A method of fusion targeted and non-targeted proteomic mass spectrometry analysis

Abstract

The application provides a fusion targeted and non-targeted proteomic mass spectrometry analysis method, a plurality of target proteins are acquired through non-targeted proteomics, polypeptide ions with relatively high relative abundance are selected from the plurality of target proteins as target polypeptide ions, a variable, non-continuous, limited retention time isolation window is constructed for the target polypeptide ions, and then quantitative analysis is carried out by using targeted proteomics, the acquired data is analyzed by using non-targeted proteomic analysis software, the advantages of targeted proteomics and non-targeted proteomics are combined, biased data acquisition is realized compared with traditional non-targeted proteomics, and single-needle hundreds of target proteins are realized in targeted identification compared with traditional targeted proteomics.

Description

Technical Field

[0001] This invention relates to the field of proteomics, and in particular to a mass spectrometry method for proteomics that integrates targeted and non-targeted approaches. Background Technology

[0002] Quantitative proteomics research can elucidate the causes and patterns of the occurrence and development of certain biological phenomena at the proteomic level, which is of great significance to life sciences and the diagnosis and treatment of human diseases. For example, quantitative proteomic studies of tumor tissues and non-tumor tissues may discover certain tumor-specific proteins as disease markers, which can be used for early diagnosis, diagnosis, and classification of tumors.

[0003] In proteomics research, there are two main mass spectrometry-based proteomics strategies: targeted and non-targeted proteomics. Targeted proteomics refers to selecting a specific set of proteins as detection targets and using mass spectrometry to perform highly sensitive and accurate quantitative analysis of these proteins. Selective or multiple reaction monitoring (S / MRM) and parallel reaction monitoring (PRM) are two of the most common targeted mass spectrometry acquisition modes. By reducing ion interference other than the target peptide precursor ions, targeted data acquisition of the target ions can be performed, resulting in higher quantitative accuracy, superior reproducibility, and sensitivity.

[0004] Non-targeted proteomics differs from targeted proteomics. It utilizes mass spectrometry to perform high-throughput identification and quantification of all proteins in a sample, without relying on prior information or specific hypotheses. It broadly screens and identifies all proteins in the sample to obtain the most comprehensive protein information possible. For non-targeted proteomics, data-dependent acquisition (DDA) and data-independent acquisition (DIA) are widely used to discover potential biomarkers in clinical specimens. Data-dependent acquisition (DDA) involves the mass spectrometer scanning all ions in the sample according to preset parameters and recording their mass-to-charge ratio (m / z) and relative intensity. Then, for ions within each scan time window, the ions with the highest relative intensity are selected as "precursor ions" for further analysis. This involves separating and fragmenting these ions to obtain more detailed mass spectrometry information. Unlike data-dependent acquisition (DDA), data-independent acquisition (DIA) pre-defines a series of mass-to-charge ratio (m / z) windows and scans and records ions within all windows. In DIA analysis, the mass spectrometer scans the entire mass-to-charge ratio range and segments the resulting mass spectrometry data into a series of overlapping mass-to-charge ratio windows. Data processing is then performed to extract all fragment mass spectra within each window to obtain comprehensive protein information. Compared to DDA, DIA can fragment peptide fragments of all proteins detected by mass spectrometry within the selected retention time and mass-to-charge ratio (m / z) range and accurately quantify them using secondary fragment ions. Therefore, it has very high repeatability and stability, achieving protein identification levels of >5000 per needle in tissue samples, making it widely applicable for high-throughput analysis of large cohorts of clinical tissue samples.

[0005] Traditional untargeted proteomics is an unbiased analysis that cannot precisely target data acquisition for ions of interest. Its quantitative accuracy and ability to uncover low-abundance proteins are lower than those of targeted proteomics. Therefore, the commonly used traditional proteomics workflow involves using untargeted proteomics methods for initial discovery and screening to identify core differentially expressed proteins, followed by targeted proteomics validation and quantification in a validation cohort. However, in mass spectrometry, targeted proteomics single-injection refers to a method where the sample is injected into the mass spectrometer only once for analysis. Single-injection targeted proteomics can quantify a limited number of target proteins, typically a few to dozens. Analyzing hundreds of target proteins using targeted proteomics requires repeated injections, increasing workflow instability and significantly increasing sample consumption and time costs. Summary of the Invention

[0006] The purpose of this invention is to provide a mass spectrometry analysis method that integrates targeted and non-targeted proteomics, combining the advantages of both. Compared with traditional non-targeted proteomics, it achieves biased data acquisition, while compared with traditional targeted proteomics, it achieves targeted identification of hundreds of target proteins with a single needle. It finds a middle ground in the workflow of non-targeted and targeted proteomics, and tries to balance the high protein identification volume of non-targeted proteomics with the quantitative accuracy of targeted proteomics.

[0007] To achieve the above objectives, this solution provides a method for proteomics mass spectrometry analysis that integrates targeted and non-targeted approaches, comprising the following steps: obtaining multiple target proteins from a protein cohort using a non-targeted proteomics method; determining the relative abundance and retention time of peptide ions of the target proteins; selecting N target peptide ions with high relative abundance for each target protein, and obtaining the mass-to-charge ratio and retention time of all target peptide ions; establishing multiple isolation windows, wherein the retention time of the same isolation window is fixed, and the mass-to-charge ratio of different isolation windows is variable and discontinuous; and using a targeted proteomics method and setting the isolation windows to acquire signals and quantitatively analyze all target peptide ions.

[0008] Compared to existing technologies, this technical solution has the following characteristics and beneficial effects: After identifying target proteins using non-targeted proteomics, several target proteins are selected. Only the top N abundant peptide ions of the target proteins are selected to construct the mass-to-charge ratio and retention time distribution maps of the corresponding peptide ions. Based on the distribution maps, a variable, discontinuous isolation window with limited protection time is established in the target proteins, achieving the effect of distributing peptide ion signals of multiple target proteins as evenly as possible within the same circulation window. Compared to traditional non-targeted proteomics, this solution can achieve biased data acquisition of target proteins, especially for low-abundance target proteins, enabling more accurate quantitative analysis. Compared to traditional targeted proteomics, it can achieve quantitative analysis of hundreds or even thousands of target proteins in a single-needle detection, eliminating the need for multiple injections. This reduces sample consumption and time costs, improves the throughput and stability of sample analysis, and achieves the advantages of high protein identification volume in non-targeted proteomics and quantitative accuracy in targeted proteomics. Attached Figure Description

[0009] Figure 1 This is a logical diagram illustrating the integration of targeted and non-targeted proteomics mass spectrometry analysis methods;

[0010] Figure 2 This is a flowchart illustrating the integration of targeted and non-targeted proteomics mass spectrometry analysis methods;

[0011] Figure 3 This is a distribution diagram of the retention time and mass-to-charge ratio of the target polypeptide ions provided in the example of this scheme;

[0012] Figure 4 This is a schematic diagram of the isolation window for the target polypeptide ion provided in the example of this scheme. Detailed Implementation

[0013] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention are within the scope of protection of the present invention.

[0014] This approach provides a method for proteomics mass spectrometry analysis that integrates targeted and non-targeted methods, including the following steps:

[0015] Multiple target proteins were obtained from a protein cohort using non-targeted proteomics methods;

[0016] Determine the relative abundance and retention time of the polypeptide ions of the target protein;

[0017] Select N target polypeptide ions with relatively high abundance for each target protein, and obtain the mass-to-charge ratio and retention time of all the target polypeptide ions;

[0018] Multiple isolation windows are established, where the retention time of the same isolation window is fixed, and the mass-to-charge ratio of different isolation windows is variable and discontinuous;

[0019] The targeted proteomics approach was used, and the isolation window was set to collect and quantify the signals of all the target peptide ions.

[0020] This approach combines a target peptide ion selection scheme from targeted proteomics methods with mass spectrometry, establishing a variable series of windows similar to non-targeted DIA. It achieves biased acquisition of target peptide ions during the mass spectrometry data acquisition stage, enabling targeted identification of hundreds of target proteins with a single needle. Compared with traditional DIA, it improves the quantitative accuracy of target proteins.

[0021] In the step of “obtaining multiple target proteins from the protein cohort using non-targeted proteomics methods”, multiple target proteins are obtained from the protein cohort using either the DIA method or the TMT-DDA method.

[0022] If the DIA method is used: conventional protein digestion methods are used to extract and digest proteins in the sample and convert them into peptides; a mass spectrometer is used with time-segmented windows for DIA measurement, and all peptides in the sample are analyzed by mass spectrometry at the same time. Within each time window, the mass spectrometer will perform multiple scans to obtain all possible ion fragments and select multiple target proteins from them.

[0023] If the TMT-DDA method is used, it refers to combining isotope labeling technology with a data-independent acquisition method. First, TMT isotope labeling technology is used to label proteins or peptides in different samples. Then, a data-independent acquisition method is used to scan the labeled samples and perform mass spectrometry analysis to obtain multiple target proteins.

[0024] In some embodiments, in the step of “obtaining multiple target proteins from a protein cohort using non-targeted proteomics methods”, multiple target proteins may also be identified based on literature review.

[0025] In the step of "determining the relative abundance and retention time of peptide ions in the target protein", the target protein is analyzed based on DIA data or targeted proteomics test data to obtain the relative abundance and retention time of peptide ions in each target protein.

[0026] If DIA is used, the DIA data is preprocessed with quality control, signal-to-noise ratio optimization, peak detection, and noise removal. Then, the peptide sequence of the target protein is determined by searching the database using bioinformatics tools. Based on the retention time of the peptide ions in the target ions, an appropriate retention time window is pre-set in the DIA data to obtain the signal of the peptide ions of the target protein. Within each retention time window, the signals of all ion fragments are superimposed to generate an ion map. In the ion map, the mass-to-charge ratio and retention time of the target ions are compared using mass spectrometry data analysis software to quantitatively calculate the relative abundance and retention time of the peptide ions of each target protein.

[0027] In the step of "selecting N target polypeptide ions with high relative abundance for each target protein and obtaining the mass-to-charge ratio and retention time of all target polypeptide ions", polypeptide ions with the highest relative abundance in the top 2-5 positions of the target protein are selected as target polypeptide ions, and the mass-to-charge ratio and retention time of all target polypeptide ions are obtained. In some embodiments, N is an integer greater than 1.

[0028] The reason this approach selects the top 2-5 peptide ions as target peptide ions is that the top few peptide ions with relatively high abundance are more meaningful for the analysis of the target protein, while too many peptide ions would lead to excessive data volume in subsequent targeted proteomics analysis. In a preferred embodiment, the top 3 peptide ions with relatively high abundance are selected from the target protein as target peptide ions.

[0029] In some embodiments, the mass-to-charge ratio and retention time of all target polypeptide ions are collected and plotted to obtain a distribution map, wherein the distribution map is plotted with retention time on the x-axis and mass-to-charge ratio on the y-axis. A schematic diagram of the distribution map obtained by this scheme is shown below. Figure 1 As shown in the image.

[0030] In the step of "establishing multiple isolation windows, where the retention time of the same isolation window is fixed and the mass-to-charge ratio of different isolation windows is variable and discontinuous", the windows are automatically divided according to the window partitioning algorithm and the window partitioning rules to obtain multiple isolation windows.

[0031] Specifically, the window division rules are as follows:

[0032] 1) The variance of the number of ions within the isolation window of the same acquisition cycle is minimized;

[0033] 2) The sum of the mass-to-charge ratios of the isolation windows within the same acquisition cycle is minimized;

[0034] 3) The isolation window of the same acquisition cycle contains the largest number of target ions;

[0035] 4) The number of isolation windows in each acquisition cycle shall not exceed the maximum number of windows limited by the gradient.

[0036] In other words, this scheme divides the windows according to the above window division rules to obtain multiple isolation windows with limited retention time, which are non-contiguous and variable.

[0037] Furthermore, in the step of "establishing multiple isolation windows, where the retention time of the same isolation window is fixed and the mass-to-charge ratio of different isolation windows is variable and discontinuous", the mass-to-charge ratio of all target peptide ions within the same retention time range is obtained as input values ​​and sorted. The maximum number of windows for each cycle is obtained, and an initial empty interval list is created, with each interval list corresponding to one isolation window. A starting value is randomly selected from the input values, and intervals with a length greater than 3 are selected from this starting value. If the selected interval does not overlap with any existing intervals in the interval list, the interval is written into the interval list. All input values ​​are traversed to generate multiple intervals in the interval list. The intervals in the interval list are optimized using a clustering method combined with the maximum number of windows.

[0038] Specifically, the windowing algorithm in this scheme is as follows:

[0039] 1) First, the mass-to-charge ratio of all target polypeptide ions within the same retention time range is sorted as input values. In this scheme, the retention time range is set to 3 minutes, so the mass-to-charge ratio within the 3-minute time window is obtained.

[0040] 2) Initialize an empty list of intervals.

[0041] 3) Iteratively generate N intervals (the maximum window size for each cycle is determined by the liquid phase gradient):

[0042] a. For each interval, randomly select a starting value from the remaining values, and a value greater than...

[0043] A random interval range equal to 3;

[0044] b. Check if the newly generated interval overlaps with an existing interval. If there is an overlap, discard the current interval.

[0045] The original interval is used to generate a new interval;

[0046] c. If the newly generated interval does not overlap with an existing interval, add it to the interval list;

[0047] 4) Use clustering methods (such as CLARA) to optimize the generated intervals. The heuristic search here is to transform the problem into a clustering problem and use existing clustering algorithms to find a better solution.

[0048] 5) Calculate the metrics for the optimized intervals, such as the standard deviation of the number of values ​​in each interval, the sum of the interval ranges, and the sum of the number of values ​​within each interval.

[0049] 6) Output the optimized interval and corresponding indicators.

[0050] It is worth noting that the retention time of each isolation window may be the same or different, and in some embodiments, the mass-to-charge ratio of each isolation window is greater than 3. In some embodiments, the retention time of each isolation window is limited to 3 minutes.

[0051] This approach combines windows with the same retention time to create isolation windows, increasing the number of target peptide ions detected while maintaining the same number of windows. Using this strategy, we constructed continuous, variable, and time-defined isolation windows, which are then used in targeted proteomics analysis to achieve a more uniform distribution of target peptide ion signals within the same cycle window.

[0052] In the step of “using targeted proteomics methods and setting the isolation window to acquire and quantify the signals of all target polypeptide ions”, the isolation window is applied to the mass spectrometry method of targeted proteomics for data acquisition. The acquired data is analyzed using the DIA data analysis mode to extract the signals of different target polypeptide ions in the same isolation window for quantification.

[0053] Furthermore, in the step of "using targeted proteomics methods and setting the series of windows to acquire and quantitatively analyze signals of all the target polypeptide ions," the targeted proteomics method detects hundreds to thousands of target proteins with a single needle. In some embodiments, a single needle detects 500-1000 target proteins, which can be 700 target proteins.

[0054] This scheme integrates targeted and non-targeted proteomics mass spectrometry analysis methods for proteomics analysis of target proteins.

[0055] To verify the significant effectiveness of this protocol in proteomics identification, the following example is provided:

[0056] Based on previously published and literature-related potential thyroid biomarkers, a total of 738 proteins, the goal is to identify as many as possible using single-needle mass spectrometry (SMS) and achieve better identification results and quantitative accuracy than DIA.

[0057] First, using single-needle DIA technology, the top three most abundant polypeptide ions of 738 proteins were identified. Retention time and mass-to-charge ratio distributions of all target polypeptide ions were then plotted, as shown below. Figure 3 As shown.

[0058] Based on the distribution of retention time and mass-to-charge ratio, a variable, discontinuous isolation window with limited retention time is calculated, such as... Figure 4 As shown.

[0059] The window division rules are as follows:

[0060] 1) The variance of the number of ions within the isolation window of the same acquisition cycle is minimized.

[0061] 2) The sum of the mass-to-charge ratios of the isolation windows within the same acquisition cycle is minimized.

[0062] 3) The isolation window of the same acquisition cycle contains the largest number of target ions.

[0063] 4) The number of windows within the isolation window of each acquisition cycle shall not exceed the maximum number of windows limited by the gradient.

[0064] A variable window was applied to the tMS2 method on a Thermo Scientific Orbitrap Exploris 480 mass spectrometer, using a 30-minute effective gradient to collect peptide samples after enzymatic digestion of thyroid tissue samples. The control group was set to the traditional DIA method, with all liquid chromatography conditions and primary mass spectrometry parameters remaining consistent except for the secondary isolation window.

[0065] Since the peptide ion retention time is based on the peptide ions identified by the control group DIA, the target protein identification rate of DIA will be slightly higher than that of this invention, as shown in Table 1. If the initial peptide ion retention time is based on the test results of targeted proteomics, the target protein identification rate of this invention may be higher than that of DIA. Regarding the evaluation of method stability, the median value of the coefficient of variation (CV) of this invention is lower, indicating better stability, as shown in Table 1.

[0066] Table 1

[0067]

[0068]

[0069] The quantitative results of this invention show a higher correlation with the quantitative results of targeted proteomics than the quantitative results of non-targeted proteomics, as shown in Table 2 below, indicating that the quantitative results of this invention are more accurate than the non-targeted quantitative results.

[0070] Table 2

[0071] Correlation Protein count Correlation number of peptides This invention 0.897 119 0.884 148 DIA 0.757 119 0.807 148 .

[0072] This invention is not limited to the preferred embodiments described above. Anyone can derive other products in various forms under the guidance of this invention. However, regardless of any changes in shape or structure, any technical solution that is the same as or similar to this application falls within the protection scope of this invention.

Claims

1. A proteomics mass spectrometry analysis method integrating targeted and non-targeted approaches, characterized in that, Includes the following steps: Multiple target proteins were obtained from a protein cohort using non-targeted proteomics methods; Determine the relative abundance and retention time of the polypeptide ions of the target protein; Select the top N target polypeptide ions with the highest relative abundance for each target protein, and obtain the mass-to-charge ratio and retention time of all the target polypeptide ions; The window is automatically divided into multiple isolation windows according to the window partitioning algorithm and the window partitioning rules. The retention time of the same isolation window is fixed, while the mass-to-charge ratio of different isolation windows is variable and non-continuous. The window partitioning algorithm is as follows: First, the mass-to-charge ratios of all target peptide ions within the same retention time range are sorted as input values ​​to obtain the maximum number of windows in each cycle; an empty interval list is initialized, and X intervals are generated iteratively: for each interval, a starting value is randomly selected from the remaining values, along with a random interval range greater than or equal to 3 m / z; it is checked whether the newly generated interval overlaps with existing intervals. If there is an overlap, the currently generated interval is discarded, and a new interval is generated; if the newly generated interval does not overlap with existing intervals, it is added to the interval list; clustering methods are used to optimize the generated intervals, the indices of the optimized intervals are calculated, and the optimized intervals and their corresponding indices are output; each interval corresponds to an isolation window; The targeted proteomics approach was used, and the isolation window was set to collect and quantify the signals of all the target peptide ions.

2. The method for fusion-targeted and non-targeted proteomics mass spectrometry analysis according to claim 1, characterized in that, The window division rules are as follows: 1) The variance of the number of target ions within the isolation window of the same acquisition cycle is minimized; 2) The sum of the mass-to-charge ratios within the isolation windows of the same acquisition cycle is minimized; 3) The isolation window of the same acquisition cycle contains the largest number of target ions; 4) The number of isolation windows in each acquisition cycle shall not exceed the maximum number of windows limited by the gradient.

3. The method for fusion-targeted and non-targeted proteomics mass spectrometry analysis according to claim 1, characterized in that, In the step of "obtaining multiple target proteins from the protein cohort using non-targeted proteomics methods", multiple target proteins are obtained from the protein cohort using either the DIA method or the TMT-DDA method.

4. The method for fusion-targeted and non-targeted proteomics mass spectrometry analysis according to claim 1, characterized in that, Select the peptide ions with the highest relative abundance in the top 2-5 positions from each target protein as the target peptide ions.

5. The method for fusion-targeted and non-targeted proteomics mass spectrometry analysis according to claim 1, characterized in that, The mass-to-charge ratio of each isolation window has a width range greater than or equal to 3 m / z.

6. The method for fusion-targeted and non-targeted proteomics mass spectrometry analysis according to claim 1, characterized in that, The retention time for each isolation window may be the same or different.

7. The method for fusion-targeted and non-targeted proteomics mass spectrometry analysis according to claim 1, characterized in that, In the step of "using targeted proteomics and setting the isolation window to acquire and quantitatively analyze the signals of all the target polypeptide ions", the targeted proteomics method detects 500-1000 target proteins with a single needle.

Images