Sample analysis based on fingerprint similarity

Abstract

In some examples, a device may include analyzing a sample and analyzing at least one reference sample. A determination may be made as to whether the sample includes a library match score. Based on determining that the sample includes the library match score, the library match score and retention time may be analyzed to determine a similarity between the sample and the at least one reference sample.

Description

Background Art

[0001] Regarding the fingerprint similarity of fingerprint chromatograms, the similarity assessment involves a comprehensive comparison of the fingerprint chromatograms of a sample and a reference sample to obtain the similarity. In this regard, characteristic peaks can be analyzed simultaneously to evaluate the stability and consistency of product comparisons between batches. The similarity assessment can be used in various fields, such as medicine, environment, food quality, etc. Brief Description of the Drawings

[0002] The features of the present disclosure are shown by way of example and are not limited to one or more of the following drawings, in which like reference numerals indicate like elements, wherein:

[0003] Figure 1 Shows the layout of a sample analysis device based on fingerprint similarity according to an example of the present disclosure;

[0004] Figure 2 Shows a flowchart of a script for checking a reference sample to illustrate Figure 1 the operation of a sample analysis device based on fingerprint similarity;

[0005] Figure 3 Shows a similarity determination script according to an example of the present disclosure to illustrate Figure 1 the operation of a sample analysis device based on fingerprint similarity;

[0006] Figure 4 Shows a flowchart of applying a sample analysis based on fingerprint similarity to a batch of data to illustrate Figure 1 the operation of a sample analysis device based on fingerprint similarity;

[0007] Figure 5 Shows a flowchart of reference type analysis according to an example of the present disclosure to illustrate Figure 1 the operation of a sample analysis device based on fingerprint similarity;

[0008] Figure 6 Shows the chromatogram and mass spectrometry (MS) spectrum of a gas chromatography - mass spectrometry (GCMS) operating in full scan (SCAN) mode according to an example of the present disclosure;

[0009] Figure 7 Shows according to an example of the present disclosure Figure 1 a schematic diagram of the software interface of a sample analysis device based on fingerprint similarity;

[0010] Figure 8 Shows a flowchart according to an example of the present disclosure to show Figure 1Examples of practical implementations of a sample analysis device based on fingerprint similarity;

[0011] Figure 9 Shows an example block diagram for sample analysis based on fingerprint similarity according to an example of the present disclosure;

[0012] Figure 10 Shows a flowchart of an example method for sample analysis based on fingerprint similarity according to an example of the present disclosure; and

[0013] Figure 11 Shows another example block diagram for sample analysis based on fingerprint similarity according to another example of the present disclosure. DETAILED DESCRIPTION

[0014] For simplicity and illustrative purposes, the present disclosure is described primarily with reference to examples. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present disclosure. However, it will be readily apparent that the practice of the present disclosure may not be limited to these specific details. In other instances, some methods and structures are not described in detail to avoid unnecessarily obscuring the present disclosure.

[0015] Throughout the present disclosure, the terms "a" and "an" are intended to denote at least one of a particular element. As used herein, the term "includes" means including but not limited to, and the term "including" means including but not limited to. The term "based on" means at least partially based on.

[0016] Disclosed herein are a sample analysis device based on fingerprint similarity, a method for sample analysis based on fingerprint similarity, and a non-transitory computer-readable medium having stored thereon machine-readable instructions for providing sample analysis based on fingerprint similarity. The disclosed device, method, and non-transitory computer-readable medium use multiple algorithms to provide for assessing the similarity of mass spectrometry fingerprint spectra. The disclosed device, method, and non-transitory computer-readable medium may be particularly relevant in the context of laboratory quality control and impurity analysis in synthetic substances, thereby addressing the need for differential assessment techniques.

[0017] Regarding the fingerprint similarity of fingerprint chromatograms, as disclosed herein, the similarity assessment involves a comprehensive comparison of the fingerprint chromatograms of a sample and a reference sample to obtain a similarity. In this regard, characteristic peaks can be analyzed simultaneously to evaluate the stability and consistency of product comparison between batches. For example, one technique for similarity determination can include analyzing the similarity between a traditional medicine sample and a reference chromatogram by using specific operating rules. However, for such a technique, the similarity determination may be limited to supporting chromatographic data formats and may not be suitable for similarity assessment of complex samples because it relies on retention time as the matching criterion for characteristic peaks shared between samples.

[0018] Some techniques for similarity assessment originated from chromatographic spectroscopy and may be insufficient to address the assessment of low-concentration components in a sample. In addition, when relying on a single reference sample for assessment, individual variations (such as sampling, background noise, and response differences) can introduce biases. Moreover, diverse application scenarios may require different assessment methods, making multiple algorithms more suitable for various samples. For example, complex samples usually contain many compounds, and relying solely on chromatographic separation may not be able to distinguish compounds with similar retention times, potentially leading to biased assessment results. Similarity assessment based on mass spectrometry data can use spectral data as another qualitative basis for assessing compounds between samples. However, implementing similarity assessment using mass spectrometer data may require manual compound identification, data export, manual alignment, and calculation, which can be time-consuming and significantly rely on the experience of analysts, resulting in potential inconsistencies in results between different individuals.

[0019] Currently, fingerprint similarity analysis supports data from gas chromatography-mass spectrometry (GCMS) instruments. GCMS can combine a gas chromatograph, which vaporizes a sample into a gas phase and separates it into various components, and a mass spectrometer, which fragments the components into ionized fragments. This allows the components to be analyzed individually for improved identification and quantification. In this regard, Figure 6 shows the chromatogram 600 and the mass spectrometry (MS) spectrum 602 of a GCMS operating in the full scan (SCAN) mode according to an example of the present disclosure text. For example, Figure 6 shows the chromatogram 600 of the GCMS, where the scanned data points are sequentially shown on the peak shape 604 of the chromatogram. The response value of each data point can be obtained by summing the responses of all ions in the full scan mass spectrum, so the chromatogram is also called the TIC (total ion chromatogram).

[0020] To at least address the above disadvantages (such as those related to the lack of a mass spectrometry data evaluation method, the need for manual verification of reference samples, manual calculations for multiple reference samples, and manual compound identification and alignment), the apparatuses, methods, and non-transitory computer-readable media disclosed herein provide software solutions for similarity assessment using multiple algorithms. The apparatuses, methods, and non-transitory computer-readable media disclosed herein automatically perform qualitative assessment of characteristic markers based on chromatographic retention times and mass spectrometry characteristics. The apparatuses, methods, and non-transitory computer-readable media disclosed herein comprehensively compare the fingerprint spectra of samples and reference samples to determine their similarity, thereby facilitating the evaluation of the similarity of a sample to one or more reference samples. Additionally, the apparatuses, methods, and non-transitory computer-readable media disclosed herein perform necessary inspections and analyses on the characteristic markers in samples and reference samples to ensure the stability and consistency of the similarity assessment process.

[0021] The apparatuses, methods, and non-transitory computer-readable media disclosed herein provide automatic (e.g., without human intervention) and practical applications and / or utilization for similarity determination. For example, a user of the apparatuses, methods, and non-transitory computer-readable media disclosed herein can specify a threshold value for the similarity score (e.g., 80%). When a sample with more than 80% similarity appears, a new selective ion monitoring (SIM) acquisition technique (e.g., only monitoring the remaining characteristic markers) will be automatically created. Compared with the SCAN method (which has lower selectivity and sensitivity), this new technique can be performed on a GCMS instrument to collect more targeted data, thereby further confirming the evaluation results. For example, if there are 3 samples in a batch with more than 80% similarity, 3 new acquisition techniques can be automatically generated. Based on the different evaluation results of the corresponding samples, the newly established techniques can be targeted only at the characteristic compounds (e.g., characteristic markers) remaining in the samples. That is, if some compounds in the sample do not pass the outlier evaluation during the similarity calculation process and are defined as having a response of 0, these compounds will not be included in the new acquisition technique.

[0022] The apparatuses, methods, and non-transitory computer-readable media disclosed herein also provide automatic (e.g., without human intervention) and practical applications and / or utilization for similarity determination through automatic control of the device. For example, if a sample is determined to be similar to a reference sample, the similarity indication can be used to trigger an alarm for the indicated similarity. Further, the triggering of the alarm can cause the device to be shut down or otherwise modify the operation of the device based on the indicated similarity and / or similarity level.

[0023] Furthermore, the devices, methods, and non-transitory computer-readable media disclosed herein encompass functions such as data import, automatic extraction of sample characteristic peaks between batches, qualitative determination of mass spectra based on these peaks, average reference sample characteristic peak response, reference sample qualitative verification, similarity determination, generation of report templates, and help resources. In one example, in an arson investigation, the devices, methods, and non-transitory computer-readable media disclosed herein can be used for forensic identification of accelerant fingerprint chromatograms and for extraction of sample characteristic peak profiles between batches.

[0024] For the devices, methods, and non-transitory computer-readable media disclosed herein, based on the full-scan mass spectrum, the user can use the library search function to search a standard spectral library to identify compounds. In one example, if the qualitative result is BHC (benzene hexachloride), if analyzed using a BHC standard solution, it should have the same retention time and mass spectrum as the sample peak.

[0025] According to the examples disclosed herein, for the devices, methods, and non-transitory computer-readable media disclosed herein, the similarity assessment workflow can support multiple algorithms and automatically perform qualitative determination of characteristic markers based on chromatographic retention time and mass spectral characteristics.

[0026] The operation of the devices, methods, and non-transitory computer-readable media disclosed herein can start by analyzing reference sample data to automatically confirm the reliability of the reference sample based on retention time and mass spectral characteristics. Subsequently, the devices, methods, and non-transitory computer-readable media disclosed herein systematically evaluate the characteristic markers in the sample based on the same rules. When, at the same specific retention time, the mass spectral characteristics of the characteristic markers in the sample match the mass spectral characteristics of the characteristic markers in the reference sample, the peak area (e.g., response area) of the characteristic marker is retained, and otherwise, the peak area is automatically set to zero. The final similarity determination includes determining the arithmetic mean of the peak areas of all the characteristic markers in the reference sample and the remaining characteristic markers in the sample when there are multiple reference samples.

[0027] According to the examples disclosed herein, the devices, methods, and non-transitory computer-readable media disclosed herein can be used for GCMS data acquired using different acquisition modes (including full scan (SCAN), selective ion monitoring (SIM), or multiple reaction monitoring (MRM)). In this regard, the devices, methods, and non-transitory computer-readable media disclosed herein perform automatic (e.g., without human intervention) qualitative determination of the characteristic peaks specified by the user, and subsequently use the responses of these peaks to assess similarity according to specific calculation rules.

[0028] For the devices, methods, and non-transitory computer-readable media disclosed herein, when using SCAN data, a deconvolution algorithm can be used to extract the mass spectra of characteristic peaks. For example, when the sample data collected in the GCMS mode is SCAN data, a deconvolution algorithm can be used to extract the full-scan mass spectral features of characteristic markers. The differences between the mass spectra of peaks with the same retention time in samples between batches can be compared with a reference fingerprint chromatogram. In this regard, the characteristic markers in the sample and the reference sample can be compared, and their differences in mass spectral features can be evaluated based on the matching score to perform qualitative determination.

[0029] The devices, methods, and non-transitory computer-readable media disclosed herein also perform automatic qualitative determination based on the matching scores of these mass spectra. For example, regarding two available qualitative techniques, one of the qualitative techniques can be automatically selected. This is because if the user sets up a reference library, there will be values in the "library matching score" column as disclosed herein. The presence of values can mean that the user has selected to use the retention time (RT) and the library matching score as the qualitative basis. Alternatively, if there are no values in the library matching score, this can mean that the user has selected to use the retention time and the ion ratio as the qualitative basis. The user can also perform a similarity analysis, in which case a qualitative judgment can be automatically performed for all compounds. Based on the rules disclosed herein, the response area of a compound with an outlier in the sample can be defined as 0, and the final response area value can be used for analysis by using the selected algorithm, where the final result is listed in the "similarity %" column as disclosed herein.

[0030] For the devices, methods, and non-transitory computer-readable media disclosed herein, when using SIM or MRM data, the degree of alignment between qualitative ions and the quantitative ion ratio of characteristic peaks can be used as the basis for automatic qualitative determination. For example, when the sample data is collected in the SIM or MRM mode or as SCAN data without specifying a reference library, the qualitative ions or ion pairs that match at the same retention time between the sample and the reference sample can be used as the basis for automatic qualitative determination. This can mean that the same characteristic markers in the sample and the reference sample should exhibit the same ion ratio. When using the SCAN mode, the retention time and the mass spectral matching score can be used as two bases for the qualitative analysis of compounds. Regarding the SIM and MRM modes, since there is no full mass spectrum for comparison, the comparison can be performed by the ratio of ions (or the ratio of ion transitions for MRM) because for a specific compound, its ion ratio is also consistent within a certain range. Therefore, when using the SIM or MRM mode, the retention time and the ion ratio can be two bases for the qualitative analysis of compounds.

[0031] To determine aspects such as same retention time, same ion ratio, or acceptable match score, these aspects can be based on user settings of outliers (e.g., quantification method). For example, the default outlier setting for retention time can be specified as 10%, and the more common unit can be minutes. For example, if the default outlier setting for retention time is set to 0.2 minutes, compounds with a retention time difference greater than 0.2 minutes can be considered different compounds. Similarly, if a compound in a fire debris sample has a library match score below the outlier, this can be considered different from the compound in the reference sample.

[0032] For the devices, methods, and non-transitory computer-readable media disclosed herein, when using SCAN data and the user does not specify a reference full-scan mass spectrum of a characteristic peak, the alignment between the qualitative ions and the quantitative ions of the characteristic peak can be used as a basis for automatic qualitative determination. In this regard, as disclosed herein, one of these two qualitative techniques can be automatically selected because if the user sets a reference library, there will be a value in the "library match score" column as disclosed herein. The presence of the value can mean that the user has chosen to use retention time and library match score as the qualitative basis. Alternatively, if there is no value in the library match score, this can mean that the user has chosen to use retention time and ion ratio as the qualitative basis.

[0033] Based on the utilization of MS data, for the devices, methods, and non-transitory computer-readable media disclosed herein, the evaluation results include improved accuracy and efficiency. In this regard, the devices, methods, and non-transitory computer-readable media disclosed herein can be used, for example, to confirm an accelerant fingerprint profile in forensic fire investigation cases within the field of public safety. Additionally, the devices, methods, and non-transitory computer-readable media disclosed herein can be used in various fields such as flavor analysis, traditional medicine fingerprint profiling, environmental pollutant analysis, food quality control, drug quality control, chemical product quality control, illegal additive detection, differential analysis, and traceability analysis, etc.

[0034] The devices, methods, and non-transitory computer-readable media disclosed herein provide similarity assessment for GCMS that goes beyond the chromatographic assessment methods seen in pharmacopoeia methods. In this regard, the similarity based on chromatographic methods can be based on whether the compounds in the sample and the reference sample are consistent based on retention time. Fingerprint similarity can be used to automatically confirm the qualitative identification of compounds based on library match score or ion ratio rather than retention time. In addition to assessing chromatographic similarity, the devices, methods, and non-transitory computer-readable media disclosed herein can utilize mass spectrometry features to automatically qualitatively determine characteristic peaks. This can include eliminating interfering peaks and ensuring the reliability of similarity results by excluding peaks with the same retention time but different mass spectrometry features.

[0035] Regarding the automatic qualitative determination from full-scan data, for the apparatuses, methods, and non-transitory computer-readable media disclosed herein, the mass spectrometry that extracts characteristic peaks using a deconvolution algorithm ensures the acquisition of spectra with rich characteristics and no interference. This precision provides accurate automatic qualitative results, ensuring the accuracy of interference peak removal and obtaining reliable similarity evaluation results.

[0036] In one example, the apparatuses, methods, and non-transitory computer-readable media disclosed herein provide for using cosine similarity for gas chromatography-mass spectrometry (GC-MS) spectral similarity. However, other techniques can be used for similarity determination. Although the algorithm can be analyzed based on the response area of compounds, some compounds determined in the sample to be inconsistent with those in the reference sample can be defined using an area of 0. In the field of instrumental analysis for evaluating similarity between mass spectra, similarity analysis can be implemented by the apparatuses, methods, and non-transitory computer-readable media disclosed herein.

[0037] When multiple reference samples are available, the peak areas (e.g., response areas) of the characteristic markers used for similarity analysis can be determined as the arithmetic mean of the peak areas from all reference samples, along with the remaining characteristic markers in the sample after automatic qualitative determination.

[0038] Characteristic markers may need to meet specific conditions to be considered eligible criteria. First, a determination can be made as to whether a reference spectrum exists (indicated by the presence of characteristic markers from the reference sample and the matching score with the reference library). If there is a matching score, a characteristic marker can be considered to meet the eligibility criteria when the matching score of each individual characteristic marker is greater than or equal to a specific value (user-defined) and the retention time deviation is less than or equal to a specific value (e.g., user-defined value). If there is no matching score in the library, a characteristic marker can be considered to meet the eligibility criteria when the ion response ratio deviation is less than or equal to a specific value (user-defined) and the retention time deviation is less than or equal to a specific value (e.g., user-defined). If any one of the retention time or mass spectrometry characteristic eligibility criteria for an individual characteristic marker is not met, the workflow indicates that the reference sample does not meet the eligibility criteria.

[0039] To remove interference peaks (e.g., set the peak area to zero), a comprehensive determination can be made based on the retention time and mass spectrometry characteristics of the characteristic markers. Specifically, if the retention time deviation is less than or equal to a specific value (user-defined) and the mass spectrometry characteristics also meet the eligibility criteria, the peak area of the characteristic marker can be retained. If any one of the retention time or mass spectrometry characteristic criteria is not met, the peak area of the characteristic marker can be set to zero.

[0040] The determination of the similarity assessment based on the similarity of the characteristic markers of the sample and the reference sample enables the use of the peak area or peak height of the quantitative ion or ion pair, or the sum of the peak areas or peak heights of multiple ions or ion pairs.

[0041] Multiple algorithms can be used to perform the similarity assessment, and the multiple algorithms can include entropy similarity, cosine similarity, Pearson correlation coefficient, Mahalanobis distance, K-L divergence, etc.

[0042] The devices, methods, and non-transitory computer-readable media disclosed herein thus provide a comprehensive multi-algorithm-based workflow for assessing the similarity of mass spectrometry fingerprints. The devices, methods, and non-transitory computer-readable media disclosed herein address the challenges of the prior art by automating the qualitative assessment of characteristic markers based on chromatographic retention time and mass spectrometry characteristics and employing various algorithms for similarity evaluation. This method enhances the accuracy and efficiency of the similarity assessment of samples and reference samples, providing valuable insights for laboratory quality control and impurity analysis in diverse applications.

[0043] For the devices, methods, and non-transitory computer-readable media disclosed herein, the elements of the devices, methods, and non-transitory computer-readable media can be any combination of hardware and programming to implement the functions of the corresponding elements. In some examples described herein, the combination of hardware and programming can be implemented in a variety of different ways. For example, the programming for an element can be processor-executable instructions stored on a non-transitory machine-readable storage medium, and the hardware for the element can include processing resources for executing those instructions. In these examples, the computing device implementing such an element can include a machine-readable storage medium storing the instructions and processing resources for executing the instructions, or the machine-readable storage medium can be stored and accessed separately by the computing device and the processing resources. In some examples, some elements can be implemented in circuitry.

[0044] Figure 1 Shows the layout of an example sample analysis device (hereinafter also referred to as "Device 100") based on fingerprint similarity.

[0045] Reference Figure 1 , Device 100 can include a reference sample type analyzer 102, which is executed by at least one hardware processor (e.g., Figure 9 hardware processor 902 and / or Figure 11 hardware processor 1104) to analyze sample 104 and analyze at least one reference sample 106.

[0046] By at least one hardware processor (e.g., Figure 9 hardware processor 902 and / or Figure 11The library match score analyzer 108 executed by at least one hardware processor (e.g., the hardware processor 1104) can determine whether the sample 104 includes a library match score 110.

[0047] Executed by at least one hardware processor (e.g., Figure 9 the hardware processor 902 and / or Figure 11 the hardware processor 1104), the library match score and retention time analyzer 112 can analyze the library match score 110 and the retention time 114 based on determining that the sample 104 includes the library match score 110 to determine the similarity 116 between the sample 104 and the at least one reference sample 106.

[0048] Executed by at least one hardware processor (e.g., Figure 9 the hardware processor 902 and / or Figure 11 the hardware processor 1104), the ion ratio and retention time analyzer 118 can analyze the ion ratio 120 and the retention time 114 based on determining that the sample 104 does not include the library match score 110 to determine the similarity 116 between the sample 104 and the at least one reference sample 106.

[0049] Executed by at least one hardware processor (e.g., Figure 9 the hardware processor 902 and / or Figure 11 the hardware processor 1104), the similarity analyzer 122 can determine the similarity 116 between the sample 104 and the at least one reference sample 106 based on cosine similarity. Alternatively or additionally, the similarity analyzer can determine the similarity 116 between the sample 104 and the at least one reference sample 106 based on entropy similarity.

[0050] According to the examples disclosed herein, the library match score and retention time analyzer 112 can analyze the library match score 110 and the retention time 114 based on determining that the sample 104 includes the library match score 110 to determine the similarity 116 between the sample 104 and the at least one reference sample 106 in the following manner: for a single reference sample entry, determine the response area of the compound associated with the single reference sample entry. Alternatively, the library match score and retention time analyzer 112 can determine the average response area of the compounds associated with the multiple reference sample entries for multiple reference sample entries.

[0051] According to the examples disclosed herein, the ion ratio and retention time analyzer 118 can analyze the ion ratio 120 and retention time 114 to determine the similarity 116 between the sample 104 and the at least one reference sample 106 by the following means based on determining that the sample 104 does not include the library match score 110: For a single reference sample entry, determine the response area of the compound associated with the single reference sample entry. Alternatively, the ion ratio and retention time analyzer 118 can determine the average response area of the compounds associated with the multiple reference sample entries for multiple reference sample entries.

[0052] Refer to Figures 1 - 8 The operation of the device 100 is described in more detail.

[0053] Figure 2 A flowchart showing a check reference sample script according to an example of the present disclosure text is shown to illustrate the operation of the device 100.

[0054] Reference Figure 2 , at block 200, the reference sample type analyzer 102 can analyze the reference sample type with respect to the sample 104 of 202 and the reference sample 106 relative to 204. Taking the analysis of gasoline residues in fire debris samples by the user as an example to illustrate the operation of the device 100, examples of the sample 104 analyzed by the GCMS instrument include fire debris samples, and examples of the reference sample 106 can include standard gasoline. There can be one reference sample, or multiple reference samples can be used, such as gasoline from different gas stations or gasoline with different octane ratings.

[0055] At block 200, based on the determination by the reference sample type analyzer 102 that these values (e.g., different standard names) are different, at block 206, the reference sample type analyzer 102 can generate an error notification. For example, the user can specify a reference sample in the "standard name" column (e.g., see Figure 7 ), such that the reference sample type analyzer 102 knows which sample or samples are reference samples. Since different types of samples may not be analyzed simultaneously, if the user enters multiple different values, subsequent runs will generate an error notification. For example, while one or more "gasoline" types can be entered and analyzed, these samples may not have one "gasoline" and one "kerosene".

[0056] Alternatively, at block 200, based on the determination by the reference sample type analyzer 102 that these values are the same (e.g., the user has entered duplicate standard names), at block 208, the library match score analyzer 108 can determine the library match score 110. In this regard, the library match score analyzer 108 can determine whether the sample includes a library match score.

[0057] Based on the library match score 110 determined by the library match score analyzer 108 having a value (e.g., the user specifies a reference library in a quantitative method, and the library match score of the compound will be displayed), at block 210, the library match score and retention time analyzer 112 can determine the library match score 110 and the retention time 114. In this regard, the library match score and retention time analyzer 112 can analyze the library match score and the retention time based on determining that the sample includes a library match score to determine the similarity between the sample and the at least one reference sample.

[0058] Based on the library match score 110 determined by the library match score analyzer 108 having no value, at block 212, the ion ratio and retention time analyzer 118 can determine the ion ratio 120 and the retention time 114. Thus, the ion ratio and retention time analyzer 118 can analyze the ion ratio and the retention time based on determining that the sample does not include a library match score to determine the similarity between the sample and the at least one reference sample. In this regard, one of these two qualitative techniques can be selected by the software because if the user sets a reference library, there will be a value in the "library match score" column in the software. The presence of the value can mean that the user has chosen to use the retention time and the library match score as the qualitative basis. On the contrary, if there is no value in the library match score, this means that the user has chosen to use the retention time and the ion ratio as the qualitative basis. An example of an interface showing the relevant parameters is shown Figure 7 in.

[0059] At block 212, based on the determination by the ion ratio and retention time analyzer 118 that a match exists (e.g., a match means that there are no outliers), at block 214, the ion ratio and retention time analyzer 118 can generate an indication of pass. In this regard, the user can execute a check script, the purpose of which is to determine that all compounds in the reference sample do not have retention time outliers and library match score outliers (or do not have retention time outliers and ion ratio outliers). This is because if there are outliers even in the reference sample, this means that the reference sample itself has problems and is not suitable for comparison. The user can correct these problems manually, or can ignore them and proceed directly to the final similarity determination.

[0060] At block 212, based on the determination by the ion ratio and retention time analyzer 118 that outliers exist, at block 216, the ion ratio and retention time analyzer 118 can generate an indication of an error notification (e.g., a reminder for the user to check the reference sample). For example, the user can run a check script, the purpose of which is to determine that all compounds in the reference sample do not have retention time outliers and library match score outliers (or do not have retention time outliers and ion ratio outliers). This is because if outliers exist even in the reference sample, it means that the reference sample itself has problems and is not suitable for comparison. In this regard, the user can manually correct these problems or can ignore them and proceed directly to the final similarity determination.

[0061] At block 218, a manual check can be implemented. In this regard, the user can choose to ignore the error notification.

[0062] Figure 3 A similarity determination script according to an example of the present disclosure is shown to illustrate the operation of the apparatus 100. At the start, a prompt indicating "Confirm the check of ion ratio and last RT" can be displayed. This prompt can request whether the user has checked the reference sample.

[0063] Reference Figure 3 , at block 300, the reference sample type analyzer 102 can determine the type of the reference sample 106. In this regard, the user can choose to directly run the analysis without checking the reference sample, in which case the reference sample type analyzer 102 can determine the type of the reference sample 106.

[0064] Based on the determination that there is no reference sample or the reference sample has different values, at block 302, the reference sample type analyzer 102 can generate an indication of an error notification. In this regard, if the user does not specify a reference sample, the reference sample type analyzer 102 cannot determine which sample to use as a reference. Further, the user may need to specify the reference sample in the "Standard Name" column (e.g., see Figure 7 ), such that the reference sample type analyzer 102 knows which sample or samples are the reference samples. Since the reference sample type analyzer 102 does not determine different types of samples simultaneously, if the user inputs multiple different values, subsequent runs will report an error notification. For example, one or more "gasoline" types can be input, but not one "gasoline" and one "kerosene".

[0065] Based on the determination that there is a reference sample or the reference sample has the same values, at block 304, the reference sample type analyzer 102 can determine the sample type. For example, the user can input "gasoline" as the reference sample for multiple samples.

[0066] Based on the determination by the reference sample type analyzer 102 in block 304 that the sample type is a reference sample, the processing can proceed to block 306. Further, the average area can be determined in block 308, and the area can be determined in block 310. For example, if the user enters multiple identical words in the "Standard Name" column (e.g., see Figure 7 ), the reference sample type analyzer 102 can consider the average response area of the same compound in these reference samples as the response area to be used in the similarity analysis. The average response area in 308 can be determined based on multiple identical values, or the response area can be determined in 310 based on a single entry.

[0067] Based on the determination by the reference sample type analyzer 102 in block 304 that the sample type is a sample (e.g., the "Standard Name" column specified by the user), the processing can proceed to block 312. At block 312, the library match score analyzer 108 can determine the library match score 110. The library match score analyzer 108 can determine whether the sample includes a library match score. In this regard, the library match score analyzer 108 can automatically select two qualitative techniques. This is because if the user sets a reference library, there will be a value in the Figure 7 "Library Match Score" column. The presence of the value can indicate that the user has selected to use retention time and library match score as the qualitative basis. Conversely, if there is no value in the library match score, this indicates that the user has selected to use retention time and ion ratio as the qualitative basis.

[0068] Based on the determination by the library match score analyzer 108 that the library match score has a value, at block 314 (similar to block 210), the library match score and retention time analyzer 112 can determine the library match score 110 and the retention time 114.

[0069] Based on the determination by the library match score analyzer 108 that the library match score has no value, at block 316 (similar to block 212), the ion ratio and retention time analyzer 118 can determine the ion ratio 120 and the retention time 114.

[0070] At block 316, based on the determination by the ion ratio and retention time analyzer 118 that there is no match, at block 318, the ion ratio and retention time analyzer 118 can zero out the peaks, as disclosed herein. In this regard, the ion ratio and retention time analyzer 118 can confirm the compounds in the sample one by one and process the responses of the compounds according to whether the compounds meet (match) the outlier requirements. If the retention time and ion ratio of the compound in the sample are "matched" (within the set outliers), the response of this compound will be retained. Conversely, if either the retention time or the ion ratio of the compound "does not match" (e.g., exceeds the set outlier range), the response of this compound will be set to 0.

[0071] At block 320, a response area can be determined as disclosed herein.

[0072] At block 322, the response area that has been qualitatively determined can be input into an algorithm selected by the user for analysis, and a similarity % result can be output.

[0073] Figure 4 A flowchart showing the application of sample analysis based on fingerprint similarity to a batch of data according to an example of the present disclosure is shown to illustrate the operation of apparatus 100.

[0074] Reference Figure 4 , at block 400, data can be imported from an icon (e.g., the "Add Sample" button in MassHunter software).

[0075] At block 402, if spectral library techniques are utilized, the spectral library match score outliers can be set to a specified value (e.g., 60%). In this regard, to determine whether there is the same retention time, the same ion ratio, or an acceptable match score, these aspects can be based on user settings of the outliers (e.g., quantitative methods). For example, the default outlier setting for retention time is 10%, and the more common unit is minutes. For example, if the outlier setting is set to 0.2 minutes, compounds with a retention time difference greater than 0.2 minutes will be considered different compounds. In the same way, if a compound in a fire debris sample has a library match score lower than the outlier, the compound is considered different from the compound in the reference sample.

[0076] At block 404, with respect to the outlier setting, for the retention time screen, a retention time window value (e.g., 0.2 minutes), also known as the retention time outlier, can be input.

[0077] At block 406, with respect to the method setting task, for the qualitative ion setting screen, an uncertainty value (e.g., 500) can be input into the qualitative ion table. This uncertainty value can serve as an outlier for a qualifier. For example, if the uncertainty value is 20%, for an ion with an ion ratio equal to 100, a range of 80 - 120 can be specified as normal.

[0078] At block 408, a quantification method can be applied to this batch of data. In this regard, after data is collected by the GCMS instrument, all data analysis tasks can be completed in quantification software (e.g., MassHunter Quantitative Analysis software). Calculations performed by MassHunter may require the user to specify a quantification method. Such a quantification method may need to include the name of the compound to be analyzed and the quantification ion (or ion transitions for MRM) used for quantification, as well as the retention time of the compound. Various techniques can be used to create a quantification method, such as the software automatically finding compounds, the user manually creating them, etc.

[0079] Figure 5 A flowchart of reference type analysis according to an example of the present disclosure is shown to illustrate the operation of apparatus 100.

[0080] Reference Figure 5 , at block 500, the reference sample type analyzer 102 can determine the reference sample type (e.g., as disclosed herein with reference to Figure 2 ).

[0081] At block 502, based on determining that the reference sample type is a standard name blank, an error notification can be generated at block 504. In this regard, as disclosed herein, the user may need to specify a reference sample in the "Standard Name" column (e.g., see Figure 7 ), such that the reference sample type analyzer 102 knows which sample or samples are reference samples. Since different types of samples may not be analyzed simultaneously, subsequent runs will generate an error notification if the user enters multiple different values. For example, while one or more "gasoline" types can be entered and analyzed, these samples may not have both a "gasoline" and a "kerosene".

[0082] At block 506, based on determining that the reference sample type is a standard name, as disclosed herein, at block 508, the library match score analyzer 108 can determine the library match score 110.

[0083] Based on the library match score analyzer 108 determining that the library match score has no value, at block 512 (similar to block 212), the ion ratio and retention time analyzer 118 can determine the ion ratio 120 and the retention time 114.

[0084] Based on the library match score analyzer 108 determining that the library match score has a value, at block 514 (similar to block 210), the library match score and retention type analyzer 112 can determine the library match score 110 and the retention time 114.

[0085] At block 512, based on a match being determined by the ion ratio and retention time analyzer 118 to exist, at block 516, the ion ratio and retention time analyzer 118 can generate an indication of passing (e.g., no outliers in retention time and library match scores as disclosed herein with respect to blocks 210 to 214).

[0086] At block 514, based on an outlier being determined by the ion ratio and retention time analyzer 118 to exist, at block 518, an indication of an error notification (e.g., a failure) can be generated (e.g., as disclosed herein with respect to blocks 210 to 216).

[0087] As disclosed herein, Figure 6 chromatograms and MS spectra of a GCMS operating in SCAN mode according to an example of the present disclosure are shown.

[0088] Reference Figure 6 , Figure 6 shows a chromatogram of a GCMS, where the peak shape of the chromatogram includes data points of a scan. The response value of each data point can be obtained by summing the responses of all ions in the full scan mass spectrum, and thus the chromatogram is also referred to as a TIC (total ion chromatogram).

[0089] Based on the full scan mass spectrum, a user can use the library search function to search a standard spectral library to identify compounds. In one example, if the qualitative result is BHC (benzene hexachloride), if analyzed using a BHC standard solution, it should have the same retention time and mass spectrum as the sample peak.

[0090] Regarding the SIM and MRM modes, since there is no full mass spectrum for comparison, the comparison can be performed by the ratio of ions (or the ratio of ion transitions for MRM), because for a specific compound, its ion ratio is also consistent within a certain range. Therefore, when using the SIM or MRM mode, the retention time and the ion ratio can be two bases for the qualitative analysis of compounds.

[0091] For a user's gasoline residue analysis of a fire debris sample, the user's purpose can include determining whether gasoline exists in the fire debris sample and thus inferring whether arson is involved. In this regard, examples of samples analyzed by a GCMS instrument can include fire debris samples, and examples of reference samples can include standard gasoline. There can be one reference sample, or multiple reference samples can be used, such as gasoline from different gas stations or gasoline with different octane numbers.

[0092] After the GCMS instrument has collected data, all data analysis tasks can be completed in quantification software such as MassHunter Quantitative Analysis Software. The operations performed by MassHunter may require the customer to specify a quantification method. Such a quantification method may need to include the name of the compound to be analyzed, the quantification ions (or ion transitions for MRM) used for quantification, and the retention time of the compound.

[0093] The MassHunter software can use an integrator to integrate the chromatographic peaks of specific quantification ions at specific retention times and obtain peak areas. In this regard, to qualitatively determine whether chromatographic peaks with the same retention time in different samples are the same compound, if the chromatographic peaks have the same retention time in the fire debris sample and the gasoline reference, two methods can be used to determine whether they are the same compound. First, if the data was collected in SCAN mode and the user specifies a reference mass spectral library in the quantification method (this reference library can be generated directly from the reference sample but may need to be manually specified in the quantification method), then the qualitative confirmation can be based on the library match score of the compounds between the fire debris sample and the reference sample at the same RT. Second, if the data was collected in SIM or MRM mode, or the user did not specify a reference library, then the qualitative confirmation can be based on the ion ratios of the compounds between the fire debris sample and the reference sample at the same RT.

[0094] To determine whether there is the same retention time, the same ion ratio, or an acceptable match score, these aspects can be based on user settings for outliers (e.g., the quantification method). For example, the default outlier setting for retention time is 10%, and the more common unit is minutes. For example, if the outlier setting is set to 0.2 minutes, compounds with a retention time difference greater than 0.2 minutes will be considered different compounds. In the same way, if a compound in the fire debris sample has a library match score below the outlier, then the compound is considered different from the compound in the reference sample.

[0095] Regarding the two available qualitative techniques, one of the qualitative techniques can be automatically selected. This is because if the user sets a reference library, there will be a value in the "Library Match Score" column as disclosed herein. The presence of a value can mean that the user has chosen to use retention time and library match score as the qualitative basis. Alternatively, if there is no value in the library match score, this can mean that the user has chosen to use retention time and ion ratio as the qualitative basis.

[0096] The user can import the collected data into the data analysis software and establish a quantification method (e.g., including a list of compounds to be compared such as compound names, retention times, quantification and qualitative ions), where the retention time outlier software has default values.

[0097] Figure 7 Shows a schematic diagram of the software interface of the device 100 according to an example of the present disclosure.

[0098] Reference Figure 7 , the user can specify a reference sample in the "Standard Name" column (for example, see Figure 7 ), so that the reference sample type analyzer 102 knows which sample or samples are reference samples. Since different types of samples may not be analyzed simultaneously, if the user enters multiple different values, a subsequent run will generate an error notification. For example, while one or more "gasoline" types can be entered and analyzed, these samples may not have both a "gasoline" and a "kerosene".

[0099] A check script can be executed to determine that all compounds in the reference sample do not have retention time outliers and library match score outliers (or do not have retention time outliers and ion ratio outliers). If there are outliers in the reference sample, the reference sample itself may have problems and may not be suitable for comparison. The user can manually correct these problems, or can ignore them and proceed directly to the final similarity analysis.

[0100] The user can also perform a similarity analysis, at which time a qualitative judgment can be automatically performed on all compounds. Based on the rules disclosed herein, the response area of a compound with an outlier in the sample can be defined as 0, and the final response area value can be used for analysis by using the selected algorithm, where the final result is listed in the "Similarity %" column as disclosed herein.

[0101] Regarding the similarity analysis between the sample and the reference sample, an example of the technique utilized herein can utilize the compound response area after an automatic process (e.g., the performed qualitative judgment).

[0102] The first example of the technique for similarity analysis herein is specified as follows: For each compound in the quantitative method, A i is the response area in the sample, while A s is the response area in the reference sample. The dot product or cosine similarity can represent the algebraic operation of vectors. In some cases, an algorithm can be utilized to treat the MS spectrum as a vector and determine the library match score between the sample compound and the standard library.

[0103] The second example of the technique for similarity analysis herein is specified as follows: I i is the normalized intensity of compound i; For the entropy similarity, S A and S B are the entropies of the sample and the reference sample, respectively.

[0104] Figure 8 The flowchart according to an example of the present disclosure is shown to illustrate an example of the actual implementation of the apparatus 100.

[0105] Referring to Figure 8 , the similarity results at 800 (e.g., similarity 116) can be analyzed at 802, where compounds with a retention time difference less than the interval value can be kept in the same time period.

[0106] The similarity results at 800 (e.g., similarity 116) can be analyzed at 804, where new time periods can be created for compounds with a retention time difference greater than the interval value.

[0107] At 806, duplicate ions can be removed from the same time period.

[0108] At 808, the number of ions in the same time period can be determined.

[0109] At 810, the total number of time periods can be determined.

[0110] At 812, the acquisition technique can be output in SIM mode. In this regard, as disclosed herein, the user can specify a critical value of the similarity score (e.g., 80%). When a sample with a similarity of more than 80% appears, a new SIM acquisition technique (e.g., only monitoring the remaining characteristic markers) will be automatically created. This new technique can be executed on a GCMS instrument to collect more target data, thereby further confirming the evaluation results. For example, if there are 3 samples in a batch with a similarity greater than 80%, 3 new acquisition techniques can be automatically generated. Based on the different evaluation results of the corresponding samples, the established new technique can be only for the characteristic compounds (e.g., characteristic markers) retained in the sample. That is to say, if some compounds in the sample do not pass the outlier evaluation during the similarity calculation and are defined as having a response of 0, these compounds will not be included in the new acquisition technique.

[0111] At 814, if the number of ions is greater than an associated value (e.g., 60) or the number of time periods is greater than an associated value (e.g., 100), a determination can be made as to whether using a retention time interval = 0.1 minute meets the following requirements: the number of ions < 60 and the number of time periods < 100; if so, re-enter the retention time interval, and if not, return an "error" message.

[0112] At 816, a similarity cutoff value and a retention time interval value can be entered.

[0113] Figures 9 - 11 Accordingly, an example block diagram 900 for sample analysis based on fingerprint similarity, a flowchart of an example method 1000, and another example block diagram 1100 are shown. Block diagram 900, method 1000, and block diagram 1100 can be implemented on the apparatus 100 described above by way of example and not limitation. Block diagram 900, method 1000, and block diagram 1100 can be practiced in other apparatuses. In addition to showing block diagram 900, Figure 1 hardware of the apparatus 100 that can execute the instructions of block diagram 900 is shown. The hardware can include a processor 902 and a memory 904 that stores machine-readable instructions, which when executed by the processor cause the processor to execute the instructions of block diagram 900. Memory 904 can represent a non-transitory computer-readable medium. Figure 9 can represent an example method for sample analysis based on fingerprint similarity and the steps of the method. Figure 10 can represent a non-transitory computer-readable medium 1102 according to an example, on which machine-readable instructions for providing sample analysis based on fingerprint similarity are stored. The machine-readable instructions, when executed, cause a processor 1104 to execute the instructions of block diagram 1100 also shown in Figure 11 The processor 902 of and / or Figure 11 The processor 1104 of can include a single or multiple processors or other hardware processing circuits to execute the methods, functions, and other processes described herein. These methods, functions, and other processes can be embodied as machine-readable instructions stored on a computer-readable medium, which can be non-transitory (e.g.,

[0114] Figure 9 the non-transitory computer-readable medium 1102 of ), such as a hardware storage device (e.g., RAM (random access memory), ROM (read-only memory), EPROM (erasable programmable ROM), EEPROM (electrically erasable programmable ROM), hard disk drive, and flash memory). Memory 904 can include RAM, where the machine-readable instructions and data for the processor can reside during operation. Figure 11 The processor 1104 of can include a single or multiple processors or other hardware processing circuits to execute the methods, functions, and other processes described herein. These methods, functions, and other processes can be embodied as machine-readable instructions stored on a computer-readable medium, which can be non-transitory (e.g., Figure 11 the non-transitory computer-readable medium 1102 of ), such as a hardware storage device (e.g., RAM (random access memory), ROM (read-only memory), EPROM (erasable programmable ROM), EEPROM (electrically erasable programmable ROM), hard disk drive, and flash memory). Memory 904 can include RAM, where the machine-readable instructions and data for the processor can reside during operation.

[0115] Reference Figures 1 - 9 and, in particular, reference is made to Figure 9 the block diagram 900 shown in, the memory 904 may include instructions 906 for analyzing the sample 104.

[0116] The processor 902 may extract, decode, and execute instructions 908 to analyze at least one reference sample 106.

[0117] The processor 902 may extract, decode, and execute instructions 910 to determine whether the sample 104 includes a library match score 110.

[0118] The processor 902 may extract, decode, and execute instructions 912 to analyze the library match score 110 and the retention time 114 based on determining that the sample 104 includes a library match score 110 to determine the similarity 116 between the sample 104 and the at least one reference sample 106.

[0119] Reference Figures 1 - 8 and Figure 10 and, in particular, reference is made to Figure 10 For the method 1000, at block 1002, the method may include analyzing the sample 104.

[0120] At block 1004, the method may include analyzing at least one reference sample 106.

[0121] At block 1006, the method may include determining whether the sample 104 includes a library match score 110.

[0122] At block 1008, the method may include analyzing the ion ratio 120 and the retention time 114 based on determining that the sample 104 does not include a library match score 110 to determine the similarity 116 between the sample 104 and the at least one reference sample 106.

[0123] Reference Figures 1 - 8 and Figure 11 and, in particular, reference is made to Figure 11 For the block diagram 1000, the non - transitory computer - readable medium 1002 may include instructions 1006 for determining whether a sample includes a library match score.

[0124] The processor 1104 may extract, decode, and execute instructions 1108 to analyze the ion ratio and the retention time based on determining that the sample does not include a library match score to determine the similarity between the sample and at least one reference sample.

[0125] The processor 1104 may extract, decode, and execute instructions 1110 to analyze the library match score and the retention time based on determining that the sample includes a library match score to determine the similarity between the sample and the at least one reference sample.

[0126] What has been described and shown herein are examples and some variations thereof. The terms, descriptions, and drawings used herein are set forth by way of illustration only and are not meant to be limiting. Many variations are possible within the spirit and scope of the subject matter, which is intended to be defined by the appended claims and their equivalents, where all terms are expressed in their broadest reasonable meaning unless otherwise stated.

[0127] The present invention also includes the following items: 1. A sample analysis device based on fingerprint similarity, comprising: A reference sample type analyzer, which is executed by at least one hardware processor to: Analyze a sample; Analyze at least one reference sample; A library match score analyzer, which is executed by the at least one hardware processor to: Determine whether the sample includes a library match score; and A library match score and retention time analyzer, which is executed by the at least one hardware processor to: Analyze the library match score and retention time based on determining that the sample includes the library match score to determine the similarity between the sample and the at least one reference sample. 2. The sample analysis device based on fingerprint similarity according to item 1, further comprising: An ion ratio and retention time analyzer, which is executed by the at least one hardware processor to: Analyze the ion ratio and the retention time based on determining that the sample does not include the library match score to determine the similarity between the sample and the at least one reference sample. 3. The sample analysis device based on fingerprint similarity according to item 1, further comprising: A similarity analyzer, which is executed by the at least one hardware processor to: Determine the similarity between the sample and the at least one reference sample based on cosine similarity. 4. The sample analysis device based on fingerprint similarity according to item 1, further comprising: A similarity analyzer, which is executed by the at least one hardware processor to: Determine the similarity between the sample and the at least one reference sample based on entropy similarity. 5. The sample analysis device based on fingerprint similarity according to item 1, wherein the library match score and retention time analyzer is executed by the at least one hardware processor to analyze the library match score and the retention time based on determining that the sample includes the library match score to determine the similarity between the sample and the at least one reference sample in the following manner: For a single reference sample entry, determine the response area of the compound associated with the single reference sample entry. 6. The sample analysis device based on fingerprint similarity according to item 1, wherein the library match score and retention time analyzer is executed by the at least one hardware processor to analyze the library match score and the retention time based on determining that the sample includes the library match score to determine the similarity between the sample and the at least one reference sample in the following manner: For multiple reference sample entries, determine the average response area of the compounds associated with the multiple reference sample entries. 7. The sample analysis device based on fingerprint similarity according to item 2, wherein the ion ratio and retention time analyzer is executed by the at least one hardware processor to analyze the ion ratio and the retention time based on determining that the sample does not include the library match score to determine the similarity between the sample and the at least one reference sample in the following manner: For a single reference sample entry, determine the response area of the compound associated with the single reference sample entry. 8. The sample analysis device based on fingerprint similarity according to item 2, wherein the ion ratio and retention time analyzer is executed by the at least one hardware processor to analyze the ion ratio and the retention time based on determining that the sample does not include the library match score to determine the similarity between the sample and the at least one reference sample in the following manner: For multiple reference sample entries, determine the average response area of the compounds associated with the multiple reference sample entries. 9. A method for sample analysis based on fingerprint similarity, the method comprising: Analyze a sample by at least one hardware processor; Analyze at least one reference sample by the at least one hardware processor; Determine by the at least one hardware processor whether the sample includes a library match score; and Analyze the ion ratio and retention time by the at least one hardware processor based on determining that the sample does not include the library match score to determine the similarity between the sample and the at least one reference sample. 10. The method according to item 9, the method further comprising: The at least one hardware processor analyzes the library match score and the retention time based on determining that the sample includes the library match score to determine the similarity between the sample and the at least one reference sample. 11. The method according to item 9, the method further comprising: The at least one hardware processor determines the similarity between the sample and the at least one reference sample based on cosine similarity. 12. The method according to item 9, the method further comprising: The at least one hardware processor determines the similarity between the sample and the at least one reference sample based on entropy similarity. 13. The method according to item 9, wherein analyzing the ion ratio and the retention time by the at least one hardware processor based on determining that the sample does not include the library match score to determine the similarity between the sample and the at least one reference sample further comprises: The at least one hardware processor determines the response area of the compound associated with the single reference sample entry for a single reference sample entry. 14. The method according to item 9, wherein analyzing the ion ratio and the retention time by the at least one hardware processor based on determining that the sample does not include the library match score to determine the similarity between the sample and the at least one reference sample further comprises: The at least one hardware processor determines the average response area of the compounds associated with the multiple reference sample entries for multiple reference sample entries. 15. The method according to item 10, wherein analyzing the library match score and the retention time by the at least one hardware processor based on determining that the sample includes the library match score to determine the similarity between the sample and the at least one reference sample further comprises: The at least one hardware processor determines the response area of the compound associated with the single reference sample entry for a single reference sample entry. 16. The method according to item 10, wherein analyzing the library match score and the retention time by the at least one hardware processor based on determining that the sample includes the library match score to determine the similarity between the sample and the at least one reference sample further comprises: The at least one hardware processor determines the average response area of the compounds associated with the multiple reference sample entries for multiple reference sample entries. 17. A non-transitory computer-readable medium having machine-readable instructions stored thereon, the machine-readable instructions, when executed by at least one hardware processor, cause the at least one hardware processor to: Determine whether a sample includes a library match score; Based on determining that the sample does not include the library match score, analyze ion ratios and retention times to determine the similarity between the sample and the at least one reference sample; and Based on determining that the sample includes the library match score, analyze the library match score and the retention time to determine the similarity between the sample and the at least one reference sample. 18. The non-transitory computer-readable medium according to item 17, wherein the machine-readable instructions, when executed by the at least one hardware processor, further cause the at least one hardware processor to: Determine the similarity between the sample and the at least one reference sample based on cosine similarity. 19. The non-transitory computer-readable medium according to item 17, wherein the machine-readable instructions, when executed by the at least one hardware processor, further cause the at least one hardware processor to: Determine the similarity between the sample and the at least one reference sample based on entropy similarity. 20. The non-transitory computer-readable medium according to item 17, wherein the machine-readable instructions for analyzing the ion ratios and the retention time based on determining that the sample does not include the library match score to determine the similarity between the sample and the at least one reference sample, and for analyzing the library match score and the retention time based on determining that the sample includes the library match score to determine the similarity between the sample and the at least one reference sample, when executed by the at least one hardware processor, further cause the at least one hardware processor to: For a single reference sample entry, determine the response area of the compound associated with the single reference sample entry; and For multiple reference sample entries, determine the average response area of the compound associated with the multiple reference sample entries.

Claims

1. A sample analysis device based on fingerprint similarity, comprising: A reference sample type analyzer executed by at least one hardware processor to: Analyze samples; analyzing at least one reference sample; a library match score analyzer executed by the at least one hardware processor to: determining whether the sample includes a library match score; as well as a library match score and retention time analyzer executed by the at least one hardware processor to: The library match score and the retention time are analyzed based on determining that the sample includes the library match score to determine similarity between the sample and the at least one reference sample.

2. The sample analysis device based on fingerprint similarity according to claim 1, further comprising: an ion ratio and retention time analyzer executed by the at least one hardware processor to: Based on determining that the sample does not include the library match score, ion ratios and the retention times are analyzed to determine similarity between the sample and the at least one reference sample.

3. The sample analysis device based on fingerprint similarity according to claim 1, further comprising: a similarity analyzer executed by the at least one hardware processor to: A similarity between the sample and the at least one reference sample is determined based on cosine similarity.

4. The sample analysis device based on fingerprint similarity according to claim 1, further comprising: a similarity analyzer executed by the at least one hardware processor to: A similarity between the sample and the at least one reference sample is determined based on the entropy similarity.

5. The sample analysis device based on fingerprint similarity according to claim 1, wherein: The library match score and retention time analyzer is executed by the at least one hardware processor to analyze the library match score and the retention time to determine similarity between the sample and the at least one reference sample based on determining that the sample includes the library match score by: For a single reference sample entry, a response area for a compound associated with the single reference sample entry is determined.

6. The sample analysis device based on fingerprint similarity according to claim 1, wherein: The library match score and retention time analyzer is executed by the at least one hardware processor to analyze the library match score and the retention time to determine similarity between the sample and the at least one reference sample based on determining that the sample includes the library match score by: For a plurality of reference sample entries, an average response area for compounds associated with the plurality of reference sample entries is determined.

7. The sample analysis device based on fingerprint similarity according to claim 2, wherein: The ion ratio and retention time analyzer is executed by the at least one hardware processor to analyze the ion ratios and the retention times to determine similarity between the sample and the at least one reference sample based on determining that the sample does not include the library match score by: For a single reference sample entry, a response area for a compound associated with the single reference sample entry is determined.

8. The sample analysis device based on fingerprint similarity according to claim 2, wherein: The ion ratio and retention time analyzer is executed by the at least one hardware processor to analyze the ion ratios and the retention times to determine similarity between the sample and the at least one reference sample based on determining that the sample does not include the library match score by: For a plurality of reference sample entries, an average response area for compounds associated with the plurality of reference sample entries is determined.

9. A method for sample analysis based on fingerprint similarity, the method comprising: analyzing the sample by at least one hardware processor; analyzing, by the at least one hardware processor, at least one reference sample; determining, by the at least one hardware processor, whether the sample includes a library match score; as well as Ion ratios and retention times are analyzed, by the at least one hardware processor, to determine similarity between the sample and the at least one reference sample based on determining that the sample does not include the library match score.

10. The method according to claim 9, further comprising: The library match score and the retention time are analyzed, by the at least one hardware processor, to determine a similarity between the sample and the at least one reference sample based on determining that the sample includes the library match score.

Images