Bidirectional simulation and difference notarization mass spectrum data conversion fidelity verification method and system

Abstract

The invention provides a two-way simulation and difference notarization mass spectrum data conversion fidelity verification method, and relates to the technical field of analysis test data governance and interoperability, the method comprises the following steps: carrying out format conversion simulation calculation on mass spectrum data in a special format to obtain mass spectrum data in a common format; field mapping and priority rules are utilized to perform reverse simulation, reconstruct a virtual head, compare with head information of mass spectrum data in a proprietary format, calculate a hash tree based on mass spectrum data in a public format according to a configured blocking rule, compare with an original hash tree, and perform fidelity scoring calculation by utilizing the calculated hash tree and the original hash tree to obtain a fidelity score. Obtaining a difference report; and when the fidelity score in the difference report does not meet the threshold value, obtaining an inconsistent result comprising the positionable difference details and the alarm, and when the fidelity score in the difference report meets the threshold value, obtaining a fidelity certificate. According to the invention, the problem that the fidelity of the exclusive-to-standard conversion is difficult to automatically and quantitatively judge in the prior art is solved.

Description

Technical Field

[0001] This invention relates to the field of analytical testing data governance and interoperability technology, and in particular to a method for bidirectional simulation and difference verification of mass spectrometry data conversion fidelity. It is applicable to data archiving and sharing on platforms such as liquid chromatography-mass spectrometry / gas chromatography-mass spectrometry (LC / GC-MS, the abbreviations in parentheses below are for supplementary explanation only) and ion mobility spectrometry-mass spectrometry (IMS-MS). Background Technology

[0002] With the widespread application of mass spectrometry in scientific research, metrology, and regulatory scenarios, cross-platform sharing and long-term storage have placed a strong demand on "interoperability" of data formats. The industry generally adopts open standard formats (such as mzML, mzXML, mzTab, etc.) to carry and exchange data; however, instruments still generate and retain proprietary formats (such as RAW, WIFF, d, etc.) during front-end acquisition.

[0003] Existing processes typically safeguard the "proprietary → standard" conversion through the following methods: 1. Structural / syntactic layer verification: For example, verifying the compliance of fields and values ​​based on XML schema (XSD) and controlled vocabulary (CV); 2. Manual sampling and visual quality inspection: For example, visually comparing total ion chromatograms (TIC) or base peak chromatograms (BPC); 3. Dual archiving: Simultaneously saving proprietary and standard formats for review.

[0004] The above approach has significant shortcomings: it can only prove "format legality," not "semantic fidelity." Even if XSD / CV passes, it cannot guarantee that key acquisition parameters and spectral segment content are consistent with the original data within the allowed equivalence classes; manual review is not scalable and is highly subjective; long-term dual archiving is costly and detrimental to compliance auditing; heterogeneous semantics are difficult to align: vendor header parameters and metadata in standard formats (such as...) <cvparam>There is a lack of a one-to-one corresponding verification path.

[0005] Therefore, the lack of an automated, quantifiable, auditable, and non-repudiable "conversion verification" technology has become the "last mile" pain point in achieving interoperability. Summary of the Invention

[0006] To address the aforementioned shortcomings in existing technologies, this invention provides a two-way simulation and difference notarization method for verifying the fidelity of mass spectrometry data conversion, which solves the problem that existing technologies struggle to automatically and quantitatively determine the fidelity of proprietary-to-standard conversions.

[0007] To achieve the aforementioned objectives, the technical solution adopted by this invention is: a method for verifying the fidelity of mass spectrometry data conversion through bidirectional simulation and difference notarization, comprising: S1: Perform format conversion simulation calculations on proprietary format mass spectrometry data to obtain common format mass spectrometry data; S2: Perform reverse simulation on the mass spectrometry data using field mapping and priority rules, reconstruct the virtual header of the common format mass spectrometry data, and compare it with the header information of the proprietary format mass spectrometry data. If the consistency comparison fails, an inconsistency result is obtained. If the consistency comparison passes, the common format mass spectrometry data is input into S3. S3: Based on the mass spectrometry data in the public format, calculate the hash tree according to the configured block rules, compare it with the original hash tree, and when the consistency comparison fails, obtain the inconsistent result; when the consistency comparison passes, input the calculated hash tree and the original hash tree into S4. S4: Calculate the fidelity score using the calculated hash tree and the original hash tree to obtain a difference report; S5: When the fidelity score in the difference report does not meet the threshold, an inconsistency result including locatable difference details and alarms is obtained. When the fidelity score in the difference report meets the threshold, a fidelity certificate is obtained, and the mass spectrometry data conversion fidelity verification is completed.

[0008] Further, S2 includes: Inverse simulation of mass spectrometry data is performed using field mapping and priority rules. Mass spectrometry data in common format is reconstructed according to the corresponding fields to obtain the corresponding virtual head. The virtual head is compared with the header information of the proprietary format mass spectrometry data. If the consistency comparison fails, an inconsistency result is obtained. If the consistency comparison passes, the common format mass spectrometry data is input into S3.

[0009] Further, S3 includes: Based on mass spectrometry data in a public format, key time series and spectral segments are normalized, quantized, and hashed according to configured block rules to obtain a hash tree including block fingerprints and Merkle trees; The hash tree is compared with the original hash tree. If the consistency comparison fails, an inconsistent result is obtained. If the consistency comparison passes, the calculated hash tree and the original hash tree are input into S4.

[0010] Furthermore, the step of comparing the hash tree with the original hash tree, obtaining an inconsistent result when the consistency comparison fails, and inputting the calculated hash tree and the original hash tree into S4 when the consistency comparison passes includes: The hash tree is compared with the original hash tree. Fingerprint generation and comparison are run in streaming / parallel mode on the data platform: block fingerprints and Merkle leaves are calculated for windows composed of multiple spectrograms; when only consistency proof is required, only Merkle roots and leaves and paths of a few disputed blocks are swapped to protect data privacy; after the certificate is generated, it is automatically archived to the audit storage and associated with the sample list / project number to achieve post-event traceability. When the consistency comparison fails, an inconsistent result is obtained. When the consistency comparison passes, the calculated hash tree and the original hash tree are input into S4.

[0011] Further, S4 includes: Define the equivalence class criterion and tolerance band; Based on the equivalence class criterion and tolerance band, the calculated hash tree and the original hash tree are used to calculate the mass-to-charge ratio tolerance, strength value and retention time to obtain the fidelity score. A discrepancy report is obtained by applying soft violations to reduce the weight of minor deviations in the fidelity score.

[0012] This invention provides a mass spectrometry data conversion fidelity verification system with two-way simulation and difference verification, comprising: Forward conversion records are used to perform format conversion simulation calculations on proprietary format mass spectrometry data to obtain common format mass spectrometry data. Reverse simulation is used to perform reverse simulation of mass spectrometry data using field mapping and priority rules, reconstruct the virtual header of common format mass spectrometry data, compare it with the header information of proprietary format mass spectrometry data, and obtain an inconsistency result when the consistency comparison fails, and input the common format mass spectrometry data into fingerprint generation when the consistency comparison passes. Fingerprint generation is used to calculate a hash tree based on public format mass spectrometry data according to the configured block rules, and compare it with the original hash tree. When the consistency comparison fails, an inconsistent result is obtained. When the consistency comparison passes, the calculated hash tree is compared with the original hash tree input difference. The difference comparison is used to calculate the fidelity score by comparing the calculated hash tree with the original hash tree and obtain a difference report; Certificate generation is used to obtain inconsistency results, including locatable difference details and alarms, when the fidelity score in the difference report does not meet the threshold. When the fidelity score in the difference report meets the threshold, a fidelity certificate is obtained, completing the fidelity verification of mass spectrometry data conversion.

[0013] This invention provides a computer-readable storage medium that stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes the mass spectrometry data conversion fidelity verification method described in any of the above-mentioned embodiments, which involves bidirectional simulation and difference notarization.

[0014] The beneficial effects of this invention are as follows: This invention provides a method for verifying the fidelity of mass spectrometry data conversion using bidirectional simulation and difference notarization. After completing the conversion between a vendor's proprietary mass spectrometry data format and an open standard format, the method verifies and notarizes the fidelity of the conversion result. By reconstructing the "virtual original header / parameter digest" from the open standard format and comparing key data blocks with stable fingerprints / hashes, a fidelity certificate is generated. This allows for automated and quantifiable determination of the fidelity of the proprietary-to-standard conversion, eliminating trust gaps in the interoperability process. (1) Introducing reverse simulation to compare proprietary header fields and standard metadata in the same semantic domain to avoid "structurally valid but semantically distorted"; (2) Through normalization-quantization-hashing and Merkle trees, it is robust to sorting disturbances, slight compression, and floating-point rounding; differences can be accurately located to the spectrum / time window; (3) Small deviations are weighted to avoid large-scale misjudgments due to small numerical differences, which is more in line with real production data; (4) Automatically generating a fidelity certificate and enabling timestamp / on-chain anchoring to achieve traceability, auditability, and non-repudiation after the fact; it can significantly reduce the cost of long-term dual archiving and manual review; (5) Mapping and criteria are configurable and adaptable to multiple vendors and multiple standard formats; fingerprints and thresholds can be adaptively calibrated according to the scenario, and have the ability to be reused across institutions and industrialized. Attached Figure Description

[0015] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting; in these embodiments, the same reference numerals denote the same structures, wherein: Figure 1 This is an exemplary flowchart of a two-way simulation and difference notarization mass spectrometry data conversion fidelity verification method according to some embodiments of this specification; Figure 2 This is a schematic diagram of a two-way simulation and difference verification mass spectrometry data conversion fidelity verification system according to some embodiments of this specification; Figure 3 This is an exemplary schematic diagram illustrating the mapping relationship between virtual raw header / parameter digests and proprietary header field fields according to some embodiments of this specification; Figure 4 This is an exemplary schematic diagram illustrating the mapping relationship between common formats and proprietary header field types according to some embodiments of this specification; Figure 5 This is an exemplary schematic diagram of a block fingerprint and Merkle tree structure shown in some embodiments of this specification; Figure 6 This is an exemplary schematic diagram illustrating the difference positioning and minimum disclosure of evidence process according to some embodiments of this specification; Figure 7 This is an exemplary schematic diagram of a fidelity certificate data structure according to some embodiments of this specification. Detailed Implementation

[0016] The specific embodiments of the present invention are described below to enable those skilled in the art to understand the present invention. However, it should be understood that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the present invention as defined and determined by the appended claims. All inventions utilizing the concept of the present invention are protected.

[0017] Example 1 Figure 1 This is an exemplary flowchart illustrating a two-way simulation and difference verification method for mass spectrometry data conversion fidelity verification, based on some embodiments of this specification. Figure 1 and Figure 6 As shown, the process includes the following steps. In some embodiments, the process may be executed by a processor.

[0018] S1: Perform format conversion simulation calculations on proprietary format mass spectrometry data to obtain common format mass spectrometry data.

[0019] Proprietary format mass spectrometry data (RAW) is mass spectrometry data in a specific format.

[0020] In some embodiments, the processor can invoke an existing converter to convert proprietary format mass spectrometry data into public format mass spectrometry data (mzML), and record the converter name, version, and parameters (for record purposes only, not for limitation). Specifically, the processor can input proprietary format mass spectrometry data (RAW) for format conversion to generate public format mass spectrometry data (mzML), and simultaneously generate block hashes according to the configured block partitioning rules.

[0021] S2: Perform reverse simulation on the mass spectrometry data using field mapping and priority rules, reconstruct the virtual header of the common format mass spectrometry data, and compare it with the header information of the proprietary format mass spectrometry data. If the consistency comparison fails, an inconsistency result is obtained. If the consistency comparison passes, the common format mass spectrometry data is input into S3.

[0022] A virtual header is a virtual raw file header and parameter summary.

[0023] The header information is a proprietary format header information.

[0024] In some embodiments, to compare the fidelity of the converted results, the processor can reconstruct the virtual original file header / parameter digest using the converted MZML file based on field mapping and priority rules. The corresponding header information extracted from the proprietary format is then compared with the virtual header.

[0025] In some embodiments, such as Figure 3 and Figure 4 As shown, the processor can parse metadata such as instrument configuration, acquisition and scanning indexes from a standard format, and reconstruct the virtual raw file header / parameter summary based on field mapping and priority rules, ensuring it is in the same semantic domain as the proprietary format header fields. Specifically, the processor can parse mzML... <filedescription> 、 <instrumentconfiguration> 、 <run> 、 <spectrumlist>The system generates virtual raw header / parameter summaries based on mapping rules, including but not limited to: instrument model, ionization mode, polarity timing, scan range, resolution / sampling point strategy, calibration / lock quality parameters, spectral counts, and time indexes. The proprietary header fields (RAW format) include a header file block and chromatographic / mass spectrometry data blocks. The header file block is stored in a binary structure: filename, generation software, generation time, instrument configuration, acquisition method, chromatographic data block offset address, and mass spectrometry data block offset address. The chromatographic / mass spectrometry data blocks are stored in a binary structure: chromatographic retention time and chromatographic intensity data. The mass spectrometry scan configuration information, scan filter, and spectral data are stored in a binary structure. The header block of the file corresponds to the original data file name, software name that generated the original data file, original data file generation time, instrument name, specifications, system composition, data acquisition method, and total number of indexes in the virtual raw header / parameter summary. The chromatography / mass spectrometry data block corresponds to the retention time of a single scan, the type of a single scan, the number of spectral data points for a single scan, and the checksum of the spectral data for a single scan in the virtual raw header / parameter summary. The common format (MZML format) is xml tags, where... <filedescription>and <softwarelist>The corresponding information in the virtual raw header / parameter summary includes the raw data file name, the software name that generated the raw data file, and the raw data file generation time. <instrumentconfigurationlist>and <dataprocessinglist>The tags correspond to the original raw data file generation time, instrument name, specifications, system composition, and data acquisition method in the virtual raw header / parameter summary. <run> 、 <spectrumlist>and <spectrum>The tags correspond to the total number of indices in the virtual raw header / parameter summary, the retention time of a single scan corresponding to an index, the type of a single scan corresponding to an index, the number of spectral data points for a single scan corresponding to an index, and the checksum of the spectral data for a single scan corresponding to an index.

[0026] Metadata includes instrument configuration, acquisition and scanning index, and raw data file description.

[0027] In some embodiments, the processor can use field mapping and priority rules to perform reverse simulation of mass spectrometry data, reconstruct the common format mass spectrometry data according to the corresponding fields, and obtain the corresponding virtual head; compare the virtual head with the header information of the proprietary format mass spectrometry data, and obtain an inconsistency result when the consistency comparison fails, and input the common format mass spectrometry data into S3 when the consistency comparison passes.

[0028] In some embodiments, the processor can generate "virtual raw file header / parameter summary" information according to the original data file description (software name that generated the raw data file, original data file name, original data file generation time); instrument configuration (instrument name, specifications, system composition); acquisition (data acquisition method); and scan index (total number of indexes, type of a single scan corresponding to the index, retention time of a single scan corresponding to the index, number of spectral data points for a single scan corresponding to the index, and checksum of spectral data for a single scan corresponding to the index). This information can be compared and verified with the corresponding information in the "proprietary format header field." When the virtual header and the proprietary format header field are in the same comparable semantic domain, consistency comparison verification is performed.

[0029] S3: Based on the mass spectrometry data in the public format, calculate the hash tree according to the configured block rules, and compare it with the original hash tree. When the consistency comparison fails, an inconsistent result is obtained. When the consistency comparison passes, input the calculated hash tree and the original hash tree into S4.

[0030] The calculated hash tree is based on mass spectrometry data in a public format.

[0031] The original hash tree was extracted from mass spectrometry data in a proprietary format.

[0032] In some embodiments, the processor can map parameters such as carrier gas programmed temperature rise / injection volume / ion source temperature; set individual tolerance bands for the conversion difference from spectral profile to centroid; and generate block fingerprints for BPC and selected characteristic ion channels.

[0033] In some embodiments, the processor can compute a hash tree according to configured block rules and compare it with the generated hash tree. (If they match, proceed to the next step.) Simultaneously, equivalence class criteria and tolerances are defined.

[0034] In some embodiments, the processor can perform normalization, quantization, and hash calculation on key time series and spectral segments based on public format mass spectrometry data and according to configured block rules to obtain a hash tree including block fingerprints and Merkle trees; compare the hash tree with the original hash tree, and if the consistency comparison fails, an inconsistent result is obtained; if the consistency comparison passes, the calculated hash tree and the original hash tree are input into S4.

[0035] In some embodiments, such as Figure 5 As shown, the processor can, based on public format mass spectrometry data, perform normalization, quantization, and hashing on key time series and spectral segments (TIC, BPC, acquisition time index, representative MS1 / MS2 subsets or block sequences, obtained from proprietary data parsing) according to configured block rules. It preferably uses block fingerprints and Merkle trees to support local difference localization and minimal evidence disclosure, resulting in a computed hash tree. Specifically, the processor can configure block rules (TIC, BPC, acquisition time index, representative MS1 / MS2 subsets or block sequences) on the mass spectrometry data file (original proprietary format), generating block data including TIC, BPC, and spectra; constructing a hash tree (Merkle tree), including TIC, BPC, and spectra; calculating block hash values; obtaining hash tree leaf nodes; and ultimately, obtaining the hash tree root node.

[0036] In some embodiments, fingerprint generation and comparison are run in a streaming / parallel manner on the data platform: block fingerprints and Merkle leaves are calculated for windows consisting of 64–256 spectra; when only consistency needs to be proven, only Merkle roots and leaves and paths of a small number of disputed blocks are exchanged to protect data privacy; after the certificate is generated, it is automatically archived to the audit storage and associated with the sample list / item number to achieve post-event traceability.

[0037] In some embodiments, the processor can perform segmented hash calculations on key time series and spectral segments (segmented hashes are hash values ​​and their metadata calculated for a single key segment) to obtain structured fingerprints. A structured fingerprint is a fingerprint that organizes segmented hash values ​​in a structured manner, such as a Merkle tree with its root. Specifically, the processor can normalize and quantize TIC, BPC, acquired time series, and sampled MS1 / MS2 blocks, calculate block fingerprints, and summarize them into a Merkle root.

[0038] In some embodiments, the processor can perform temperature / equivalence regression to calibrate the threshold based on historical "pass / fail" samples to reduce false negatives; DDA (Data Dependent Acquisition) and DIA (Data Independent Acquisition) employ different sampling and block strategies; when systematic deviations occur due to new compression or differences in vendor firmware, the "suspicious but allowed" label is retained and rectification suggestions are recorded.

[0039] Data extraction and rule-based block segmentation (such as TIC, BPC, time-of-collection index, representative MS1 / MS2 subsets) are performed on key blocks in proprietary formats. A hash tree is constructed using block hashing. The consistency of all blocks can be directly compared through the root node, while the specific blocks that have been modified can be located based on the leaf nodes.

[0040] S4: Calculate the fidelity score using the calculated hash tree and the original hash tree to obtain a difference report.

[0041] The fidelity score is a rating of how well the hash tree is faithful to the original hash tree.

[0042] The discrepancy report includes a breakdown of discrepancies, alerts, and fidelity score data.

[0043] In some embodiments, the processor can use a defined equivalence class criterion and tolerance to compare the m / z, intensity values ​​(raw / quantized), and retention times of corresponding spectra in RAW and MZML, and provide a fidelity score.

[0044] In some embodiments, the processor can define equivalence class criteria and tolerance bands; based on the equivalence class criteria and tolerance bands, the computed hash tree and the original hash tree are used to calculate the mass-to-charge ratio tolerance, strength value and retention time to obtain a fidelity score; and a difference report is obtained by performing soft violation downgrading on slight deviations in the fidelity score.

[0045] Equivalence criteria include mass-to-charge ratio tolerance, intensity quantization, and retention time (RT) micro-alignment.

[0046] Tolerance bands include compression / floating-point error tolerance, etc.

[0047] In some embodiments, the processor can define equivalence class criteria and tolerance bands, specifically setting m / z tolerance (e.g., ±5ppm or ±0.01Da), intensity quantization step size, RT alignment window, etc.

[0048] In some embodiments, the processor can, under equivalence class and threshold constraints (comparison of original parsed values ​​and transformed values ​​(less than the tolerance value)), compare the virtual original header with the real original header field obtained from lightweight parsing of the proprietary format, perform fingerprint / hash pair verification calculations, obtain a fidelity score, perform soft violation weighting (not a veto) for minor deviations, and output a difference report. Specifically, the processor can compare the virtual original header with the real header field obtained from lightweight RAW parsing; compare the corresponding block fingerprints with Merkle root data to generate a difference list (which can locate specific spectral segments); calculate the fidelity score and perform soft violation weighting (e.g., if it only occurs in a few low-intensity blocks, the deduction weight is reduced).

[0049] S5: When the fidelity score in the difference report does not meet the threshold, an inconsistency result including locatable difference details and alarms is obtained. When the fidelity score in the difference report meets the threshold, a fidelity certificate is obtained, and the mass spectrometry data conversion fidelity verification is completed.

[0050] Inconsistency results are detailed breakdowns of differences and alerts when the fidelity score fails to meet the threshold.

[0051] A fidelity certificate is a certification result that the fidelity score meets a threshold.

[0052] In some embodiments, the processor may consider data points consistent if the fidelity score meets a threshold (the number and order of data points must be consistent, otherwise inconsistency is considered). Quality values ​​exceeding the tolerance range are considered inconsistent. Intensity values ​​exceeding the tolerance by 5% are considered inconsistent, otherwise inconsistent. Retention times exceeding the tolerance range but within 2% are considered inconsistent, otherwise inconsistent. If the original intensity values ​​in the mzml are consistent, but the quantized values ​​are not equal to the original values ​​after quantization, inconsistency is considered present.

[0053] In some embodiments, the processor can generate a FidelityCertificate containing tool version / parameters, equivalence class definition, difference summary, timestamp or digital signature, and optional on-chain anchoring information when the score meets a threshold, and bind or store it alongside a standard format data file; if the score does not meet the threshold, it outputs a locatable difference detail and an alarm. Specifically, when the score meets the threshold, a FidelityCertificate is generated as follows: Figure 7 The fidelity certificate, in the format shown, includes the name and version of the transformation and verification tool, parameter and equivalence class definitions, difference summary, score and threshold, timestamp or digital signature, Merkle root, and optional on-chain hash. The certificate is saved as a concurrent JSON (or JSON-LD) file or written as a comment in the mzML additional metadata area. If the score fails to meet the requirements, an alert and a re-transformation suggestion are output. The fidelity certificate specifically includes: fidelity certificate number, fidelity evaluation result (consistent) (inconsistent) (difference exists), list of difference spectra (id number), and annotations (describing the fidelity evaluation), etc., in the original proprietary format; file name, file type, file generation time, file hash value, file hash algorithm type, and name of the software that generated the file, etc. / / converted result (public format file information); file name, file type, file generation time, file hash value, file hash algorithm type, and name of the software that generated the file, etc., difference tolerance, quality number tolerance tolerance, intensity tolerance tolerance, and time tolerance tolerance, etc., block hashing rules ("tic", "bpc", "specm-n", where m is the starting spectra ID and n is the ending spectra ID); hash tree (stored in a list format by level, including the root node, left subtree, and right subtree), virtual header data (Base64 encoded), timestamp of the fidelity certificate file's generation (rfc3161) (Base64 encoded), and signature value of the fidelity certificate file (Base64 encoded).

[0054] In some embodiments of this specification, a method for verifying the fidelity of mass spectrometry data conversion using bidirectional simulation and difference notarization is provided. After the conversion between a vendor-proprietary mass spectrometry data format and an open standard format is completed, the method verifies the fidelity and credible notarization of the conversion result. This method generates a fidelity certificate by reconstructing the "virtual original header / parameter digest" from the open standard format and comparing key data blocks with a stable fingerprint / hash. This allows for automated and quantifiable determination of the fidelity of the conversion from proprietary to standard, eliminating trust gaps in the interoperability process. (1) Introducing reverse simulation to compare proprietary header fields and standard metadata in the same semantic domain to avoid "structurally valid but semantically distorted"; (2) Through normalization-quantization-hashing and Merkle trees, it is robust to sorting disturbances, slight compression, and floating-point rounding; differences can be accurately located to the spectrum / time window; (3) Small deviations are weighted to avoid large-scale misjudgments due to small numerical differences, which is more in line with real production data; (4) Automatically generating a fidelity certificate and enabling timestamp / on-chain anchoring to achieve traceability, auditability, and non-repudiation after the fact; it can significantly reduce the cost of long-term dual archiving and manual review; (5) Mapping and criteria are configurable and adaptable to multiple vendors and multiple standard formats; fingerprints and thresholds can be adaptively calibrated according to the scenario, and have the ability to be reused across institutions and industrialized.

[0055] Example 2 Figure 2 This is a schematic diagram of a mass spectrometry data conversion fidelity verification system based on some embodiments of this specification, using bidirectional simulation and difference verification.

[0056] In some embodiments, the mass spectrometry data conversion fidelity verification system with bidirectional simulation and difference notarization may include forward conversion recording, reverse simulation, fingerprint generation, difference comparison, and certificate generation.

[0057] Forward conversion records are used to perform format conversion simulation calculations on proprietary format mass spectrometry data to obtain common format mass spectrometry data.

[0058] Reverse simulation is used to perform reverse simulation of mass spectrometry data using field mapping and priority rules. It reconstructs the virtual header of the common format mass spectrometry data and compares it with the header information of the proprietary format mass spectrometry data. When the consistency comparison fails, an inconsistency result is obtained. When the consistency comparison passes, the common format mass spectrometry data is input into fingerprint generation.

[0059] Fingerprint generation is used to calculate a hash tree based on public format mass spectrometry data according to the configured block rules, and compare it with the original hash tree. When the consistency comparison fails, an inconsistent result is obtained. When the consistency comparison passes, the difference between the calculated hash tree and the original hash tree input is compared.

[0060] The difference comparison is used to calculate the fidelity score by comparing the calculated hash tree with the original hash tree, and to obtain a difference report.

[0061] Certificate generation is used to obtain inconsistency results, including locatable difference details and alarms, when the fidelity score in the difference report does not meet the threshold. When the fidelity score in the difference report meets the threshold, a fidelity certificate is obtained, completing the fidelity verification of mass spectrometry data conversion.

[0062] In some embodiments, the mass spectrometry data conversion fidelity verification system based on two-way simulation and difference verification can be used to perform the mass spectrometry data conversion fidelity verification method based on two-way simulation and difference verification, including: S1: performing format conversion simulation calculation on proprietary format mass spectrometry data to obtain common format mass spectrometry data; S2: performing reverse simulation on the mass spectrometry data using field mapping and priority rules to reconstruct the virtual header of the common format mass spectrometry data, comparing it with the header information of the proprietary format mass spectrometry data, obtaining an inconsistency result when the consistency comparison fails, and inputting the common format mass spectrometry data into S3 when the consistency comparison passes; S3: Based on the public format mass spectrometry data, calculate the hash tree according to the configured block rules and compare it with the original hash tree. If the consistency comparison fails, an inconsistency result is obtained. If the consistency comparison passes, input the calculated hash tree and the original hash tree into S4. S4: Use the calculated hash tree and the original hash tree to calculate the fidelity score and obtain a difference report. S5: If the fidelity score in the difference report does not meet the threshold, obtain an inconsistency result including locatable difference details and alarms. If the fidelity score in the difference report meets the threshold, obtain a fidelity certificate and complete the mass spectrometry data conversion fidelity verification.

[0063] In some embodiments of this specification, the processor uses a bidirectional simulation and difference notarization mass spectrometry data conversion fidelity verification system to perform a bidirectional simulation and difference notarization mass spectrometry data conversion fidelity verification method. In this way, (1) reverse simulation is introduced to compare proprietary header fields and standard metadata in the same semantic domain, avoiding "structurally valid but semantically distorted"; (2) through normalization-quantization-hashing and Merkle trees, it is robust to sorting disturbances, slight compression, and floating-point rounding; the difference can be accurately located to the spectral segment / time window; (3) small deviations are weighted to avoid large-scale misjudgments due to small numerical differences, which is more in line with real production data; (4) fidelity certificates are automatically generated and can be timestamped / on-chain anchored to achieve traceability, auditability, and non-repudiation after the fact; the cost of long-term dual archiving and manual review can be significantly reduced; (5) mapping and criteria are configurable to adapt to multiple vendors and multiple standard formats; fingerprints and thresholds can be adaptively calibrated according to the scenario, and have the ability to be reused across institutions and industrialized.

[0064] Example 3 In some embodiments, a computer-readable storage medium stores computer instructions, and when a computer reads the computer instructions from the storage medium, the computer can execute a substation asset management method.

[0065] It should be noted that different embodiments may produce different beneficial effects. In different embodiments, the beneficial effects may be any one or a combination of the above, or any other possible beneficial effects.< / spectrum> < / spectrumlist> < / run> < / dataprocessinglist> < / instrumentconfigurationlist> < / softwarelist> < / filedescription> < / spectrumlist> < / run> < / instrumentconfiguration> < / filedescription> < / cvparam>

Claims

1. A method for verifying the fidelity of mass spectrometry data conversion using two-way simulation and difference notarization, characterized in that, include: S1: Perform format conversion simulation calculations on proprietary format mass spectrometry data to obtain common format mass spectrometry data; S2: Perform reverse simulation on the mass spectrometry data using field mapping and priority rules, reconstruct the virtual header of the common format mass spectrometry data, and compare it with the header information of the proprietary format mass spectrometry data. If the consistency comparison fails, an inconsistency result is obtained. If the consistency comparison passes, the common format mass spectrometry data is input into S3. S3: Based on the mass spectrometry data in the public format, calculate the hash tree according to the configured block rules, compare it with the original hash tree, and when the consistency comparison fails, obtain the inconsistent result; when the consistency comparison passes, input the calculated hash tree and the original hash tree into S4. S4: Calculate the fidelity score using the calculated hash tree and the original hash tree to obtain a difference report; S5: When the fidelity score in the difference report does not meet the threshold, an inconsistency result including locatable difference details and alarms is obtained. When the fidelity score in the difference report meets the threshold, a fidelity certificate is obtained, and the mass spectrometry data conversion fidelity verification is completed.

2. The method for verifying the fidelity of mass spectrometry data conversion using two-way simulation and difference notarization according to claim 1, characterized in that, S2 includes: Inverse simulation of mass spectrometry data is performed using field mapping and priority rules. Mass spectrometry data in common format is reconstructed according to the corresponding fields to obtain the corresponding virtual head. The virtual head is compared with the header information of the proprietary format mass spectrometry data. If the consistency comparison fails, an inconsistency result is obtained. If the consistency comparison passes, the common format mass spectrometry data is input into S3.

3. The method for verifying the fidelity of mass spectrometry data conversion using two-way simulation and difference notarization according to claim 1, characterized in that, S3 includes: Based on mass spectrometry data in a public format, key time series and spectral segments are normalized, quantized, and hashed according to configured block rules to obtain a hash tree including block fingerprints and Merkle trees; The hash tree is compared with the original hash tree. If the consistency comparison fails, an inconsistent result is obtained. If the consistency comparison passes, the calculated hash tree and the original hash tree are input into S4.

4. The method for verifying the fidelity of mass spectrometry data conversion using two-way simulation and difference notarization according to claim 3, characterized in that, The step of comparing the hash tree with the original hash tree, obtaining an inconsistent result when the consistency comparison fails, and inputting the calculated hash tree and the original hash tree into S4 when the consistency comparison succeeds includes: The hash tree is compared with the original hash tree. Fingerprint generation and comparison are run in streaming / parallel mode on the data platform: block fingerprints and Merkle leaves are calculated for windows composed of multiple spectra; when only consistency proof is required, only Merkle roots and leaves and paths of a few disputed blocks are swapped to protect data privacy; after the certificate is generated, it is automatically archived to the audit storage and associated with the sample list / project number to achieve post-event traceability. When the consistency comparison fails, an inconsistent result is obtained. When the consistency comparison passes, the calculated hash tree and the original hash tree are input into S4.

5. The method for verifying the fidelity of mass spectrometry data conversion using two-way simulation and difference notarization according to claim 1, characterized in that, S4 includes: Define the equivalence class criterion and tolerance band; Based on the equivalence class criterion and tolerance band, the calculated hash tree and the original hash tree are used to calculate the mass-to-charge ratio tolerance, strength value and retention time to obtain the fidelity score. A discrepancy report is obtained by applying soft violations to reduce the weight of minor deviations in the fidelity score.

6. A mass spectrometry data conversion fidelity verification system based on two-way simulation and difference notarization, used to execute the mass spectrometry data conversion fidelity verification method based on two-way simulation and difference notarization as described in claims 1-5, characterized in that, include: Forward conversion records are used to perform format conversion simulation calculations on proprietary format mass spectrometry data to obtain common format mass spectrometry data. Reverse simulation is used to perform reverse simulation of mass spectrometry data using field mapping and priority rules, reconstruct the virtual header of common format mass spectrometry data, compare it with the header information of proprietary format mass spectrometry data, and obtain an inconsistency result when the consistency comparison fails, and input the common format mass spectrometry data into fingerprint generation when the consistency comparison passes. Fingerprint generation is used to calculate a hash tree based on public format mass spectrometry data according to the configured block rules, and compare it with the original hash tree. When the consistency comparison fails, an inconsistent result is obtained. When the consistency comparison passes, the calculated hash tree is compared with the original hash tree input difference. The difference comparison is used to calculate the fidelity score by comparing the calculated hash tree with the original hash tree and obtain a difference report; Certificate generation is used to obtain inconsistency results, including locatable difference details and alarms, when the fidelity score in the difference report does not meet the threshold. When the fidelity score in the difference report meets the threshold, a fidelity certificate is obtained, completing the fidelity verification of mass spectrometry data conversion.

Images