with a weak peak protection mechanism is as follows: First, divide the spectrum into individual cluster peaks, and then process each cluster peak as follows: Subtract the identified weak peak Gaussian components from each cluster peak, and optimize the remaining non-weak peak Gaussian components using the LM algorithm; After completing the optimization of the non-weak peak Gaussian parameters, subtract the fitted non-weak peak Gaussian components from the cluster peak, and optimize and fit the estimated weak peak Gaussian components using the LM algorithm to obtain the peak information in the spectrum.

6. The rapid separation method for isotope overlapping peaks of a small ion trap mass spectrometer according to claim 5, characterized in that, When the distance between the central positions of two Gaussian components is less than the 2Th equivalent sampling time length, these two Gaussian components will be classified into the same cluster peak.

7. The rapid separation method for isotope overlapping peaks of a small ion trap mass spectrometer according to claim 5, wherein The definition of the weak peak is as follows: A i <0.05A MAX , where A MAX represents the amplitude of the Gaussian component with the largest amplitude in the packet cluster peak.

8. The rapid separation method for isotope overlapping peaks of a small ion trap mass spectrometer according to claim 5, wherein The peak information is the Gaussian parameters in the Gaussian component, and the isotopes in the overlapping peaks can be separated according to the peak information.

Images