A method and device for detecting microsatellite instability based on second-generation sequencing

Abstract

The present invention belongs to the technical field of bioinformatics, and specifically discloses a method and device for detecting microsatellite instability based on second-generation sequencing. The detection method of the present invention uses sequencing reads obtained by second-generation sequencing technology to first calculate the average number of missing units of each candidate microsatellite site in the sample to be tested, detect the MSI state of each microsatellite site based on this, and then evaluate the MSI state of the sample to be tested. The present invention uses the average number of missing units of the microsatellite site as a characteristic value, does not need to use a reference sample, and does not need to compare the polymorphic distribution of each site. The detection principle is simple, the calculation is simple, the requirement for sequencing depth is low, the clinical feasibility is good, the stability is high, the detection result is accurate and reliable, and the consistency with the PCR gold standard is 100%.

Description

Technical Field

[0001] The invention belongs to the technical field of bioinformatics, and specifically discloses a microsatellite instability detection method and device based on second-generation sequencing. Background Art

[0002] Microsatellite (MS), also known as short tandem repeats (STRs) or simple tandem repeats (SSR), is widely present in the genomes of prokaryotes and eukaryotes. Its repeating unit is 1 to 6 nucleotides, and is generally composed of 10 to 50 repeating units in series. Due to the difference in the number of repetitions of the core repeating unit in the sequence, microsatellites have population polymorphism, and their total length is often less than 400bp. Commonly found are mononucleotide, dinucleotide, trinucleotide, and tetranucleotide repeat sequences, accounting for about 5% of the eukaryotic genome.

[0003] In the human genome, microsatellite mononucleotide repeats are mainly polyA and polyT, dinucleotide repeats are commonly found in (CA)n, (GT)n, (AA)n, (GG)n, and (GC / CG) are rare, because C with G at the 3' end (i.e., CPG) is easily methylated; trinucleotide repeats are mainly (CXG)n, and because trinucleotide repeats are highly polymorphic, they are often used as DNA markers. Studies have shown that when the human body's mismatch repair (MMR) mechanism is defective, microsatellite sequences will change in length, becoming microsatellite instability (MSI). Microsatellite instability is closely related to the occurrence of tumors, especially gastrointestinal cancers such as colorectal cancer, and plays an important role in the treatment and prognosis of tumors. For example, among colorectal cancer patients, patients with high microsatellite instability (MSI-H) have a better prognosis than those with low microsatellite instability (MSI-L) or microsatellite stability (MSS).

[0004] In order to achieve the assessment of microsatellite instability, Chinese patent application CN112955570A (Lexent) discloses a method, including: obtaining quantitative measurements of multiple microsatellite repeat units from a subject's plasma sample (cfDNA), processing the multiple quantitative measurements to obtain statistical measurements of their deviations, and detecting the presence of microsatellite instability when the statistical measurements of the deviations of the multiple quantitative measurements meet a predetermined standard, and vice versa, detecting the absence of microsatellite instability. This method integrates all microsatellite sites in the sample to calculate the statistical measurement of the deviation of the average length (relative to the average z score of the reference sample, and the sequencing depth is less than 10×), which can assess the MSI status of the subject to a certain extent, but requires reference samples, has poor clinical feasibility, and does not detect the MSI status of a single site, so the accuracy is not high. In addition, Chinese patent application CN112639983A (Roche) discloses a method for detecting microsatellite instability, including: detecting sequencing reads from multiple microsatellite sites of human genomic DNA, determining the metric of the repeat length distribution (RLD) of each microsatellite site, comparing the metric with the threshold of the corresponding site, counting the number of microsatellite sites exceeding the threshold, comparing the number with the set microsatellite instability ratio threshold, and determining that the person has microsatellite instability if it exceeds the threshold. The metric of this method is a similarity metric between the RLD of the sample to be tested and the RLD of the control, such as a t statistic determined according to the average read length and variance of the RLD of the site (that is, comparing the polymorphic distribution of each site), which is complex to operate and has a large amount of calculation, and there is a certain inconvenience in practical application.

[0005] Therefore, it is necessary to develop a method for detecting microsatellite instability with simple principle, strong stability and high readiness. Summary of the invention

[0006] The main technical problem solved by the present invention is to provide a microsatellite instability detection method based on second-generation sequencing. The method has simple principle, simple calculation, high stability, strong repeatability, and accurate and reliable detection results.

[0007] At the same time, the present invention also provides a device for detecting microsatellite instability.

[0008] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0009] In a first aspect, the present invention provides a method for detecting microsatellite instability based on second-generation sequencing, comprising the following steps:

[0010] Obtaining sequencing reads of candidate microsatellite loci in the sample to be tested;

[0011] Obtain the average number of missing units at the candidate microsatellite loci in the sample to be tested;

[0012] Detect the MSI status of candidate microsatellite loci in the sample to be tested;

[0013] Evaluate the MSI status of the sample to be tested.

[0014] As a preferred embodiment of the present invention, the sample to be tested is derived from a human or a non-human mammal, preferably a tissue sample or a blood sample.

[0015] Specifically, the tissue sample is a tumor sample, such as a fresh tissue sample, a frozen tissue sample, a FFPE sample, etc.

[0016] Specifically, the blood sample includes a whole blood sample, a plasma sample, etc.

[0017] As a preferred embodiment of the present invention, the candidate microsatellite locus is any one or more of a mononucleotide repeat site, a dinucleotide repeat site, and a trinucleotide repeat site.

[0018] Preferably, the candidate microsatellite site is a mononucleotide repeat site.

[0019] As a preferred embodiment of the present invention, the selection of the candidate microsatellite loci has specific requirements, specifically:

[0020] Providing a training set, wherein the training set includes a plurality of samples with microsatellite instability and a plurality of samples with microsatellite stability;

[0021] Based on the training set, the classification performance of each microsatellite locus was obtained, and the classification performance was represented by an auc value. Microsatellite loci with an auc value ≥ 0.65 were selected as candidate microsatellite loci.

[0022] The auc value is between 0 and 1, and the closer to 1, the better the classification performance of the microsatellite locus. In practical applications, microsatellite loci with mononucleotide repeats with an auc value ≥ 0.65 are preferred. Of course, the auc values ​​of each microsatellite locus can also be sorted by size, and multiple loci with high auc values ​​can be selected as candidate microsatellite loci.

[0023] As a preferred embodiment of the present invention, when calculating the auc value, the average number of missing units is used as the characteristic value.

[0024] Specifically, the auc value is obtained by the following method:

[0025] Based on the training set, the sequencing reads of each microsatellite locus in the sample are obtained;

[0026] The average number of missing units per microsatellite locus in the sample was obtained;

[0027] The average number of missing units at the same microsatellite locus in all samples was fed into the machine learning model as a feature value to obtain the AUC value of each microsatellite locus.

[0028] Preferably, the machine learning model uses a random forest model, and other suitable machine learning models may also be used.

[0029] As a preferred embodiment of the present invention, the number of the candidate microsatellite loci is at least 50. The greater the number, the more accurate the detection method can be.

[0030] As a preferred embodiment of the present invention, the average number of missing units is calculated based on the sequencing reads of the microsatellite loci and the number of repetitions of the repetitive units at the corresponding loci on the reference genome. The corresponding average number of missing units can be calculated for each microsatellite locus.

[0031] Specifically, the average number of missing units The calculation formula is:

[0032]

[0033] Where:

[0034] R is the number of repetitions of the repeat unit of the microsatellite locus on the reference genome.

[0035] N is the total number of sequencing reads covering the microsatellite locus in the sample (ie, the total number of reads).

[0036] i is a certain number of repetitions of the repeating unit of the microsatellite locus in the sample.

[0037] For example, the number of repetitions of repeat unit A at position chr1:10177680 of the reference genome is 12, and the number of repetitions of repeat unit A measured in the sample may be 9, 10, or 11.

[0038] n i It is the number of sequencing reads corresponding to a certain number of repetitions of the repeat unit of the microsatellite locus in the sample.

[0039] For example, for the above-mentioned repeating unit A, the measured number of sequencing reads corresponding to 9 repeat times is 5, the number of sequencing reads corresponding to 10 repeat times is 8, and the number of sequencing reads corresponding to 11 repeat times is 20.

[0040] i is an integer, 1≤i<R.

[0041] n i is an integer, n i ≤N.

[0042] Preferably, the sequencing depth of the microsatellite loci is at least 20×. The average number of missing units is calculated on the premise of meeting the sequencing depth.

[0043] Specifically, the position, repeat unit sequence, repeat unit length, and number of repeats of each microsatellite locus on the reference genome can be obtained using the scan function of MSIsensor1, or other software with similar functions, such as SSRHunter, GMATO, etc.

[0044] As a preferred embodiment of the present invention, the sequencing reads are obtained based on the second-generation sequencing technology, such as the second-generation sequencing technology and platform provided by Illumina, Life, Singular genomics, BGI, etc.

[0045] As a preferred embodiment of the present invention, the calculation of the average number of missing units at the candidate microsatellite loci in the sample to be tested is the same as above, requiring the sequencing depth to be at least 20×.

[0046] As a preferred embodiment of the present invention, the method for detecting the MSI status of the candidate microsatellite loci in the sample to be tested is:

[0047] The average number of missing units at each candidate microsatellite locus in the sample to be tested is compared with the threshold set for the corresponding locus. Loci exceeding the threshold are judged to be microsatellite unstable; otherwise, they are judged to be microsatellite stable.

[0048] Preferably, the threshold is set based on the average μ and standard deviation σ of the average number of missing units of candidate microsatellite loci obtained from multiple microsatellite stable samples. Each candidate microsatellite locus has a separate threshold.

[0049] Specifically, the threshold value may be set by the following method:

[0050] providing a background set, wherein the background set only includes samples that are stable for multiple microsatellites;

[0051] Obtaining sequencing reads of candidate microsatellite loci in the sample;

[0052] The average number of missing units at the candidate microsatellite loci in the samples was obtained;

[0053] The mean value μ and standard deviation σ of the average number of missing units for each candidate microsatellite locus in the sample were obtained, and the threshold of each candidate microsatellite locus was set accordingly.

[0054] The calculation of the average number of missing units at the candidate microsatellite loci in the sample is the same as above, requiring the sequencing depth to be at least 20×.

[0055] The threshold is set to the sum of μ (mean value) and 3σ (standard deviation), ie, μ+3σ.

[0056] As a preferred embodiment of the present invention, the method for evaluating the MSI status of the sample to be tested is:

[0057] The number of candidate microsatellite loci with microsatellite instability in the sample to be tested is obtained, and their proportion in the total number of candidate microsatellite loci (i.e., MSIscore) is counted. The ratio is compared with the set ratio threshold. If the ratio exceeds the threshold, the sample to be tested is judged to have microsatellite instability; otherwise, the sample is judged to be microsatellite stable.

[0058] Preferably, the ratio threshold is set based on the classification performance of the sample MSIscore in the training set. The sample MSIscore is determined based on the MSI status evaluation of the candidate microsatellite loci.

[0059] Specifically, the ratio threshold may be set by the following method:

[0060] Based on the training set, the average number of missing units for each candidate microsatellite locus in the sample was obtained and compared with the threshold set for the corresponding locus. Loci exceeding the threshold were judged to be microsatellite unstable; otherwise, they were judged to be microsatellite stable.

[0061] Obtain the number of candidate microsatellite loci with microsatellite instability in each sample, and calculate their proportion in the total number of candidate microsatellite loci, i.e., the sample MSIscore;

[0062] Based on the classification performance of the MSIscore of the samples in the training set, the ratio threshold is determined.

[0063] The ratio threshold is set to be below 25% and above 15%, and more preferably 20%.

[0064] In a second aspect, the present invention provides a device for detecting microsatellite instability, comprising:

[0065] An analysis module, used to analyze the sequencing reads of the candidate microsatellite loci in the sample to be tested;

[0066] A calculation module, used to calculate the average number of missing units at the candidate microsatellite loci in the sample to be tested;

[0067] A detection module, used to detect the MSI status of candidate microsatellite loci in the sample to be tested;

[0068] The evaluation module is used to evaluate the MSI status of the sample to be tested.

[0069] As a preferred embodiment of the present invention, one or more of the analysis module, calculation module, detection module and evaluation module can be optionally assembled to form a system with corresponding functions in the device.

[0070] As a preferred embodiment of the present invention, the device for detecting microsatellite instability may further include one or more of the following modules:

[0071] The data quality control module is used to select sequencing data of microsatellite loci that meet the quality control requirements.

[0072] For example, sequencing reads of microsatellite loci with a sequencing depth of at least 20× are selected.

[0073] The site screening module is used to select microsatellite sites that meet the requirements (i.e., candidate microsatellite sites).

[0074] For example, a microsatellite locus with a single nucleotide repeat having an auc value ≥ 0.65 is selected.

[0075] The threshold building module is used to set the threshold for each candidate microsatellite locus.

[0076] For example, the setting of the threshold is determined based on the average value μ and standard deviation σ of the average number of deletion units at the candidate microsatellite loci obtained from multiple microsatellite stable samples, preferably μ+3σ.

[0077] The ratio threshold building module is used to set the ratio threshold for evaluating the MSI status of the sample to be tested.

[0078] Beneficial effects of the invention:

[0079] The microsatellite instability detection method provided by the present invention uses the sequencing reads obtained by the second-generation sequencing technology to first calculate the average number of missing units of the candidate microsatellite loci in the sample to be tested, detect the MSI status of the candidate microsatellite loci based on this, and then evaluate the MSI status of the sample to be tested. The detection method of the present invention has a simple principle, simple calculation, high stability, good repeatability, and accurate and reliable detection results.

[0080] The detection method of the present invention calculates the average number of missing units at the candidate sites in the sample to be tested based on the information of the microsatellite sites on the reference genome. There is no need to use a reference sample (i.e., a control sample) or compare the polymorphic distribution of each site. The calculation process is simple, the clinical feasibility is good, the requirement for sequencing depth is low, and the detection cost and manpower can be greatly saved.

[0081] The device for detecting microsatellite instability provided by the present invention takes the average number of missing units at the microsatellite locus as the characteristic value, first detects the MSI status of the locus, and then evaluates the MSI status of the sample. The detection principle is simple, the calculation is simple, the detection result is accurate and reliable, and the consistency with the PCR gold standard is 100%. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] Figure 1 This is a demonstration of the classification performance of the test set in Example 1 of the present invention. The dotted line in the figure is the ratio threshold line. DETAILED DESCRIPTION

[0083] The technical scheme of the present invention will be clearly and completely described below in conjunction with specific embodiments, but those skilled in the art should understand that the embodiments described below are only used to illustrate the present invention and should not be regarded as limiting the scope of protection of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in the field without making creative work, such as modified or replaced embodiments, belong to the scope of protection of the present invention.

[0084] Unless otherwise specified, the experimental methods used in the following examples are all conventional methods; the materials and reagents used are all commercially available unless otherwise specified.

[0085] Example 1

[0086] This embodiment provides a method for detecting microsatellite instability based on second-generation sequencing, comprising the following steps:

[0087] 1. Selection of microsatellite loci

[0088] (1) Provide training set

[0089] From the total sample set, 19 MSI-H (human tissue samples) and 56 MSS (human tissue samples and human leukocyte samples) were randomly selected, with a total of 75 samples as the training set.

[0090] (2) Selection of sequencing data

[0091] Illumina second-generation sequencing technology was used to obtain sequencing reads for each microsatellite locus in each sample, and the sequencing data of microsatellite loci with a sequencing depth below 20× were screened and removed.

[0092] (3) Calculation of the average number of missing units

[0093] Use the scan function of MSIsensor1 to obtain the location, repeat unit sequence, repeat unit length, and repeat number of all microsatellite loci in the entire human genome, and calculate the average number of missing units for each microsatellite locus in each sample. The calculation formula is as follows:

[0094]

[0095] Where:

[0096] R is the number of repetitions of the repeating unit of the microsatellite locus on the reference genome;

[0097] N is the total number of sequencing reads covering the microsatellite locus in the sample;

[0098] i is a certain number of repetitions of the repeating unit of the microsatellite locus in the sample;

[0099] n i The number of sequencing reads corresponding to a certain number of repetitions of the repeating unit of the microsatellite locus in the sample;

[0100] i is an integer, 1≤i<R;

[0101] n i is an integer, n i ≤N.

[0102] (4) Obtaining the auc value

[0103] The average number of missing units at the same microsatellite locus in all samples was sent as the feature value into the machine learning random forest model to obtain the AUC value of each microsatellite locus.

[0104] (5) Determination of candidate microsatellite loci

[0105] The AUC values ​​of all microsatellite loci were sorted in order of size, and the top 100 microsatellite loci (AUC values ​​≥ 0.65) were selected as the locus set for subsequent MSI detection. The locus information is shown in Table 1.

[0106] Table 1 Candidate microsatellite loci

[0107]

[0108]

[0109]

[0110] 2. Setting the threshold of candidate microsatellite loci

[0111] Using 30 human leukocyte samples as the background set, WES sequencing data of each sample was obtained (sequencing depth was required to be at least 20×), and the average number of missing units of 100 candidate microsatellite loci in each sample (as shown in Table 1) was calculated according to the above formula, and then the average value μ and standard deviation σ of the average number of missing units of each microsatellite locus were calculated based on this, and the thresholds were set to μ+3σ. The thresholds of 100 candidate microsatellite loci are shown in Table 1 above.

[0112] 3. Setting the ratio threshold

[0113] Based on the training set, the average number of missing units at each candidate microsatellite locus in the 75 samples was compared with the threshold set for the corresponding locus. Loci exceeding the threshold were judged to be microsatellite unstable; otherwise, they were judged to be microsatellite stable.

[0114] Obtain the number of candidate microsatellite loci with microsatellite instability in each sample, and calculate their proportion in the total number of candidate microsatellite loci, i.e., the sample MSIscore;

[0115] Based on the classification performance of MSIscore of 75 samples in the training set, the ratio threshold was determined to be 20%.

[0116] 4. Acquisition of sequencing reads of candidate microsatellite loci in the sample to be tested

[0117] From the total sample set, 6 MSI-H (human tissue samples) and 19 MSS (human tissue samples and human leukocyte samples) were randomly selected, a total of 25 samples as the test set (different from the training set samples). Among them, the microsatellite status of the test set samples was based on the PCR gold standard.

[0118] Using Illumina second-generation sequencing technology, sequencing reads of 100 candidate microsatellite loci in the samples to be tested (a total of 25 cases) were obtained, and the sequencing data of microsatellite loci with a sequencing depth below 20× were screened and removed.

[0119] 5. Obtaining the average number of missing units at candidate microsatellite loci in the sample to be tested

[0120] The average number of missing units at each candidate microsatellite locus in the sample to be tested was calculated according to the above formula.

[0121] 6. Detection of MSI status of candidate microsatellite loci in the sample to be tested

[0122] The average number of missing units of each candidate microsatellite locus in the sample to be tested was compared with the threshold set for the corresponding locus (as shown in Table 1), and the loci exceeding the threshold were judged to be microsatellite instability; otherwise, they were judged to be microsatellite stable. The test results are shown in Table 2.

[0123] Table 2 Detection results of MSI status of candidate microsatellite loci in the tested samples

[0124]

[0125]

[0126] 7. Evaluation of MSI status of samples to be tested

[0127] The number of candidate microsatellite loci with microsatellite instability in each sample to be tested is obtained from Table 2, and its proportion in the total number of candidate microsatellite loci is calculated, that is, the sample MSIscore (as shown in Table 3). The ratio is compared with the set ratio threshold (20%). If the ratio exceeds the threshold, the sample to be tested is judged to have microsatellite instability; otherwise, it is judged to be microsatellite stable. The evaluation results are shown in Tables 3 and Figure 1 shown.

[0128] Table 3 Detection results of MSI status of samples to be tested

[0129]

[0130] From Table 3 and Figure 1 It can be seen that the classification accuracy of the detection method of the present invention for the test set is 100%. Since the microsatellite status of the test set samples is based on the PCR gold standard, the consistency between the detection method of the present invention and the PCR gold standard is 100%.

[0131] Experimental example

[0132] The stability of the detection method of the present invention was evaluated using standard products. The standard products used were purchased from Nanjing Kebai Biotechnology Co., Ltd. Each standard product includes a tissue DNA standard product and a paired DNA standard product, wherein the tissue DNA standard product has an MSI state or an MSS state, and the paired DNA standards are all in the MSS state. In this experimental example, the MSI state of the standard product was detected using the method of Example 1, and 3 replicates were set for each standard product. The test results are shown in Table 4.

[0133] Table 4 Stability evaluation results of the detection method of the present invention

[0134]

[0135] As can be seen from Table 4, the detection results of the present invention are completely consistent with the PCR results, and the results are stable between different repetitions.

[0136] The above comparison results show that the detection method of the present invention has high stability and good repeatability, and the detection results are accurate and reliable, with a consistency of 100% with the PCR gold standard.

[0137] Although the present invention has been described in detail above by means of general description, specific implementation methods and experimental examples, it is obvious to those skilled in the art that some modifications or improvements can be made to the present invention. Therefore, these modifications or improvements made without departing from the spirit of the present invention all fall within the scope of protection claimed by the present invention.

Claims

1. A method for detecting microsatellite instability based on second-generation sequencing, characterized in that: The method comprises the following steps: Obtaining sequencing reads of candidate microsatellite loci in the sample to be tested; Obtain the average number of missing units at the candidate microsatellite loci in the sample to be tested; Detect the MSI status of candidate microsatellite loci in the sample to be tested; Evaluate the MSI status of the sample to be tested; The average number of missing units is calculated based on the sequencing reads of the microsatellite loci and the number of repetitions of the repeat units at the corresponding loci on the reference genome; The calculation formula for the average number of missing units is: ； Where: R is the number of repetitions of the repeating unit of the microsatellite locus on the reference genome; N is the total number of sequencing reads covering the microsatellite locus in the sample; i is a certain number of repetitions of the repeating unit of the microsatellite locus in the sample; n i The number of sequencing reads corresponding to a certain number of repetitions of the repeating unit of the microsatellite locus in the sample; i is an integer, 1≤ i < R ; n i is an integer, n i ≤ N ; The method for detecting the MSI status of the candidate microsatellite loci in the sample to be tested is: The average number of missing units of each candidate microsatellite locus in the sample to be tested is compared with the threshold set for the corresponding locus, and the locus exceeding the threshold is judged to be microsatellite unstable; otherwise, it is judged to be microsatellite stable; The setting of the threshold is determined based on the average value μ and standard deviation σ of the average number of missing units of the candidate microsatellite loci obtained from multiple microsatellite stable samples; The threshold is set in the following way: providing a background set, wherein the background set only includes samples that are stable for multiple microsatellites; Obtaining sequencing reads of candidate microsatellite loci in the sample; The average number of missing units at the candidate microsatellite loci in the samples was obtained; The mean value μ and standard deviation σ of the average number of missing units of each candidate microsatellite locus in the sample were obtained, and the threshold of each candidate microsatellite locus was set as the sum of μ and 3σ; The ratio threshold is set below 25% and above 15%; The method for evaluating the MSI status of the sample to be tested is: The number of candidate microsatellite loci with microsatellite instability in the sample to be tested is obtained, and their proportion in the total number of candidate microsatellite loci is counted. The ratio is compared with the set ratio threshold. If the ratio exceeds the threshold, the sample to be tested is judged to have microsatellite instability; otherwise, the sample is judged to be microsatellite stable.

2. The method for detecting microsatellite instability based on second-generation sequencing according to claim 1, characterized in that: The ratio threshold is set to 20%.

3. The method for detecting microsatellite instability based on second generation sequencing according to claim 1, characterized in that: The sample to be tested is derived from a human or a non-human mammal; The sample to be tested is a tissue sample or a blood sample; The tissue sample is a tumor sample; The blood sample includes a whole blood sample and a plasma sample.

4. The method for detecting microsatellite instability based on second generation sequencing according to claim 1, characterized in that: The candidate microsatellite loci are selected by the following method: Providing a training set, wherein the training set includes a plurality of samples with microsatellite instability and a plurality of samples with microsatellite stability; Based on the training set, the classification performance of each microsatellite locus is obtained, wherein the classification performance is represented by an auc value, and microsatellite loci with an auc value ≥ 0.65 are selected as candidate microsatellite loci; The auc value is obtained by the following method: Based on the training set, the sequencing reads of each microsatellite locus in the sample are obtained; The average number of missing units per microsatellite locus in the sample was obtained; The average number of missing units at the same microsatellite locus in all samples was fed into the machine learning model as a feature value to obtain the AUC value of each microsatellite locus.

5. The method for detecting microsatellite instability based on second generation sequencing according to claim 1, characterized in that: The candidate microsatellite loci are one or more of a mononucleotide repeat site, a dinucleotide repeat site, and a trinucleotide repeat site.

6. The method for detecting microsatellite instability based on second generation sequencing according to claim 5, characterized in that: The candidate microsatellite locus is a single nucleotide repeat site.

7. A device dedicated to implementing the microsatellite instability detection method according to any one of claims 1 to 6, characterized in that: The device comprises: An analysis module, used to analyze the sequencing reads of the candidate microsatellite loci in the sample to be tested; A calculation module, used to calculate the average number of missing units at the candidate microsatellite loci in the sample to be tested; A detection module, used to detect the MSI status of candidate microsatellite loci in the sample to be tested; The evaluation module is used to evaluate the MSI status of the sample to be tested.

8. The device according to claim 7, characterized in that: The device also includes one or more of the following modules: Data quality control module, used to select sequencing data of microsatellite loci that meet quality control requirements; A site screening module is used to select microsatellite sites that meet the requirements; A threshold building module is used to set the threshold for each candidate microsatellite locus; The ratio threshold building module is used to set the ratio threshold for evaluating the MSI status of the sample to be tested.

Images