A method for targeted detection and quantification of protein post-translational modifications based on spectral library-driven and parallel reaction monitoring

Abstract

The application belongs to the technical field of proteomics and mass spectrometry, and discloses a kind of post-translational modification targeted detection and quantification method for protein based on spectral library driving and parallel reaction monitoring.The method generates initial PRM target list or initial PRM target panel based on modified peptide reference spectral library, assigns candidate precursors to one or more PRM acquisition methods using precursor ion dynamic scheduling, forms verified PRM target panel through actual PRM acquisition and detectability confirmation of verification sample, and can further form compressed PRM target panel through one or several rounds of quality screening, so as to realize high sensitivity, high reproducibility targeted quantitative analysis and horizontal comparison of lysine acylation modification in low input biological sample in the premise of not depending on antibody enrichment and isotope labeling in the detection stage of sample to be tested.

Description

Technical Field

[0001] This invention belongs to the field of proteomics and mass spectrometry analysis technology, specifically relating to a targeted detection and quantification method for post-translational modification (PTM) of proteins based on a modified peptide reference spectrum library and parallel reaction monitoring (PRM). In particular, it relates to a targeted quantitative analysis method for lysine acylation modification suitable for low-input biological samples, including but not limited to one or more of lysine lactylation, acetylation, benzoylation, and nicotinylation. Background Technology

[0002] Post-translational modifications of proteins are important molecular mechanisms regulating protein structure, activity, localization, interaction, and stability, playing a role in cellular metabolic regulation, signal transduction, and epigenetic regulation. Zhang et al. reported histone lysine lactylation and its association with lactate metabolism and gene expression regulation; subsequently, lactylation, as well as non-classical lysine acylation modifications such as nicotinylation and benzoylation, have gradually become the focus of proteomics research, but the biological mechanisms, abundance ranges, and detection resource maturity of different modifications vary.

[0003] Currently, mass spectrometry detection methods for protein post-translational modifications mainly include data-dependent acquisition (DDA), data-independent acquisition (DIA), and targeted mass spectrometry. However, for post-translational modification types with low abundance, low occupancy, or a lack of high-quality publicly available reference resources, existing methods still have the following shortcomings.

[0004] First, while enrichment strategies based on antibodies or other specific affinity reagents can improve the detection probability of target modified peptides, they can still be affected by factors such as antibody specificity, epitope bias, batch variability, elution efficiency, and nonspecific adsorption. For some non-classical lysine acylation modifications, mature and well-covered commercial antibodies or affinity reagents are still relatively limited. At the same time, enrichment procedures usually require a high initial protein amount and may cause sample loss, making them difficult to directly apply to low-input scenarios such as clinical tissue sections, needle biopsies, microsurgical tissue dissections, and rare cell and single-cell samples.

[0005] Secondly, the detection sensitivity and coverage of existing methods still fall short of the requirements. The abundance of target modified peptides in complex biological samples is usually low and is often affected by unmodified peptides, co-eluting interfering ions, and background noise. Discovery-based detection methods based on DDA have randomness in the selection of low-abundance modified peptides, making it difficult to obtain stable detection results across different batches. The success rate of directly identifying low-abundance modified peptides in trace, unenriched samples is extremely low.

[0006] Third, publicly available modified peptide spectra resources and their coverage remain limited. For emerging lysine acylation modifications such as lactylation, nicotinylation, and benzoylation, the maturity of existing publicly available spectra libraries, standards, or affinity enrichment resources varies, making it difficult for researchers to directly obtain targeting methods that combine high coverage, transferability, and low input applicability.

[0007] Fourth, the development of large-scale PRM methods faces significant scheduling and capacity limitations. When there are a large number of modified peptide precursors to be detected, traditional methods often struggle to effectively cover a large number of targets while ensuring the number of chromatographic peak sampling points, minimum residence time, and sampling stability, thus affecting quantitative accuracy and reproducibility.

[0008] Fifth, existing technologies generally lack a technical route that can organically integrate the construction of modified peptide spectral libraries, the development of PRM methods, quality screening, and application with low-input samples. Especially in the formation of target sets, existing methods often lack flexibility and are difficult to accommodate the following two types of application needs: one is the formation of target protein sets based on pre-defined biological questions, candidate pathways, or candidate biomarkers; the other is the formation of high-quality target sets and their corresponding protein sets through reference spectral library generation, pre-screening, and quality screening. For the latter, existing technologies particularly lack a method for screening compressed target panels suitable for subsequent cross-sectional comparisons of multiple small samples from a large number of candidate modified peptides.

[0009] In addition, some existing methods rely on tandem mass tags of equal weight and internal standard mixtures rich in target modified peptides to achieve inter-sample comparison or improve the probability of target peptide identification. However, such schemes are complex, costly, and have strict requirements on sample processing conditions and experimental design, which is not conducive to the direct targeted detection of unlabeled trace samples.

[0010] In summary, for some emerging lysine acylation modifications, current technologies still lack a systematic approach that can reduce dependence on specific antibody enrichment and labeled internal standards during the sample detection stage, and integrate reference spectral library generation, dynamic scheduling of precursor ions, actual PRM validation, optional one or more rounds of quality screening, and quantitative analysis of low-input samples. Therefore, a technical solution is needed to establish a method that can efficiently generate and schedule large-scale PRM targets based on modified peptide reference spectral libraries, forming empirically validated or compressed PRM target panels suitable for low-input sample detection. Summary of the Invention

[0011] The technical terms used in the technical solutions of this specification are common technical content and scope known to those skilled in the art, specifically including: "Lysine acylation modification": refers to post-translational modification formed by acylation of the amino group on the lysine side chain. Lactylation, acetylation, benzoylation, and nicotinylation are representative examples used in this invention to illustrate aliphatic acyl groups, aromatic acyl groups, heterocyclic acyl groups, and their derived acyl groups, and do not constitute a limiting definition. The lysine acylation modification described in this invention may also include modifications formed by other aliphatic acyl groups, aromatic acyl groups, heterocyclic acyl groups, or their derived acyl groups; as long as a data resource containing the modified peptide sequence, modification site, precursor ion, product ion, and chromatographic retention time information can be obtained or established, the initial PRM target generation, dynamic scheduling, PRM validation, optional quality screening, and targeted quantification process described in this invention can be adopted.

[0012] "Modified Peptide Reference Spectrum Library" or "Reference Spectrum Library" refers to a dataset containing mass spectrometry information related to the target post-translational modified peptide. This information includes at least the precursor ion mass-to-charge ratio, product ion spectrum information, chromatographic retention time information, and the position of the target modification within the modified peptide. It may also include one or more candidate protein accessions and their corresponding modification sites and / or protein grouping information as protein mapping annotations. The reference spectrum library may be derived from one or more data sources, including simulated post-translational modified peptides, naturally occurring post-translational modified peptides, chemically synthesized modified peptides, or publicly available spectral databases.

[0013] "Target post-translational modified peptide": refers to a peptide containing a post-translational modification site to be detected. The post-translational modification can be a naturally occurring modification or a simulated modification that has the same or corresponding mass increment, structural characteristics, chromatographic behavior, or fragmentation characteristics as the natural target modification. In some embodiments, the target post-translational modification includes acylation modification at lysine sites, including but not limited to one or more of lactylation, acetylation, benzoylation, and nicotinylation.

[0014] "Target set": refers to a group of target modified peptides, precursor ions, or combinations thereof used for PRM method development, pre-screening validation, quality screening, or subsequent quantitative analysis. The target set can be a pre-determined set based on a preset biological question, functional pathway, disease mechanism, or candidate biomarker, or it can be a set generated from a reference spectral library and retained after quality screening.

[0015] "Target protein set": refers to a group of proteins corresponding to the target protein set. The target protein set may be a set of proteins related to the same biological pathway, the same cellular function, the same disease mechanism, the same organelle location, or the same candidate biomarker, or it may be a set of proteins obtained by reverse induction through pre-screening and quality screening.

[0016] "Initial PRM Target List": This refers to a list of candidate targets extracted from a reference spectral library for establishing the PRM acquisition method. The initial PRM target list typically includes information on the precursor ion mass-to-charge ratio, charge state, product ion type, product ion charge state, and retention time window.

[0017] "Initial PRM Target Panel": This refers to the set of candidate PRM targets determined before actual PRM acquisition verification or quality screening, based on a reference spectral library, DDA identification results of modified peptides, publicly available modification data resources, or a combination thereof. The membership of the initial PRM target panel is determined at least by the target modified peptide sequence and the position of the target modification within the peptide. The panel may also record or associate one or more acquisition attributes or annotation information for each member, including precursor ion mass-to-charge ratio, precursor ion charge state, product ion information, chromatographic retention time, retention time window, and protein mapping annotation. The initial PRM target panel can be converted into one or more PRM acquisition methods through dynamic precursor ion scheduling; the initial PRM target list and the initial PRM target panel can be mutually converted according to data expression formats. The precursor ion mass-to-charge ratio and precursor ion charge state can be recorded as explicit fields, or calculated or converted based on the modified peptide sequence, the mass increment of the target post-translational modification, and preset charge state rules; the chromatographic retention time used for dynamic scheduling can be measured, predicted, calibrated, or obtained through alignment and conversion.

[0018] A "validated PRM target panel" or "large-scale PRM target panel" refers to a collection of multiple PRM targets that have been validated through actual PRM acquisition, yielding identifiable chromatographic peaks and / or meeting predetermined detectability requirements. Membership in the validated PRM target panel is determined based on the target modified peptide sequence and the position of the target modification within the peptide. Precursor ion mass-to-charge ratio, precursor ion charge state, product ion information, retention time, and retention time window are acquisition attributes. One or more candidate protein accessions and their corresponding modification sites and / or protein grouping information constitute protein mapping annotations. These protein mapping annotations are not a necessary condition for distinguishing members. Actual PRM detection results in validation or pre-screening samples are used to evaluate the feasibility of the initial PRM target panel, to form a validated PRM target panel, or to further form a compressed PRM target panel, without reverse altering the initial PRM target list or the composition of the initial PRM target panel. Adjustments to product ions, retention time, precursor ion charge state, collision energy, mass-to-charge ratio calibration, liquid chromatography conditions, mass spectrometry platform, PRM acquisition method allocation, software data format, database version, protein merging rules, or peptide-protein mapping relationships do not alter the membership determined by the modified peptide sequence and the modification location within the peptide.

[0019] "Retention time window": refers to the chromatographic acquisition time interval set for a single precursor ion, used to limit the monitoring period of the precursor in PRM acquisition, so as to improve method capacity and sampling efficiency.

[0020] "Precursor ion dynamic scheduling" refers to the process of allocating multiple candidate precursor ions to one or more independent PRM acquisition methods based on the chromatographic retention time information of each candidate precursor ion. The dynamic scheduling can be further executed based on one or more of the following: acquisition time window, minimum residence time, chromatographic peak width, number of sampling points per peak, and acquisition load within different time intervals, and optionally employs a load balancing approach. This process can be implemented through software, algorithms, programs, or modules.

[0021] "Balance load" refers to the balanced distribution of the load of each PRM acquisition method in the time dimension during the dynamic scheduling of precursor ions or the generation of compressed PRM panels, based on the distribution and overlap of precursor ions in different time periods. This reduces excessive accumulation of precursors in local retention periods and optimizes sampling efficiency and quantitative stability.

[0022] "Pre-screened samples" refer to samples used to validate the detectability, peak quality, interference levels, and repeatability of candidate precursor ions in a multi-method PRM acquisition panel before formal testing of the biological sample to be tested. These pre-screened samples can be from the same source as the sample to be tested, or they can be comparable to the sample in terms of biological type, sample processing procedures, or target modification background. Pre-screened samples can be used to further screen high-quality targets suitable for subsequent actual sample testing from a large-scale PRM target panel, forming a compressed PRM target panel.

[0023] "Quality screening" refers to the process of screening candidate precursor ions based on pre-screened samples or repeated data collection to retain target precursor ions suitable for subsequent PRM quantitative analysis.

[0024] In this invention, quality screening can be based on one or more of the following indicators: absolute peak area, signal-to-noise ratio or signal-to-background ratio, relative abundance, time correlation, number of effective product ions and coefficient of variation, and may include one or more rounds of screening.

[0025] Signal-to-noise ratio (S / N): This refers to the ratio of the target signal strength to the noise signal strength, and is one of the indicators for evaluating PRM signal quality.

[0026] Signal-to-background ratio (S / B) refers to the ratio of the target signal strength to the background signal strength, and is one of the indicators for evaluating the peak quality and background interference level of the PRM. In different implementations, the signal-to-noise ratio, signal-to-background ratio, or both can be used for quality screening based on the output indicators of the instrument platform or data analysis software.

[0027] "Relative abundance" refers to the proportion of the peak area of ​​a certain product ion in each product ion corresponding to the same precursor ion to the sum of the peak areas of all or major product ions of the same precursor ion, or the degree of matching with the relative intensity of the corresponding product ions in the reference spectrum library. It is used to evaluate the rationality of the target product ion signal.

[0028] "Time correlation" refers to the degree of co-elution correlation of multiple product ions corresponding to the same precursor ion in the chromatographic dimension, or the consistency index of retention time / peak shape information between the chromatographic peak of the product ion extracted from the sample and the corresponding spectrum in the reference spectrum library.

[0029] "Effective product ions" refer to product ions that meet the accuracy requirements in PRM acquisition, have complete peak shapes, reasonable co-elution relationships, and match the reference spectral library. Preferably, this invention requires that at least three effective product ions be retained for each quantifiable precursor.

[0030] "Coefficient of variation (CV)" refers to the ratio of standard deviation to mean, used to evaluate the reproducibility of repeated detection results. This invention preferably uses a CV of no more than 30% as one of the quality screening criteria for precursor ion retention.

[0031] "Compressed PRM target panel": refers to a high-quality subset of PRM targets retained after one or more rounds of quality screening from a validated PRM target panel or a large-scale PRM target panel. The PRM target subset may include a set of target precursor ions or a set of corresponding modified peptides, and is suitable for highly reproducible targeted detection in low-input samples or complex clinical samples.

[0032] “Low-input samples” refer to biological samples with low protein input, including but not limited to trace amounts of cell lysates, clinical tissue sections, biopsy samples, rare cell samples, and single-cell samples with a protein input of no more than 200 ng.

[0033] "Single-cell level sample": refers to a sample that corresponds to or is close to a single-cell input level in terms of analytical sensitivity, sample source quantity, or sample processing scale. The single-cell level sample can be a sample directly prepared from a single cell, or a sample obtained through extremely low cell number dilution, micro-separation, or equivalent low input methods.

[0034] "Single-cell source sample": refers to a sample directly derived from a single cell or obtained from a single cell through lysis, enzymatic digestion, separation, transfer, or other trace processing procedures. In this specification, "single-cell level sample" and "single-cell source sample" can be used separately depending on the specific experimental design. The former emphasizes the input quantity and analytical scale, while the latter emphasizes the sample source.

[0035] "Simulated post-translational modified peptides" refer to peptides or protein hydrolysates obtained under in vitro conditions through chemical reactions, derivatization, simulated modification, or other methods, which are identical or corresponding to the natural target post-translational modification in terms of mass increment, structural characteristics, chromatographic behavior, or fragmentation characteristics. These can serve as a data source for constructing reference spectral libraries. It should be noted that the focus of this invention is on constructing reference spectral libraries based on the modified peptides and developing, scheduling, screening, and applying PRM methods, and does not involve the preparation methods of the simulated modified peptides themselves.

[0036] "Horizontal comparative analysis" refers to the comparative analysis of the same target post-translational modified peptides in multiple samples from different sources but with the same detection target using the same compressed PRM target panel. The samples may come from different individuals, different disease states, different treatment conditions, different time points, or different sample types.

[0037] To overcome the shortcomings of the prior art, the present invention provides the following technical solution: In a first aspect, the present invention provides a method for targeted detection and quantification of protein post-translational modifications (PTM) based on spectral library-driven and parallel reaction monitoring (PRM), characterized by comprising the following steps: (1) Construct a reference spectral library of the target post-translational modified peptide, wherein the reference spectral library includes at least the precursor ion mass-to-charge ratio, product ion spectrum information, chromatographic retention time information and the position of the target modification in the modified peptide, and may include protein mapping annotation information; (2) The reference spectral library is derived from one or more of the following: natural post-translational modified peptides, simulated post-translational modified peptides, chemically synthesized modified peptides, or publicly available spectral databases; (3) Generate an initial PRM target list or an initial PRM target panel based on the reference spectral library; (4) Perform dynamic scheduling of precursor ions on the initial PRM target list or the initial PRM target panel to reasonably allocate multiple candidate precursor ions to one or more independent PRM acquisition methods. The dynamic scheduling satisfies the constraints of retention time window, minimum residence time, chromatographic peak width and minimum number of sampling points per peak, and optionally adopts a balanced load mode. (5) Using validation samples, perform actual PRM acquisition, peak extraction and detectability confirmation on one or more PRM acquisition methods to form a validated PRM target panel. (6) Optionally, the validated PRM target panel is subjected to one or more rounds of quality screening using pre-screened samples to retain PRM targets that meet the predetermined quality requirements, thereby forming a compressed PRM target panel. (7) Using one or more of the initial PRM target list, initial PRM target panel, validated PRM target panel or compressed PRM target panel, targeted PRM data acquisition and quantitative analysis are performed on biological samples.

[0038] Further, the post-translational modification is lysine acylation. Lysine lactylation, acetylation, benzoylation, and nicotinylation are representative examples and do not constitute a limiting factor; the lysine acylation modification can also be a lysine acylation modification formed by other aliphatic acyl groups, aromatic acyl groups, heterocyclic acyl groups, or their derived acyl groups. Preferably, the lysine acylation modification is lysine lactylation, and the reference spectral library is constructed by in vitro chemical derivatization modification, simulated modification introduction, or chemical synthesis of proteins or peptides.

[0039] Furthermore, the quality screening criteria include one or more of the following: absolute peak area > 500, signal-to-background ratio (S / B) > 3, relative abundance > 0.05, time correlation > 0.8, effective product ion number ≥ 3, and coefficient of variation (CV) ≤ 30%; the quality screening can be performed in one or more rounds to obtain a high-quality compressed PRM target panel suitable for low input samples.

[0040] Furthermore, the dynamic scheduling of precursor ions is implemented by a computer program, algorithm, or software module, and can be adjusted according to liquid chromatography conditions, column type, instrument platform, retention time calibration results, sample complexity, or target precursor ion density to control the acquisition time window, minimum residence time, peak width, and number of sampling points per peak. The dynamic scheduling can optionally employ a balanced load mode to balance the load of each PRM acquisition method in the time dimension. Furthermore, the biological sample is a low-input sample, with a protein input not exceeding 200 ng, selected from one or more of cell lysates, clinical tissue sections, biopsy samples, rare cell samples, single-cell level samples, or single-cell source samples. The method achieves targeted quantification and cross-sectional comparative analysis between multiple samples without relying on specific antibody enrichment of modified peptides, or on isoweight tandem mass labeling or a mixture of internal standards rich in target modified peptides during the PRM detection and quantitative analysis of the sample.

[0041] Furthermore, the lysine acylation modification is lysine lactylation (K(lac), mass increment +72.0211 Da), and the reference spectral library is constructed by in vitro chemical derivatization modification, simulated modification introduction, or chemical synthesis of proteins or peptides.

[0042] Furthermore, the biological samples were selected from HeLa cells treated with sodium lactate, the H1299 non-small cell lung cancer cell line, and tumor-infiltrating CD8 cells. + One or more of the following: T cells, non-small cell lung cancer tumor tissue sections and their paired adjacent normal tissues, or single-cell-derived samples; the method described is for non-disease diagnostic applications.

[0043] Furthermore, the lysine acylation modification is one of lysine acetylation (K(ac), mass increment +42.0106 Da), lysine benzoylation (K(bz), mass increment +104.0262 Da), or lysine nicotinylation (K(nic), mass increment +105.0215 Da), and the reference spectral library is accordingly constructed by in vitro chemical derivatization modification or simulated modification.

[0044] Secondly, this invention provides a method for establishing and validating an initial PRM target panel based on DDA identification results of modified peptides. The method extracts modified peptides and their mass spectrometry and retention time information from the DDA identification results, forms a list of modified peptide search results, and directly generates an initial PRM target panel. One or more PRM acquisition methods are dynamically scheduled. Subsequent PRM detection is used to evaluate the feasibility or detection coverage of the initial panel, without altering the composition of the initial panel based on actual detection results in specific validation samples.

[0045] Thirdly, this invention provides a method for generating grouped PRM acquisition methods based on joint analysis of a modified peptide reference spectral library and publicly available modification site data. The method determines the spectral library target members by identifying the modified peptide sequence and the location of the target modification within the peptide, and compares whether one or more protein accession-protein sequence-site mapping pairs associated with each member are recorded in the publicly available modification site data, thereby forming a target group supported by existing publicly available annotations and a supplementary target group to the spectral library. This grouping is only used to record the publicly available annotation support status and to generate PRM acquisition methods separately, without changing the target member identity.

[0046] Both the second and third aspects mentioned above can be implemented independently of the complete development chain of the first aspect; wherein, the generation and dynamic scheduling of the initial PRM target panel are not necessary conditions for the formation of a validated PRM target panel or a compressed PRM target panel, unless further corresponding PRM validation or quality screening is performed according to the specific testing purpose.

[0047] Fourthly, the present invention provides a system for implementing the method described in the first, second, or third aspect, characterized in that the system comprises: (1) Data resource acquisition module, configured to construct or read one or more of the following: reference spectral library of target post-translational modified peptides, DDA identification results of modified peptides, or publicly available modification site data; (2) PRM target generation and grouping module, configured to generate an initial PRM target list or an initial PRM target panel based on the data resources, and optionally mark the targets as targets supported by existing public annotations or supplementary targets of the spectral library according to the matching status of the publicly available modified site data. (3) The precursor ion dynamic scheduling module is configured to perform precursor ion dynamic scheduling based on chromatographic retention time and target density in different time intervals, so as to allocate candidate precursors to one or more PRM acquisition methods; (4) Validation and quality screening module, configured to perform peak extraction, target identification and detectability evaluation based on the actual PRM collection data of the validation sample, and optionally form a validated PRM target panel or perform one or more rounds of quality screening based on the pre-screened sample data to form a compressed PRM target panel. (5) PRM data acquisition control module, configured to control the mass spectrometer to perform data acquisition according to the scheduled PRM method or compressed PRM target panel; (6) Quantitative and analytical module, configured to extract peak area, normalize, perform statistical tests and output results for PRM data.

[0048] Furthermore, the system further includes a mass spectrometer configured to support adaptive retention time, PRM mode data acquisition, and a high-resolution Orbitrap or Astral analyzer.

[0049] Fifthly, the present invention provides a computer program product, characterized in that it includes computer-executable instructions stored on a non-transitory computer-readable medium; when the instructions are executed by a computer processor, the processor performs at least the generation of precursor ion dynamic scheduling and mass spectrometer acquisition control instructions, and optionally performs one or more of the following: data resource acquisition or reading, initial PRM target generation, public annotation support state grouping, peak extraction and detectability or coverage evaluation of validation data, one or more rounds of quality screening, and quantitative analysis and result output of PRM data.

[0050] In a sixth aspect, the present invention provides a non-transitory computer-readable storage medium having stored thereon computer program instructions; when the instructions are executed by a processor, the processor performs computer-executable data processing and mass spectrometry control steps in the methods of the first, second, or third aspects, and generates mass spectrometer control instructions for one or more PRM acquisition methods, or performs target generation or classification, dynamic scheduling, peak extraction, detectability or coverage evaluation, optional quality screening, quantitative analysis, and result output on the received PRM data.

[0051] The beneficial effects of this invention include: 1. Enables large-scale PRM method development based on reference spectral libraries. This invention uses a modified peptide reference spectral library as a priori information source, which can extract precursor ions, product ions, and retention time information of the target post-translational modified peptide, helping to improve the efficiency of targeted method development and reduce dependence on a single publicly available spectral resource.

[0052] 2. Supports dynamic scheduling of large-scale target precursors. This invention, through dynamic scheduling of precursor ions, allocates a large number of candidate targets to multiple independent PRM acquisition methods under constraints such as minimum residence time, chromatographic peak width, and minimum number of sampling points per peak, thereby improving detection throughput and maintaining predetermined sampling requirements.

[0053] 3. Establish a standardized one- or multi-round quality screening process. This invention systematically screens candidate precursor ions based on absolute peak area, signal-to-noise ratio or signal-to-background ratio, relative abundance, time correlation, number of effective product ions, and coefficient of variation of repeated acquisitions. This effectively eliminates unstable, low-quality, or non-reproducible targets, forming a high-quality compressed PRM target panel suitable for actual sample analysis.

[0054] 4. Applicable to low-input samples and complex clinical samples. The compressed PRM target panel of this invention, after quality screening, can be applied to low-input samples with protein input as low as 200 ng. It is suitable for cell samples, clinical tissue sections, puncture biopsy samples, rare cell samples, and single-cell derived samples, significantly expanding the application boundaries of post-translational modification targeted quantification methods.

[0055] 5. Compatible with multiple methods of forming target sets. The target set in this invention can be determined based on preset biological questions, functional pathways, disease mechanisms or candidate biomarkers, or it can be generated through reference spectral libraries and formed after quality screening. Therefore, it is suitable for both pathway-oriented and candidate biomarker-oriented targeted detection, as well as panel-based detection obtained by stepwise screening from a large number of candidate targets.

[0056] 6. Supports the construction of reference spectral libraries through in vitro derivatization or simulated modification. In some embodiments, this invention allows the construction of reference spectral libraries from proteins, peptides, or their enzymatic digestion products corresponding to the target set through in vitro derivatization, simulated post-translational modification, or chemical synthesis modification, thereby providing an operable library construction path for modification types lacking natural high-abundance samples or with insufficient publicly available spectral library resources.

[0057] 7. Supports cross-sectional comparative analysis between multiple samples. The compressed PRM target panel developed in this invention can be used for cross-sectional comparative analysis of target post-translational modifications between multiple trace samples, low-input samples, or clinical samples from different sources but with the same detection target, which is beneficial for building a stable and transferable detection system.

[0058] 8. No specific antibody enrichment of modified peptides is required in the sample detection stage. The method of this invention does not rely on antibodies targeting specific post-translational modifications for target peptide enrichment in the PRM detection and quantitative analysis stages of the sample, overcoming the limitations of antibody reagents in terms of specificity, coverage, batch variation, and cost. In particular, it provides an accessible targeted analysis method for the study of novel post-translational modifications in the absence of commercially available antibodies.

[0059] 9. No need for isoweight tandem mass labeling or modified internal standard mixtures. This invention does not rely on isoweight tandem mass labeling or reporter ion quantification in the sample detection process, nor does it rely on adding internal standard mixtures rich in target post-translational modified peptides to the sample. It is suitable for direct targeted detection of unlabeled samples, with a simpler process, more controllable costs, and more flexible applications. Attached Figure Description

[0060] Figure 1 This document presents a schematic diagram and results of the establishment and validation of a PRM-targeted detection method based on a chemically induced lactylation reference spectral library. The diagram includes: a) the construction of the lactylation reference spectral library, PRM method generation, validation injection and redistribution procedures, and the corresponding number of precursor ions / methods; b) the process for generating a lactylated compressed PRM target panel after quality screening, and the number of peptides, modification sites, and proteins on this panel; c) the reproducibility assessment of the lactylated compressed PRM target panel at a protein input of 200 ng (coefficient of variation distribution); and d) a volcano plot showing significant changes in lactylated peptides after sodium lactate treatment of HeLa cells.

[0061] Figure 2 The following is a diagram showing the two-round screening method and results of lactylated PRM for clinical samples based on cell line quality screening; where: a. the two-round screening process and the corresponding number of retained precursors and methods (H1299 cell line and CD8). + b. Reproducibility assessment of the second round of screening at a protein input of 200 ng (H1299 cell line and CD8) + c. Volcano plot of differentially lactated peptides in non-small cell lung cancer tumor and adjacent normal tissue sections; d. Tumor invasion CD8 + T cells and adjacent CD8 cells + Volcano plot of differentially lactylated peptides in T cells.

[0062] Figure 3 Figures show the results of establishing a lactylated PRM targeted detection method suitable for ultra-low input scenarios and validating it with single-cell samples. These include: a. Construction of an ultra-low input lactylated reference spectral library, PRM method scheduling, validation, and compression process and number of samples; b. Trends in the total signal intensity of lactylated peptides under different cell input levels; c. Proportion of lactylated peptides with CV < 30% under different cell input levels; d. Proportion of lactylated peptides with CV < 30% detected under different cell input levels in the total detected peptides; e. Quantitative signal graphs of the representative peptide AEVEGK(lac)DLPEHAVLK under different cell input levels; f. Fragment ion extraction ion chromatograms of the representative peptide AEVEGK(lac)DLPEHAVLK at single-cell levels and with varying input levels.

[0063] Figure 4 Figures show the results of establishing and applying a PRM-targeted detection method for lysine acetylation; including: a. the construction of an acetylation reference spectrum library, the generation, confirmation, and redistribution of the PRM method, and the number of samples; b. the process of generating an acetylation compressed PRM target panel after quality screening, and the number of peptides, modification sites, and proteins in the panel; c. the number of quantifiable peptides, modification sites, and proteins in the acetylation compressed PRM target panel; d. the reproducibility assessment of the acetylation compressed PRM target panel (coefficient of variation distribution); e. a volcano plot of acetylated peptides showing significant changes after TSA treatment of HeLa cells.

[0064] Figure 5 Figures show the results of establishing and applying a PRM-targeted detection method for lysine benzoylation; including: a. the construction of a benzoylation reference spectral library, the generation, confirmation, and redistribution of the PRM method, and the number of samples; b. the process of generating a benzoylation compressed PRM target panel after quality screening, and the number of peptides, modification sites, and proteins in the panel; c. the number of quantifiable peptides, modification sites, and proteins in the benzoylation compressed PRM target panel; d. the reproducibility assessment of the benzoylation compressed PRM target panel (coefficient of variation distribution); e. a volcano plot of significantly changed benzoylation peptides after sodium benzoate treatment of HeLa cells.

[0065] Figure 6 Figures show the results of the establishment and application of the nicotinylation PRM targeted detection method for lysine; including: a. Construction of the nicotinylation reference spectrum library, generation, confirmation, and redistribution of the PRM method and its number; b. Immunoblot images of antibenzoylated lysine in HeLa cell lysates before and after sodium nicotinate treatment, showing upregulation of nicotinylation signal; c. The process of generating a compressed nicotinylation PRM target panel after quality screening, and the number of peptides, modification sites, and proteins in the panel; d. Reproducibility assessment of the compressed nicotinylation PRM target panel (coefficient of variation distribution); e. Volcano plot of nicotinylated peptides that were significantly upregulated after sodium nicotinate treatment of HeLa cells.

[0066] Figure 7 The diagram shows the initial PRM target panel based on DDA identification results of lactylated enriched samples and the PRM validation results without enrichment. Specifically: a) the workflow for lactylated peptide enrichment, DDA data acquisition and identification, initial PRM target panel generation, dynamic scheduling, and PRM acquisition methods; b) the number of PRM detected and 92.16% detection coverage of the initial PRM target panel containing 1,874 lactylated peptides in 200 ng of corresponding HeLa cell whole-protein digested peptides without lactylated antibody enrichment.

[0067] Figure 8This figure shows the grouped PRM target detection and validation results based on the joint analysis of acetylation reference spectral library and published acetylation sites; where: a. the workflow of generating existing publicly annotated target groups and supplementary target groups from the spectral library according to the matching status of publicly available modification site data, and generating initial PRM target panels and dynamic scheduling methods respectively; b. the PRM detection coverage of the two initial PRM target panels in acetyl-NHS chemically modified HeLa samples; c. the PRM detection coverage of the two initial PRM target panels in TSA-induced HeLa samples that have not been enriched with acetylation antibodies. Detailed Implementation

[0068] The following detailed embodiments further illustrate the concept and technical effects of the present invention to fully understand its purpose, features, and effects. Unless otherwise specified, all methods described are conventional methods. Unless otherwise specified, all materials are available from publicly available commercial sources. The illustrative embodiments and descriptions of the present invention are used to explain the invention and do not constitute an undue limitation thereof. It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0069] It should be noted that the core of this invention lies in the development of PRM methods based on a reference spectral library, dynamic scheduling of precursor ions, one or more rounds of quality screening, and application to low-input samples. The reference spectral library can be derived from one or more of natural post-translational modified peptides, simulated post-translational modified peptides, chemically synthesized modified peptides, or publicly available spectral resources. In the following embodiments, if the target modified peptide is obtained through in vitro chemical modification, derivatization modification, or other methods to construct the reference spectral library, this is merely one implementation method for obtaining the reference spectral library, used to support subsequent PRM method development and targeted detection, and does not constitute a limitation on the scope of protection of this invention.

[0070] It should also be noted that, in some embodiments, the compressed PRM target panel formed by the present invention can further correspond to a target protein set and can be used for cross-sectional comparative analysis between samples from different sources. However, the target protein set can be pre-defined based on specific biological problems before method development, or it can be formed by reverse induction based on the proteins corresponding to the retained target targets after the generation of a reference spectral library and one or more rounds of quality screening. Therefore, even if the following embodiments do not explicitly include the step of "pre-selecting a protein set," it does not affect the validity of the technical solution of the present invention regarding the formation of the target target set, the establishment of the compressed PRM target panel, and its subsequent applications. The parameters such as the acquisition time window, minimum residence time, chromatographic peak width, and number of sampling points per peak listed in this specification are example parameters applicable to the chromatographic conditions and mass spectrometer platform in the corresponding embodiments. Those skilled in the art can adjust the above parameters according to the chromatographic column, chromatographic gradient, flow rate, mass spectrometer platform, and number of candidate precursor ions.

[0071] It should be further clarified that the focus of this invention is on the development and application process of the targeted detection method itself, and not limited to the source of a specific modified peptide reference spectral library, a specific dynamic scheduling software, a specific data analysis software, or a specific post-translational modification type. Any equivalent scheme based on the technical ideas described in this invention that uses a spectral library-driven PRM method for the development, scheduling, validation, screening, and low-input sample quantitative analysis of the target post-translational modified peptide should fall within the scope of protection of this invention. The following examples use lactylation, acetylation, benzoylation, and nicotinylation as examples to illustrate the implementation of this invention, but should not be construed as this invention being limited to the above-mentioned modification types. For other lysine acylation modifications, as long as a reference spectral library of the corresponding modified peptide can be obtained, an initial PRM target list or initial PRM target panel can be generated according to the process described in this invention, precursor ion dynamic scheduling and actual PRM validation can be performed, and optionally one or more rounds of quality screening can be conducted to form an empirical PRM target panel or compressed PRM target panel for targeted detection.

[0072] In some embodiments of the present invention, cell lysis buffer, tissue lysis buffer, chemically modified products, enzyme-digested peptide mixtures, naturally enriched samples, or chemically synthesized modified peptides can be used as input materials for reference spectral library construction or PRM method development. For reference spectral library construction, samples from natural sources, in vitro simulated modified samples, chemically derivatized samples, publicly available spectral resources, or combinations thereof can be selected based on the target post-translational modification type. For subsequent PRM method validation and compressed PRM target panel establishment, samples with higher input amounts, population cell samples, or samples with relatively high target modification abundance can be used as pre-screening samples; for final application validation, trace samples, low input samples, clinical tissue sections, puncture biopsy samples, rare cell samples, or single-cell derived samples can be used.

[0073] The method of this invention obtains detection and / or quantitative results of target post-translational modified peptides, which can be used for scientific research, drug evaluation, sample comparison, biomarker research, or to provide intermediate detection information for subsequent analysis. The method of this invention itself does not include steps for directly determining whether a subject has a disease, determining a disease state, or developing a treatment plan based on the detection and / or quantitative results.

[0074] It should be noted that the pre-screened samples are used to verify the detectability, peak quality, and reproducibility of candidate targets, and based on this, a compressed PRM target panel suitable for subsequent sample analysis is formed. In some embodiments, the modified peptides and their corresponding proteins covered by the compressed PRM target panel can be further formed into a target set or a target protein set for cross-sectional comparative analysis between multiple samples from different sources but with the same detection target.

[0075] For low-input cell samples, in one embodiment, the induced cells can be first prepared into a single-cell suspension and then aliquoted according to a predetermined cell number. Specifically, HeLa cells treated with 100 mM sodium lactate for 48 hours can be digested with 0.25% trypsin-EDTA buffer at 37°C for about 4 minutes to obtain a single-cell suspension. After counting using an automated cell counter, the suspension is diluted according to a predetermined cell number gradient. The cells are then resuspended in a lysis buffer, which may contain 1% sodium deoxycholate (SDC), 10 mM tris(2-carboxyethyl)phosphine (TCEP), 40 mM 2-chloroacetamide (CAA), and 100 mM Tris-HCl at pH 8.5. The sample can be homogenized after heating at 95°C for 5 minutes, and the supernatant can be collected by centrifugation at 14,000 g for 5 minutes. The supernatant is enzymatically hydrolyzed overnight at 37°C, and then formic acid is added to a final concentration of about 0.5% to terminate the reaction. After removing SDC by centrifugation at 16,000 g, the sample is desalted using a C18 pipette tip and stored at -80°C for later use.

[0076] For ultra-low input samples based on a single-cell sorting platform, in one embodiment, the cellenONE single-cell sorting system can be used to dispense cells into multi-well plates pre-loaded with a lysis / enzymatic digestion mixture. The lysis / enzymatic digestion mixture may contain 0.2% dodecyl maltodextrin (DDM), 100 mM triethylammonium bicarbonate (TEAB), and 20 ng / μL trypsin. Cells can be screened based on morphological parameters, such as cell diameter of approximately 15–35 μm and elongation not exceeding 1.8, and sorted into 384-well plates at 1–100 cells / well. After sorting, the samples can be incubated at 37°C and approximately 70% humidity for approximately 2 hours, then cooled to approximately 20°C, acidified with 0.1% trifluoroacetic acid (TFA), and a portion of the sample is used for LC-MS analysis.

[0077] For clinical tissue section samples, in one embodiment, tumor tissue and its paired adjacent normal tissue can be embedded in OCT embedding medium, and serial sections can be prepared using a cryostat. Thicker sections can be used for parallel validation experiments such as immunoblotting, while thinner sections can be used for proteomics or targeted mass spectrometry analysis. This applies to tissue section samples and isolated and purified tumor-infiltrating CD8+. + T-cell samples can be processed using the same or similar procedures as those for the low-input samples described above, including lysis, reduction, alkylation, enzymatic digestion, desalting, and LC-MS detection.

[0078] It should be understood that the above sample preparation process is only a preferred embodiment applicable to the low-input application scenario of the present invention. Those skilled in the art can make equivalent adjustments to the lysis system, enzymatic digestion system, desalting method and sample loading method according to the sample type, sample volume and detection platform, without departing from the essence of the present invention.

[0079] General instructions for obtaining immune cell samples In some embodiments of the present invention, the pre-screening sample or the sample to be tested may be derived from peripheral blood CD8. + T cells or tumor-infiltrating CD8 + T cells.

[0080] CD8 from peripheral blood + In one embodiment, T cells can be obtained by collecting peripheral blood from a healthy donor, separating peripheral blood mononuclear cells (PBMCs) through density gradient centrifugation, removing residual red blood cells through erythrocyte lysis, and further employing CD8+... + CD8+ enriched using a T-cell isolation kit for negative sorting or other suitable methods. + T cells. Freshly isolated CD8 cells. +T cells can be activated using anti-human CD3 antibodies and anti-human CD28 antibodies and cultured in a T cell expansion medium; the medium may further contain penicillin, streptomycin and recombinant human interleukin-2 (IL-2) to obtain primary immune cell samples suitable for quality screening or method validation.

[0081] For tumor-infiltrating CD8 + In one embodiment, fresh tumor tissue and paired adjacent normal tissue from a lung adenocarcinoma patient are taken, washed with ice-cold phosphate buffer, weighed, and mechanically minced into small pieces. The tissue is then digested in a culture medium containing collagenase IV and deoxyribonuclease I, for example, at 37°C with shaking for approximately 30 minutes. The digested tissue suspension is then filtered through a cell filter, centrifuged to collect cells, and residual red blood cells are removed by erythrocyte lysis. The resulting cell pellet can be further processed using CD8+, a method suitable for tumor-infiltrating lymphocytes. + T-cell magnetic bead sorting kits or magnetic bead sorting methods were used to enrich and obtain tumor-infiltrating CD8 cells. + T cell samples.

[0082] It should be understood that peripheral blood CD8 + T cells and tumor-infiltrating CD8 + The method of T cell isolation is not limited to the specific reagent systems or equipment platforms mentioned above, as long as CD8 can be obtained to meet the requirements of subsequent low-input proteomics or PRM detection. + T cell samples can all be used to implement this invention.

[0083] Example 1: Establishment and validation of a PRM-targeted detection method driven by a lysine lactylation reference spectral library 1.1 Chemically induced lactylated modified peptides In this embodiment, to obtain a reference spectral library for subsequent PRM method development, samples containing the target modified peptides were prepared using chemical modification, and a reference spectral library of the modified peptides was constructed by acquiring high-resolution LC-MS / MS data. Figure 1 a). It should be noted that the chemical modification library construction method described is only a preferred method for obtaining the reference spectral library, and the development and application of the PRM method of the present invention are not limited to this library construction method.

[0084] HeLa cells were harvested and lysed using lysis buffer containing 200 mM HEPES (pH 8.0) with 8 M urea. The supernatant was collected after centrifugation at 16,000 g for 10 minutes. Protein concentration was determined using the BCA method. The protein solution was diluted to a concentration of 5 μg / μL. 10 μL (50 μg of protein) of the lysis buffer was added to a final concentration of 10 mM with 1 μL of 130 mM TCEP, and incubated at 56 °C for 30 minutes. Then, 1 μL of 260 mM CAA was added to a final concentration of 20 mM, and the mixture was incubated at room temperature for 20 minutes. A laboratory-synthesized lactyl-NHS compound was accurately weighed and prepared as a 130 mM stock solution using anhydrous DMSO. This stock solution was added to the protein lysis buffer to a final concentration of 10 mM. The reaction mixture was incubated at 25 °C with shaking at 400 rpm for 1.5 hours. Add 1 volume of methanol (13 μL) and 1 / 4 volume of chloroform (3.25 μL), mix well and vortex for 10 min. Centrifuge at 16,000 g for 10 min and discard the supernatant. Wash the precipitate with methanol. Redissolve the precipitate in 8 M urea / 200 mM HEPES (pH 8.0) solution. Digest with LysC (1:100, w / w) and trypsin (1:100, w / w) sequentially at 37 °C and 400 rpm for 16 h. Desalt the obtained peptides using C18 ZipTip, dry by vacuum centrifugation, and store at -80 °C for later use.

[0085] 1.2 DDA Data Acquisition: Acquisition of Spectral Data from a Proof-of-Concept Lactylation Reference Spectrum Library HPLC parameters: The HPLC system was used in Trap-elute mode for separation. A nano-level ultra-high performance liquid chromatography (nanoUHPLC) system was employed, first loaded onto a PepMap Neo capture column (300 μm × 5 mm, 5 μm), and then separated on an Aurora Elite CSI C18 column (75 μm × 150 mm, 1.7 μm, IonOpticks). The column oven temperature was set to 50 °C. The mobile phase A was 0.1% formic acid aqueous solution, and the mobile phase B was 0.1% formic acid acetonitrile solution. The gradient elution time was set to 30 minutes. Mass spectrometry parameters: The Orbitrap Astral (Thermo Fisher Scientific) ultra-high resolution mass spectrometry platform was used for data acquisition in DDA mode. The first-stage mass spectrometry (MS1) was acquired in the Orbitrap analyzer with a resolution of 180,000 (FWHM, @200 m / z), a scan range of m / z 300 to 1,600, and an automatic gain control target (AGC target) of 300%. The candidate precursor ions with the highest intensity were selected for secondary fragmentation sequentially, with an isolation window of 1.6 Th, high-energy collisional dissociation (HCD) as the fragmentation mode, and a normalized collision energy (NCE) of 30%. The second-stage mass spectrometry (MS2) fragment ions were detected on the Astral analyzer with a scan range of m / z 100 to 2000 and an AGC target of 200%.

[0086] Data Acquisition: DDA data were collected from chemically induced lactylated modified peptides. Each sample was collected three times and the data were combined for data search.

[0087] Data Search: The obtained raw data was searched against the UniProt human protein database using the SEQUEST search engine built into Proteome Discoverer 3.1. Lysine lactylation was defined as a variable modification, cysteine ​​aminomethylation as a fixed modification, and methionine oxidation as a variable modification. Peptides were identified at a false discovery rate (FDR) of 1%. A proof-of-concept lactylation reference spectral library was constructed, including precursor ion m / z, secondary fragment ion spectra, chromatographic retention times, and modification site annotations. In this embodiment, the reference spectral library is named ChemIntelligence Klac Library, containing 22,388 unique lactylated peptides, corresponding to 19,413 lactylation modification sites on 2,826 proteins, for subsequent development of PRM methods. Figure 1 a).

[0088] 1.3 Import the reference spectrum library into the targeted mass spectrometry analysis software and generate a PRM target list. A PRM target list was generated using targeted mass spectrometry analysis software. In this embodiment, Skyline (v24.1.1.398) was used. The parameter settings are as follows: Peptide parameters: set the protease to Trypsin [KR / P], allowing a maximum of 3 missed cleavage sites; set the peptide length range to 6 to 45 amino acids; enable the "Auto-select all matching peptides" option.

[0089] Modification parameters: The fixed modification is carbamidomethylation of cysteine; the variable modification is lactylation of lysine (+72.0211 Da) and oxidation of methionine. A maximum of 3 variable modifications are allowed, and a maximum of 3 neutral losses are allowed.

[0090] Product ion parameters: precursor ion charge states are set to +2 and +3; product ion types are b ions and y ions; product ion charge states are set to +1 and +2; product ion matching tolerance is set to 20 ppm; the full scan is set using DDA parameters: MS1 screening is performed using the Orbitrap instrument with a resolution of 180,000@200 m / z, and MS / MS screening is performed using 20 ppm; the retention time window is set to 2 minutes, and the measured value is used first when the actual retention time is available.

[0091] Product ion screening: 15 product ions were selected for each precursor, and the library matching tolerance was set to 0.5 Da.

[0092] Data Import: After completing the above parameter configuration, export the DDA data search results corresponding to the proof-of-concept lactylation reference spectrum library in pepXML format, screen for peptides with a Percolator q-value < 0.01, and import them into the targeted mass spectrometry analysis software (Skyline v24.1.1.398 in this embodiment) to construct the reference library (BiblioSpec, i.e., the .blib file in this embodiment). Perform peak shape extraction on all candidate precursors, exclude precursor ions with unidentifiable chromatographic peaks, and obtain a list of effective candidate precursors for subsequent dynamic scheduling of precursor ions.

[0093] 1.4 Generation of precursor ions through dynamic scheduling and PRM acquisition method The aforementioned reference library file is invoked, and the precursor ion dynamic scheduling module is used to dynamically schedule all valid precursors in the standard dosage lactylation reference spectrum library. In this embodiment, the precursor ion dynamic scheduling module is implemented using PRMConductor (v1.0.4344).

[0094] The parameters are set as follows: Select "Keep All Precs"; set the acquisition time window for each precursor to 1.4 minutes; set the minimum residence time to 10 ms; set the peak width to 10 seconds; ensure at least 7 acquisition points for each peak; retain only a single charge state for each precursor.

[0095] Following the above scheduling, 17,543 precursor ions from the proof-of-concept lactylation reference spectral library were assigned to 18 independent PRM methods, each of which can be imported into the mass spectrometer control software. Figure 1 a) Since these methods are based on computer simulation scheduling generated from the DDA spectral library, they need to be verified and calibrated through actual PRM acquisition.

[0096] 1.5 PRM Data Acquisition: Validation Injection and Sample Detection Liquid chromatography parameters: The liquid chromatography method used for PRM acquisition is completely consistent with that used for DDA acquisition. Mass spectrometry parameters: PRM data acquisition was performed using a Stellar mass spectrometer. Adaptive Retention Time (RT) was enabled. Under different sample matrices, processing methods, or formulations, it is preferable to first generate an rtbin file for retention time calibration and then use this rtbin file in subsequent relative quantification data acquisitions.

[0097] MS1 scans were performed in the m / z range of 300–1,600 at a scan rate of 33 kDa / s. Maximum Injection Time was set to custom mode, corresponding to 10 ms. Automatic Gain Control (AGC) was set to standard mode, with an absolute AGC target of 3 × 10⁻⁶. 4 .

[0098] MS2 scanning used HCD for fragmentation, with NCE set to 30%. The MS2 scan range was 200–1,500 m / z, and the scan rate was 125 kDa / s. The AGC target was set to standard, and the absolute AGC target was 1×10⁻⁶. 4 The maximum ion implantation time is set to dynamic mode. Loop control is set to "All".

[0099] Data acquisition: Chemically induced lactylated peptide samples (i.e., HeLa samples chemically modified with lactyl-NHS, the same as those used in library construction) were individually validated by single injections using 18 newly generated PRM methods on a Stellar mass spectrometer. The initial PRM acquisition methods generated based on the DDA reference spectral library were validated and calibrated on the actual platform.

[0100] 1.6 PRM Data Import and Quality Control Screening: Method Confirmation and Compression The raw data acquired by PRM is imported into the targeted mass spectrometry analysis software for peak extraction. The software parameters in this stage remain consistent with those in the target list generation stage, and are adjusted appropriately based on the characteristics of the PRM data, as follows: Peptide and modification parameters: The same peptide and modification parameters as in step 1.2 are used, i.e. the protease is set to Trypsin [KR / P], and a maximum of 3 missed cleavage sites are allowed; the fixed modification is cysteine ​​aminomethylation, and the variable modification is lysine lactylation and methionine oxidation.

[0101] Product ion parameter adjustments: Precursor ion charge states are set to +2 and +3; product ion types are b ions and y ions; product ion charge states are set to +1 and +2; product ion quantity is set to retain 15 product ions per precursor; ion matching tolerance is adjusted from 20 ppm in the DDA stage to 0.5 Da (to adapt to the actual resolution requirements of Orbitrap detection in PRM acquisition); full scan settings are switched to PRM mode: MS1 screening is disabled, MS2 screening is set to PRM mode, matching tolerance is set to 0.5 Da; retention time window is set to 2 minutes.

[0102] Data import and screening: After importing the raw data obtained from PRM into the targeted mass spectrometry analysis software to complete peak extraction, precursor ions with no identifiable chromatographic peaks are first excluded, and then the precursors are initially screened according to the preset quality standards. For precursors that pass the screening, the method allocation is further re-utilized using the precursor ion dynamic scheduling module to generate an optimized PRM method suitable for the actual detection platform.

[0103] Parameter settings: Select "Keep All Precs"; set the acquisition time window for each precursor to 1.4 minutes; set the minimum residence time to 10 ms; set the peak width to 9 seconds; ensure at least 8 acquisition points per peak; retain only a single charge state for each precursor. After a single validation injection and peak extraction screening, the detectable precursors were reassigned, and the proof-of-concept lactylated ChemIntelligence Library ultimately identified 17,494 precursor ions, which were reassigned to 21 PRM methods (…). Figure 1 a).

[0104] 1.7 Method Application Validation: Dynamic Quantification of Lactylation in HeLa Cells under Lactate Stimulation PRM detection and first-round quality screening: HeLa cell positive samples treated with sodium lactate (final concentration 100 mM) for 48 hours were used as pre-screening samples, and the above 21 PRM methods were run on a Stellar mass spectrometer.

[0105] Data Import and Screening: The collected raw data was imported into Skyline. Peak shape identification was performed according to the parameters of the PRM data import and peak extraction stage in Example 1.6, and precursors with no identifiable chromatographic peaks were excluded. The quality screening results were evaluated according to the following criteria: absolute peak area > 500; signal-to-background ratio (S / B) > 3; relative abundance > 0.05; time correlation > 0.8; number of effective product ions ≥ 3.

[0106] Parameter settings: Based on this, PRM Conductor rescheduled these precursors, setting the acquisition time window to 1.4 minutes, the minimum residence time to 10 ms, the peak width to 10 seconds, and at least 7 acquisition points per peak. A balanced load mode was used for compression, resulting in a compressed PRM target panel containing 1,122 lactylated peptides for subsequent quantitative analysis. Figure 1 b).

[0107] Quantitative analysis: HeLa cells were treated with sodium lactate (L-lactate, final concentration 100 mM) for 48 hours as a positive stimulation group, and untreated cells were used as a control group. Each group was a triple biological replicate, and the protein input per replicate was 200 ng (this amount is only an example, and the present invention is also applicable to lower or slightly higher protein input amounts. The 200 ng amount demonstrates its effectiveness in low input scenarios). Peptides were prepared according to the standard trypsin digestion process, purified by C18 desalting column, and data were collected using a compressed PRM target panel finally established in Example 1.

[0108] Maintaining the PRM settings in Skyline, the raw mass spectrometry data were imported into Skyline, peak areas were extracted, and normalization was performed using total ion current (TIC). Records of peptides that were stably detected across all paired samples, had no truncation peaks, and were uniquely gene-specific peptides were further selected only when peptide signals were assigned to specific genes or proteins and downstream differential statistics were performed; this unique peptide limitation was not used as a constraint on the composition or membership of the PRM target panel. For differentially modified peptides in cellular samples, a two-tailed Student t-test combined with Benjamini-Hochberg correction was used for statistical analysis.

[0109] Repeated assays on sodium lactate-treated HeLa samples showed that 762 out of 1,122 peptides targeted by the compressed PRM target panel exhibited excellent analytical reproducibility (CV < 30%), demonstrating the high precision and reliability of the method. Figure 1 c).

[0110] Quantitative analysis showed that, compared with the control group, 48 lactylated peptides underwent significant changes after lactate treatment (65 upregulated and 3 downregulated, with fold changes >2.0 or <0.5, and corrected P<0.05). Figure 1 d).

[0111] This embodiment demonstrates that, for amplifiable cell line samples, the PRM method developed in this invention can effectively establish a compressed PRM target panel through a single round of quality screening without labeling or antibody enrichment of the test samples, achieving high-throughput and highly reproducible targeted quantitative analysis. This method features high sensitivity (requiring only 200 ng protein input), high reproducibility (CV < 30%), and high throughput (>800 quantifiable peptides), providing a powerful and reliable technical platform for studying lactylation modifications in low-input samples and rare clinical samples.

[0112] Example 2: A two-round screening method for lactylated PRM in clinical samples based on two rounds of cell line quality screening 2.1 Method Design In actual clinical research, protein materials derived from tumor tissue sections or immune cells in the tumor microenvironment are usually very limited, and conventional antibody enrichment strategies often require high protein input, making them unsuitable for low-input clinical samples. To address these issues, this invention constructs a targeted PRM detection strategy based on a reference spectral library. By combining a first-round quality screening of amplifiable cells with a second-round quality screening of clinical samples, high-throughput and stable detection of lactylation modifications can be achieved under extremely low sample input conditions.

[0113] The core idea of ​​this strategy is as follows: First, representative cell line samples are used to perform PRM validation on candidate precursor ions in the reference spectral library to screen for candidate peptides with stable signal responses in biological samples. Then, a precursor ion dynamic scheduling module is used to reschedule the candidate precursors and construct an optimized PRM method. Finally, one or more rounds of quality screening are used to obtain a high-quality target list suitable for clinical samples. In this embodiment, two rounds of screening are preferably used as an example for illustration; the reference spectral library used is named ChemIntelligence Klac Library; the precursor ion dynamic scheduling module is implemented through PRM Conductor.

[0114] 2.2 First round of screening: based on expandable cells The H1299 human non-small cell lung cancer cell line was used as the representative cell line for the first round of pre-quality screening of clinical lung cancer tissue samples. After inducing H1299 cells with 80 mM sodium lactate for 48 hours, a simulated lactylated digest of H1299 cells was obtained. After enzymatic desalting according to the standard procedure, the full-library PRM method (21 methods, covering 17,494 precursors) established in Example 1 was used for PRM detection on Stellar. The protein input was 200 ng (as an example of a low-input sample).

[0115] Data Import and Screening: Raw data was imported into Skyline and peak shape identification was performed using the same parameters as PRM, excluding precursors with unidentifiable chromatographic peaks. PRM Conductor was run to evaluate the pre-screening results using the following criteria: absolute peak area > 500; signal-to-background ratio (S / B) > 3; relative abundance > 0.05; time correlation > 0.8; effective product ion count ≥ 3. After the first round of screening, a total of 1,628 precursors meeting the above criteria were found in H1299 cells. Figure 2 a).

[0116] Parameter settings: Based on this, PRM Conductor rescheduled 1,628 precursors, set the peak width to 9 seconds, the residence time to 10 ms, at least 8 acquisition points per peak, and the acquisition window to 1.4 minutes, and assigned them to 3 independent PRM methods as detection methods for the subsequent second round of screening.

[0117] For primary immune cell pre-screening samples, in one embodiment, peripheral blood from healthy donors is collected, and PBMCs are obtained by density gradient centrifugation, followed by further separation and purification of CD8. + T cells were activated and expanded in vitro using an anti-CD3 / anti-CD28 stimulation system and a culture system containing IL-2. Subsequently, the peripheral blood-derived CD8 cells were treated with 40 mM sodium lactate. + T cells, after 48 hours of enzymatic desalting according to standard procedures, were used as representative primary immune cell samples for pre-quality screening. Following the same PRM detection and quality screening process as the H1299 samples described above, 2,050 valid precursors were obtained and allocated to three PRM methods, serving as the basis for subsequent clinical immune cell sample screening and detection. Figure 2 a).

[0118] 2.3 Second round of screening: Screening based on technical repeatability and reproducibility A second round of testing was performed on tumor cell line samples using a triple-technology repeat method to screen for precursor ions with a CV < 30%.

[0119] For samples from the H1299 source, the raw data obtained from three technical repetitions were imported into Skyline, peak shape identification was performed using the same parameters, and the following two quality control criteria were added: (1) Precursors with a CV > 30% in three technical replicates were excluded; (2) Exclude precursors that have truncated peaks.

[0120] In the second round of screening, the parameters of the precursor ion dynamic scheduling module can be further adjusted to: minimum residence time of 15 ms, peak width of 8 seconds, at least 7 collection points per peak, and collection window of 1.4 minutes. After the second round of fine screening, 1,255 high-quality, quantifiable precursors without truncation peaks were ultimately retained from the H1299-derived clinical samples and allocated to 3 final clinical PRM methods, which can be used for batch detection of large cohorts of clinical samples. Figure 2 ab).

[0121] For CD8 + For T cell samples, the parameters of the precursor ion dynamic scheduling module can be adjusted as follows: minimum residence time 12ms, peak width 8 seconds, at least 7 collection points per peak, and collection window 1.4 minutes. Using a balanced loading mode, a compressed PRM target panel was finally obtained, containing 771 quantifiable precursors without truncation peaks. Figure 2 ab).

[0122] 2.4 Detection of PRM for Lactylation in Clinical Samples Using the PRM method obtained from the above screening, clinical lung cancer tissue, adjacent normal tissue, and tumor-infiltrating CD8 were analyzed. + T cells and adjacent CD8 cells + T cell samples were subjected to lactylation detection.

[0123] Sample preparation: In one embodiment, paired tumor tissue and adjacent normal tissue from three patients with non-small-cell lung cancer (NSCLC) were taken, embedded in optimal cutting temperature (OCT) embedding medium, and serial tissue sections were prepared using a cryostat.

[0124] In one implementation, tumor-infiltrating CD8 + T cells are derived from fresh tumor tissue and paired adjacent normal tissue. Specifically, tissue samples are washed with ice-cold PBS, minced, and then enzymatically digested at 37°C in a medium containing collagenase IV and deoxyribonuclease I. The digested cell suspension is filtered, centrifuged, and residual red blood cells are removed before further processing using CD8+. +T-cell magnetic bead sorting method enriches tumor-infiltrating CD8 cells. + T cells were used for subsequent low-input sample preparation and PRM detection.

[0125] 10 μm thick tissue sections and purified tumor-infiltrating CD8 + T cell samples are preferably processed using the same or similar preparation procedures as low-input samples.

[0126] For low-input tissue section samples and CD8 + In one embodiment, T cell samples were added to a lysis buffer at pH 8.5, the lysis buffer containing 1% SDC, 10 mM TCEP, 40 mM CAA, and 100 mM Tris-HCl. The samples were homogenized after heating at 95°C for 5 minutes and centrifuged at 14,000 g for 5 minutes to remove insoluble matter. The supernatant was collected and enzymatically digested overnight at 37°C. After digestion, formic acid was added to a final concentration of approximately 0.5% to terminate the reaction, and the samples were centrifuged at 16,000 g to remove SDC. The resulting supernatant was desalted using a C18 electrolyte and used for subsequent PRM assays.

[0127] In some implementations, to ensure consistency in testing between different clinical samples, tissue section samples and CD8... + The protein input for T-cell samples was controlled at approximately 200 ng per injection. Raw mass spectrometry data were imported into Skyline, peak area was extracted, and TIC normalization was performed. Only when peptide signals were assigned to specific genes or proteins and downstream differential statistics were performed, records of peptides that were stably detected in paired samples, had no truncation peaks, and were uniquely gene-specific were further selected; this unique peptide restriction was not used as a limitation on the composition or membership of the PRM target panel. For differential analysis between tumor and adjacent normal tissue samples, a two-tailed paired t-test was performed on the log2-transformed tumor / adjacent normal tissue ratio; when the fold change in tumor / adjacent normal tissue ratio was >2 or <0.5 and P <0.05, it was considered a candidate for significantly changed lactylated peptides.

[0128] Quantitative analysis: Lactylation analysis of frozen tissue sections from non-small cell lung cancer (NSCLC) showed that targeted quantitative analysis of three pairs of paired NSCLC tumor and adjacent normal tissue sections stably quantified 890 unique lactylated peptides across 593 proteins, containing 1,115 modification sites. Of these, 163 peptides showed significantly upregulated lactylation levels, while 7 peptides showed significantly decreased lactylation levels. Figure 2 c).

[0129] Targeting tumor-infiltrating CD8 + Lactylation analysis of T cells showed that, through the analysis of three pairs of paired NSCLC tumors and adjacent invasive CD8 cells,+ Targeted quantitative analysis of T cells stably quantified 476 unique lactylated peptides across 588 proteins, containing 1,141 modification sites. Among these, 28 peptides showed significantly upregulated lactylation levels, corresponding to 31 modification sites on these 28 proteins. Figure 2 d).

[0130] 2.5 Summary of Methods This embodiment demonstrates the significant advantages of the "two-round quality screening" strategy constructed in this invention for clinical tissue sections and rare immune cell samples with extremely limited protein input. The first round of screening utilizes expandable cell lines or expanded primary cells as pre-screening samples, selecting candidate precursors with reliable signals in biological samples from a large-scale initial target pool based on criteria such as signal intensity, signal-to-noise ratio, and co-elution correlation. The second round of screening involves technically repeating the pre-screening samples, using CV no higher than 30% and peak integrity as criteria to further screen for high-quality precursors with high reproducibility. The compressed PRM target panel generated through these two rounds of screening enables stable and highly reproducible targeted quantitative analysis of lactylation modifications in tissue sections and rare immune cell samples with protein inputs as low as 200 ng, without antibody enrichment or isotope labeling of the clinical samples to be tested. This "two-round screening" process is a typical implementation of the "multi-round screening" described in this invention, and is crucial for ensuring the detection success rate and data reliability of low-input and complex clinical samples.

[0131] Example 3: Establishment and Validation of a Lactylation PRM Targeted Detection Method Applicable to Ultra-Low Input Scenarios 3.1 Establishment of the ChemIntelligence Library with Ultra-Low Investment Liquid chromatography parameters for DDA data acquisition in ultra-low input samples: To support the targeted detection of emulsification in ultra-low input samples, the data acquisition method during the reference spectral library construction phase was optimized. Liquid chromatography was performed in direct injection mode, i.e., without using a trapping column; other settings were essentially the same as conventional DDA settings.

[0132] Mass spectrometry parameters: The Orbitrap Astral Zoom mass spectrometer is connected to the FAIMS Pro module, and the compensation voltage (CV) is set to... 48 V. DDA acquisition mode, other parameters remain the same as constant settings. Enable pre-accumulation mode.

[0133] Data acquisition: DDA data were acquired on the platform described above using the same chemically modified lactylated samples as in Example 1 to construct a lactylated reference spectrum library optimized for ultra-low input scenarios.

[0134] Data search: Similar to Example 1, the SEQUEST engine was used for searching. The final result was a lactylation reference spectral library suitable for ultra-low-input scenarios; in this embodiment, this reference spectral library is named the ultra-low-input ChemIntelligence Klac Library, containing 26,835 lactylated peptides, corresponding to 24,588 lactylation sites of 3,721 proteins. Figure 3 a).

[0135] 3.2 Establishing a PRM Method for Extremely Low Investment Scenarios Based on the ultra-low input lactylation ChemIntelligence Library, the spectral library was imported according to the Skyline parameters described in Example 1 to generate a PRM target list. The target list was imported into PRM Conductor, and precursor ions were assigned according to the following parameters: minimum residence time 10 ms; peak width 10 seconds; at least 7 acquisition points per peak; acquisition window 1.4 minutes; each precursor retaining only a single charge state. After scheduling, 18,157 precursors from the low-input lactylation reference spectral library were assigned to 15 independent PRM methods.

[0136] 3.3 PRM Method Confirmation and Reallocation Subsequently, single-shot validation injections were performed on these PRM methods using lactyl-NHS-modified HeLa lysis buffer on a Stellar mass spectrometer, followed by peak extraction and quality screening using Skyline. After excluding precursors with no identifiable chromatographic peaks under the validation conditions, the target list was imported into PRM Conductor for reassignment; ultimately, 18,012 precursors yielded identifiable chromatographic peaks in this validation data and were reassigned to 19 PRM methods. Figure 3 a).

[0137] 3.4 PRM Method Validation PRM Detection and First-Round Quality Screening: For ultra-low input scenarios, due to the extremely limited available materials for a single sample, it is usually difficult to repeatedly collect a large number of independent PRM methods on the same sample. Therefore, it is preferable to first use a reference sample with a higher input volume to perform quality screening on candidate targets. In this embodiment, HeLa cell enzymatic digestion peptides treated with 100 mM sodium lactate for 48 hours were used with a protein input of 200 ng (as an example of a low input sample) and the 19 PRM methods established in Example 3.2 were run on a Stellar mass spectrometer to validate candidate precursors in the ChemIntelligence Library with ultra-low input volume lactylation.

[0138] Data import and screening: Import the data into Skyline for analysis and screen according to the following criteria: absolute peak area > 500, signal-to-background ratio (S / B) > 3, relative abundance > 0.05, time correlation > 0.8 and effective product ion number ≥ 3.

[0139] Parameter settings: Minimum residence time 10 ms, peak width 9 seconds, at least 7 acquisition points per peak, acquisition window 1.4 minutes, each precursor retaining only a single charge state, and a balanced loading mode were used to ultimately form a compressed PRM target panel containing 1,224 lactylated peptides (covering 1,533 modification sites, 1,003 proteins) and its corresponding single PRM acquisition method. Figure 3 a).

[0140] 3.5 Suitability Validation of Single-Cell Derivative Samples To evaluate the applicability of the method of the present invention in samples with extremely low input and samples derived from single cells, it was validated using gradient cell number samples and samples prepared based on a single cell sorting platform.

[0141] Sample preparation: In one embodiment, HeLa cells treated with 100 mM sodium lactate for 48 hours were digested with 0.25% trypsin-EDTA buffer at 37°C for approximately 4 minutes to prepare a single-cell suspension, and counted using an automated cell counter. Subsequently, the cells were diluted to 10³, 10⁻⁶, and 10⁻⁶ cells per cell number gradient according to a predetermined cell number gradient. 4 10 5Cell samples were collected. Next, the cells were resuspended in 100 μL of lysis buffer, pH 8.5, containing 1% SDC, 10 mM TCEP, 40 mM CAA, and 100 mM Tris-HCl. The samples were homogenized after heating at 95°C for 5 minutes, centrifuged at 14,000 g for 5 minutes, and the supernatant was collected and enzymatically digested overnight at 37°C. After digestion, formic acid was added to a final concentration of approximately 0.5% to terminate the reaction. The cells were then centrifuged at 16,000 g to remove SDC, desalted using C18 chromatography, and stored at -80°C for later use.

[0142] In another embodiment, for ultra-low input samples of 1–100 cells, the cellenONE single-cell sorting system can be used to sort HeLa cells treated with sodium lactate into 384-well plates. The 384-well plates are pre-filled with 1 μL of master mix containing 0.2% DDM, 100 mM TEAB, and 20 ng / μL trypsin. Sorting can be performed based on cell morphology parameters, for example, selecting cells with a diameter of approximately 15–35 μm and an elongation not exceeding 1.8, and allocating them to the wells according to a predetermined cell number. After sorting, the samples are incubated at 37°C and approximately 70% humidity for approximately 2 hours, then cooled to approximately 20°C, acidified with 3.5 μL of 0.1% TFA, and approximately 4.5 μL of the sample is taken for LC-MS analysis.

[0143] Method application: For 10 4 and 10 5 For cell samples, the protein input for a single test can be controlled to approximately 200 ng using conventional quantitative methods (as an example of a low-input sample); for serially diluted and desalted samples, a conventional LC-MS loading procedure is used; for ultra-low-input samples obtained by cellenONE sorting, loading and detection are performed after acidification.

[0144] In PRM detection, the overall signal intensity gradually decreases as the number of cells decreases. Figure 3 b) The overall CV of the peptide signal increases ( Figure 3 c). Under the single-cell-derived sample preparation and detection conditions of this embodiment, a total of 1,085 lactylated peptides were detected, of which 295 peptides had a CV < 30% ( Figure 3 d). Taking the lactylated peptide AEVEGK(lac)DLPEHAVLK at the K626 site of HNRNPU protein as an example, this peptide was detected in all cell input groups. Figure 3 e), its fragment ion extraction ion chromatogram increases with increasing cell input and remains identifiable in single-cell source samples. Figure 3 f).

[0145] In summary, by optimizing the construction of the reference spectral library and the scheduling of the PRM method for ultra-low input scenarios, this invention detected 1,085 lactylated peptides under the single-cell-derived sample preparation and detection conditions, of which 295 peptides had repeatability with a CV of <30%. These results demonstrate the potential of the method for targeted detection of lactylation in single-cell-derived samples; however, the detection coverage may vary under different sample preparation platforms and mass spectrometry conditions.

[0146] Example 4: Establishment and application of a PRM-targeted detection method for lysine acetylation. 4.1 Construction of the acetylation reference spectral library Following the construction procedure of the lactylated reference sample in Example 1, the sample was chemically modified using acetyl-NHS, with a final concentration of 3 mM. The acetylated peptides in the spectral library exhibit a characteristic cyclic iminium ion (m / z 126.0913). After enzyme digestion, desalting, and high-resolution LC-MS / MS acquisition, spectral data of the acetylated modified peptides were obtained. Database searching was performed using the SEQUEST search engine, targeting the UniProt human protein database. Lysine acetylation was set as a variable modification, and peptide identification and annotation were completed at a 1% FDR level to construct an acetylated reference spectral library. In this embodiment, this reference spectral library is named ChemIntelligence KacLibrary, containing 23,702 acetylated peptides, corresponding to 2,921 acetylated proteins and 20,047 acetylation sites (…). Figure 4 a).

[0147] 4.2 Generation of acetylated PRM target list Referring to the parameter settings in step 1.3 of Example 1, the acetylation reference spectrum library was imported into the Skyline targeted mass spectrometry analysis software: the protease was set to Trypsin [KR / P], allowing a maximum of 3 missed cleavage sites; the precursor ion charge states were set to +2 and +3; the product ion types were b ions and y ions, with 15 product ions selected for each precursor; the retention time window was set to 2 minutes. Based on this, the acetylation modification was set to a mass increment of +42.0106 Da on lysine residues, generating an acetylated PRM target list.

[0148] 4.3 Dynamic scheduling of precursor ions and establishment of methods Referring to step 1.4 of Example 1, the list of acetylated PRM targets was imported into PRM Conductor. Precursor ion dynamic scheduling was performed using the same scheduling parameters as in Example 1 (acquisition time window 1.4 minutes, minimum residence time 10 ms, peak width 12 seconds, at least 8 acquisition points per peak), resulting in the allocation of 16,662 peptides to 17 PRM methods. Figure 4 a).

[0149] 4.4 PRM Method Confirmation and Reallocation The liquid chromatography and mass spectrometry parameters were completely consistent with those in step 1.5 of Example 1. HeLa cell enzymatically digested peptides modified with acetyl-NHS were used as validation samples. The 17 PRM methods generated in step 4.3 were used for single-injection validation on a Stellar mass spectrometer to verify and calibrate the initial PRM acquisition method generated based on the DDA reference spectral library on a real-world platform.

[0150] The collected raw data was imported into Skyline for peak extraction, with parameter settings identical to step 1.6 of Example 1. Precursor ions with unidentifiable chromatographic peaks were excluded. Then, the method allocation was recalculated using the precursor ion dynamic scheduling module (acquisition time window 1.4 minutes, minimum residence time 10 ms, peak width 9 seconds, at least 8 acquisition points per peak). After a single validation injection and peak extraction screening, a total of 16,662 peptides were identified and allocated to 22 PRM methods (…). Figure 4 a).

[0151] 4.5 PRM Method Validation PRM detection and first-round quality screening: To validate the target and select high-response peptides suitable for low-input samples, HeLa cell enzymatically digested peptides treated with 1 μM TSA for 24 hours were used as positive samples, and the above 22 acetylation PRM detection methods were run under the condition of protein loading of 200 ng (as an example of low-input samples).

[0152] Data import and screening: Import the data into Skyline for analysis and perform quality control screening according to the following criteria: absolute peak area > 500, signal-to-background ratio (S / B) > 3, relative abundance > 0.05, time correlation > 0.8 and effective product ion number ≥ 3.

[0153] Parameter settings: Based on this, PRM Conductor rescheduled these precursors, setting the acquisition time window to 1.4 minutes, the minimum residence time to 10 ms, the peak width to 15 seconds, and at least 8 acquisition points per peak. It then used a balanced load mode for compression, ultimately resulting in a compressed PRM target panel containing 1,369 acetylated peptides (covering 1,683 modification sites, 925 proteins). Figure 4 bc).

[0154] Method reproducibility evaluation: The compressed PRM target panel described above was used to perform three technical replicates on TSA-treated HeLa samples. The results showed that 1,168 of the 1,369 peptides targeted by the compressed PRM detection method exhibited excellent analytical reproducibility (CV < 30%), demonstrating the high accuracy and reliability of the method. Figure 4 d).

[0155] Quantitative analysis: HeLa cells were treated with TSA (final concentration 1 μM) for 24 hours as the positive stimulation group, and untreated cells were used as the control group. Each group was a triple biological replicate, with a protein input of 200 ng (as an example of a low-input sample). The acetylation compressed PRM target panel established above was used for detection. Quantitative data were statistically compared after TIC normalization in Skyline. For quantifying endogenous acetylation levels, a two-sided Student t-test combined with BH correction was used for statistical analysis. The results showed that compared with the control group, TSA treatment significantly upregulated 103 acetylated peptides and significantly downregulated 49 acetylated peptides (fold change >1.5 or <0.67, corrected P<0.05). Figure 4 e).

[0156] This embodiment demonstrates that, for the classic acylation modification type of lysine acetylation, the present invention can also achieve rapid development of a PRM targeting method based on a reference spectral library. Through a single round of quality screening, a large-scale target covering more than 20,000 modification sites was successfully compressed into a single PRM detection method containing 1,369 high-confidence acetylated peptides. This method exhibits excellent quantitative reproducibility at a protein input of 200 ng (>85% of targeted peptides CV < 30%), and successfully identified 152 acetylated peptides that underwent significant changes before and after TSA treatment. These results prove that the present invention is suitable for high-throughput, high-sensitivity targeted quantitative analysis of classic modifications, providing a powerful tool for the study of acetylation dynamics in low-input samples.

[0157] Example 5: Establishment and application of a PRM-targeted detection method for lysine benzoylation. 5.1 Benzoylation: Construction of the ChemIntelligence Library Following the construction procedure of the lactylated reference sample in Example 1, the sample was chemically modified with laboratory-synthesized benzoyl-NHS at a final concentration of 10 mM. After enzyme digestion, desalting, and high-resolution LC-MS / MS data acquisition, spectral data of benzoylated peptides were obtained. Using the SEQUEST search engine, a search was conducted on the UniProt human protein database. Lysine benzoylation was set as a variable modification, and peptide identification and annotation were completed at the 1% FDR level to construct a benzoylated reference spectral library. In this embodiment, this reference spectral library is named ChemIntelligence KbzLibrary, containing 16,874 benzoylated peptides (corresponding to 2,756 benzoylated proteins and 15,031 benzoylation sites), forming a high-coverage benzoylated reference spectral library. Figure 5 a).

[0158] 5.2 Skyline Parameter Settings and PRM Target List Generation Referring to the parameter settings in step 1.3 of Example 1, the benzoylation reference spectrum library was imported into the Skyline targeted mass spectrometry analysis software: peptide length 6-45 amino acids, precursor charge states +2 and +3, product ions b ions and y ions (charge states +1 and +2), 15 product ions were selected for each precursor, and the retention time window was 2 minutes. The benzoylation modification was set to a mass increment of +104.0262 Da on lysine, and a list of benzoylation PRM targets was generated.

[0159] 5.3 PRM Conductor Scheduling and Method Establishment Referring to step 1.4 of Example 1, the list of benzoylation PRM targets was imported into PRM Conductor. Precursor ion dynamic scheduling was performed using the same scheduling parameters as in Example 1 (acquisition time window 1.4 minutes, minimum residence time 10 ms, peak width 12 seconds, at least 8 acquisition points per peak, and each precursor retaining only a single charge state). A total of 11,940 peptides were assigned to 19 PRM methods. Figure 5 a).

[0160] 5.4 PRM Method Confirmation and Reallocation The liquid chromatography and mass spectrometry parameters were the same as in step 1.5 of Example 1. HeLa cell enzymatically digested peptides modified with benzoyl-NHS were used as validation samples. Multiple PRM methods generated in step 5.3 were used for single-injection validation on a Stellar mass spectrometer to verify and calibrate the initial PRM acquisition method generated based on the DDA reference spectral library on a real-world platform.

[0161] The collected raw data was imported into Skyline for peak extraction, with parameter settings identical to step 1.6 of Example 1. Precursor ions with unidentifiable chromatographic peaks were excluded. Then, the method allocation was recalculated using the precursor ion dynamic scheduling module (acquisition time window 1.4 minutes, minimum residence time 10 ms, peak width 9 seconds, at least 8 acquisition points per peak). After a single validation injection and peak extraction screening, a total of 11,940 peptides were identified and allocated to 24 PRM methods (…). Figure 5 a).

[0162] 5.5 PRM Method Validation PRM detection and first-round quality screening: After treating HeLa cells with 60 mM sodium benzoate for 24 hours, the enzymatically digested peptides were used as positive samples. The above 24 benzoylation PRM detection methods were run under the condition of a protein loading amount of 200 ng (as an example of a low input sample).

[0163] Data import and screening: Import the data into Skyline for peak extraction and quantitative analysis, and perform quality control screening according to the following criteria: absolute peak area > 500, signal-to-background ratio (S / B) > 3, relative abundance > 0.05, time correlation > 0.8, and number of effective product ions ≥ 3.

[0164] Parameter settings: Based on this, PRM Conductor rescheduled these precursors, setting the acquisition time window to 1.4 minutes, the minimum residence time to 10 ms, the peak width to 14 seconds, and at least 8 acquisition points per peak. It also used a balanced load mode for compression, ultimately resulting in a compressed PRM target panel containing 1,020 benzoyl peptides. Figure 5 bc).

[0165] Method reproducibility evaluation: The compressed PRM target panel described above was used to perform three technical reproducibility analyses on HeLa samples treated with sodium benzoate, with a CV < 30% used as the reproducibility evaluation criterion. The results showed that among the 1,020 peptides targeted by the compressed PRM detection method, 877 peptides exhibited excellent analytical reproducibility (CV < 30%), demonstrating the high precision and reliability of the method. Figure 5 d).

[0166] Quantitative analysis: HeLa cells were treated with sodium benzoate (final concentration 60 mM) for 48 hours as the positive stimulation group, and untreated cells were used as the control group. Each group was a triple biological replicate, with a protein input of 200 ng (as an example of a low-input sample). The benzoylation compressed PRM target panel established above was used for detection. Quantitative data were statistically compared after TIC normalization in Skyline. For quantifying endogenous benzoylation levels, a two-sided Student t-test combined with BH correction was used for statistical analysis. The results showed that, compared with the control group, sodium benzoate treatment significantly upregulated 63 benzoylation peptides and significantly downregulated 26 benzoylation peptides (fold change >1.5 or <0.67, corrected P <0.05). Figure 5 e).

[0167] This embodiment demonstrates that the benzoylation PRM targeting detection method established in this invention, starting from a reference spectral library containing 16,874 benzoylation peptides, successfully established a compressed PRM target panel containing 1,020 high-quality benzoylation peptides after a single round of quality screening. Method validation results show that, with a protein input of 200 ng, the quantitative reproducibility of the targeted peptides reached 86.0% (877 / 1,020), and 106 benzoylation peptides that underwent significant changes upon sodium benzoate stimulation were successfully identified. This embodiment further verifies the versatility and reliability of the technical solution of this invention in various lysine acylation modifications (lactylation, acetylation, benzoylation, nicotinylation), demonstrating its potential as a universal platform for lysine acylation modification omics.

[0168] Example 6: Establishment and application of a lysine nicotinylation PRM targeted detection method 6.1 Construction of the nicotinylation reference spectral library Nicotinylation (Knicotinyl) is a non-classical lysine acylation modification, and currently available reference spectral resources and commercially available affinity reagents are relatively limited. Some studies have proposed that nicotinic acid can first generate nicotinoyl-CoA, and ACSS2 may be involved in the formation of this metabolic intermediate; the specific enzymatic process of further transfer of the nicotinyl group to lysine residues in proteins still needs further elucidation. This implementation method uses a reference spectral library-driven PRM method development workflow, which does not require specific antibody enrichment as a necessary step in the PRM detection stage of the test sample, thus achieving targeted analysis of nicotinyl peptides.

[0169] Following the method described in Example 1 for constructing the lactylated modified peptide library, the sample was chemically modified using nicotinyl-NHS to a final concentration of 10 mM. After enzyme digestion, desalting, and high-resolution LC-MS / MS data acquisition, spectral data of the nicotinyl modified peptides were obtained. Figure 6 a). Using the SEQUEST search engine, a search was conducted on the UniProt human protein database. Lysine nicotinylation was defined as a variable modification, and peptide identification and annotation were performed at the 1% FDR level to construct a nicotinylation reference library. In this embodiment, this reference library is named the ChemIntelligence Knic Library, containing 26,605 nicotinylated peptides, corresponding to 3,273 nicotinylated proteins and 22,624 nicotinylation sites ( Figure 6 a).

[0170] 6.2 Skyline Parameter Settings and PRM Target List Generation Referring to step 1.3 of Example 1, the nicotinylation reference spectrum library was imported into Skyline software. Parameter settings: protease Trypsin [KR / P], maximum of 3 missed cleavage sites; precursor charge states +2 and +3; product ions b and y ions (charge states +1 and +2), 15 product ions selected for each precursor; retention time window of 2 minutes. The nicotinylation modification was set to a mass increment of +105.0215 Da on lysine residues to generate a list of nicotinylation PRM targets.

[0171] 6.3 PRM Conductor Scheduling and Method Establishment Referring to step 1.4 of Example 1, the list of nicotinic PRM targets was imported into PRM Conductor. Precursor ion dynamic scheduling was performed using the same scheduling parameters as in Example 1 (acquisition time window 1.4 minutes, minimum residence time 10 ms, peak width 10 seconds, at least 7 acquisition points per peak, and only one charged precursor ion retained per peptide). This assigned 16,811 nicotinic peptides to 20 independent PRM methods. Figure 6 a).

[0172] 6.4 PRM Method Confirmation and Reallocation The liquid chromatography and mass spectrometry parameters were consistent with those in step 1.5 of Example 1. HeLa cell enzymatically digested peptides modified with nicotinyl-NHS were used as validation samples. Multiple PRM methods generated in step 6.3 were used for single-injection validation on a Stellar mass spectrometer to verify and calibrate the initial PRM acquisition method generated based on the DDA reference spectral library on a real-world platform.

[0173] The collected raw data was imported into Skyline for peak extraction, with parameter settings identical to step 1.6 of Example 1. Precursor ions with unidentifiable chromatographic peaks were first excluded, and then the method allocation was recalculated using the precursor ion dynamic scheduling module. After a single validation injection and peak extraction screening, a total of 16,759 peptides were identified and reassigned to 24 PRM methods (…). Figure 6 a).

[0174] 6.5 PRM Method Validation and Quality Control Screening Immunoblot-assisted assessment: HeLa cell lysates before and after sodium nicotinic acid treatment were immunoblotted using an experimentally validated anti-benzoylated lysine antibody that cross-reacts with nicotinic epitopes. Compared to the untreated control group, the cross-reactivity signal was enhanced after 48 hours of treatment with 100 mM sodium nicotinic acid. This result is consistent with the increased intracellular nicotinic acid esterification level and can serve as an auxiliary basis for selecting samples for subsequent PRM validation, but it is not the sole evidence for confirming nicotinic acid esterification sites; specific nicotinic peptides will still be identified through subsequent PRM peak extraction and fragment ion matching. Figure 6 b).

[0175] PRM assay and first-round quality screening: To validate the target and select high-response peptides suitable for low-input samples, HeLa cell enzymatically digested peptides treated with 100 mM sodium nicotinate for 48 hours were used as positive samples, and the above 24 PRM assay methods were run under the condition of a protein loading amount of 200 ng (as an example of a low-input sample).

[0176] Data import and screening: Import the data into Skyline for analysis and screen it according to the following criteria: absolute peak area > 500, signal-to-background ratio (S / B) > 3, relative abundance > 0.05, time correlation > 0.8 and number of effective product ions ≥ 3.

[0177] Parameter settings: Based on this, PRM Conductor rescheduled these precursors, setting the acquisition time window to 1.4 minutes, the minimum residence time to 10 ms, the peak width to 9 seconds, and at least 7 acquisition points per peak. It also used a balanced load mode for compression, ultimately resulting in a compressed PRM target panel containing 797 nicotinic peptide fragments. Figure 6 c).

[0178] Method reproducibility assessment: The compressed PRM target panel described above was used to perform three technical replicates on HeLa samples treated with sodium nicotinic acid. Of the 797 targeted peptides, 713 peptides showed a CV < 30%, indicating that the panel has good technical reproducibility under the conditions of this embodiment. Figure 6 d).

[0179] 6.6 Method Application: Dynamic Quantification of Nicotinylation under Sodium Nicotinate Stimulation HeLa cells were treated with sodium nicotinic acid (final concentration 100 mM) for 48 hours as the positive stimulation group, with untreated cells serving as the control group. Each group was a triple biological replicate, with a protein input of 200 ng (as an example of a low-input sample). The nicotinylation compressed PRM target panel established above was used for detection. Quantitative data were normalized to TIC in Skyline before statistical comparison. For quantification of endogenous nicotinylation levels, a two-sided Student t-test combined with BH correction was used for statistical analysis.

[0180] The analysis results showed that, compared with the control group, sodium nicotinic acid treatment significantly upregulated 456 nicotinyl peptides (corresponding to 384 proteins) (fold change > 1.5, corrected P < 0.05). Figure 6 e).

[0181] This embodiment demonstrates that the PRM method development process described in this invention, which combines reference spectral library-driven approach, dynamic precursor ion scheduling, and one or more rounds of quality screening, can complete the entire process from constructing a target post-translational modification reference spectral library, generating initial targets, establishing a multi-method PRM acquisition panel, to targeted quantitative analysis of low-input samples. This process does not rely on specific antibody enrichment, nor on isoweighted tandem mass tagging, or internal standard mixtures rich in target post-translational modification peptides during sample detection. It is suitable for studies of lysine acylation modifications, including nicotinylation, which lack commercially available affinity reagents. The modified peptides and their corresponding proteins retained after one or more rounds of quality screening can be further used to form a target target set or target protein set suitable for cross-sectional comparative analysis of multiple samples.

[0182] Example 7: Establishment and Validation of PRM Targeted Detection Method Based on DDA Identification Results of Lactylated Enriched Samples 7.1 Enrichment of Lactylated Peptides and DDA Data Acquisition This embodiment illustrates that the liquid chromatography-mass spectrometry data resources used to generate the initial PRM target can come not only from a reference spectral library constructed from chemically modified samples, but also from the identification results of naturally modified peptide DDA enriched by antibodies or other affinity materials. The overall process is described in [link to documentation]. Figure 7 a.

[0183] In this embodiment, HeLa cells treated with sodium lactate were used as lactylated positive samples. HeLa cells were treated with sodium lactate at a final concentration of 100 mM for 48 hours. After harvesting the cells, they were lysed, protein quantified, reduced, alkylated, digested with trypsin, and desalted at C18 to obtain fully digested peptides.

[0184] A portion of the enzymatically digested peptides is enriched using an anti-lactylated lysine antibody or other affinity material that can specifically bind to the lactylated peptides. The lactylated peptides are then bound to the affinity material. After washing to remove unbound peptides, the bound lactylated peptides are eluted to obtain a lactylated enriched peptide sample.

[0185] The affinity material can be a free antibody, antibody-conjugated magnetic beads, antibody-conjugated agarose material, or other affinity materials capable of recognizing lactylated lysine epitopes. Lactylated enrichment samples are not limited to HeLa cells, but can also be derived from other cells, tissues, body fluids, clinical samples, or processed biological samples.

[0186] The obtained lactylated enriched peptide samples were acquired using a DDA mode on an Orbitrap Astral mass spectrometer. The liquid chromatography system was configured in Trap-elute mode; the peptides were first loaded onto a PepMap Neo trapping column, and then separated using an Aurora Elite C18 analytical column. Mobile phase A was an aqueous solution containing 0.1% formic acid, and mobile phase B was an acetonitrile solution containing 0.1% formic acid. The liquid chromatography gradient time was set to 30 minutes.

[0187] MS1 scan range was set to m / z 300 to 1,600, resolution was set to 180,000, and AGC target was set to 300%; candidate precursor ion isolation window was set to 1.6 Th, fragmentation was performed using HCD, and normalized collision energy was set to 30%; MS2 scan range was set to m / z 100 to 2,000, and AGC target was set to 200%.

[0188] 7.2 Search for DDA data in emulsified enriched samples Raw DDA data from lactylated enriched samples were imported into database search software and searched the UniProt human protein database. The protease was set to trypsin, allowing a maximum of three missed cleavage sites; cysteine ​​aminomethylation was set as a fixed modification; lysine lactylation and methionine oxidation were set as variable modifications; a target-decoy strategy was employed to control the peptide mapping results to a false detection rate of less than 1%.

[0189] For lactylated peptides that have passed quality control, extract their modified peptide sequences, the position of the lactylation modification in the peptide, the precursor ion mass-to-charge ratio, the precursor ion charge state, the product ion spectrum, and the chromatographic retention time information. Optionally record one or more candidate protein accessions and their corresponding lactylation sites and / or protein grouping information in their protein sequences as protein mapping annotations to form a search result list of DDA-modified peptides in lactylated enriched samples.

[0190] The modified peptide search results list can be directly or after format conversion imported into the targeted mass spectrometry analysis software to generate an initial PRM target list or an initial PRM target panel.

[0191] 7.3 Generation of Initial PRM Target Panel and PRM Acquisition Method Import the DDA-modified peptide search results from the lactylated enriched sample obtained in step 7.2 into the Skyline targeted mass spectrometry analysis software. The parameter settings are as follows: Peptide parameters: The protease is set to Trypsin [KR / P], allowing a maximum of 3 missed cleavage sites; the peptide length is set to 6-45 amino acids.

[0192] Modification parameters: Cysteine ​​aminomethylation was set as a fixed modification; lysine lactylation was set as a variable modification of +72.0211 Da; methionine oxidation was set as a variable modification.

[0193] Ion parameters: the precursor ion charge states are set to +2 and +3; the product ion types are set to b ion and y ion; the product ion charge states are set to +1 and +2; a maximum of 15 product ions are selected for each precursor ion; the measured product ions and chromatographic retention time information in the DDA data are preferred.

[0194] In this embodiment, 1,874 lactylated peptides meeting the PRM target generation conditions were obtained from the DDA identification results of the lactylated enriched samples. The PRM targets corresponding to these 1,874 lactylated peptides collectively constitute the enriched source lactylated initial PRM target panel.

[0195] The initial PRM target panel is generated directly from the DDA identification results of the lactylated enriched samples, without undergoing pre-PRM quality screening, and without using the actual detection results in specific validation samples as the retention conditions for members of the initial PRM target panel.

[0196] The initial PRM target panel of the lactylated target is imported into PRM Conductor, dynamically scheduled according to the chromatographic retention time of each PRM target and the target density in different time intervals, and assigned to one or more PRM acquisition methods.

[0197] In one implementation, the acquisition time window is set to 1.4 minutes, the minimum residence time is set to 10 ms, the peak width is set to 9-12 seconds, and at least 7-8 acquisition points are guaranteed for each peak. These parameters can also be adjusted according to the liquid chromatography conditions, instrument platform, sample complexity, and target precursor ion density.

[0198] 7.4 PRM Validation in Unenriched Samples To verify the feasibility of the initial PRM target panel and PRM acquisition method described in step 7.3, the sodium lactate-treated HeLa cell protease digested peptides corresponding to the lactylated enrichment samples were used as verification samples. The verification samples were not subjected to lactylated antibody enrichment after enzymatic digestion and desalting, and the amount of peptide injected for each PRM injection was 200 ng.

[0199] PRM data acquisition was performed on a Stellar mass spectrometer using one or more PRM acquisition methods generated in step 7.3. The liquid chromatography conditions were essentially the same as the DDA liquid chromatography conditions in step 7.1. Adaptive retention time was enabled, and an rtbin file was generated based on the sample matrix and processing conditions for retention time calibration.

[0200] MS1 scan range was set to m / z 300 to 1,600, scan rate to 33 kDa / s, and maximum ion implantation time to 10 ms. MS2 used HCD fragmentation, normalized collision energy was set to 30%, scan range was set to m / z 200 to 1,500, scan rate to 125 kDa / s, maximum ion implantation time was set to dynamic mode, and loop control was set to "All".

[0201] The acquired raw PRM data was imported into Skyline, and peak extraction and target identification were performed based on the modified peptide sequence, precursor ion mass-to-charge ratio, chromatographic retention time, and product ion co-elution relationship.

[0202] The results showed that, of the 1,874 lactylated peptides contained in the initial PRM target panel, 1,727 were detectable in 200 ng of the corresponding HeLa cell enzymatically digested peptides without enrichment, with a detection coverage of 92.16%. Figure 7 b).

[0203] The PRM assays described above were used to verify the feasibility and detection coverage of the initial PRM target panel, without altering the composition of the 1,874 initial PRM targets. Targets not detected in the samples used in this validation can still be detected in other sample types, under different processing conditions, liquid chromatography conditions, or mass spectrometry platforms.

[0204] This embodiment demonstrates that the data resources used to directly generate the initial PRM target panel are not limited to reference spectral libraries constructed from chemically modified samples, but can also originate from the DDA identification results of modified peptides enriched by antibodies or other affinity materials. This enrichment step is used for front-end data resource acquisition or target discovery, rather than as a necessary step for PRM detection and quantitative analysis of the test sample; therefore, this invention neither requires antibody enrichment as a necessary condition for implementation nor excludes the use of data obtained from antibody enrichment as the data source for the initial PRM target panel.

[0205] Example 8: A grouped PRM-targeted detection method based on joint analysis of acetylation reference spectral library and published acetylation sites 8.1 Joint analysis of acetylation reference spectral library and published acetylation sites This embodiment illustrates how, by jointly analyzing an acetylation reference spectral library with published acetylation site data, library members can be grouped according to their published annotation support status, and PRM acquisition methods can be established for each group. This grouping is only used for data source annotation, method organization, and coverage comparison, and is not a necessary limitation on the general PRM development process or panel member identities of this invention. The overall process is described in [link to full documentation]. Figure 8 a.

[0206] The ChemIntelligence Kac Library constructed in Example 4 was used as the acetylation reference spectrum library. The acetylation reference spectrum library contains 23,702 acetylated peptides, corresponding to 20,047 acetylation sites and 2,921 acetylated proteins.

[0207] Collect data on human lysine acetylation sites from publicly available databases and published studies. This data may originate from one or more of the following resources: iPTMnet, CPLM, dbPTM, published acetylation affinity enrichment studies, acetylated peptide search results lists, or other publicly available acetylation data resources.

[0208] The acetylation reference spectral library and published acetylation site data were standardized. Spectral library target members were determined based on the acetylated peptide sequence and the location of the acetylation modification within the peptide, and one or more standardized protein accession-protein sequence-site mapping pairs were recorded for each member. Each mapping pair was compared with publicly available modification site data; when at least one valid mapping pair for the same member was recorded in the publicly available data, the member was marked as a target supported by existing publicly available annotations; when no valid mapping pairs were recorded, the member was marked as a supplementary target for the spectral library. Changes in database version or protein mapping relationships do not alter the membership determined by the modified peptide sequence and the modification location within the peptide.

[0209] Based on the above comparison results, members of the spectral library whose effective protein mapping pairs are not found in the publicly available data are classified into the spectral library supplementary target group; members of the spectral library whose effective protein mapping pairs are found in the publicly available data are classified into the target group with existing publicly available annotations.

[0210] 8.2 Generation of two initial PRM target panels Based on peptide length, intrapeptide modification location, precursor ion charge state, product ion information, and retention time information, 913 acetylated peptides suitable for generating PRM acquisition methods were obtained from the supplementary target group of the spectral library, forming the initial PRM target panel for supplementing the spectral library.

[0211] 1,054 acetylated peptides suitable for generating PRM acquisition methods were obtained from the target set supported by existing public annotations, forming an initial PRM target panel supported by existing public annotations.

[0212] The two target classes were imported into Skyline. The protease was set to Trypsin [KR / P], allowing a maximum of 3 missed cleavage sites; lysine acetylation was set to a variable modification of +42.0106 Da; the precursor ion charge states were set to +2 and +3; the product ions were selected as b and y ions; the product ion charge states were set to +1 and +2; a maximum of 15 product ions were selected for each precursor ion; and the retention time window was set to 2 minutes.

[0213] The two initial PRM target panels were generated directly from the joint analysis results of the acetylation reference spectrum library and published acetylation site data, without prior PRM quality screening, and without using the actual detection results in specific samples as the retention conditions for the initial panel members.

[0214] 8.3 Generation of two sets of PRM data collection methods Import the initial PRM target panel supplemented by the spectral library and the initial PRM target panel supported by existing public annotations into PRM Conductor respectively. Dynamically schedule the data according to the measured or predicted retention time of each PRM target to generate one or more grouped PRM acquisition methods respectively.

[0215] In one implementation, the acquisition time window is set to 1.4 minutes, the minimum residence time is set to 10 ms, the peak width is set to 12 seconds, and at least 8 acquisition points are guaranteed for each peak. A balanced load mode can be used to assign the two sets of targets to the corresponding PRM acquisition methods.

[0216] By scheduling the two sets of targets separately, the coverage of targets with existing public site annotation support and targets supplemented by the current spectral library in PRM detection can be compared. The two sets of PRM acquisition methods can also be merged, split, or rescheduled according to the detection purpose.

[0217] 8.4 Validation of PRM in acetyl-NHS chemically modified samples First, HeLa cell enzymatically digested peptides modified with acetyl-NHS were used as reference validation samples. Sample preparation followed Example 4, with the final concentration of acetyl-NHS set at 3 mM. After chemical modification, enzymatic digestion, and desalting, the amount of peptide injected into each PRM was 200 ng.

[0218] The PRM acquisition methods for the existing publicly available annotation-supported target group and the spectral library supplementary target group generated in step 8.3 were run separately. PRM data acquisition was performed on a Stellar mass spectrometer, with liquid chromatography conditions and mass spectrometry parameters as described in Example 1.

[0219] The collected data was imported into Skyline, and peak extraction and target identification were performed based on the acetylated peptide sequence, precursor ion mass-to-charge ratio, chromatographic retention time, and product ion co-elution relationship.

[0220] The results showed that in acetyl-NHS-modified HeLa samples, the detection coverage of the initial PRM target panel, supported by publicly available annotations, was 90.89%, and the detection coverage of the initial PRM target panel supplemented by the spectral library was 91.35%. Figure 8 b).

[0221] The above results indicate that both types of initial PRM target panels generated from the joint analysis of reference spectral libraries and published acetylation sites can be converted into practically executable PRM acquisition methods.

[0222] 8.5 Validation of PRM in TSA-induced biological samples To further verify the detectability of the two types of PRM collection methods in endogenously acetylated biological samples, HeLa cells treated with Trichostatin A (TSA) were used as positive verification samples.

[0223] In this embodiment, HeLa cells were treated with TSA at a final concentration of 1 μM for 24 hours. After cell harvesting, the cells were lysed, reduced, alkylated, digested with trypsin, and desalted using C18 enzymes. The resulting samples were not enriched with acetylation antibodies, and the amount of peptide injected per PRM injection was 200 ng.

[0224] The PRM acquisition methods for TSA-treated HeLa samples were tested using both publicly available annotated target groups and spectral library-supplemented target groups. Results showed that the detection coverage of the panel supporting the initial PRM targets with existing public annotations was 89.94%, while the detection coverage of the panel supplemented by the spectral library for the initial PRM targets was 91.57%. Figure 8 c).

[0225] The above PRM detection results are used to evaluate the practical feasibility and detection coverage of the two initial PRM target panels, without changing the composition of the two initial panels. PRM targets not detected in the samples in this validation can still be detected in other sample types, under different processing conditions, or on other mass spectrometry platforms.

[0226] This embodiment demonstrates that the target set used in PRM method development can be grouped and organized by combining a reference spectral library and publicly available modified data resources. Initial PRM target panels are generated and dynamically scheduled according to the publicly available annotation support status, which can be used to compare the PRM detectability of targets with different annotation statuses. This grouping analysis does not change the target membership and is not a necessary step in forming a general PRM development process.

[0227] The grouping analysis method is not limited to acetylation, but can also be applied to lactylation, benzoylation, nicotinylation, and other post-translational modifications of proteins. Depending on the detection purpose, one or more rounds of quality screening can be performed on any initial PRM target panel to form a corresponding compressed PRM target panel; quality screening is not a necessary step for the initial PRM target panel grouping and direct detection described in this embodiment.

Claims

1. A method for targeted detection and quantification of protein post-translational modification (PTM) based on spectral library-driven and parallel reaction monitoring (PRM), characterized in that, Includes the following steps: (1) Construct a reference spectral library of the target post-translational modified peptide, wherein the reference spectral library includes at least the precursor ion mass-to-charge ratio, product ion spectrum information, chromatographic retention time information and the position of the target modification in the modified peptide, and may include protein mapping annotation information; (2) The reference spectral library is derived from one or more of the following: natural post-translational modified peptides, simulated post-translational modified peptides, chemically synthesized modified peptides, or publicly available spectral databases; (3) Generate an initial PRM target list or an initial PRM target panel based on the reference spectral library; (4) Based on the chromatographic retention time information of candidate precursor ions in the initial PRM target list or initial PRM target panel, perform dynamic scheduling on the candidate precursor ions to allocate multiple candidate precursor ions to one or more independent PRM acquisition methods; wherein, the dynamic scheduling may optionally adopt a balanced load method. (5) Using validation samples, perform actual PRM data acquisition, peak extraction and detectability confirmation on one or more PRM acquisition methods to form a validated PRM target panel. (6) Optionally, the validated PRM target panel is subjected to one or more rounds of quality screening using pre-screened samples to form a compressed PRM target panel. (7) Using one or more of the initial PRM target list, initial PRM target panel, validated PRM target panel or compressed PRM target panel, targeted PRM data acquisition and quantitative analysis are performed on biological samples.

2. The method according to claim 1, characterized in that, The post-translational modification is lysine acylation modification, which may be lysine acylation modification formed by aliphatic acyl groups, aromatic acyl groups, heterocyclic acyl groups or their derived acyl groups.

3. The method according to claim 1, characterized in that, The quality screening criteria in step (6) include one or more of the following: absolute peak area > 500, signal-to-background ratio (S / B) > 3, relative abundance > 0.05, time correlation > 0.8, effective product ion number ≥ 3, and coefficient of variation (CV) ≤ 30%; the quality screening is performed in one or more rounds to obtain a high-quality compressed PRM target panel suitable for low input samples.

4. The method according to any one of claims 1 to 3, characterized in that, The precursor ion dynamic scheduling is based on one or more of the following: acquisition time window, minimum residence time, chromatographic peak width, number of sampling points per peak, and acquisition load in different time intervals, and can optionally adopt a balanced load mode.

5. The method according to claim 4, characterized in that, The acquisition time window, minimum residence time, peak width, or number of sampling points per peak are adjusted according to the liquid chromatography conditions, column type, instrument platform, retention time calibration results, sample complexity, or target precursor ion density.

6. The method according to any one of claims 1 to 3, characterized in that, The biological sample is selected from one or more of the following: cell sample, tissue sample, clinical tissue section, puncture biopsy sample, rare cell sample, single-cell level sample, or single-cell source sample.

7. A method for establishing and validating an initial PRM target panel based on the identification results of modified peptide DDA, characterized in that, include: (1) Obtain DDA data containing the target post-translational modified peptide and perform database search, extract the modified peptide sequence, the position of the target modification in the modified peptide, the precursor ion mass-to-charge ratio, the precursor ion charge state, the product ion spectrum and chromatographic retention time information, and optionally record one or more candidate protein accessions and the modification sites and / or protein grouping information in their corresponding protein sequences as protein mapping annotations to form a modified peptide search result list. (2) Generate an initial PRM target panel based on the modified peptide search results list, and do not use the actual PRM detection results in a specific validation sample as the retention condition for the members of the initial PRM target panel. (3) Dynamically schedule the initial PRM target panel according to the chromatographic retention time and the target density in different time intervals to generate one or more PRM acquisition methods; (4) Perform PRM data acquisition, peak extraction and target identification using validation samples to evaluate the feasibility or detection coverage of the initial PRM target panel; The DDA data may be selected from naturally modified peptide samples enriched with antibodies or other affinity materials, and the actual detection results in the validation samples do not change the composition of the initial PRM target panel.

8. A method for generating grouped PRM acquisition based on joint analysis of modified peptide reference spectral libraries and publicly available modification site data, characterized in that, include: (1) Obtain a reference spectrum library of modified peptides and publicly available modification site data for the target post-translational modifications; The target members of the spectral library are determined by the modified peptide sequence and the position of the target modification in the peptide, and the one or more protein accession-protein sequence site mapping pairs associated with each member are normalized. (2) Compare the mapping pairs with the publicly available modified site data. When at least one valid mapping pair of the same spectral library target member is recorded in the publicly available modified site data, the member is marked as a target supported by existing publicly available annotations. When none of its valid mapping pairs are recorded in the publicly available modified site data, the member is marked as a supplementary target in the spectral library. (3) Based on the markers, existing publicly available annotation-supported target groups and spectral library supplementary target groups are formed respectively, and one or more grouped PRM acquisition methods are dynamically scheduled according to the measured or predicted chromatographic retention time. (4) Optionally use the verification sample to perform PRM data acquisition, peak extraction and target identification; wherein, the grouping is only used to record the public annotation support status and generate the grouped PRM acquisition method, and does not change the target membership determined by the modified peptide sequence and the position of the target modification in the peptide, and the actual detection results in the verification sample do not change the composition of the two initial PRM targets.

9. A system for implementing the method according to any one of claims 1 to 8, characterized in that, The system includes: (1) Data resource acquisition module, configured to construct or read one or more of the following: reference spectral library of target post-translational modified peptides, DDA identification results of modified peptides, or publicly available modification site data; (2) PRM target generation and grouping module, configured to generate an initial PRM target list or an initial PRM target panel based on the data resources, and optionally mark the targets as targets supported by existing public annotations or supplementary targets of the spectral library according to the matching status of the publicly available modified site data. (3) The precursor ion dynamic scheduling module is configured to perform precursor ion dynamic scheduling based on chromatographic retention time and target density in different time intervals, so as to allocate candidate precursors to one or more PRM acquisition methods; (4) Validation and quality screening module, configured to perform peak extraction, target identification and detectability evaluation based on the actual PRM collection data of the validation sample, and optionally form a validated PRM target panel or perform one or more rounds of quality screening based on the pre-screened sample data to form a compressed PRM target panel. (5) PRM data acquisition control module, configured to control the mass spectrometer to perform data acquisition according to the scheduled PRM method or compressed PRM target panel; (6) Quantitative and analytical module, configured to extract peak area, normalize, perform statistical tests and output results for PRM data.

10. A non-transitory computer-readable storage medium storing computer program instructions thereon, which, when executed by a processor, cause the processor to perform the computer-executable data processing and mass spectrometry control steps of the method according to any one of claims 1 to 8; the steps comprising at least: Dynamic scheduling is performed based on the chromatographic retention time information of candidate precursor ions to generate one or more PRM acquisition methods, and acquisition control instructions are generated to control the mass spectrometer to perform data acquisition according to the PRM acquisition methods; the steps also include one or more of the following: acquiring or reading a modified peptide reference spectrum library, modified peptide DDA identification results or publicly available modification site data; generating an initial PRM target list or an initial PRM target panel. Group targets according to the status supported by publicly available annotations; receive PRM data and perform peak extraction, target identification, detectability confirmation or detection coverage evaluation; optionally perform one or more rounds of quality screening; It also performs peak area extraction, quantitative analysis, and result output on PRM data.

11. The method according to any one of claims 1, 7, or 8, characterized in that, The precursor ion mass-to-charge ratio and precursor ion charge state in the initial PRM target list or initial PRM target panel are recorded as explicit fields, or are calculated or converted from the modified peptide sequence, the target post-translational modification type, and charge state rules; the chromatographic retention time information is the measured retention time, the predicted retention time, the calibrated retention time, or the retention time information obtained by converting the above retention times.

12. The method according to any one of claims 1, 7, or 8, characterized in that, The sequence of the modified peptide containing the target modification and the position of the target modification in the modified peptide are used as the basis for determining the identity of the initial PRM target members. The same member can be associated with one or more candidate protein accessions and the modification sites and / or protein grouping information in their corresponding protein sequences. The association information is used as a protein mapping annotation and is not a necessary condition for distinguishing members. Adjustments to the database version, protein merging rules, or peptide-protein mapping relationship do not change the membership of the corresponding initial PRM target. The actual PRM detection results in the validation samples or pre-screened samples are used to form the validated PRM target panel or compressed PRM target panel, without changing the membership composition of the initial PRM target list or initial PRM target panel. Adjustments to the number or composition of product ions, retention time or retention time window, precursor ion charge state, collision energy, precursor ion or product ion mass-to-charge ratio calibration, liquid chromatography conditions, mass spectrometry platform, PRM acquisition method allocation, or software data format do not change the membership of the corresponding initial PRM target.

13. The method according to claim 2, characterized in that, The lysine acylation modification is selected from one or more of lysine lactylation, lysine acetylation, lysine benzoylation, and lysine nicotinylation.

Images