Method for detecting microsatellite site stability and electronic device thereof

Abstract

The application provides a microsatellite site stability detection method and an electronic device thereof. The microsatellite site stability detection method comprises the following steps: S1, obtaining characteristic data in a preset microsatellite site set by using high-throughput targeted sequencing data of a to-be-detected sample, wherein the characteristic data at least comprises the following: length variation of a short tandem repeat sequence and structural variation occurring at the microsatellite site; S2, establishing a prediction model by using the characteristic data; and S3, outputting a microsatellite stability result of the to-be-detected sample by using the prediction model. The method can solve the limitation problem of the microsatellite instability detection method in the prior art and is suitable for the tumor detection field.

Description

Technical Field

[0001] This invention relates to the field of tumor detection, and more specifically, to a method and electronic device for detecting microsatellite locus stability. Background Technology

[0002] Microsatellite instability (MSI), characterized by variations in microsatellite sequence length caused by defects in DNA mismatch repair mechanisms, is an important biomarker for certain cancers, particularly colorectal and endometrial cancers. MSI plays a central role not only in diagnosis and prognostic assessment but also in genetic screening and personalized treatment selection, such as in the identification of Lynch syndrome. Precise detection based on microsatellite instability has profound implications for clinical practice and scientific research in oncology.

[0003] Currently, clinical tumor gene sequencing products primarily employ Next-Generation Sequencing (NGS) technology, focusing on target regions of specific cancer types. The aim is to acquire relevant sequencing information by designing sequencing protocols that include MSI sites and then using algorithms to predict MSI status. However, the clinical reality is that the emergence of various tumor-targeting products, coupled with different sequencing platforms and library preparation strategies, makes it difficult for single MSI prediction algorithms to adapt to the current complex environment, resulting in variability and uncertainty in prediction results. Especially in the "gray zone" of MSI status assessment—between MSI-H and MSS—or in specific stages of tumor evolution, the consistency between NGS prediction and PCR validation (polymerase chain reaction based on capillary electrophoresis, considered the gold standard) significantly decreases.

[0004] Existing technologies include methods that attempt to improve the performance of MSI prediction through machine learning of large-scale feature matrices or broad-coverage targeted sequencing and whole-exome sequencing. However, these methods have extremely high requirements for sample size and sequencing data depth, which limits their application in the development of small and medium-sized targeted sequencing products or new products. Summary of the Invention

[0005] The main objective of this invention is to provide a method and electronic device for detecting microsatellite site stability, thereby addressing the limitations of existing methods for detecting microsatellite instability.

[0006] To achieve the above objectives, according to a first aspect of the present invention, a method for detecting microsatellite site stability is provided, the method comprising: S1) using high-throughput targeted sequencing data of the sample to be tested to obtain feature data in a preset set of microsatellite sites, the feature data including at least the following: length variations of short tandem repeat sequences and structural variations occurring at microsatellite sites; S2) using the feature data to establish a prediction model; S3) using the prediction model to output the microsatellite stability result of the sample to be tested.

[0007] Further, S1) includes: S1-1) By comparing the high-throughput targeted sequencing data of the sample to be tested with the reference genome data of the sample to be tested, and combining the sequencing platform information and the information of the corresponding cancer type of the sample to be tested, a set of preset microsatellite loci and characteristic data are obtained; S1-2) Based on the length variation of short tandem repeat sequences, a count file of repeating short fragments of microsatellite loci is obtained; S1-3) Based on the structural variations occurring at the microsatellite loci, the peak distribution information of the microsatellite loci of the sample to be tested is obtained.

[0008] Further, in S2), the method for establishing the prediction model includes: using feature data from a preset microsatellite locus set of training set samples obtained through different sequencing platforms to train a mathematical model and obtain a prediction model; the training set samples include training set tumor samples and control samples of training set tumor samples; both training set tumor samples and control samples include microsatellite stable samples and microsatellite unstable samples; the prediction model includes a first threshold, a second threshold, a third threshold, a correction coefficient, a baseline, and a gray zone value; the correction coefficient is: for different sequencing platforms, the standard deviation of the read length corresponding to each microsatellite locus in the microsatellite stable samples of the peak distribution information of the training set samples is obtained, and the standard deviation is corrected to obtain the corresponding correction coefficient; the baseline includes a first baseline value and a second baseline value, the first baseline value being the average of the standard deviations of the peak distribution of all microsatellite loci in the microsatellite stable samples; the second baseline value is the average of the number of peaks in the peak distribution of each microsatellite locus in the microsatellite stable samples; the gray zone value is 0.05-0.1;

[0009] The first threshold is: based on the cancer type corresponding to the training set samples, calculate the difference in the repetitive sequence length value corresponding to the vertex of the simulated main peak between the tumor samples and control samples, and record it as the first threshold; the second threshold includes a single-sample threshold and a paired-sample threshold; the single-sample threshold includes the first single-sample threshold and the second single-sample threshold; the first single-sample threshold is: for different sequencing platforms, calculate the standard deviation of the peak distribution of each microsatellite locus in the training set samples, determine whether each microsatellite locus is a microsatellite unstable locus, and read the total value of the microsatellite unstable locus; calculate the number of microsatellite unstable loci in each training set sample. The total number of microsatellite unstable sites divided by the total number of sites that have passed quality control is denoted as the first single-sample threshold. The second single-sample threshold is the sum of the first single-sample threshold and the gray zone value for different sequencing platforms. The paired-sample threshold is calculated by extracting the count of repeated short fragments of sites that have passed quality control in the tumor samples and control samples of the training set according to the cancer type corresponding to the training set samples, and calculating the average value of the step difference. The third threshold is the number of microsatellite unstable sites determined based on the PCR peak diagram of the training set samples for different sequencing platforms.

[0010] Further, when the sample to be tested is a single tumor sample, S3) includes: S3-A1) performing quality filtering on the peak distribution information to obtain qualified peak distribution information, and recording the total number of microsatellite loci in the qualified peak distribution information as the total number of qualified microsatellite loci; S3-A2) calculating the standard deviation of the repeat sequence length difference of each microsatellite locus in the single tumor sample based on the qualified peak distribution information to obtain a first value; multiplying the first baseline value of the corresponding sequencing platform by the correction coefficient and adding the second baseline value to obtain the site baseline value corresponding to each microsatellite locus; comparing the first value with the site baseline value, and if it exceeds the site baseline value, marking the corresponding microsatellite locus as a microsatellite unstable site and recording it in the first microsatellite unstable site set; dividing the total number of microsatellite unstable sites in the first microsatellite unstable site set by the total number of qualified microsatellite loci to obtain the first tumor single sample evaluation value;

[0011] Based on the peak distribution information, the PCR validation sites of each microsatellite locus in the test sample are determined using the method of S3-A2) to determine whether the PCR validation sites are microsatellite unstable sites. The PCR validation sites that are determined to be microsatellite unstable sites are summarized to form the first PCR microsatellite unstable site set. The total value of the first PCR microsatellite unstable site set is recorded as the second tumor single sample assessment value.

[0012] (S3-A4) The first tumor single sample assessment value is compared with the first single sample threshold and the second single sample threshold to obtain the first result; the second tumor single sample assessment value is compared with the third threshold to obtain the second result; the first result and the second result are combined to determine and output the microsatellite site stability result of the tumor single sample.

[0013] Furthermore, the judgment methods in S3-A4) include: a. When both the first result and the second result are microsatellite-stable, the final result output is the microsatellite-stable judgment result; b. When the first result is microsatellite-stable and the second result is microsatellite-unstable, the final result output is the microsatellite-stable judgment result; c. When the first tumor single-sample assessment value is between the first single-sample threshold and the second single-sample threshold, and the second result is microsatellite-stable, the final result output is the microsatellite-stable judgment result; d. When the first tumor single-sample assessment value is between the first single-sample threshold and the second single-sample threshold, and both second results are microsatellite-unstable, the final result output is the microsatellite-unstable judgment result.

[0014] e. If the first tumor single-sample assessment value is greater than or equal to the second single-sample threshold, the second result is the determination result of microsatellite stability, and the final result output is the determination result of microsatellite instability; f. If the first tumor single-sample assessment value is greater than or equal to the second single-sample threshold, the second result is the determination result of microsatellite instability, and the final result output is the determination result of microsatellite instability.

[0015] Furthermore, when the test sample includes tumor samples and control samples of tumor samples, S3) includes: S3-B1) performing quality filtering on the duplicate short fragment count files of tumor samples and control samples to obtain qualified duplicate short fragment count files; performing quality filtering on the peak group distribution information to obtain qualified peak group distribution information, and recording the total number of microsatellite loci in the qualified peak group distribution information as the total number of qualified microsatellite loci.

[0016] (S3-B2) Traverse the qualified repeated short fragment count files, and obtain the repeated short fragment count information of tumor samples and control samples at the same microsatellite locus. Use asymptotic regression difference calculation to obtain the stepwise difference value of the microsatellite locus. The average of the stepwise difference values ​​of all loci is recorded as the first paired sample evaluation value. The first paired sample evaluation value is compared with the paired sample threshold and gray zone value of the corresponding cancer type to obtain the first result.

[0017] (S3-B3) Traverse the qualified repeat short fragment counting files, simulate capillary PCR peak diagrams for the PCR validation sites of each microsatellite locus in the test sample, calculate the difference in repeat sequence length values ​​corresponding to the apex of the main peak of the tumor sample and the control sample, and record it as the second paired sample evaluation value; compare the second paired sample evaluation value with the first threshold of the corresponding cancer type, and if it exceeds the first threshold, it is determined to be a microsatellite unstable site, and the first paired PCR microsatellite unstable site set is obtained; compare the total value of microsatellite unstable sites in the first paired PCR microsatellite unstable site set with the third threshold of the corresponding sequencing platform to obtain the second result;

[0018] (S3-B4) Based on the qualified peak distribution information, calculate the standard deviation of the repeat sequence length difference of the tumor sample at each microsatellite locus to obtain the first paired value; multiply the first baseline value of the corresponding sequencing platform by the correction coefficient, and finally add the second baseline value to obtain the locus baseline value corresponding to each microsatellite locus; compare the first paired value with the locus baseline value, and if it exceeds the locus baseline value, mark the corresponding microsatellite locus as a microsatellite unstable locus and record it in the first paired microsatellite unstable locus set; divide the total number of microsatellite unstable loci in the first paired microsatellite unstable locus set by the total number of qualified microsatellite loci to obtain the first paired tumor single sample evaluation value; compare the first paired tumor single sample evaluation value with the first single sample threshold and gray zone value respectively to obtain the third result;

[0019] (S3-B5) Based on the peak distribution information, the PCR validation sites for each microsatellite locus in the test sample are determined using the method in (S3-B4) to determine whether the PCR validation sites are microsatellite unstable sites. The PCR validation sites determined to be microsatellite unstable sites are summarized to form the first paired PCR microsatellite unstable site set. The total value of the first paired PCR microsatellite unstable site set is recorded as the second paired tumor single sample evaluation value. The second paired tumor single sample evaluation value is compared with the third threshold of the corresponding sequencing platform to obtain the fourth result.

[0020] S3-B6) Combines the first, second, third, and fourth results to determine and output the microsatellite site stability results of the tumor sample.

[0021] Furthermore, the judgment method in S3-B6) includes: the gray zone judgment rule includes: the lower limit of the paired sample threshold minus the gray zone value; the upper limit of the paired sample threshold plus the gray zone value; if the first paired sample evaluation value is between the lower limit and the upper limit of the paired sample threshold, the result output is that the corresponding microsatellite site is in the gray zone, otherwise it is not in the gray zone.

[0022] The first single-sample threshold minus the gray area value is the lower limit of the single-sample threshold; the first single-sample threshold plus the gray area value is the upper limit of the single-sample threshold; if the first single-sample evaluation value is between the lower limit and the upper limit of the first single-sample threshold, the result output is that the microsatellite locus is in the gray area, otherwise it is not in the gray area.

[0023] a. When the first, second, third, and fourth results are consistent, the final output is a consistent judgment result; b. When the second and fourth results are consistent, and the result is consistent with either the first or third result: b1. If consistent with the third result, the final output is the third result as the judgment result; b2. If consistent with the first result, and the first result is a judgment result indicating microsatellite instability, if the third result is in the gray area, the final output is a judgment result indicating microsatellite instability; if the third result is not in the gray area, the final output is a judgment result indicating microsatellite stability.

[0024] b3. If the result is consistent with the first result, and the first result is the determination result of microsatellite stability, and the third result is in the gray area, the final result output is the determination result of microsatellite stability; if the third result is not in the gray area, the final result output is the determination result of microsatellite instability.

[0025] c. When the second and fourth results are consistent, but inconsistent with the first and third results: c1. The second and fourth results are the determination results for microsatellite stability. If the third result is in the gray area, the final output is the determination result for microsatellite stability; if the third result is not in the gray area, the final output is the determination result for microsatellite instability. c2. The second and fourth results are the determination results for microsatellite instability. If the third result is in the gray area, the final output is the determination result for microsatellite instability; if the third result is not in the gray area, the final output is the determination result for microsatellite stability.

[0026] d. If the second and fourth results are inconsistent, but the first and third results are consistent, the final output will be the judgment result corresponding to the consistency between the first and third results;

[0027] e. When the second and fourth results are inconsistent, and when they are inconsistent with the first and third results:

[0028] e1.1. Both the third and fourth results are the determination results of microsatellite instability. The first single sample evaluation value exceeds the upper limit of the first single sample threshold. The number of microsatellite unstable sites in the fourth result is greater than or equal to 3. The final result output is the determination result of microsatellite instability.

[0029] e1.2. The third and fourth results are both the determination results of microsatellite stability. The first single sample evaluation value is lower than the lower limit of the first single sample threshold. The fourth result has 0 microsatellite unstable sites. The final result output is the determination result of microsatellite stability.

[0030] e1.3. If the first and second results are consistent, and the third and fourth results are consistent, and the first result is determined to be a microsatellite instability, the evaluation value of the first paired sample is greater than or equal to the sum of the upper limit of the paired sample threshold and the gray area value, and the number of microsatellite instability sites in the second result is greater than or equal to 4, the final result output is the microsatellite instability determination result; otherwise, the final result output is the determination result of the third result.

[0031] e2. If the first and second results are inconsistent, and the third and fourth results are inconsistent, the final output will be the judgment result of the third result.

[0032] To achieve the above objectives, according to a second aspect of the present invention, an electronic device for detecting the stability of microsatellite sites is provided, the electronic device comprising a feature data acquisition unit, a prediction model establishment unit, and a result output unit;

[0033] The feature data acquisition unit is used to acquire feature data from a preset set of microsatellite loci using high-throughput targeted sequencing data of the sample to be tested. The feature data includes at least the following: length variations of short tandem repeat sequences and structural variations occurring at microsatellite loci; the prediction model building unit is used to build a prediction model using the feature data; and the result output unit is used to output the microsatellite stability results of the sample to be tested using the prediction model.

[0034] Furthermore, the feature data acquisition unit includes a microsatellite set acquisition unit, a repeating short fragment count file acquisition unit, and a peak group distribution information acquisition unit;

[0035] The microsatellite set acquisition unit is used to obtain a preset microsatellite locus set and feature data by comparing the high-throughput targeted sequencing data of the sample to be tested with the reference genome data of the sample to be tested, and combining the sequencing platform information and the corresponding cancer type information of the sample to be tested.

[0036] The repeating short fragment count file acquisition unit is used to obtain the repeating short fragment count file of microsatellite loci based on the length variation of short tandem repeat sequences.

[0037] The peak distribution information acquisition unit is used to obtain the peak distribution information of the microsatellite sites of the sample under test based on the structural variations that occur at the microsatellite sites.

[0038] Furthermore, the prediction model building unit includes a model training unit;

[0039] The model training unit is used to train a mathematical model using feature data from a pre-defined set of microsatellite loci in the training set samples to obtain a prediction model. The training set samples include training set tumor samples and control samples of the training set tumor samples. Both the training set tumor samples and control samples include microsatellite stable samples and microsatellite unstable samples. The mathematical model is trained using feature data from a pre-defined set of microsatellite loci in the training set samples obtained through different sequencing platforms to obtain a prediction model. The prediction model includes a first threshold, a second threshold, a third threshold, a correction coefficient, a baseline, and gray area values.

[0040] The correction factor is: for different sequencing platforms, the standard deviation of the read length corresponding to each microsatellite locus in the microsatellite stable sample of the peak distribution information of the training set samples, and the value after correcting the standard deviation is the correction factor of the corresponding sequencing platform; the baseline includes a first baseline value and a second baseline value. The first baseline value is the average of the standard deviations of the peak distribution of all microsatellite loci in the microsatellite stable sample; the second baseline value is the average of the number of peaks in the peak distribution of each microsatellite locus in the microsatellite stable sample, which is the first baseline value.

[0041] The gray zone value is 0.05-0.1; the first threshold is: based on the cancer type corresponding to the training set samples, the difference in the length of the repeat sequence corresponding to the peak of the simulated main peak between the training set tumor samples and the control samples is calculated, and this is recorded as the first threshold; the second threshold includes a single-sample threshold and a paired-sample threshold; the single-sample threshold includes a first single-sample threshold and a second single-sample threshold; the first single-sample threshold is: for different sequencing platforms, the standard deviation of the peak distribution of each microsatellite locus in the training set samples is calculated, and it is determined whether each microsatellite locus is a microsatellite unstable locus, and the total value of the microsatellite unstable locus is read; the total value of the microsatellite unstable locus in each training set sample is divided by the total number of quality-controlled loci, and this is recorded as the first single-sample threshold; the second single-sample threshold is: for different sequencing platforms, the sum of the first single-sample threshold and the gray zone value is recorded as the second single-sample threshold; the paired-sample threshold is: based on the cancer type corresponding to the training set samples, the average value of the step difference calculated by extracting the repeat short fragment counts of the quality-controlled loci in the training set tumor samples and the control samples is recorded as the paired-sample threshold; The third threshold is: for different sequencing platforms, the third threshold is the number of microsatellite unstable sites determined based on the PCR peak diagram of the training set samples.

[0042] Furthermore, when the sample to be tested is a single tumor sample, the result output unit includes a first single tumor sample evaluation value acquisition unit, a second single tumor sample evaluation value acquisition unit, and a result comparison unit;

[0043] The first tumor single-sample evaluation value acquisition unit includes a qualified peak distribution information acquisition unit, a first numerical value acquisition unit, a site baseline value acquisition unit, a first microsatellite unstable site set acquisition unit, and a first tumor single-sample evaluation value output unit. The qualified peak distribution information acquisition unit is used to perform quality filtering on the peak distribution information to obtain qualified peak distribution information. The total number of microsatellite sites in the qualified peak distribution information is recorded as the total number of qualified microsatellite sites.

[0044] The first numerical acquisition unit is used to calculate the standard deviation of the repetitive sequence length difference at each microsatellite locus in a single tumor sample using qualified peak distribution information, and obtain a first numerical value. The locus baseline value acquisition unit is used to multiply the first baseline value of the corresponding sequencing platform by a correction coefficient and add a second baseline value to obtain the locus baseline value corresponding to each microsatellite locus. The first microsatellite unstable locus set acquisition unit is used to compare the first numerical value with the locus baseline value. If the value exceeds the locus baseline value, the corresponding microsatellite locus is marked as a microsatellite unstable locus and recorded in the first microsatellite unstable locus set. The first tumor single sample evaluation value output unit is used to divide the total number of microsatellite unstable loci in the first microsatellite unstable locus set by the total number of qualified microsatellite loci to obtain the first tumor single sample evaluation value.

[0045] The second tumor single-sample evaluation value acquisition unit includes a first PCR microsatellite unstable site set acquisition unit and a second tumor single-sample evaluation value output unit. The first PCR microsatellite unstable site set acquisition unit is used to determine whether the PCR validation site of each microsatellite site in the test sample is a microsatellite unstable site based on the peak distribution information, and to summarize the PCR validation sites determined to be microsatellite unstable sites to form the first PCR microsatellite unstable site set. The second tumor single-sample evaluation value output unit is used to read the total value of the first PCR microsatellite unstable site set and record it as the second tumor single-sample evaluation value. The result comparison unit compares the first tumor single-sample evaluation value with the first single-sample threshold and the second single-sample threshold to obtain the first result; compares the second tumor single-sample evaluation value with the third threshold to obtain the second result; and combines the first result and the second result to determine and output the microsatellite site stability result of the tumor single sample.

[0046] Furthermore, when the sample to be tested is a single tumor sample, the methods for determination include:

[0047] a. When both the first and second results are determinations of microsatellite stability, the final output will be the determination of microsatellite stability.

[0048] b. When the first result is that the microsatellite is stable and the second result is that the microsatellite is unstable, the final output result is the determination result of microsatellite stability;

[0049] c. The first tumor single-sample assessment value is between the first single-sample threshold and the second single-sample threshold. The second result is the microsatellite stability determination result. The final result output is the microsatellite stability determination result.

[0050] d. The first tumor single-sample assessment value is between the first single-sample threshold and the second single-sample threshold. The second result is the determination result of microsatellite instability. The final output result is the determination result of microsatellite instability.

[0051] e. If the first tumor single-sample assessment value is greater than or equal to the second single-sample threshold, the second result is the determination result of microsatellite stability, and the final output result is the determination result of microsatellite instability;

[0052] f. If the first tumor single-sample assessment value is greater than or equal to the second single-sample threshold, the second result is the determination result of microsatellite instability, and the final output result is the determination result of microsatellite instability.

[0053] Furthermore, when the test sample includes a tumor sample and a control sample of the tumor sample, the result output unit includes a repeating short fragment count file acquisition unit, a qualified peak group distribution information acquisition unit, a first result acquisition unit, a second result acquisition unit, a third result acquisition unit, a fourth result acquisition unit, and a result comparison unit;

[0054] The duplicate short fragment count file acquisition unit is used to perform quality filtering on the duplicate short fragment count files of tumor samples and control samples to obtain qualified duplicate short fragment count files.

[0055] The qualified peak group distribution information acquisition unit is used to filter the peak group distribution information to obtain qualified peak group distribution information. The total number of microsatellite sites in the qualified peak group distribution information is recorded as the total number of qualified microsatellite sites.

[0056] The first result acquisition unit includes a first paired sample evaluation value acquisition unit and a first result output unit;

[0057] The first paired sample evaluation value acquisition unit is used to traverse the qualified repeated short fragment count files, the repeated short fragment count information of tumor samples and control samples at the same microsatellite locus, and use progressive regression difference calculation to obtain the stepwise difference value of microsatellite locus. The average of the stepwise difference values ​​of all loci is recorded as the first paired sample evaluation value.

[0058] The first result output unit is used to compare the evaluation value of the first paired sample with the threshold and gray value of the paired sample of the corresponding cancer type to obtain the first result;

[0059] The second result acquisition unit includes a second paired sample evaluation value acquisition unit, a paired first PCR microsatellite unstable site set acquisition unit, and a second result output unit. The second paired sample evaluation value acquisition unit is used to traverse qualified repetitive short fragment count files, simulate capillary PCR peak diagrams for the PCR validation sites of each microsatellite site in the test sample, and calculate the difference in repetitive sequence length values ​​corresponding to the apex of the main peak of the tumor sample and the control sample, which is recorded as the second paired sample evaluation value. The paired first PCR microsatellite unstable site set acquisition unit is used to compare the second paired sample evaluation value with a first threshold corresponding to the cancer type; if it exceeds the first threshold, it is determined to be a microsatellite unstable site, and the values ​​are summarized to obtain the paired first PCR microsatellite unstable site set. The second result output unit is used to compare the total value of microsatellite unstable sites in the paired first PCR microsatellite unstable site set with a third threshold of the corresponding sequencing platform to obtain the second result.

[0060] The third result acquisition unit includes a first paired value acquisition unit, a paired site baseline value acquisition unit, a first paired microsatellite unstable site set acquisition unit, a first paired tumor single sample evaluation value output unit, and a third result output unit. The first paired value acquisition unit is used to calculate the standard deviation of the repeat sequence length difference of the tumor single sample at each microsatellite site based on qualified peak distribution information to obtain the first paired value.

[0061] The system comprises the following components: a paired-site baseline value acquisition unit, which multiplies the first baseline value of the corresponding sequencing platform by a correction coefficient and adds the second baseline value to obtain the site baseline value for each microsatellite site; a first paired microsatellite unstable site set acquisition unit, which compares the first paired value with the site baseline value, and if the value exceeds the site baseline value, marks the corresponding microsatellite site as a microsatellite unstable site and adds it to the first paired microsatellite unstable site set; a first paired tumor single-sample evaluation value output unit, which divides the total number of microsatellite unstable sites in the first paired microsatellite unstable site set by the total number of qualified microsatellite sites to obtain the first paired tumor single-sample evaluation value; a third result output unit, which compares the first paired tumor single-sample evaluation value with the first single-sample threshold and gray zone value respectively to obtain the third result; and a fourth result acquisition unit, which includes the first paired PCR microsatellite unstable site set acquisition unit, the second paired tumor single-sample evaluation value output unit, and the fourth result output unit.

[0062] The first paired PCR microsatellite unstable site set is used to determine whether the PCR validation site of each microsatellite site in the test sample is a microsatellite unstable site based on the peak distribution information using the S3-B4 method. The PCR validation sites identified as microsatellite unstable sites are summarized to form the first paired PCR microsatellite unstable site set. The second tumor single sample evaluation value output unit is used to read the total value of the first paired PCR microsatellite unstable site set and record it as the second paired tumor single sample evaluation value. The fourth result output unit is used to compare the second paired tumor single sample evaluation value with the third threshold of the corresponding sequencing platform to obtain the fourth result. The stability of the microsatellite sites in the tumor sample is determined by combining the first result, the second result, the third result and the fourth result.

[0063] Furthermore, when the test samples include tumor samples and control samples of tumor samples, the determination methods include: the lower limit of the paired sample threshold minus the gray area value; the upper limit of the paired sample threshold plus the gray area value; if the first paired sample evaluation value is between the lower limit and the upper limit of the paired sample threshold, the result output is that the corresponding microsatellite locus is in the gray area, otherwise it is not in the gray area; the lower limit of the first single sample threshold minus the gray area value; the upper limit of the first single sample threshold plus the gray area value.

[0064] If the first single-sample evaluation value is between the lower limit and the upper limit of the single-sample threshold, the result output is that the microsatellite locus is in the gray area; otherwise, it is not in the gray area.

[0065] a. When the first, second, third, and fourth results are consistent, the final output will be a consistent judgment result;

[0066] b. When the second and fourth results are consistent, and the result is consistent with either the first or the third result:

[0067] b1. Consistent with the third result, the final output is the third result as the judgment result;

[0068] b2. If the result is consistent with the first result, and the first result is the determination result of microsatellite instability, and the third result is in the gray area, the final result output is the determination result of microsatellite instability; if the third result is not in the gray area, the final result output is the determination result of microsatellite stability.

[0069] b3. If the result is consistent with the first result, and the first result is the determination result of microsatellite stability, and the third result is in the gray area, the final result output is the determination result of microsatellite stability.

[0070] If the third result is not in the gray area, the final output will be the determination result of the microsatellite instability.

[0071] c. When the second and fourth results are consistent, but inconsistent with the first and third results:

[0072] c1. The second and fourth results are the determination results of microsatellite stability. If the third result is in the gray area, the final result output is the determination result of microsatellite stability; if the third result is not in the gray area, the final result output is the determination result of microsatellite instability.

[0073] c2. The second and fourth results are the determination results of microsatellite instability. If the third result is in the gray area, the final result output is the determination result of microsatellite instability; if the third result is not in the gray area, the final result output is the determination result of microsatellite stability.

[0074] d. If the second and fourth results are inconsistent, but the first and third results are consistent, the final output will be the judgment result corresponding to the consistency between the first and third results;

[0075] e. When the second and fourth results are inconsistent, and when they are inconsistent with the first and third results:

[0076] e1.1. Both the third and fourth results are the determination results of microsatellite instability. The first single sample evaluation value exceeds the upper limit of the first single sample threshold. The number of microsatellite unstable sites in the fourth result is greater than or equal to 3. The final result output is the determination result of microsatellite instability.

[0077] e1.2. The third and fourth results are both the determination results of microsatellite stability. The first single sample evaluation value is lower than the lower limit of the first single sample threshold. The fourth result has 0 microsatellite unstable sites. The final result output is the determination result of microsatellite stability.

[0078] e1.3. If the first and second results are consistent, and the third and fourth results are consistent, and the first result is determined to be a microsatellite instability, the evaluation value of the first paired sample is greater than or equal to the sum of the upper limit of the paired sample threshold and the gray area value, and the number of microsatellite instability sites in the second result is greater than or equal to 4, the final result output is the microsatellite instability determination result; otherwise, the final result output is the determination result of the third result.

[0079] e2. If the first and second results are inconsistent, and the third and fourth results are inconsistent, the final output will be the judgment result of the third result.

[0080] To achieve the above objectives, according to a third aspect of the present invention, a computer device is provided, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the above-described method for detecting microsatellite site stability.

[0081] To achieve the above objectives, according to a fourth aspect of the present invention, a computer-readable storage medium is provided, on which computer program instructions are stored, which, when executed by a processor, implement the steps of the above-described method for detecting microsatellite site stability.

[0082] To achieve the above objectives, according to a fifth aspect of the present invention, a computer program product is provided, comprising a computer program / instructions that, when executed by a processor, implement the steps of the above-described method for detecting microsatellite site stability.

[0083] By applying the technical solution of this invention, a quantifiable judgment threshold is established. The microsatellite stability of the test sample is evaluated by combining the repetitive sequences and kurtosis values ​​of the sample. The evaluation results are then compared with the judgment threshold, resulting in a more accurate assessment of the microsatellite state of the test sample compared to existing technologies. Furthermore, the detection method of this application is applicable to different sequencing platforms, generating accurate microsatellite judgment results, and solves the problem of gray region judgment in existing technologies, making it more suitable for clinical application. Attached Figure Description

[0084] The accompanying drawings, which form part of this application, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings:

[0085] Figure 1 A flowchart illustrating the method for constructing the tumor single-sample contamination judgment model in this application specification is shown.

[0086] Figure 2 A schematic diagram of the electronic device used to construct a judgment model for single-sample contamination of tumors, as described in this application specification, is shown.

[0087] Figure 3 The present invention illustrates the method for constructing a model for determining single-sample contamination of tumors and a hardware block diagram for determining single-sample contamination of tumors.

[0088] Figure 4 The diagram shows an NGS-based pseudocapillary PCR pattern from Example 2 of this application, wherein... Figure 4 Image A shows the NGS capillary PCR result of the control sample. Figure 4 Image B in the image shows an NGS-based capillary PCR of a tumor sample.

[0089] Figure 5 This application specification shows an infographic illustrating different tumor types in Embodiment 5, wherein... Figure 5 Image A in the image is a schematic diagram of the results based on TCGA (Cancer Genome Atlas) data for 39 different tumor types. Figure 5 The diagram in B shows the percentage of MSI-H cases across 39 different tumor types.

[0090] Figure 6 An analysis diagram of 12,019 tumor samples based on the Foundation Medicine database is shown in Embodiment 5 of this application specification. Detailed Implementation

[0091] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The present invention will now be described in detail with reference to the embodiments.

[0092] Terminology Explanation:

[0093] MSI-H (Microsatellite Instability High): This refers to a state of high microsatellite instability, characterized by a large number of microsatellite sequence length variations in the genome of tumor cells.

[0094] MSS (Microsatellite Stable): Microsatellite stable state refers to the state in which the length of microsatellite sequences in tumor cells remains relatively stable without significant changes.

[0095] Gray Zone: In the context of MSI detection, the gray zone refers to the area where there is uncertainty or borderline variation in the assessment of MSI status. The gray zone exists because the assessment of MSI status is based on a series of quantitative indicators. When these indicators fall near certain thresholds, the status of a sample may be difficult to clearly distinguish between MSI-H and MSS. For example, some samples may show slight changes in microsatellite sequence length, but these changes are insufficient to meet the criteria for MSI-H, nor can they be simply classified as MSS. The management of gray zone samples is particularly important in clinical decision-making because they may require more detailed analysis or additional validation experiments (such as PCR validation) to determine their exact MSI status, which directly affects patient diagnosis, prognostic evaluation, and the choice of treatment strategy.

[0096] As mentioned in the background section, existing microsatellite instability detection methods have certain limitations. Based on this, the inventors in this application attempt to develop a new method for detecting microsatellite site stability, and thus propose a series of protection schemes in this application.

[0097] In a first typical embodiment of this application, a method for detecting microsatellite site stability is provided, the flowchart of which is shown below. Figure 1As shown, the detection method includes: S1) using high-throughput targeted sequencing data of the sample to be tested to obtain feature data from a preset set of microsatellite loci, the feature data including at least the following: length variation of short tandem repeat sequences and structural variation occurring at microsatellite loci; S2) using the feature data to establish a prediction model; S3) using the prediction model to output the microsatellite stability results of the sample to be tested.

[0098] Currently available open-source software for detecting tumor microsatellite stability faces numerous challenges due to the complex production processes involving different sequencing platforms, library preparation methods, and probes. These challenges result in numerous detection gray areas and discrepancies between different methods. Furthermore, existing solutions lack specific processing and optimization for particular cancer types (such as colorectal cancer and endometrial cancer), leading to significant differences in results between different methods. Further capillary PCR validation is required for bidirectional correction, a complex and time-consuming process.

[0099] For patients in the advanced stages of tumor development, there is a gradual process of dysfunction in the DNA mismatch repair (MMR) system. MSI site mutations also exhibit temporal differences and individual patient variations. Simple NGS or capillary PCR validation methods have inherent biases and limitations. For example, some PCR-validated sites may have significant mutations, but NGS analysis of a large number of MSI sites may not show a significant overall mutation signal; or NGS analysis may predict MSI-H, MMR-related genes may show mutations, and clinical immunohistochemistry may also indicate MSI-H, but PCR validation may show MSI-L. PCR validation requires personalized interpretation of peak patterns in conjunction with clinicopathological findings, lacking a clear threshold, which hinders the determination of subsequent treatment plans. Other existing methods require large datasets for feature matrix machine learning, or screening for microsatellite site combinations in broad-coverage targeted sequencing or whole-exome sequencing for microsatellite stability prediction. These methods are not well-suited for small-to-medium-sized targeted sequencing products or newly developed products.

[0100] This application, based on high-throughput DNA sequencing data of the sample to be tested, combined with information on the corresponding cancer type and sequencing platform, simulates the electrophoretic peak diagram of PCR verification sites by analyzing the length variation patterns of short tandem repeat sequences in microsatellite sequences and the structural variation patterns occurring at microsatellite sites. It comprehensively considers repeat fragment counting, peak distribution, and electrophoretic diagram simulation to establish a predictive model. The model outputs the microsatellite stability results of the sample to be tested through comprehensive evaluation, aiming to detect the microsatellite stability of the sample. It can detect the gray area of ​​microsatellite sites, reducing reliance on PCR verification and improving prediction accuracy and efficiency.

[0101] In a preferred embodiment, S1) includes: S1-1) comparing the high-throughput targeted sequencing data of the sample to be tested with the reference genome data of the sample to be tested, and combining the sequencing platform information and the information of the corresponding cancer type of the sample to be tested to obtain a preset set of microsatellite loci and feature data; S1-2) obtaining a count file of repeating short fragments of microsatellite loci based on the length variation of short tandem repeat sequences; S1-3) obtaining the peak distribution information of microsatellite loci of the sample to be tested based on the structural variations occurring at the microsatellite loci.

[0102] The sequencing platform information mentioned above includes, but is not limited to, SURF5000, T7 or Xplus. This application does not impose any restrictions. Those skilled in the art can obtain information based on the sequencing platform of the high-throughput targeted sequencing data of the sample to be tested.

[0103] In a preferred embodiment, the method for establishing the prediction model includes: using feature data from a preset microsatellite locus set of training set samples obtained through different sequencing platforms to train a mathematical model and obtain a prediction model; the training set samples include training set tumor samples and control samples of the training set tumor samples; both the training set tumor samples and the control samples include microsatellite stable samples and microsatellite unstable samples.

[0104] The prediction model includes a first threshold, a second threshold, a third threshold, correction coefficients, a baseline, and gray area values;

[0105] The correction factor is: for different sequencing platforms, the standard deviation of the read length corresponding to each microsatellite locus in the stable microsatellite sample of the peak distribution information of the training set sample is obtained, and the standard deviation is corrected to obtain the correction factor for the corresponding sequencing platform.

[0106] The baseline includes a first baseline value and a second baseline value. For different sequencing platforms, the software msings is used to extract the standard deviation of the peak distribution of each microsatellite locus in the training set samples that are determined to be microsatellite stable. The average of all standard deviations is calculated, and the obtained value is the first baseline value. The average number of peaks in the peak distribution of each microsatellite locus is the second baseline value.

[0107] The gray zone value is 0.05-0.1. Preferably, if the sample to be tested is a paired sample, the gray zone value is 0.05 for all sequencing platforms; if the sample to be tested is a single tumor sample, the gray zone value range is 0.05-0.1, which can be limited according to different sequencing platforms. The gray zone value is a constant value, representing the range extending before and after the single-sample threshold and the paired-sample threshold, i.e., the gray zone value range. It is determined by the gray zone range where the instability scores obtained from the analysis of microsatellite stable samples and microsatellite unstable samples in the training set cannot be distinguished by MSINS and MANTIS. Alternatively, those skilled in the art can limit this gray zone value based on common knowledge.

[0108] The first threshold is: based on the cancer type corresponding to the training set samples, calculate the difference in the length of the repeat sequence corresponding to the vertex of the simulated main peak of the tumor sample and the control sample, and record it as the first threshold; the second threshold includes the single sample threshold and the paired sample threshold; the single sample threshold includes the first single sample threshold and the second single sample threshold;

[0109] The first single-sample threshold is as follows: For different sequencing platforms, calculate the standard deviation of the peak distribution of each microsatellite locus in the training set samples, and determine whether each microsatellite locus is a microsatellite unstable locus. Read the total value of the microsatellite unstable locus determined to be a microsatellite unstable locus. Calculate the total value of microsatellite unstable loci in each training set sample and divide it by the total number of loci that have passed quality control. This value is recorded as the first single-sample threshold.

[0110] The second single-sample threshold is the sum of the first single-sample threshold and the gray zone value for different sequencing platforms.

[0111] The paired sample threshold is as follows: based on the cancer type corresponding to the training set samples, the count of repeated short fragments of quality-controlled sites in the tumor samples and control samples of the training set is extracted, and the average value of the step difference is calculated, which is recorded as the paired sample threshold; qualified sites refer to: the qualified microsatellite unstable sites remaining after quality control filtering are called qualified sites (passing_loci).

[0112] The third threshold is: for different sequencing platforms, the third threshold is the number of microsatellite unstable sites determined based on the PCR peak diagram of the training set samples.

[0113] This application obtains alignment information between microsatellite loci and the reference genome from training set samples, performs counting analysis on repetitive short fragments in these samples, detects mutations, and extracts peak distribution information. Baselines for different sequencing platforms are obtained by processing microsatellite-stable samples. Capillary PCR peak diagrams are simulated using NGS data, and an ROC curve model is constructed using data from multiple samples. This allows for the determination of multiple thresholds and correction coefficients for judging the microsatellite status of the sample to be tested. The methods for training the above mathematical model include, but are not limited to, the construction of ROC curves; other mathematical training methods known to those skilled in the art can also be used, and this application makes no limitation. The training set samples are all microsatellite loci samples selected through PCR validation and bioinformatics analysis (containing both stable and unstable microsatellite loci; the PCR validation and bioinformatics analysis methods are all detection methods known to those skilled in the art). The number can be selected from approximately 80 to 120 samples, and those skilled in the art can choose according to actual needs; this application makes no limitation.

[0114] The correction coefficient is derived from the standard deviation of the read length corresponding to microsatellite loci in the stable microsatellite samples of the reference training set, within the range of variation of different sequencing platforms. It is then corrected by taking a value from the upper and lower quartiles of all loci, rounded to one decimal place. This final correction coefficient corresponds to the sequencing platform of the sample to be tested. The method for obtaining the correction coefficient (CF) includes: using the ROC curve method with Youden's Index to evaluate the accuracy of the positive decision threshold for different sequencing platforms; the model with the largest Youden's Index is optimal. In the specific embodiments of this application, the CF of the Genius sequencing instrument SURFSeq 5000 is 2, and the Youden's Index is 0.987; the CF of the BGI sequencing instrument T7 is 1.5, and the Youden's Index is 0.972; the CF of the Illumina sequencing instrument NovaSeq Xplus is 2, and the Youden's Index is 1. Those skilled in the art can flexibly select the appropriate model to obtain the corresponding correction coefficient according to actual needs.

[0115] The baselines described above are files obtained using the open-source software msings with microsatellite-stabilized samples. These files include microsatellite locus location information and the average standard deviation of the peak distribution at the microsatellite loci, used to determine whether changes in the test sample are significant. The degree of instability near the microsatellite loci represents the degree of change in base distribution at microsatellite repeat units in the training set tumor samples (including length and base variations), which can be observed and its standard deviation calculated through peak distribution.

[0116] The first threshold is derived by comparing the difference in repeat sequence length values ​​corresponding to the apex of the simulated main peak between tumor samples and control samples in the training set. This simulates the PCR peak diagram, categorizing it by specific cancer type and reflecting the changes in MSI status between tumor and control samples. Using the repeat fragment count file of the training set samples, with repeat base signals at microsatellite sites as the x-axis and the clustering statistics read count (kmer_counts) of the corresponding repeats as the y-axis, a simulated peak curve is generated. The main peak is found by the maximum coverage area, and its apex is taken as the signal peak position. The corresponding repeats value is the simulated electrophoretic distance (MD). The difference in MD between tumor samples and control samples is calculated and becomes the first threshold.

[0117] Methods for obtaining the first threshold include, but are not limited to: dividing specific cancer types (e.g., endometrial cancer) and other cancer types into two categories for separate calculation; comparing each locus with prior PCR results; using the ROC (Receiver Operating Characteristic) curve method to plot curves by calculating the true positive rate (TPR) and false positive rate (FPR) at different thresholds; and using the Youden's Index to determine the positive determination threshold, with loci below the threshold being MSS loci and those above the threshold being MSI-H loci. In a specific embodiment of this application, referring to the PCR validation gold standard and expert guidelines, the model uses the maximum Youden's Index, ultimately obtaining MD_th = 1.6 and Youden's Index = 1 for endometrial cancer; and MD_th = 2.5 and Youden's Index = 1 for other cancer types. Those skilled in the art can select the model for obtaining the first threshold according to actual needs.

[0118] The aforementioned Repeats refer to the number of times a specific base sequence (such as CA, AT, etc.) is repeated consecutively at an MSI site; Counts data refers to the observation frequency of a specific number of base repeats at an MSI site during sequencing, i.e., how many reads cover the base sequence with that number of repeats; kmer_counts refers to the counts obtained by K-mer counting (a bioinformatics sequence analysis technique commonly used in genome assembly and sequencing, and those skilled in the art can also use other common software for counting).

[0119] Special cancer types (such as endometrial cancer) and other cancer types were divided into two categories for separate calculation. Each locus was compared with the prior PCR results. The ROC (Receiver Operating Characteristic) curve method was used to calculate the true positive rate (TPR) and false positive rate (FPR) at different thresholds to plot the curves. The Youden's Index was used to determine the positive judgment threshold. Loci below the threshold were MSS loci, and loci above the threshold were MSI-H loci.

[0120] The first single-sample threshold is determined by comparing the standard deviation of the repetitive sequence length changes at each microsatellite locus in the training set tumor samples and control samples to determine whether microsatellite instability has occurred. Each sequencing platform has its own unique base sequencing preference and systematic error. The degree of microsatellite instability of the sample is measured by dividing the number of unstable sites by the number of sites that pass quality control. A cutoff threshold is defined by the prior results, which is the first single-sample threshold.

[0121] The aforementioned quality control sites refer to the precise combination of the reference genome pairing file corresponding to the sample and the microsatellite instability site file in the sample, summarizing comprehensive information on the microsatellite sites of the sample to be tested, including location, average depth, base quality, alignment quality, length variation, and other relevant parameters, to obtain the quality control file. Further, all data with a coverage depth lower than 30× in the empty file are removed, as low coverage depth may introduce unnecessary errors and uncertainties. Further filtering removes data with a minimum read quality lower than 25, a minimum average locus quality lower than 30, a minimum read length lower than 35, and a coverage depth lower than 30×, obtaining the final file used for analysis. The microsatellite instability sites contained in this file are the sites that have passed quality control.

[0122] Based on different sequencing platforms, the second single-sample threshold is a larger value than the first single-sample threshold, which together are used to specify the range of stability and instability. Its value corresponds to the sequencing platform. The second single-sample threshold is the sum of the first single-sample threshold and the gray zone value.

[0123] Paired sample threshold: Microsatellite instability is determined by comparing the step differences calculated from the counts of repeated short fragments at each microsatellite locus in the training set tumor samples and control samples. The degree of microsatellite instability is measured by calculating the average of the step differences of quality-controlled loci (the remaining qualified microsatellite unstable loci after quality control filtering are called qualified loci (passing_loci)). Clinically, the electrophoretic distance standard for judging microsatellite instability using the PCR gold standard varies for specific cancer types. Here, cancer types are also classified, and a cutoff threshold is defined based on the prior results, which is the paired sample threshold.

[0124] The third threshold is determined by referencing the PCR gold standard judgment criteria and the systematic error of the actual sequencing platform. It is the number of unstable sites determined by PCR peak diagrams simulated using high-throughput sequencing data, and corresponds to the sequencing platform.

[0125] In a preferred embodiment, when the sample to be tested is a single tumor sample, S3) includes: S3-A1) performing quality filtering on the peak distribution information to obtain qualified peak distribution information, and the total number of microsatellite sites in the qualified peak distribution information is recorded as the total number of qualified microsatellite sites;

[0126] (S3-A2) Based on the qualified peak distribution information, calculate the standard deviation of the repeat sequence length difference of each microsatellite locus in a single tumor sample to obtain a first value; multiply the first baseline value of the corresponding sequencing platform by the correction coefficient, and add the second baseline value to obtain the site baseline value corresponding to each microsatellite locus; compare the first value with the site baseline value, and if it exceeds the site baseline value, mark the corresponding microsatellite locus as a microsatellite unstable site and record it in the first microsatellite unstable site set; divide the total number of microsatellite unstable sites in the first microsatellite unstable site set by the total number of qualified microsatellite sites to obtain the first tumor single sample evaluation value;

[0127] Based on the peak distribution information, the PCR validation sites of each microsatellite locus in the test sample are determined using the method of S3-A2) to determine whether the PCR validation sites are microsatellite unstable sites. The PCR validation sites that are determined to be microsatellite unstable sites are summarized to form the first PCR microsatellite unstable site set. The total value of the first PCR microsatellite unstable site set is recorded as the second tumor single sample assessment value.

[0128] (S3-A4) The first tumor single sample assessment value is compared with the first single sample threshold and the second single sample threshold to obtain the first result; the second tumor single sample assessment value is compared with the third threshold to obtain the second result; the first result and the second result are combined to determine and output the microsatellite site stability result of the tumor single sample.

[0129] By calculating the normalized number of peaks at each microsatellite locus that passed quality control in a single tumor sample, the standard deviation of the repeat sequence length difference at microsatellite variation sites is obtained. This standard deviation is then compared with the baseline value after correction. If the first value exceeds the baseline tolerance range, the instability of the corresponding microsatellite locus significantly exceeds the microsatellite stable sample set used to establish the baseline, and it can be marked as a microsatellite unstable site, forming the first microsatellite unstable site set. Subsequently, the total number of the first microsatellite unstable site set is divided by the total number of qualified microsatellite loci in the tumor sample to obtain the first tumor single-sample assessment value, which is used to assess the overall MSI status of the sample.

[0130] Simultaneously, based on peak distribution information, a simulated PCR peak diagram is generated. Similarly, using a first single-sample threshold and a second single-sample threshold, the state of the baseline microsatellite stable sample set is located, and a second tumor single-sample assessment value is calculated. Finally, the microsatellite instability state of the test sample is judged based on the first structure and the second result, and the final microsatellite site stability result of the tumor single sample is output.

[0131] In a preferred embodiment, the determination method includes: a. When both the first result and the second result are microsatellite stable, the final result output is microsatellite stable; b. When the first result is microsatellite stable and the second result is microsatellite unstable, the final result output is microsatellite stable; c. When the first tumor single-sample assessment value is between the first single-sample threshold and the second single-sample threshold, and the second result is microsatellite stable, the final result output is microsatellite stable; d. When the first tumor single-sample assessment value is between the first single-sample threshold and the second single-sample threshold, and the second result is microsatellite unstable, the final result output is microsatellite unstable; e. When the first tumor single-sample assessment value is greater than or equal to the second single-sample threshold, and the second result is microsatellite stable, the final result output is microsatellite unstable; f. When the first tumor single-sample assessment value is greater than or equal to the second single-sample threshold, and the second result is microsatellite unstable, the final result output is microsatellite unstable.

[0132] When the test sample is only a single tumor sample, if the first result indicates that the microsatellite state of the sample is stable, but the second result indicates that the microsatellite state is unstable, the final result will still be judged as a stable microsatellite site. Clinically used immunohistochemical identification methods only focus on whether mismatch repair genes (MMR genes) are expressed. PCR validation, because it only focuses on a small number of classic microsatellite sites, can have certain discrepancies. High-throughput sequencing, compared to PCR validation, can accommodate more unstable microsatellite candidate sites and can indirectly detect the patient's mismatch repair gene function. If the overall stability is comparable to the baseline sample set, the data results are more reliable.

[0133] In handling complex situations, if the evaluation value of the first tumor sample falls between the pre-set first single-sample threshold and the second single-sample threshold, and the second results are all microsatellite stable sites, given that the first result does not have a strong microsatellite instability signal and the second result has consistent feedback, the MSI status of the sample is determined to be microsatellite stable. The results of the previous research in this application prove that the judgment rule is feasible.

[0134] When the evaluation value of the first tumor sample falls between the pre-set first single-sample threshold and the second single-sample threshold, but the second result consistently points to the microsatellite instability site, considering that this range usually represents the transition from stable to unstable MSI status, and combined with the signal of the second result indicating that the classical microsatellite site is unstable, the MSI status of the sample is finally determined to be microsatellite instability.

[0135] When the evaluation value of the first tumor sample exceeds the second single-sample threshold, given that the first result already has a strong microsatellite instability signal and because the algorithm reflects the state of a large number of microsatellite sites, regardless of the second result, the sample MSI state will be ultimately determined to be microsatellite instability.

[0136] When both the first and second results point to a microsatellite instability site, the final result is determined to be a microsatellite instability site.

[0137] In a preferred embodiment, when the sample to be tested includes a tumor sample and a control sample of the tumor sample, S3) includes:

[0138] S3-B1) Quality filtering is performed on the duplicate short fragment count files of tumor samples and control samples to obtain qualified duplicate short fragment count files; quality filtering is performed on the peak group distribution information to obtain qualified peak group distribution information, and the total number of microsatellite sites in the qualified peak group distribution information is recorded as the total number of qualified microsatellite sites.

[0139] (S3-B2) Traverse the qualified repeated short fragment count files, and obtain the repeated short fragment count information of tumor samples and control samples at the same microsatellite locus. Use asymptotic regression difference calculation to obtain the stepwise difference value of the microsatellite locus. The average of the stepwise difference values ​​of all loci is recorded as the first paired sample evaluation value. The first paired sample evaluation value is compared with the paired sample threshold and gray zone value of the corresponding cancer type to obtain the first result.

[0140] (S3-B3) Traverse the qualified repeat short fragment counting files, simulate capillary PCR peak diagrams for the PCR validation sites of each microsatellite locus in the test sample, calculate the difference in repeat sequence length values ​​corresponding to the apex of the main peak of the tumor sample and the control sample, and record it as the second paired sample evaluation value; compare the second paired sample evaluation value with the first threshold of the corresponding cancer type, and if it exceeds the first threshold, it is determined to be a microsatellite unstable site, and the first paired PCR microsatellite unstable site set is obtained; compare the total value of microsatellite unstable sites in the first paired PCR microsatellite unstable site set with the third threshold of the corresponding sequencing platform to obtain the second result;

[0141] (S3-B4) Based on the qualified peak distribution information, calculate the standard deviation of the repeat sequence length difference of the tumor sample at each microsatellite locus to obtain the first paired value; multiply the first baseline value of the corresponding sequencing platform by the correction coefficient, and add the second baseline value to obtain the locus baseline value corresponding to each microsatellite locus; compare the first paired value with the locus baseline value, and if it exceeds the locus baseline value, mark the corresponding microsatellite locus as a microsatellite unstable locus and record it in the first paired microsatellite unstable locus set; divide the total number of microsatellite unstable loci in the first paired microsatellite unstable locus set by the total number of qualified microsatellite loci to obtain the first paired tumor single sample evaluation value;

[0142] The first paired tumor single sample evaluation value is compared with the first single sample threshold and gray area value to obtain the third result;

[0143] (S3-B5) Based on the peak distribution information, the PCR validation sites for each microsatellite locus in the test sample are determined using the method in (S3-B4) to determine whether the PCR validation sites are microsatellite unstable sites. The PCR validation sites determined to be microsatellite unstable sites are summarized to form the first paired PCR microsatellite unstable site set. The total value of the first paired PCR microsatellite unstable site set is recorded as the second paired tumor single sample evaluation value. The second paired tumor single sample evaluation value is compared with the third threshold of the corresponding sequencing platform to obtain the fourth result.

[0144] S3-B6) Combines the first, second, third, and fourth results to determine and output the microsatellite site stability results of the tumor sample.

[0145] Preferably, before obtaining the distribution difference value, K-mer counting is performed on the repeating short fragment count file of the above-mentioned test sample, and the data in the test sample that supports a specific repeat count with a minimum read count of 3 are filtered out, so as to obtain the Counts data corresponding to each base repeat number within the position range of each MSI site of the two samples.

[0146] Preferably, before simulating the PCR peak diagram, K-mer counting is performed on the repeat short fragment count file of the above-mentioned test sample to obtain the counts data corresponding to the number of repeats of each base in each MSI site range of the two samples, and capillary PCR diagram is simulated to obtain the above-mentioned difference (the method is actually the same as the calculation method of the first threshold in the training set samples).

[0147] Through the above-described detection method, this invention not only solves many problems in the existing technology of MSI detection, such as inconsistencies in prediction caused by platform differences, differences in tumor pathology, and challenges in detection gray areas, but also successfully pushes the accuracy of MSI detection to a higher level, providing technical support for early cancer diagnosis, genetic risk assessment, and treatment plan formulation, and can promote the progress of oncology research and clinical practice.

[0148] In a preferred embodiment, the method for determining S3-B6) includes: the gray zone determination rule includes: the lower limit of the paired sample threshold minus the gray zone value, and the upper limit of the paired sample threshold plus the gray zone value. If the first paired sample evaluation value is between the lower limit and the upper limit of the paired sample threshold, the result output is that the corresponding microsatellite locus is in the gray zone; otherwise, it is not in the gray zone. The lower limit of the first single sample threshold minus the gray zone value, and the upper limit of the first single sample threshold plus the gray zone value, are also included. If the first single sample evaluation value is between the lower limit and the upper limit of the first single sample threshold, the result output is that the microsatellite locus is in the gray zone; otherwise, it is not in the gray zone.

[0149] a. When the first, second, third, and fourth results are consistent, the final output will be a consistent judgment result;

[0150] b. When the second and fourth results are consistent, and the result is consistent with either the first or the third result:

[0151] b1. Consistent with the third result, the final output is the third result as the judgment result;

[0152] b2. If the result is consistent with the first result, and the first result is the determination result of microsatellite instability, and the third result is in the gray area, the final result output is the determination result of microsatellite instability; if the third result is not in the gray area, the final result output is the determination result of microsatellite stability.

[0153] b3. If the result is consistent with the first result, and the first result is the determination result of microsatellite stability, and the third result is in the gray area, the final result output is the determination result of microsatellite stability; if the third result is not in the gray area, the final result output is the determination result of microsatellite instability.

[0154] c. When the second and fourth results are consistent, but inconsistent with the first and third results:

[0155] c1. The second and fourth results are the determination results of microsatellite stability. If the third result is in the gray area, the final result output is the determination result of microsatellite stability; if the third result is not in the gray area, the final result output is the determination result of microsatellite instability.

[0156] c2. The second and fourth results are the determination results of microsatellite instability. If the third result is in the gray area, the final result output is the determination result of microsatellite instability; if the third result is not in the gray area, the final result output is the determination result of microsatellite stability.

[0157] d. If the second and fourth results are inconsistent, but the first and third results are consistent, the final output will be the judgment result corresponding to the consistency between the first and third results;

[0158] e. When the second and fourth results are inconsistent, and when they are inconsistent with the first and third results:

[0159] e1.1. Both the third and fourth results are the determination results of microsatellite instability. The first single sample evaluation value exceeds the upper limit of the first single sample threshold. The number of microsatellite unstable sites in the fourth result is greater than or equal to 3. The final result output is the determination result of microsatellite instability.

[0160] e1.2. The third and fourth results are both the determination results of microsatellite stability. The first single sample evaluation value is lower than the lower limit of the first single sample threshold. The fourth result has 0 microsatellite unstable sites. The final result output is the determination result of microsatellite stability.

[0161] e1.3. If the first and second results are consistent, and the third and fourth results are consistent, and the first result is determined to be a microsatellite instability, the evaluation value of the first paired sample is greater than or equal to the sum of the upper limit of the paired sample threshold and the gray area value, and the number of microsatellite instability sites in the second result is greater than or equal to 4, the final result output is the microsatellite instability determination result; otherwise, the final result output is the determination result of the third result.

[0162] e2. If the first and second results are inconsistent, and the third and fourth results are inconsistent, the final output will be the judgment result of the third result.

[0163] The detection method proposed in this application improves the accuracy and reliability of MSI status assessment, fully considers sample characteristics, threshold ranges, and the mutual verification between multiple assessment methods, and provides a technological foundation and innovative solution for research and clinical applications in the biomedical field.

[0164] In this application, when the test sample includes tumor samples and their corresponding control samples, the comprehensive judgment strategy of this application can further improve the accuracy and reliability of microsatellite instability (MSI) state assessment, as follows: when the four assessment results show complete consistency, it means that the detection method of this application has reached the same MSI state conclusion in different dimensions, and the output is a consistent result.

[0165] As previously described, when using high-throughput data to calculate the MSI status of multiple microsatellite loci in a sample, significantly stable states (MSS) or significantly unstable states (MSI-H) are more reliable. Results calculated using a large set of MSS samples based on a baseline of a specific cancer type and a specific sequencing platform are more reliable in judging MSS and MSI-H than results from a single control. PCR simulation results can serve as supporting evidence. When judging MSS and MSI-H, the reliability ranking is third result > first result > fourth result > second result. For moderately unstable states (MSI-L), the gray zone range and data consistency need to be considered. If the underlying data algorithms (standard deviation of repeat sequence length difference, stepwise difference value of repeat short fragment count) are the same, the more consistent the conclusions, the more reliable they are. At the same time, if more reliable results are within the gray zone, the output is microsatellite stable; if more reliable results are outside the gray zone, the output is microsatellite unstable. Through the analysis of a large number of validation set samples, the comprehensive evaluation of this application has been validated.

[0166] The most complex situation occurs when the first, second, third, and fourth results differ even when based on the same algorithm (standard deviation of the repeat sequence length difference, step-by-step difference in the repeating short segment count). In this case, following the aforementioned result reliability rules and gray zone determination rules, it can be seen that even in complex and variable data environments, the detection method and judgment rules of this application can provide a stable MSI state judgment.

[0167] In a preferred embodiment, the method for obtaining a repeating short fragment file includes: using Mantis software to obtain the repeating short fragment file.

[0168] In a preferred embodiment, the method for obtaining peak distribution information of repetitive short fragment files includes: using msings analysis software to obtain peak distribution information.

[0169] Before detecting microsatellite instability in the test samples, this application performed quality control on both the test sample files and the high-throughput targeted data files of the training set samples. Specifically, the reference genome pairing file corresponding to the sample was first precisely combined with the microsatellite instability site file in the sample to obtain the first file, which summarizes comprehensive information on the microsatellite sites in the test sample, including location, length variations, and other relevant parameters.

[0170] In the first file, this application excluded all data with a coverage depth lower than 30×, as low coverage depth may introduce unnecessary errors and uncertainties. Furthermore, data with a minimum read quality lower than 25, a minimum average locus quality lower than 30, a minimum read length lower than 35, and a coverage depth lower than 30× were filtered out. Through quality control, a file containing the counts of repeated short fragments from the test samples was obtained for subsequent analysis. Simultaneously, based on the first file, reads with a coverage depth lower than 6× were filtered to obtain the corresponding peak distribution information, thereby improving the accuracy and reliability of the peak distribution information and providing more accurate support for subsequent microsatellite instability assessment.

[0171] In a second typical embodiment of this application, an electronic device for detecting microsatellite site stability is provided. This electronic device includes a feature data acquisition unit 01, a prediction model building unit 02, and a result output unit 03. A schematic diagram of this electronic device is shown below. Figure 2 As shown;

[0172] The feature data acquisition unit is used to acquire feature data from a preset microsatellite locus set using high-throughput targeted sequencing data of the sample to be tested and information on the cancer type and sequencing platform corresponding to the high-throughput targeted sequencing data. The feature data includes at least the following: length variations of short tandem repeat sequences and structural variations occurring at microsatellite loci; the prediction model building unit is used to build a prediction model using the feature data; and the result output unit is used to output the microsatellite stability results of the sample to be tested using the prediction model.

[0173] In a preferred embodiment, the feature data acquisition unit includes a microsatellite set acquisition unit, a repetitive short fragment count file acquisition unit, and a peak distribution information acquisition unit. The microsatellite set acquisition unit is used to obtain a preset microsatellite locus set and feature data by comparing the high-throughput targeted sequencing data of the sample to be tested with the reference genome data of the sample to be tested, and combining platform information and information on the corresponding cancer type. The repetitive short fragment count file acquisition unit is used to obtain the repetitive short fragment count file of the microsatellite locus based on the length variation of the short tandem repeat sequence. The peak distribution information acquisition unit is used to obtain the peak distribution information of the microsatellite locus of the sample to be tested based on the structural variations occurring at the microsatellite locus.

[0174] In a preferred embodiment, the prediction model building unit includes a model training unit;

[0175] The model training unit is used to train a mathematical model using feature data from a preset set of microsatellite loci in the training set samples, and to obtain a prediction model. The training set samples include training set tumor samples and control samples of the training set tumor samples. Both the training set tumor samples and control samples include microsatellite stable samples and microsatellite unstable samples.

[0176] Using feature data from a pre-defined set of microsatellite loci obtained from training samples acquired through different sequencing platforms, a mathematical model is trained to obtain a prediction model; the prediction model includes a first threshold, a second threshold, a third threshold, a correction coefficient, a baseline, and gray zone values;

[0177] The correction factor is: for different sequencing platforms, the standard deviation of the read length corresponding to each microsatellite locus in the stable microsatellite sample of the peak distribution information of the training set samples, and the value after correcting the standard deviation is the correction factor of the corresponding sequencing platform.

[0178] The baseline includes a first baseline value and a second baseline value. The first baseline value is the average of the standard deviations of the peak group distribution of all microsatellite sites in the microsatellite stable sample; the gray area value is 0.05-0.1.

[0179] The second baseline value is the average number of peaks in the peak cluster distribution of each microsatellite site in the microsatellite stable sample, which is the first baseline value.

[0180] The first threshold is: based on the cancer type corresponding to the training set samples, calculate the difference in the length of the repeat sequence corresponding to the vertex of the simulated main peak of the tumor sample and the control sample, and record it as the first threshold; the second threshold includes the single sample threshold and the paired sample threshold; the single sample threshold includes the first single sample threshold and the second single sample threshold;

[0181] The first single-sample threshold is as follows: For different sequencing platforms, calculate the standard deviation of the peak distribution of each microsatellite locus in the training set samples, and determine whether each microsatellite locus is a microsatellite unstable locus. Read the total value of the microsatellite unstable locus determined to be a microsatellite unstable locus. Calculate the total value of microsatellite unstable loci in each training set sample and divide it by the total number of loci that have passed quality control. This value is recorded as the first single-sample threshold.

[0182] The second single-sample threshold is the sum of the first single-sample threshold and the gray zone value for different sequencing platforms. The paired-sample threshold is the average of the step difference calculated by extracting the repeating short fragment counts of the quality-controlled sites in the training set tumor samples and control samples according to the cancer type corresponding to the training set samples.

[0183] The third threshold is: for different sequencing platforms, the third threshold is the number of microsatellite unstable sites determined based on the PCR peak diagram of the training set samples.

[0184] In a preferred embodiment, when the sample to be tested is a single tumor sample, the result output unit includes a first single tumor sample evaluation value acquisition unit, a second single tumor sample evaluation value acquisition unit, and a result comparison unit; wherein, the first single tumor sample evaluation value acquisition unit includes a qualified peak distribution information acquisition unit, a first numerical value acquisition unit, a site baseline value acquisition unit, a first microsatellite unstable site set acquisition unit, and a first single tumor sample evaluation value output unit;

[0185] The qualified peak group distribution information acquisition unit is used to filter the peak group distribution information to obtain qualified peak group distribution information. The total number of microsatellite sites in the qualified peak group distribution information is recorded as the total number of qualified microsatellite sites.

[0186] The first numerical acquisition unit is used to calculate the standard deviation of the repeat sequence length difference of a single tumor sample at each microsatellite locus using qualified peak group distribution information, and obtain the first numerical value.

[0187] The site baseline value acquisition unit is used to multiply the first baseline value of the corresponding sequencing platform by the correction coefficient and add the second baseline value to obtain the site baseline value corresponding to each microsatellite site; the first microsatellite unstable site set acquisition unit is used to compare the first value with the site baseline value. If it exceeds the site baseline value, the corresponding microsatellite site is marked as a microsatellite unstable site and recorded in the first microsatellite unstable site set.

[0188] The first tumor single-sample assessment value output unit is used to divide the total number of microsatellite unstable sites in the first microsatellite unstable site set by the total number of qualified microsatellite sites to obtain the first tumor single-sample assessment value. The second tumor single-sample assessment value acquisition unit includes a first PCR microsatellite unstable site set acquisition unit and a second tumor single-sample assessment value output unit. The first PCR microsatellite unstable site set acquisition unit is used to determine whether a PCR validation site is a microsatellite unstable site based on peak distribution information and the method of the first microsatellite unstable site set acquisition unit. It then summarizes the PCR validation sites determined to be microsatellite unstable sites to form the first PCR microsatellite unstable site set. The second tumor single-sample assessment value output unit is used to read the total value of the first PCR microsatellite unstable site set and record it as the second tumor single-sample assessment value.

[0189] The result comparison unit compares the first tumor single-sample assessment value with the first single-sample threshold and the second single-sample threshold to obtain the first result; it compares the second tumor single-sample assessment value with the third threshold to obtain the second result; and it combines the first result and the second result to determine and output the microsatellite site stability result of the tumor single sample.

[0190] In a preferred embodiment, when the sample to be tested is a single tumor sample, the determination method includes: a. If both the first result and the second result are microsatellite stable, the final result output is microsatellite stable; b. If the first result is microsatellite stable and the second result is microsatellite unstable, the final result output is microsatellite stable; c. If the first tumor single sample assessment value is between the first single sample threshold and the second single sample threshold, and the second result is microsatellite stable, the final result output is microsatellite stable; d. If the first tumor single sample assessment value is between the first single sample threshold and the second single sample threshold, and the second result is microsatellite unstable, the final result output is microsatellite unstable; e. If the first tumor single sample assessment value is greater than or equal to the second single sample threshold, and the second result is microsatellite stable, the final result output is microsatellite unstable; f. If the first tumor single sample assessment value is greater than or equal to the second single sample threshold, and the second result is microsatellite unstable, the final result output is microsatellite unstable.

[0191] In a preferred embodiment, when the sample to be tested is a paired sample, the determination method includes: when the sample to be tested includes a tumor sample and a control sample of the tumor sample, the result output unit includes a first result acquisition unit, a second result acquisition unit, a third result acquisition unit, a fourth result acquisition unit, and a result comparison unit.

[0192] The first result acquisition unit includes a first paired sample evaluation value acquisition unit and a first result output unit;

[0193] The first paired sample evaluation value acquisition unit is used to traverse qualified duplicate short fragment count files, use progressive regression difference calculation to obtain the stepwise difference value of the microsatellite locus by combining the duplicate short fragment count information of tumor samples and control samples at the same microsatellite locus, calculate the average of the stepwise difference values ​​of all loci, and obtain the first paired sample evaluation value.

[0194] The first result output unit is used to compare the evaluation value of the first paired sample with the threshold and gray value of the paired sample of the corresponding cancer type to obtain the first result;

[0195] The second result acquisition unit includes a second paired sample evaluation value acquisition unit, a paired first PCR microsatellite unstable site set acquisition unit, and a second result output unit.

[0196] The second paired sample evaluation value acquisition unit is used to traverse the qualified repeat short fragment count file, simulate the capillary PCR peak diagram for the PCR verification site of each microsatellite locus of the sample to be tested, calculate the difference in repeat sequence length value corresponding to the apex of the main peak of the tumor sample and the control sample, and obtain the second paired sample evaluation value.

[0197] The first PCR microsatellite unstable site set acquisition unit is used to compare the evaluation value of the second paired sample with the first threshold of the corresponding cancer type. If the value exceeds the threshold, it is determined to be a microsatellite unstable site, and the first PCR microsatellite unstable site set is acquired.

[0198] The second result output unit is used to compare the total value of the paired first PCR microsatellite unstable site set with the third threshold of the corresponding sequencing platform to obtain the second result;

[0199] The third result acquisition unit includes a first numerical acquisition unit, a single-sample first microsatellite unstable site set acquisition unit, a first tumor single-sample evaluation value output unit, and a third result output unit.

[0200] The first paired value acquisition unit is used to calculate the standard deviation of the repeat sequence length difference of a single tumor sample at each microsatellite locus using qualified peak group distribution information, and obtain the first paired value.

[0201] The paired site baseline value acquisition unit multiplies the first baseline value of the corresponding sequencing platform by a correction coefficient and adds the second baseline value to obtain the site baseline value corresponding to each microsatellite site. The first paired microsatellite unstable site set acquisition unit is used to compare the first paired value with the site baseline value. If it exceeds the site baseline value, the corresponding microsatellite site is marked as a microsatellite unstable site and recorded in the first paired microsatellite unstable site set. The first tumor single sample evaluation value output unit is used to divide the total value of the first microsatellite unstable site set of a single sample by the total number of microsatellite unstable sites in the repeated short fragment count file of the tumor single sample to obtain the first tumor single sample evaluation value.

[0202] The third result output unit is used to compare the first tumor single sample evaluation value with the first single sample threshold and the second single sample threshold to obtain the third result.

[0203] The fourth result acquisition unit includes the first single-sample PCR microsatellite unstable site set acquisition unit, the second tumor single-sample evaluation value output unit, and the fourth result output unit.

[0204] The first single-sample PCR microsatellite unstable site set acquisition unit is used to obtain the first single-sample PCR microsatellite unstable site set for PCR validation sites using a method similar to the single-sample first microsatellite unstable site set acquisition unit.

[0205] The second tumor single-sample assessment value output unit is used to calculate the number of sites in the first single-sample PCR microsatellite unstable site set and obtain the second tumor single-sample assessment value.

[0206] The fourth result output unit is used to compare the second tumor single sample evaluation value with the third threshold corresponding to the sequencing platform to obtain the fourth result.

[0207] The result comparison unit is used to integrate the first, second, third, and fourth results to determine the stability of microsatellite sites in tumor samples.

[0208] In a preferred embodiment, the determination method includes: subtracting the gray area value from the paired sample threshold to obtain the lower limit of the paired sample threshold, and adding the gray area value to the paired sample threshold to obtain the upper limit of the paired sample threshold. If the first paired sample evaluation value is between the lower limit and the upper limit of the paired sample threshold, it is said that the first result is in the gray area; otherwise, it is said that it is not in the gray area. Subtracting the gray area value from the first single sample threshold to obtain the lower limit of the single sample threshold, and adding the gray area value to the first single sample threshold to obtain the upper limit of the single sample threshold, if the first single sample evaluation value is between the lower limit and the upper limit of the single sample threshold, it is said that the third result is in the gray area; otherwise, it is said that it is not in the gray area.

[0209] a. When the first, second, third, and fourth results are consistent, the final output will be a consistent judgment result;

[0210] b. When the second and fourth results are consistent, and the result is consistent with either the first or the third result:

[0211] b1. If the result is consistent with the third result, output the judgment result of the third result;

[0212] b2. Consistent with the first result, and the first result indicates microsatellite instability. If the third result is in the gray area, the final output result is microsatellite instability; if the third result is not in the gray area, the final output result is microsatellite stable.

[0213] b3. Consistent with the first result, and the first result indicates that the microsatellite is stable. If the third result is in the gray area, the final output result is that the microsatellite is stable; if the third result is not in the gray area, the final output result is that the microsatellite is unstable.

[0214] c. When the second and fourth results are consistent, but inconsistent with the first and third results:

[0215] c1. The second and fourth results indicate that the microsatellite is stable. If the third result is in the gray area, the final output result is that the microsatellite is stable; if the third result is not in the gray area, the final output result is that the microsatellite is unstable.

[0216] c2. The second and fourth results indicate that the microsatellite is unstable. If the third result is in the gray area, the final output result is that the microsatellite is unstable; if the third result is not in the gray area, the final output result is that the microsatellite is stable.

[0217] d. If the second and fourth results are inconsistent, but the first and third results are consistent, the final output will be the result corresponding to the consistency between the first and third results;

[0218] e. When the second and fourth results are inconsistent, and when they are inconsistent with the first and third results:

[0219] e1.1. Both the third and fourth results indicate microsatellite instability. The first single-sample evaluation value exceeds the upper limit of the single-sample threshold, and the number of microsatellite instability sites in the fourth result is greater than or equal to 3. The final output result is microsatellite instability.

[0220] e1.2. Both the third and fourth results indicate that the microsatellites are stable. The first single-sample evaluation value is lower than the lower limit of the single-sample threshold, and the fourth result shows that there are 0 unstable microsatellite sites. The final output result is that the microsatellites are stable.

[0221] e1.3. If the first and second results are consistent, and the third and fourth results are consistent, and the first result determines that the microsatellite is unstable, the evaluation value of the first paired sample is greater than or equal to the sum of the upper limit of the paired sample threshold and the gray area value, and the number of microsatellite unstable sites in the second result is greater than or equal to 4, the final result output is microsatellite unstable; otherwise, the final result output is the third result.

[0222] e2. If the first and second results are inconsistent, and the third and fourth results are inconsistent, the final output will be the judgment result of the third result.

[0223] Before detecting microsatellite instability in the sample to be tested, the reference genome pairing file corresponding to the high-throughput targeted sequencing data of the sample to be tested is combined with the microsatellite instability site file in the sample to be tested to obtain the first file; reads with a coverage depth of less than 30× in the first file are filtered to obtain the peak distribution information in the first file and obtain the repeating short fragment count file; in the first file, reads with a minimum read quality of less than 25, a minimum average locus quality of less than 30, a minimum read length of less than 35, and data with a coverage depth of less than 30× are filtered to obtain the peak distribution information.

[0224] In a third typical embodiment of this application, a computer device is provided, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the above-described method for detecting microsatellite site stability.

[0225] In a fourth typical embodiment of this application, a computer-readable storage medium is provided, on which computer program instructions are stored, and when the computer program / instructions are executed by a processor, the steps of the above-described method for detecting the stability of microsatellite sites are implemented.

[0226] In a fifth typical embodiment of this application, the computer program product includes a computer program / instructions that, when executed by a processor, implement the steps of the above-described method for detecting microsatellite site stability.

[0227] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, because according to this invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions involved are not necessarily essential to this application.

[0228] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus hardware devices such as detection devices. Based on this understanding, the data processing part of the technical solution of this application can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments or some parts of the embodiments of this application.

[0229] This application can be used in a wide range of general-purpose or special-purpose computing system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics devices, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices.

[0230] The method provided in this application can be executed on a terminal, computer terminal, or similar computing device. Taking running on a terminal as an example, Figure 3 This is a hardware structure block diagram of a method for detecting microsatellite site stability. For example... Figure 3 As shown, a terminal may include one or more ( Figure 3Only one is shown in the diagram. A processor A1 (processor A1 may include, but is not limited to, a microprocessor (MCU) or a programmable logic device (FPGA), etc.) and a memory B1 for storing data are also shown. Optionally, the terminal may further include a transmission device C1 for communication functions and an input / output device D1. Those skilled in the art will understand that... Figure 3 The structure shown is for illustrative purposes only and does not limit the structure of the terminal described above. For example, the terminal may also include components that are more... Figure 3 The more or fewer components shown, or having the same Figure 3 The different configurations shown.

[0231] Memory B1 can be used to store computer programs, such as application software programs and modules, like the computer programs corresponding to the methods of segment concatenation, clustering, and consistency processing in this embodiment of the invention. Processor A1 executes various functional applications and data processing by running the computer programs stored in memory B1, thereby implementing the methods described above. Memory B1 may include high-speed random access memory and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, memory B1 may further include memory remotely located relative to processor A1, and these remote memories can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0232] Transmission device C1 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the terminal's communication provider. In one example, transmission device C1 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, transmission device C1 may be a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0233] Obviously, those skilled in the art should understand that some modules or steps of this application described above can be implemented on general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, thereby storing them in a storage device for execution by a computing device, or fabricating them separately as individual integrated circuit modules, or fabricating multiple modules or steps as a single integrated circuit module. Thus, this application is not limited to any specific hardware and software combination. The beneficial effects of this application will be further explained in detail below with reference to specific embodiments.

[0234] Example 1: Threshold Calculation

[0235] NGS simulation of capillary PCR peak diagrams to determine the threshold parameters for the computational model:

[0236] 224 paired samples (i.e., training set samples, including tumor samples and their control normal tissue samples, of which 48 were MSI-H and 176 were MSS) were selected from the submitted samples. The location information of these microsatellite loci on the hg19 reference genome was verified by PCR as follows:

[0237] "Bat25": '4:55546550-55546571';

[0238] "Bat26": '2:47641558-47641586';

[0239] "NR21": '14:23652345-23652367';

[0240] "NR24": '2:95849360-95849384';

[0241] "NR27": '11:102193507-102193534'.

[0242] The alignment results file of the paired samples (i.e. the file after the tumor sample and its control sample are aligned) is analyzed by Mantis software and Misings software to obtain the count results file of repeated short fragments of each microsatellite locus and the peak distribution file of each microsatellite locus.

[0243] Counting of repeating short segments and calculation of their thresholds: Using repeating base sequence signals Repeatats at microsatellite sites as the x-axis and the clustering statistics read count kmer_counts of the corresponding Repeatats as the y-axis, a peak curve is simulated, and the main peak is found by the maximum coverage area.

[0244] The apex of the main peak is taken as the position of the signal peak, and its corresponding Repeats value is the simulated electrophoretic distance (MD). The MD values ​​of the tumor sample and the control sample are obtained, and the difference in MD between the tumor sample and the control sample is calculated.

[0245] Endometrial cancer was divided into two categories, one with other cancers, and a first threshold was calculated for each category. Each microsatellite locus was compared with its prior PCR results. The ROC (Receiver Operating Characteristic) curve method was used to calculate the true positive rate (TPR) and false positive rate (FPR) at different thresholds to plot the curves. The Youden's Index was used to determine the positive determination threshold. Loci below the threshold were MSS loci, and loci above the threshold were MSI-H loci.

[0246] Referring to the gold standard of PCR validation, the model, using the maximum Youden index, ultimately yielded MD_th (first threshold) = 1.6 for endometrial cancer and Youden index = 1; for other cancer types, MD_th (first threshold) = 2.5 and Youden index = 1.

[0247] By utilizing peak distribution information, correction coefficients are obtained for data from different sequencing platforms, and standard deviations are calculated. Combining CF and standard deviations, the unstable state of microsatellite loci is determined.

[0248] The ROC curve method was used to construct a model with the Youden index to determine the positive decision threshold, and finally the correction coefficients for different sequencing platforms were obtained.

[0249] Peak distribution and its threshold calculation: The upper limit of site instability is obtained by multiplying the baseline peak + standard deviation by the correction coefficient CF. If the number of sample peaks is lower than the upper limit of site instability, it is considered an MSS site, and if it exceeds the upper limit of site instability, it is considered an MSI-H site.

[0250] Depending on the sequencing platform, the accuracy of the positive decision threshold was evaluated using the ROC curve method with the Youden's Index. The model with the largest Youden's Index was the optimal one.

[0251] The final result for the Genmax sequencer SURFSeq 5000 was CF=2 and Youden index=0.987.

[0252] The BGI Genomics T7 sequencer has a CF of 1.5 and a Youden index of 0.972.

[0253] The Illumina NovaSeq Xplus sequencer has a CF of 2 and a Youden index of 1.

[0254] The second threshold includes a single-sample threshold and a paired-sample threshold;

[0255] The first single-sample threshold is determined by the standard deviation of microsatellite length variation, which is the number of unstable sites divided by the number of sites that pass quality control, and corresponds to the sequencing platform.

[0256] The first single-sample threshold is determined by comparing the standard deviation of the repetitive sequence length change at each microsatellite locus in the training set tumor samples and control samples to determine whether microsatellite instability has occurred. Each sequencing platform has its own unique base sequencing preference and systematic error. The degree of microsatellite instability of the sample is measured by dividing the number of unstable sites by the number of sites that pass quality control. A cutoff threshold is defined based on the prior results, which is the first single-sample threshold. It needs to be classified according to the sequencing platform.

[0257] The second single-sample threshold is a larger value based on the first single-sample threshold. Together, they are used to specify the range of stability and instability. Corresponding to the sequencing platform, it is the sum of the first single-sample threshold and the gray zone value.

[0258] The paired sample threshold is calculated by taking the average value of the quality-controlled sites after calculating the step difference based on the count of repeated short fragments between tumor samples and control samples, and then assigning it to the cancer type of the sample.

[0259] Different sequencing platforms and their thresholds are shown in Tables 1 and 2.

[0260] The third threshold is the number of unstable sites identified in the PCR peak diagram simulated using high-throughput sequencing data, and it corresponds to the sequencing platform. This third threshold takes into account the systematic error between the PCR gold standard judgment criteria and the actual sequencing platform. The number of unstable sites identified in the PCR peak diagram simulated using high-throughput sequencing data is the third threshold, corresponding to the sequencing platform. The correction coefficient is derived by referring to the variation range of the standard deviation of the read length corresponding to the microsatellite site in the stable microsatellite samples of the training set across different sequencing platforms. It is then combined with a value taken from the upper and lower quartiles of all sites, rounded to one decimal place, and is also corresponding to the sequencing platform.

[0261] The threshold information obtained from the training set samples in this embodiment, based on different sequencing platforms, is shown in Tables 1 and 2. Table 1 shows the threshold information for tumor samples and their corresponding control samples, while Table 2 shows the threshold information when the test sample is only a single tumor sample.

[0262] Table 1

[0263]

[0264] Table 2

[0265]

[0266] Note: In Tables 1 and 2, the NGS-MSI prediction results for control samples are referred to as pNGS, and the NGS-MSI prediction results for single tumor samples are referred to as sNGS.

[0267] In Table 2, the second single-sample threshold is equal to the first single-sample threshold plus the gray area width value.

[0268] Example 2

[0269] Pretreatment and quality filtering of test samples:

[0270] The first file is obtained by combining the reference genome pairing file of the sample to be tested with the MSI location information file of the sample to be tested;

[0271] MSI location information files are BED (Browser Extensible Data) format files in the field of bioinformatics, a concise text format used to store genomic features and describe interval information in the genome. In this embodiment, the targeted sequencing probes used contain location information of known microsatellite unstable sites.

[0272] When calculating the Peaks distribution of the test sample, the data with a coverage depth of less than 30× in the first file are filtered out, and msings analysis is used. Combined with the baseline obtained above and the platform optimization threshold, the Peaks distribution file of the test sample is obtained, which is the second file.

[0273] When calculating the count of duplicate short fragments, reads with a minimum read quality (minimum) of less than 25, data with a minimum average locus quality of less than 30, reads with a minimum read length of less than 35, and data with a coverage depth of less than 30× are filtered out. Using Mantis analysis, the count file of duplicate short fragments of the sample to be tested is obtained, which is the third file.

[0274] When the test samples are a single tumor sample from the same subject and a paired sample (normal tissue sample) corresponding to the single tumor sample, the MSI prediction method is as follows:

[0275] 1) Analyze and predict based on the MSI of two samples from the test sample:

[0276] 1.1) Obtaining NGS-MSI prediction results through asymptotic regression differencing

[0277] The repeat short fragment count files of tumor samples and their paired control samples were organized and K-mer counted. Data with a minimum read count of 3 supporting a specific repeat count were filtered out to obtain the data read for each MSI site and the number of repeats for each base in each sample. The site and count results for the first 20 rows are shown in Table 3.

[0278] Table 3

[0279]

[0280] The system iterates through the qualified repeat short fragment count files, counts the repeat short fragments of tumor samples and control samples at the same microsatellite locus, and calculates the step-wise difference (DIF) value of the microsatellite locus using asymptotic regression difference. The average of the step-wise difference values ​​of all loci is recorded as the first paired sample evaluation value, which is regarded as msi2_score.

[0281] The first paired sample evaluation value is compared with the paired sample threshold and gray zone value in Table 1 of Example 1 for the corresponding cancer type to obtain the first result.

[0282] 1.2) Obtaining PCR-MSI prediction results based on PCR simulation

[0283] The repeat short fragment count files of tumor samples and their paired control samples were compiled and K-mer counted to obtain the counts data for each repeat number of each base at each MSI site for both samples, and a simulated capillary PCR map was generated (consistent with the "NGS simulated capillary PCR peak map" method in Example 1). Combining the cancer type information with the corresponding distance threshold (i.e., MD_th, the first threshold calculated in Example 1), the microsatellite instability of each PCR validation site was calculated.

[0284] The NGS simulated capillary PCR diagram uses repeat base signals at microsatellite sites as the x-axis and the cluster statistics read count (kmer_counts) of the corresponding repeats as the y-axis to simulate peak curves. The main peak is found by the maximum coverage area, and the vertex of the main peak is taken as the signal peak position. The corresponding repeats value is the simulated electrophoretic distance (MD), which is recorded as the second paired sample evaluation value.

[0285] The NGS-based capillary PCR diagram in this embodiment is shown below. Figure 4 As shown, where, Figure 4 Image A shows the NGS capillary PCR result of the control sample. Figure 4 In the middle B, there is an NGS-based capillary PCR image of the tumor sample (with repeat base signals at microsatellite sites as the x-axis and the cluster count of repeats (kmer_counts) as the y-axis).

[0286] The evaluation value of the second paired sample is compared with the first threshold of the corresponding cancer type. If it exceeds the first threshold, it is determined to be a microsatellite unstable site. The paired first PCR microsatellite unstable site set is obtained by summarizing.

[0287] The total value of microsatellite unstable sites in the paired first PCR microsatellite unstable site set is compared with the third threshold of the corresponding sequencing platform to obtain the second result.

[0288] 2) Analyze and predict the MSI of a single tumor sample in the test sample.

[0289] 2.1) Obtaining NGS-MSI prediction results from the standard deviation of microsatellite length variation

[0290] The remaining qualified microsatellite unstable sites after quality control filtering are called qualified sites (passing_loci).

[0291] Calculate the average depth of the tumor sample at each MSI site and obtain the normalized peak count.

[0292] Based on the qualified peak group distribution information, calculate the standard deviation (i.e., the first value) of the difference in repeat sequence length at each microsatellite locus in a single tumor sample.

[0293] The correction coefficient is obtained from the sequencing platform information, and the upper limit of site instability (i.e., the corresponding site baseline value) is calculated using the following formula.

[0294] The upper limit of site instability = baseline peak number + standard deviation × correction factor CF.

[0295] A sample peak number below the upper limit of site instability is identified as an MSI site (unstable_loci), and a peak number above the upper limit of site instability is also identified as an MSI site (unstable_loci). The result msi1_score (first paired tumor single sample evaluation value) is obtained by dividing the number of unstable_loci by the number of passing_loci sites.

[0296] 2.2) Obtaining PCR-MSI prediction results based on PCR simulation

[0297] By combining the sequencing platform information with the threshold obtained in Example 1, the microsatellite instability of each PCR validation site was calculated.

[0298] Based on the peak distribution information, the PCR validation sites of each microsatellite locus in the test sample are determined using the method in 2.1) to determine whether the PCR validation sites are microsatellite unstable sites. The PCR validation sites that are determined to be microsatellite unstable sites are summarized to form the first paired PCR microsatellite unstable site set. The total value of the first paired PCR microsatellite unstable site set is recorded as the second paired tumor single sample evaluation value.

[0299] The second paired tumor single-sample assessment value is compared with the third threshold of the corresponding sequencing platform to obtain the fourth result.

[0300] 3) Final judgment rules of the comprehensive scoring model

[0301] By combining the first, second, third, and fourth results, the stability of microsatellite sites in tumor samples is determined and output.

[0302] The lower limit of the paired sample threshold is the value of the gray area, and the upper limit of the paired sample threshold is the value of the value of the gray area. If the evaluation value of the first paired sample is between the lower limit and the upper limit of the paired sample threshold, it is said that the first result is in the gray area; otherwise, it is said that it is not in the gray area.

[0303] The first single-sample threshold minus the gray area value is the lower limit of the single-sample threshold, and the first single-sample threshold plus the gray area value is the upper limit of the single-sample threshold. If the first single-sample evaluation value is between the lower limit and the upper limit of the single-sample threshold, it is said that the third result is in the gray area; otherwise, it is said that it is not in the gray area.

[0304] The gray zone determination rules include: the lower limit of the paired sample threshold minus the gray zone value, and the upper limit of the paired sample threshold plus the gray zone value. If the evaluation value of the first paired sample is between the lower limit and the upper limit of the paired sample threshold, the result output is that the corresponding microsatellite locus is in the gray zone; otherwise, it is not in the gray zone. The lower limit of the single sample threshold minus the gray zone value, and the upper limit of the single sample threshold plus the gray zone value, are also specified. If the evaluation value of the first single sample is between the lower limit and the upper limit of the single sample threshold, the result output is that the microsatellite locus is in the gray zone; otherwise, it is not in the gray zone.

[0305] a. When the first, second, third, and fourth results are consistent, the final output will be a consistent judgment result;

[0306] b. When the second and fourth results are consistent, and the result is consistent with either the first or the third result:

[0307] b1. Consistent with the third result, the final output is the third result as the judgment result;

[0308] b2. If the result is consistent with the first result, and the first result is the determination result of microsatellite instability, and the third result is in the gray area, the final result output is the determination result of microsatellite instability; if the third result is not in the gray area, the final result output is the determination result of microsatellite stability.

[0309] b3. If the result is consistent with the first result, and the first result is the determination result of microsatellite stability, and the third result is in the gray area, the final result output is the determination result of microsatellite stability; if the third result is not in the gray area, the final result output is the determination result of microsatellite instability.

[0310] c. When the second and fourth results are consistent, but inconsistent with the first and third results:

[0311] c1. The second and fourth results are the determination results of microsatellite stability. If the third result is in the gray area, the final result output is the determination result of microsatellite stability; if the third result is not in the gray area, the final result output is the determination result of microsatellite instability.

[0312] c2. The second and fourth results are the determination results of microsatellite instability. If the third result is in the gray area, the final result output is the determination result of microsatellite instability; if the third result is not in the gray area, the final result output is the determination result of microsatellite stability.

[0313] d. If the second and fourth results are inconsistent, but the first and third results are consistent, the final output will be the judgment result corresponding to the consistency between the first and third results;

[0314] e. When the second and fourth results are inconsistent, and when they are inconsistent with the first and third results:

[0315] e1.1. Both the third and fourth results are the determination results of microsatellite instability. The first single sample evaluation value exceeds the upper limit of the single sample threshold, and the number of microsatellite unstable sites in the fourth result is greater than or equal to 3. The final output is the determination result of microsatellite instability.

[0316] e1.2. The third and fourth results are both the determination results of microsatellite stability. The first single sample evaluation value is lower than the lower limit of the single sample threshold, and the fourth result has 0 microsatellite unstable sites. The final result output is the determination result of microsatellite stability.

[0317] e1.3. If the first and second results are consistent, and the third and fourth results are consistent, and the first result is determined to be a microsatellite instability, the evaluation value of the first paired sample is greater than or equal to the sum of the upper limit of the paired sample threshold and the gray area value, and the number of microsatellite instability sites in the second result is greater than or equal to 4, the final result output is the microsatellite instability determination result; otherwise, the final result output is the determination result of the third result.

[0318] e2. If the first and second results are inconsistent, and the third and fourth results are inconsistent, the final output will be the judgment result of the third result.

[0319] Example 3

[0320] When the test sample is a single tumor sample without a paired control sample, the difference between this method and the method used when the test sample is a single tumor sample and a corresponding paired sample is only:

[0321] The content of step “1)” in Example 2 is omitted; the rest of the analysis is the same as step “1)”.

[0322] In the final judgment method of the comprehensive scoring model, the threshold is the same as Table 2 in Example 1. The difference between the final judgment rules of the comprehensive scoring model and those in Example 2 "3)" is as follows:

[0323] sNGS threshold 1 (first tumor single sample threshold) is abbreviated as th1, sNGS threshold 2 (second tumor single sample threshold) is abbreviated as th2, and the pseudo-PCR site threshold is abbreviated as pth (third threshold).

[0324] The judgment rules are as follows:

[0325] a. msi1_score <= th1 and the number of MSI-H sites in sPCR <= pth;

[0326] The output is MSS.

[0327] b. msi1_score <= th1 and the number of MSI-H sites in sPCR > pth;

[0328] The output is MSS.

[0329] c. The msi1_score is between th1 and th2, and the number of MSI-H sites in sPCR is < pth;

[0330] The output is MSS.

[0331] d. The msi1_score is between th1 and th2, and the number of MSI-H sites in sPCR is >= pth;

[0332] The output is MSI-H.

[0333] e. msi1_score >= th2 and the number of MSI-H sites in sPCR < pth;

[0334] The output is MSI-H.

[0335] f. msi1_score >= th2 and the number of MSI-H sites in sPCR >= pth;

[0336] The output is MSI-H.

[0337] Example 4

[0338] Using single tumor samples and their paired samples with clinical priors different from the training set samples, Mantis analysis, Msings analysis, and the detection methods of Examples 2 and 3 were performed, along with capillary PCR analysis. The results are shown in Table 4. In Table 4, "paired sample" refers to two samples, namely, a single tumor sample from the same subject and its corresponding paired sample (normal white blood cell sample).

[0339] Currently, this invention only classifies and optimizes endometrial cancer. The "other" cancer types in Table 4 include non-endometrial cancer types, including samples of lung cancer, stomach cancer, liver cancer, and colorectal cancer.

[0340] Table 4

[0341]

[0342]

[0343]

[0344]

[0345]

[0346]

[0347] Samples numbered "P1" consist of 22 samples sequenced on the Xplus platform, samples numbered "P2" consist of 16 samples sequenced on the SURFSeq5000 platform, and samples numbered "P3" consist of 4 samples sequenced on the T7 platform. These were clinically paired samples selected from the gray zone near the detection threshold of open-source software. Samples numbered "P4", "P5", and "P6" consist of 28 clinically paired samples sequenced on Xplus, T7, and SURFSeq5000 platforms to cross-validate differences between sequencing platforms. Sample numbered "P7" consisted of 3 replicates each of the MSI-H standard (catalog number CA4645) and MSS standard (catalog number CA4644) purchased from Jingliang Company to verify the reproducibility of the results. The preliminary results were obtained by combining PCR validation, immunohistochemical detection, bioinformatics NGS analysis, medical interpretation, and clinical judgment.

[0348] According to the results in Table 4, the accuracy rate of Mantis was 69.7%, the accuracy rate of Msings was 93.2%, while the accuracy rate of the paired sample judgment method of this application was 100%, the accuracy rate of the single sample judgment method of this application was 100%, and the method of this application solved the problem of determining the MSI-L of the gray area of ​​two samples in PCR verification.

[0349] Example 5

[0350] For samples without prior results, this embodiment selected a large number of lung cancer samples for indirect verification.

[0351] This includes 2820 lung cancer samples, comprising paired samples and single samples from three sequencing platforms. Specific sample information is as follows: Figure 5 As shown and Figure 6 As shown.

[0352] Figure 5 (Source: Landscape of Microsatellite Instability Across 39 Cancer Types, DOI: 10.1200 / PO.17.00086), where... Figure 5 Panel A displays the percentage of MSI-H (Microsatellite Instability-High) cases across 39 different tumor types based on TCGA (Cancer Genome Atlas) data. It can be seen that UCEC (Endometrial Cancer), COAD (Colorectal Adenocarcinoma), and STAD (Gastric Adenocarcinoma) have the highest proportions of MSI-H.

[0353] Figure 5 Panel B shows the distribution of MANTIS scores for different tumor types. MANTIS is a computational tool for detecting microsatellite instability (MSI). The black line (y=0.4) in the figure is often used as the threshold for determining MSI-H. This article provides a detailed analysis of the microsatellite instability landscape in the context of pan-cancer and is an important reference for research on MSI and biomarkers for immunotherapy.

[0354] The literature provides the MSI-H ratio (1%) for a large sample of lung cancer (squamous cell carcinoma LUSC, adenocarcinoma LUAD). This embodiment uses the methods of Embodiments 2 and 3 of this application to detect a large number of clinical samples, calculates the actual MSI-H ratio, and compares it with the literature value to evaluate the accuracy of the detection method of this application. In this embodiment, there are approximately 514 LUAD (lung adenocarcinoma) samples. Figure 5 The results showed that the MSI-H ratio was close to 0%; LUSC (lung squamous cell carcinoma): approximately 492 samples. Figure 5 The data shows that its MSI-H ratio is also very low (<1%).

[0355] Figure 6The image is from an official meeting document from the U.S. Food and Drug Administration (FDA), specifically the briefing document from the Oncologic Drugs Advisory Committee (ODAC) meeting on May 10, 2017 (FDA Briefing Document: Oncologic Drugs Advisory Committee (ODAC) Meeting - Pembrolizumab (Keytruda) for the Treatment of Patients with Microsatellite Instability-High (MSI-H) or Mismatch Repair Deficient (dMMR) Solid Tumors).

[0356] Analysis based on 12,019 tumor samples from the Foundation Medicine database. Figure 6 It is particularly famous in the history of tumor immunotherapy because it provided crucial epidemiological data to support Keytruda's (K drug) acquisition of the first-ever "tissue-agnostic" indication. Specifically, Figure 6 Mismatch repair deficiencies in 12,019 tumors were disclosed. The proportion of mismatch repair deficient tumors in each cancer subtype is expressed as a percentage. Mismatch repair deficient tumors were found in 24 of the 32 tumor subtypes examined, and these tumors were more common in early-stage disease (defined as stage IV or below).

[0357] The literature provides the MSI-H ratio (1-2%) for large-sample lung cancer (small cell lung cancer, non-small cell lung cancer, NSCLC). This embodiment utilizes the methods in Embodiments 2 and 3 of this application to detect a large number of clinical samples, calculate the actual MSI-H ratio, and compare it with the literature values ​​to evaluate the accuracy of the method in this application. The sample size for NSCLC (non-small cell lung cancer) in this embodiment is 1388 cases. The sample size for Small Cell Lung Cancer is 164 cases.

[0358] Figure 5 and Figure 6 These two figures show the MSI-H ratio of lung cancer under two taxonomic categories. The sample size is large, so the theoretical values ​​from the literature are quite reliable.

[0359] Using the calculation methods of Examples 2 and 3 of this application, the above samples meet the expected value of MSI.

[0360] The results calculated using Examples 2 and 3 of this application are shown in Table 5. In Table 5, "paired sample" refers to two samples: a single tumor sample from the same subject and its corresponding paired sample (normal tissue sample). A single sample refers to a single tumor sample.

[0361] Table 5

[0362]

[0363] As can be seen from the above description, the embodiments of the present invention achieve the following technical effects: The method of this application has achieved a technological breakthrough in the field of microsatellite instability (MSI) detection, improving the accuracy and consistency of MSI state prediction, reducing reliance on traditional PCR validation, thereby reducing detection costs, accelerating result reporting, and further improving the adaptability to individual differences in tumors, especially providing more reliable MSI state analysis at specific stages of tumor evolution. This application establishes a unified prediction threshold framework by building a comprehensive computational model, reducing errors from subjective interpretation, and strengthening the reliability and objectivity of detection results, providing support for clinical oncology and genetic research, as well as subsequent treatment guidance.

[0364] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for detecting microsatellite site stability, characterized in that, The detection method includes: S1) Using high-throughput targeted sequencing data of the sample to be tested, obtain characteristic data from a preset set of microsatellite loci, wherein the characteristic data includes at least the following: length variations of short tandem repeat sequences and structural variations occurring at microsatellite loci; S2) Establish a prediction model using the feature data; S3) Utilize the prediction model to output the microsatellite stability results of the sample to be tested; The sample to be tested is either a) a single tumor sample, or b) a tumor sample and a control sample of the tumor sample; In step S2), the method for establishing the prediction model includes: Using the feature data from the preset microsatellite locus set of training set samples obtained through different sequencing platforms, a mathematical model is trained to obtain the prediction model; The training set samples include training set tumor samples and control samples of the training set tumor samples; Both the training set tumor samples and the control samples include microsatellite stable samples and microsatellite unstable samples. The prediction model includes a first threshold, a second threshold, a third threshold, a correction coefficient, a baseline, and gray area values; The correction coefficient is: for different sequencing platforms, the standard deviation of the read length corresponding to each microsatellite locus in the microsatellite stable sample for obtaining the peak distribution information of the training set sample is obtained, and the standard deviation is corrected to obtain the corresponding correction coefficient; The baseline includes a first baseline value and a second baseline value, wherein the first baseline value is the average of the standard deviations of the peak group distribution of all the microsatellite sites in the microsatellite stable sample; The second baseline value is the average number of peaks in the peak cluster distribution of each microsatellite locus of the microsatellite stable sample, where the first baseline value is the average number of peaks. The gray area value is 0.05-0.1; The first threshold is: based on the cancer type corresponding to the training set sample, the difference between the length of the repeat sequence corresponding to the vertex of the simulated main peak of the training set tumor sample and the control sample is calculated and denoted as the first threshold; The second threshold includes a single-sample threshold and a paired-sample threshold; The single-sample threshold includes a first single-sample threshold and a second single-sample threshold; The first single-sample threshold is as follows: For different sequencing platforms, calculate the standard deviation of the peak distribution of each microsatellite locus in the training set sample, determine whether each microsatellite locus is a microsatellite unstable locus, and read the total value of the microsatellite unstable locus; calculate the total value of microsatellite unstable loci in each training set sample divided by the total number of loci that have passed quality control, and record it as the first single-sample threshold. The second single-sample threshold is: the sum of the first single-sample threshold and the gray zone value for different sequencing platforms, which is denoted as the second single-sample threshold; The paired sample threshold is: based on the cancer type corresponding to the training set sample, extract the count of repeated short fragments at quality-controlled sites in the training set tumor sample and the control sample, calculate the average value of the step difference, and record it as the paired sample threshold. The third threshold is, for different sequencing platforms, the number of microsatellite unstable sites determined based on the PCR peak diagram of the training set samples; S1) includes: S1-1) By comparing the high-throughput targeted sequencing data of the sample to be tested with the reference genome data of the sample to be tested, and combining the sequencing platform information and the information of the corresponding cancer type of the sample to be tested, the preset microsatellite locus set and the feature data are obtained. S1-2) Obtain the repeating short fragment count file of the microsatellite locus based on the length variation of the short tandem repeat sequence; S1-3) Based on the structural variations occurring at the microsatellite sites, obtain the peak group distribution information of the microsatellite sites of the sample to be tested.

2. The detection method according to claim 1, characterized in that, When the sample to be tested is a single tumor sample, step S3) includes: S3-A1) The peak group distribution information is filtered for quality to obtain qualified peak group distribution information. The total number of microsatellite sites in the qualified peak group distribution information is recorded as the total number of qualified microsatellite sites. S3-A2) Based on the qualified peak group distribution information, calculate the standard deviation of the repeat sequence length difference of each microsatellite locus in the single tumor sample to obtain a first value; Multiply the first baseline value of the corresponding sequencing platform by the correction coefficient, and then add the second baseline value to obtain the site baseline value corresponding to each microsatellite site; The first value is compared with the site baseline value. If it exceeds the site baseline value, the corresponding microsatellite site is marked as a microsatellite unstable site and recorded in the first microsatellite unstable site set. The first tumor single-sample assessment value is obtained by dividing the total number of microsatellite unstable sites in the first microsatellite unstable site set by the total number of qualified microsatellite sites. S3-A3) Based on the peak distribution information, for each microsatellite site in the sample to be tested, the PCR verification site is determined to be a microsatellite unstable site using the method of S3-A2). The PCR verification sites determined to be microsatellite unstable sites are summarized to form the first PCR microsatellite unstable site set. The total value of the first PCR microsatellite unstable site set is denoted as the second tumor single-sample assessment value; (S3-A4) The first tumor single-sample evaluation value is compared with the first single-sample threshold and the second single-sample threshold respectively to obtain the first result; The second tumor single-sample assessment value is compared with the third threshold to obtain a second result; Combining the first and second results, the stability of the microsatellite loci of the tumor single sample is determined and output.

3. The detection method according to claim 2, characterized in that, The determination method in S3-A4 includes: a. When both the first result and the second result are determinations of microsatellite stability, the final output result is the determination of microsatellite stability; b. When the first result is that the microsatellite is stable and the second result is that the microsatellite is unstable, the final output result is the determination result of microsatellite stability; c. If the first tumor single-sample assessment value is between the first single-sample threshold and the second single-sample threshold, the second result is the microsatellite stability determination result, and the final result output is the microsatellite stability determination result; d. The first tumor single-sample assessment value is between the first single-sample threshold and the second single-sample threshold. The second result is the determination result of microsatellite instability. The final result output is the determination result of microsatellite instability. e. If the first tumor single-sample assessment value is greater than or equal to the second single-sample threshold, the second result is the determination result of microsatellite stability, and the final result output is the determination result of microsatellite instability; f. If the first tumor single-sample assessment value is greater than or equal to the second single-sample threshold, the second result is the determination result of microsatellite instability, and the final result output is the determination result of microsatellite instability.

4. The detection method according to claim 1, characterized in that, When the sample to be tested includes a tumor sample and a control sample of the tumor sample, step S3) includes: S3-B1) Perform quality filtering on the duplicate short fragment count files of the tumor sample and the control sample to obtain qualified duplicate short fragment count files; The peak group distribution information is quality filtered to obtain qualified peak group distribution information, and the total number of microsatellite sites in the qualified peak group distribution information is recorded as the total number of qualified microsatellite sites. S3-B2) Traverse the qualified repeated short fragment count files, and obtain the repeated short fragment count information of tumor samples and control samples at the same microsatellite locus using asymptotic regression difference calculation. The average of the stepwise difference values ​​of all loci is recorded as the first paired sample evaluation value. The first paired sample evaluation value is compared with the paired sample threshold and the gray zone value of the corresponding cancer type to obtain a first result; S3-B3) Traverse the qualified repeat short fragment counting file, simulate capillary PCR peak diagram for the PCR verification site of each microsatellite locus of the sample to be tested, calculate the difference in repeat sequence length value corresponding to the apex of the main peak of the tumor sample and the control sample, and record it as the second paired sample evaluation value. The evaluation value of the second paired sample is compared with the first threshold of the corresponding cancer type. If it exceeds the first threshold, it is determined to be a microsatellite unstable site. The paired first PCR microsatellite unstable site set is obtained by summarizing the results. The total value of microsatellite unstable sites in the paired first PCR microsatellite unstable site set is compared with the third threshold of the corresponding sequencing platform to obtain the second result; (S3-B4) Based on the qualified peak group distribution information, calculate the standard deviation of the repeat sequence length difference of the tumor sample at each of the microsatellite sites to obtain the first paired value; Multiply the first baseline value of the corresponding sequencing platform by the correction coefficient, and finally add the second baseline value to obtain the site baseline value corresponding to each microsatellite site; The first paired value is compared with the site baseline value. If it exceeds the site baseline value, the corresponding microsatellite site is marked as a microsatellite unstable site and recorded in the first paired microsatellite unstable site set. Divide the total number of microsatellite unstable sites in the first paired microsatellite unstable site set by the total number of qualified microsatellite sites to obtain the first paired tumor single sample evaluation value. The first paired tumor single-sample evaluation value is compared with the first single-sample threshold and the gray area value to obtain a third result; S3-B5) Based on the peak distribution information, for each microsatellite site in the sample to be tested, the PCR verification site is determined to be a microsatellite unstable site using the method of S3-B4). The PCR verification sites determined to be microsatellite unstable sites are summarized to form the first set of paired PCR microsatellite unstable sites. The total value of the first paired PCR microsatellite unstable site set is denoted as the second paired tumor single-sample assessment value; The second paired tumor single-sample evaluation value is compared with the third threshold of the corresponding sequencing platform to obtain the fourth result; (S3-B6) Combining the first result, the second result, the third result, and the fourth result, determine and output the microsatellite site stability result of the tumor sample.

5. The detection method according to claim 4, characterized in that, The determination method in S3-B6 includes: The gray area determination rule includes: the lower limit of the paired sample threshold minus the gray area value; The upper limit of the paired sample threshold plus the gray area value is the paired sample threshold; If the first paired sample evaluation value is between the lower limit and the upper limit of the paired sample threshold, the result output is that the corresponding microsatellite site is in the gray area; otherwise, it is not in the gray area. The first single-sample threshold minus the gray area value is the lower limit of the single-sample threshold; The first single-sample threshold plus the gray area value is the upper limit of the single-sample threshold; If the first single-sample evaluation value is between the lower limit of the first single-sample threshold and the upper limit of the first single-sample threshold, the result output is that the microsatellite site is in the gray area; otherwise, it is not in the gray area. a. When the first result, the second result, the third result, and the fourth result are consistent, the final result output is a consistent judgment result; b. When the second result is consistent with the fourth result, and is also consistent with one of the first result and the third result: b1. Consistent with the third result, the final output is the third result as the judgment result; b2. Consistent with the first result, where the first result is the determination of microsatellite instability. If the third result is in the gray area, the final output is the determination result of microsatellite instability; If the third result is not in the gray area, the final output is the determination result of microsatellite stability; b3. Consistent with the first result, and the first result is the determination of microsatellite stability. If the third result is in the gray area, the final output is the determination result of microsatellite stability; If the third result is not in the gray area, the final output will be the determination result of the microsatellite instability. c. When the second result is consistent with the fourth result, but inconsistent with both the first and third results: c1. The second result and the fourth result are the determination results of microsatellite stability. If the third result is in the gray area, the final output is the determination result of microsatellite stability; If the third result is not in the gray area, the final output is the determination result of microsatellite instability; c2. The second result and the fourth result are the determination results of microsatellite instability. If the third result is in the gray area, the final output is the determination result of microsatellite instability; If the third result is not in the gray area, the final output is the determination result of microsatellite stability; d. If the second result and the fourth result are inconsistent, but the first result and the third result are consistent, the final result output is the judgment result corresponding to the consistency between the first result and the third result; e. When the second result is inconsistent with the fourth result, and inconsistent with the first result and the third result: e1.

1. Both the third and fourth results are the determination results of microsatellite instability. The first single sample evaluation value exceeds the upper limit of the first single sample threshold. The number of microsatellite unstable sites in the fourth result is greater than or equal to 3. The final result output is the determination result of microsatellite instability. e1.

2. Both the third and fourth results are microsatellite stability determination results. The first single-sample evaluation value is lower than the lower limit of the first single-sample threshold. The fourth result has 0 microsatellite unstable sites. The final result output is the microsatellite stability determination result. e1.

3. The first result and the second result are consistent, the third result and the fourth result are consistent, and when the first result is determined to be a microsatellite instability, the first paired sample evaluation value is greater than or equal to the sum of the upper limit of the paired sample threshold and the gray area value, and the microsatellite instability sites of the second result are greater than or equal to 4, and the final result output is the microsatellite instability determination result; in other cases, the final result output is the determination result of the third result. e2. If the first result and the second result are inconsistent, and the third result and the fourth result are inconsistent, the final result output is the judgment result of the third result.

6. An electronic device for detecting the stability of microsatellite sites, characterized in that, The electronic device includes a feature data acquisition unit, a prediction model establishment unit, and a result output unit; The feature data acquisition unit is used to acquire feature data from a preset set of microsatellite loci using high-throughput targeted sequencing data of the sample to be tested. The feature data includes at least the following: length variations of short tandem repeat sequences and structural variations occurring at microsatellite loci. The prediction model building unit is used to build a prediction model using the feature data; The result output unit is used to output the microsatellite stability result of the sample to be tested using the prediction model; The prediction model building unit includes a model training unit; The model training unit is used to train a mathematical model using feature data from a preset set of microsatellite loci in the training set samples, thereby obtaining the prediction model. The training set samples include training set tumor samples and control samples of the training set tumor samples; Both the training set tumor samples and the control samples include microsatellite stable samples and microsatellite unstable samples. Using the feature data from the preset microsatellite locus set of training set samples obtained through different sequencing platforms, a mathematical model is trained to obtain the prediction model; The prediction model includes a first threshold, a second threshold, a third threshold, a correction coefficient, a baseline, and gray area values; The correction coefficient is: for different sequencing platforms, the standard deviation of the read length corresponding to each microsatellite locus in the microsatellite stable sample of the peak group distribution information of the training set sample, and the value after correcting the standard deviation is the correction coefficient of the corresponding sequencing platform. The baseline includes a first baseline value and a second baseline value, wherein the first baseline value is the average of the standard deviations of the peak group distribution of all the microsatellite sites in the microsatellite stable sample; The second baseline value is the average number of peaks in the peak cluster distribution of each microsatellite locus of the microsatellite stable sample, where the first baseline value is the average number of peaks. The gray area value is 0.05-0.1; The first threshold is: based on the cancer type corresponding to the training set sample, the difference between the length of the repeat sequence corresponding to the vertex of the simulated main peak of the training set tumor sample and the control sample is calculated and denoted as the first threshold; The second threshold includes a single-sample threshold and a paired-sample threshold; The single-sample threshold includes a first single-sample threshold and a second single-sample threshold; The first single-sample threshold is: for different sequencing platforms, calculate the standard deviation of the peak distribution of each microsatellite locus in the training set sample, determine whether each microsatellite locus is a microsatellite unstable locus, and read the total value of the microsatellite unstable locus; The total number of microsatellite unstable sites in each training set sample is calculated and divided by the total number of sites that have undergone quality control, and this value is recorded as the first single-sample threshold. The second single-sample threshold is: the sum of the first single-sample threshold and the gray zone value for different sequencing platforms, which is denoted as the second single-sample threshold; The paired sample threshold is: based on the cancer type corresponding to the training set sample, the average value of the step difference is calculated by extracting the count of repeated short fragments at quality-controlled sites in the training set tumor sample and the control sample, and is recorded as the paired sample threshold. The third threshold is, for different sequencing platforms, the number of microsatellite unstable sites determined based on the PCR peak diagram of the training set samples; The feature data acquisition unit includes a microsatellite set acquisition unit, a repeating short fragment count file acquisition unit, and a peak group distribution information acquisition unit; The microsatellite set acquisition unit is used to obtain the preset microsatellite locus set and the feature data by comparing the high-throughput targeted sequencing data of the sample to be tested with the reference genome data of the sample to be tested, and combining the sequencing platform information and the information of the corresponding cancer type of the sample to be tested. The repeating short fragment count file acquisition unit is used to obtain the repeating short fragment count file of the microsatellite locus based on the length variation of the short tandem repeat sequence. The peak distribution information acquisition unit is used to obtain the peak distribution information of the microsatellite sites of the sample to be tested based on the structural variations that occur at the microsatellite sites.

7. The electronic device according to claim 6, characterized in that, When the sample to be tested is a single tumor sample, the result output unit includes a first single tumor sample evaluation value acquisition unit, a second single tumor sample evaluation value acquisition unit, and a result comparison unit; The first tumor single-sample evaluation value acquisition unit includes a qualified peak distribution information acquisition unit, a first numerical value acquisition unit, a site baseline value acquisition unit, a first microsatellite unstable site set acquisition unit, and a first tumor single-sample evaluation value output unit. The qualified peak group distribution information acquisition unit is used to perform quality filtering on the peak group distribution information to obtain qualified peak group distribution information, and the total number of microsatellite sites in the qualified peak group distribution information is recorded as the total number of qualified microsatellite sites. The first numerical acquisition unit is used to calculate the standard deviation of the repeat sequence length difference of the tumor single sample at each of the microsatellite sites using the qualified peak group distribution information, and obtain the first numerical value; The site baseline value acquisition unit is used to multiply the first baseline value of the corresponding sequencing platform by the correction coefficient, and then add the second baseline value to obtain the site baseline value corresponding to each microsatellite site; the first microsatellite unstable site set acquisition unit is used to compare the first value with the site baseline value, and if it exceeds the site baseline value, then the corresponding microsatellite site is marked as a microsatellite unstable site and recorded in the first microsatellite unstable site set. The first tumor single-sample evaluation value output unit is used to divide the total number of microsatellite unstable sites in the first microsatellite unstable site set by the total number of qualified microsatellite sites to obtain the first tumor single-sample evaluation value. The second tumor single-sample evaluation value acquisition unit includes a first PCR microsatellite unstable site set acquisition unit and a second tumor single-sample evaluation value output unit. The first PCR microsatellite unstable site set acquisition unit is used to determine whether the PCR verification site of each microsatellite site in the sample to be tested is a microsatellite unstable site according to the peak distribution information, and to summarize the PCR verification sites that are determined to be microsatellite unstable sites to form the first PCR microsatellite unstable site set. The second tumor single-sample evaluation value output unit is used to read the total value of the first PCR microsatellite unstable site set and record it as the second tumor single-sample evaluation value; The result comparison unit compares the first tumor single-sample evaluation value with the first single-sample threshold and the second single-sample threshold respectively to obtain the first result. The second tumor single-sample assessment value is compared with the third threshold to obtain a second result; Combining the first and second results, the stability of the microsatellite loci of the tumor single sample is determined and output.

8. The electronic device according to claim 7, characterized in that, When the sample to be tested is a single tumor sample, the method for determination includes: a. When both the first result and the second result are determinations of microsatellite stability, the final output result is the determination of microsatellite stability; b. When the first result is that the microsatellite is stable and the second result is that the microsatellite is unstable, the final output result is the determination result of microsatellite stability; c. If the first tumor single-sample assessment value is between the first single-sample threshold and the second single-sample threshold, the second result is the microsatellite stability determination result, and the final result output is the microsatellite stability determination result; d. The first tumor single-sample assessment value is between the first single-sample threshold and the second single-sample threshold. The second result is the determination result of microsatellite instability. The final result output is the determination result of microsatellite instability. e. If the first tumor single-sample assessment value is greater than or equal to the second single-sample threshold, the second result is the determination result of microsatellite stability, and the final result output is the determination result of microsatellite instability; f. If the first tumor single-sample assessment value is greater than or equal to the second single-sample threshold, the second result is the determination result of microsatellite instability, and the final result output is the determination result of microsatellite instability.

9. The electronic device according to claim 6, characterized in that, When the test sample includes a tumor sample and a control sample of the tumor sample, the result output unit includes a repeating short fragment count file acquisition unit, a qualified peak group distribution information acquisition unit, a first result acquisition unit, a second result acquisition unit, a third result acquisition unit, a fourth result acquisition unit, and a result comparison unit. The repeated short fragment count file acquisition unit is used to perform quality filtering on the repeated short fragment count files of the tumor sample and the control sample to obtain qualified repeated short fragment count files. The qualified peak group distribution information acquisition unit is used to perform quality filtering on the peak group distribution information to obtain qualified peak group distribution information, and the total number of microsatellite sites in the qualified peak group distribution information is recorded as the total number of qualified microsatellite sites. The first result acquisition unit includes a first paired sample evaluation value acquisition unit and a first result output unit; The first paired sample evaluation value acquisition unit is used to traverse the qualified repeated short fragment count files, the repeated short fragment count information of tumor samples and control samples at the same microsatellite locus, and use progressive regression difference calculation to obtain the step difference value of the microsatellite locus. The average of the step difference values ​​of all loci is recorded as the first paired sample evaluation value. The first result output unit is used to compare the first paired sample evaluation value with the paired sample threshold and the gray area value of the corresponding cancer type to obtain a first result; The second result acquisition unit includes a second paired sample evaluation value acquisition unit, a paired first PCR microsatellite unstable site set acquisition unit, and a second result output unit. The second paired sample evaluation value acquisition unit is used to traverse the qualified repeat short fragment counting file, simulate capillary PCR peak diagram for the PCR verification site of each microsatellite locus of the test sample, calculate the difference in repeat sequence length value corresponding to the apex of the main peak of the tumor sample and the control sample, and record it as the second paired sample evaluation value. The paired first PCR microsatellite unstable site set acquisition unit is used to compare the evaluation value of the second paired sample with the first threshold of the corresponding cancer type. If the value exceeds the first threshold, it is determined to be a microsatellite unstable site, and the paired first PCR microsatellite unstable site set is obtained by summarizing the results. The second result output unit is used to compare the total value of microsatellite unstable sites in the paired first PCR microsatellite unstable site set with the third threshold of the corresponding sequencing platform to obtain a second result; The third result acquisition unit includes a first paired value acquisition unit, a paired site baseline value acquisition unit, a first paired microsatellite unstable site set acquisition unit, a first paired tumor single sample evaluation value output unit, and a third result output unit. The first paired value acquisition unit is used to calculate the standard deviation of the repeat sequence length difference of the tumor single sample at each of the microsatellite sites based on the qualified peak group distribution information, and obtain the first paired value; The paired site baseline value acquisition unit multiplies the first baseline value of the corresponding sequencing platform with the correction coefficient, and then adds the second baseline value to obtain the site baseline value corresponding to each microsatellite site. The first paired microsatellite unstable site set acquisition unit is used to compare the first paired value with the site baseline value. If it exceeds the site baseline value, the corresponding microsatellite site is marked as a microsatellite unstable site and recorded in the first paired microsatellite unstable site set. The first paired tumor single-sample evaluation value output unit is used to divide the total number of microsatellite unstable sites in the first paired microsatellite unstable site set by the total number of qualified microsatellite sites to obtain the first paired tumor single-sample evaluation value. The third result output unit is used to compare the first paired tumor single sample evaluation value with the first single sample threshold and the gray area value respectively to obtain a third result; The fourth result acquisition unit includes the first paired PCR microsatellite unstable site set acquisition unit, the second paired tumor single sample evaluation value output unit, and the fourth result output unit. The first paired PCR microsatellite unstable site set is used to determine whether the PCR validation site of each microsatellite site in the sample to be tested is a microsatellite unstable site according to the peak distribution information, using the S3-B4 method, and to summarize the PCR validation sites that are determined to be microsatellite unstable sites to form the first paired PCR microsatellite unstable site set. The second paired tumor single-sample evaluation value output unit is used to read the total value of the first paired PCR microsatellite unstable site set, and record it as the second paired tumor single-sample evaluation value; The fourth result output unit is used to compare the second paired tumor single sample evaluation value with the third threshold of the corresponding sequencing platform to obtain the fourth result; By combining the first result, the second result, the third result, and the fourth result, the stability of microsatellite sites in the tumor sample is determined.

10. The electronic device according to claim 9, characterized in that, When the sample to be tested includes a tumor sample and a control sample of the tumor sample, the method for determination includes: The lower limit of the paired sample threshold minus the gray area value is the paired sample threshold; The upper limit of the paired sample threshold plus the gray area value is the paired sample threshold; If the first paired sample evaluation value is between the lower limit and the upper limit of the paired sample threshold, the result output is that the corresponding microsatellite site is in the gray area; otherwise, it is not in the gray area. The first single-sample threshold minus the gray area value is the lower limit of the first single-sample threshold; The first single-sample threshold plus the gray area value is the upper limit of the first single-sample threshold; If the first single-sample evaluation value is between the lower limit of the single-sample threshold and the upper limit of the single-sample threshold, the result output is that the microsatellite site is in the gray area; otherwise, it is not in the gray area. a. When the first result, the second result, the third result, and the fourth result are consistent, the final result output is a consistent judgment result; b. When the second result is consistent with the fourth result, and is also consistent with one of the first result and the third result: b1. Consistent with the third result, the final output is the third result as the judgment result; b2. Consistent with the first result, where the first result is the determination of microsatellite instability. If the third result is in the gray area, the final output is the determination result of microsatellite instability; If the third result is not in the gray area, the final output is the determination result of microsatellite stability; b3. Consistent with the first result, and the first result is the determination of microsatellite stability. If the third result is in the gray area, the final output is the determination result of microsatellite stability; If the third result is not in the gray area, the final output will be the determination result of the microsatellite instability. c. When the second result is consistent with the fourth result, but inconsistent with both the first and third results: c1. The second result and the fourth result are the determination results of microsatellite stability. If the third result is in the gray area, the final output is the determination result of microsatellite stability; If the third result is not in the gray area, the final output is the determination result of microsatellite instability; c2. The second result and the fourth result are the determination results of microsatellite instability. If the third result is in the gray area, the final output is the determination result of microsatellite instability; If the third result is not in the gray area, the final output is the determination result of microsatellite stability; d. If the second result and the fourth result are inconsistent, but the first result and the third result are consistent, the final result output is the judgment result corresponding to the consistency between the first result and the third result; e. When the second result is inconsistent with the fourth result, and inconsistent with the first result and the third result: e1.

1. Both the third and fourth results are the determination results of microsatellite instability. The first single sample evaluation value exceeds the upper limit of the first single sample threshold. The number of microsatellite unstable sites in the fourth result is greater than or equal to 3. The final result output is the determination result of microsatellite instability. e1.

2. Both the third and fourth results are microsatellite stability determination results. The first single-sample evaluation value is lower than the lower limit of the first single-sample threshold. The fourth result has 0 microsatellite unstable sites. The final result output is the microsatellite stability determination result. e1.

3. The first result and the second result are consistent, the third result and the fourth result are consistent, and when the first result is determined to be a microsatellite instability, the first paired sample evaluation value is greater than or equal to the sum of the upper limit of the paired sample threshold and the gray area value, and the microsatellite instability sites of the second result are greater than or equal to 4, and the final result output is the microsatellite instability determination result; in other cases, the final result output is the determination result of the third result. e2. If the first result and the second result are inconsistent, and the third result and the fourth result are inconsistent, the final result output is the judgment result of the third result.

11. A computer device comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the method for detecting the stability of microsatellite sites according to any one of claims 1 to 5.

12. A computer-readable storage medium storing computer program instructions thereon, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method for detecting the stability of microsatellite sites according to any one of claims 1 to 5.

13. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method for detecting the stability of microsatellite sites according to any one of claims 1 to 5.

Images