Method and device for distinguishing single gene pure and heterozygous deletion

Abstract

The invention provides a method and a device for detecting exogene deletion. The method comprises the following steps: S1, obtaining sequencing data; s2, carrying out partition on the hybridization capture region, and respectively obtaining a log2 value of the sequencing coverage degree in each bin partition; s3, calculating the median of the log2 values of all bin partitions obtained in the step S2 as a median value; s4, based on the log2 values obtained in the step S2, calculating the median of the log2 values of all bin partitions corresponding to the to-be-detected gene as a logR value; and S5, judging the deletion state of the to-be-detected gene of the to-be-detected sample according to the difference value between the logR and the median and the threshold value of the logR. According to the method, whether the target gene is subjected to heterozygous deletion or homozygous deletion can be accurately detected.

Description

Technical Field

[0001] This disclosure relates to the field of bioinformatics analysis technology, specifically to a method and apparatus for distinguishing between homozygous and heterozygous deletions of a single gene. Background Technology

[0002] Loss of heterozygosity (LOH) refers to the deletion of one allele from two alleles on a pair of homologous chromosomes, while the other allele remains intact. It is usually caused by chromosomal breakage and recombination, somatic mutations, or genomic instability, and is commonly seen in the monoallelic inactivation of tumor suppressor genes during tumorigenesis. Homozygous deletion, on the other hand, refers to the simultaneous deletion or complete loss of function of both alleles on a pair of homologous chromosomes, such as the homozygous mutation of CCR5Δ32 leading to the HIV-1 immunophenotype.

[0003] At the biological functional level, LOH events may lead to the penetrance of recessive pathogenic genes or the monoallelic inactivation of tumor driver genes, while homozygous deletions directly cause complete loss of gene function; for example, homozygous deletion of the TP53 gene is highly associated with cancer progression. In clinical applications, LOH analysis is used to assess cancer susceptibility (e.g., BRCA1 / 2 genes) and preimplantation genetic diagnosis (PGD), while homozygous deletion detection aids in the molecular subtyping of genetic diseases (e.g., cystic fibrosis) and the study of targeted therapy resistance mechanisms. Accurate identification of these two types of deletion events is of significant value for elucidating the genetic mechanisms of diseases and developing precision medicine strategies.

[0004] Methods for detecting loss of heterozygosity (LOH) and homozygosity (LOH) encompass the integrated application of high-throughput sequencing, molecular biology, and bioinformatics technologies. For LOH detection, a mid-depth whole-genome strategy based on next-generation sequencing (NGS) uses a sliding window to segment chromosomal regions, statistically analyzes the Z-score distribution of variant allele frequencies (VAFs), and calculates the ratio value of heterozygous sites on chromosome arms to achieve high-resolution LOH localization. The Hampel identifier method is then used to screen outliers to improve accuracy. Simultaneously, SNP microarray hybridization technology analyzes informative SNP sites in non-matched tumor tissue control samples, calculates the heterozygosity difference value Δ and the threshold ΔT, expanding the scope of clinical applications and reducing dependence on normal samples.

[0005] Chinese invention patent application No. 202211522110.5 discloses a method for detecting loss of heterozygosity. This method calculates the population heterozygosity of SNP sites and detects loss of heterozygosity based on the difference in population heterozygosity between control samples and tumor tissue samples. However, this method is not applicable to single tumor tissue samples.

[0006] Chinese invention patent application No. 201510050352.2 discloses a method for detecting loss of heterozygosity. This method is based on single-cell sequencing and uses the dbSNP database as a background. It determines the region's loss of heterozygosity by calculating the frequency of homozygous SNP sites in each detection region of each single cell. However, this method relies on single-cell sequencing technology. Summary of the Invention

[0007] This invention provides a method and apparatus for detecting homozygous and heterozygous deletions of genes in a single sample.

[0008] According to one aspect of this disclosure, a method for detecting gene deletion states is provided, the method comprising the following steps: S1, Obtain sequencing data: Extract DNA from the sample to be tested, perform hybridization capture and sequencing to obtain sequencing data; S2, the region captured by hybridization is divided into partitions, and the log2 value of the sequencing coverage in each bin partition is obtained; S3, calculate the median of the log2 values ​​of all bin partitions obtained in step S2 as the median value; S4, based on the log2 value obtained in step S2, calculate the median of the log2 values ​​of all bin partitions corresponding to the gene to be tested as the logR value; and S5. Based on the threshold of the difference between logR and median (logR - median) and the threshold of logR, determine the deletion status of the gene to be tested in the sample to be tested.

[0009] In some embodiments, in step S1, the sequencing data may be sequencing data of multiple genes obtained by those skilled in the art using conventional sequencing methods. For example, the sequencing data may be sequencing data of multiple genes including the gene to be tested, or whole-genome sequencing data. In a specific embodiment, the sequencing data may be sequencing data obtained using a conventional sequencing panel (e.g., the 1021 gene mutation detection kit).

[0010] In some embodiments, the hybridization capture includes hybridization capture using probes targeting multiple SNP sites of the gene to be tested and NGS detection reagents or kits. In some embodiments, the multiple SNP sites of the gene to be tested can be obtained through the following steps: (a1) Loci with frequencies between 0.05 and 0.95 in Asian populations were selected from the database; (a2) Filter out sites in simple repeat regions, centromere regions, and / or sites that cannot be accurately mapped to human genome regions during sequencing; and / or (a3) Select the SNP sites according to the start position and end position of the gene to be detected.

[0011] In some embodiments, the sequencing is high-throughput sequencing, preferably NGS sequencing.

[0012] In some embodiments, in step S2, obtain the log2 value of the sequencing reads coverage (depth) within each bin partition, that is, log2 = log2depth. Those skilled in the art should understand that the coverage (depth) is the coverage of the sequencing reads for the target region bin of the gene to be detected.

[0013] In some embodiments, in step S3, use the cumulative distribution function (CDF) of kernel density estimation (Kernel Density Estimation, KDE) in the R language package stats to calculate the median of the log2 values of all bin partitions obtained in step S2 as the median value.

[0014] In some embodiments, in step S5, determining the deletion status of the gene to be detected in the test sample includes: a) if logR < median - 0.2, it is determined that the gene to be detected has a deletion; b) if median - 0.2 ≤ logR ≤ median + 0.2, it is determined that the gene to be detected is wild type.

[0015] In some embodiments, the method further includes the following steps: S6, based on the sequencing data obtained in step S1, obtain the minor allele mutation frequency BAF value of the multiple SNP sites; S7, calculate the average BAF value BAF_mean of all SNP sites in germline mutations; and S8, according to BAF_mean and logR, determine whether the gene to be detected in the test sample is heterozygous deletion or homozygous deletion.

[0016] Those skilled in the art should understand that S6, S7, and S8 in this disclosure are only for identifying steps and do not constitute a limitation on the execution order between steps.

[0017] In some embodiments, step S6 includes: obtaining the mutation frequency VAF of the SNP site, and calculating the minor allele mutation frequency BAF of the SNP site according to the VAF value, BAF = min(VAF, 1 - VAF).

[0018] In some embodiments, step S7 includes: classifying the BAF values obtained in step S6 into two categories of somatic mutations and germline mutations, and calculating the average BAF value BAF_mean of all SNP sites in the germline mutations.

[0019] In some embodiments, the mutation frequency VAF of the SNP sites can be obtained by conventional methods in the art. For example, but not limited to, Bayesian statistical models, UMI correction techniques, cross-validation strategies, etc. can be used to calculate the mutation frequency VAF.

[0020] In some embodiments, if logR < median - 0.2, then in step S8, determining that the gene to be tested in the test sample is a heterozygous deletion or a homozygous deletion includes: i. If logR is less than -1.6, it is determined that the gene to be tested is a homozygous deletion; ii. If 0.35 ≤ BAF_mean ≤ 0.5, it is determined that the gene to be tested is a homozygous deletion; iii. If BAF_mean < 0.35 or BAF_mean > 0.5, it is determined that the gene to be tested is a heterozygous deletion.

[0021] According to another aspect of the present disclosure, there is provided a device for detecting the gene deletion status, the device including: A sequencing data acquisition module, configured to extract DNA of a test sample, perform hybridization capture and sequencing to obtain sequencing data; A log2 value determination module, configured to partition the hybridized capture region to obtain the log2 value of the sequencing coverage in each bin partition; A median value determination module, configured to calculate the median of the log2 values of all bin partitions obtained in step S2 as the median value; A logR value determination module, configured to calculate the median of the log2 values of all bin partitions corresponding to the gene to be tested based on the log2 values obtained in step S2 as the logR value; A BAF_mean determination module, configured to obtain the BAF value of the minor allele gene mutation frequency of the multiple SNP sites based on the sequencing data; calculate the average BAF value BAF_mean of all SNP sites in the germline mutations; and A judgment module, configured to judge the deletion status of the gene to be tested in the test sample according to the difference between logR and median (logR - median) and the threshold of logR.

[0022] In some embodiments, in the sequencing data acquisition module, the hybridization capture includes hybridization capture using probes targeting multiple SNP sites of the gene to be tested, as well as NGS detection reagents or kits; and / or The multiple SNP sites of the gene to be tested can be obtained through the following steps: (a1) Loci with frequencies between 0.05 and 0.95 in Asian populations were selected from the database; (a2) Filter out sites in simple repeat regions, centromere regions, and / or sites that cannot be accurately mapped to human genome regions during sequencing; and / or (a3) Select the SNP sites based on the start and end positions of the gene to be tested.

[0023] In some embodiments, in step S1, the sequencing data can be sequencing data of multiple genes obtained by those skilled in the art using conventional sequencing methods. For example, the sequencing data can be sequencing data of multiple genes including the gene to be tested, or whole genome sequencing data. In a specific embodiment, the sequencing data can be sequencing data obtained using a conventional sequencing panel (e.g., the 1021 gene mutation detection kit).

[0024] In some implementations, the log2 value determination module obtains the log2 value of the sequencing read coverage (depth) within each bin partition, i.e., log2 = log2depth. Those skilled in the art will understand that coverage (depth) refers to the coverage of the target region bin of the gene to be tested by the sequencing reads.

[0025] In some implementations, in the median value determination module, the cumulative distribution function (CDF) of the kernel density estimation (KDE) in the R language package stats is used to calculate the median of the log2 values ​​of all bin partitions obtained in step S2 as the median value.

[0026] In some implementations, the BAF_mean determination module includes: using K-means clustering to classify the obtained BAF values ​​into two categories, somatic mutations and embryonic mutations, and calculating the average BAF value BAF_mean of all SNP sites in the embryonic mutations.

[0027] In some implementations, the determination module determines the deletion status of the test gene in the test sample according to the following rules: 1) If logR <median - 0.2; i. If logR < -1.6, it indicates homozygous missing; ii. BAF_mean in the range [0.35, 0.5] indicates homozygous deletion; iii. BAF_mean is not in the range [0.35, 0.5], indicating heterozygous missing values; 2) If median - 0.2 ≤ logR ≤ median + 0.2, it is wild type.

[0028] According to another aspect of this disclosure, a system for detecting gene deletions is provided, the system comprising: a memory and a processor; The memory is used to store program instructions; The processor is used to invoke program instructions, which, when executed, are used to perform the method for detecting homozygous and heterozygous deletions of genes as described in this disclosure.

[0029] According to another aspect of this disclosure, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the method for detecting gene deletions described in this disclosure.

[0030] This disclosure has the following beneficial effects: This disclosure presents a method and apparatus for detecting gene deletions based on high-throughput sequencing technology. By screening SNP sites, the accuracy of BAF calculation is significantly improved. Furthermore, by jointly determining the logR value, median value, and BAF_mean value, the accuracy of detecting homozygous and heterozygous deletions is significantly improved. Attached Figure Description

[0031] Figure 1 The diagram shows the distribution of homozygous deletion logR and BAF_mean values. In Figure A, the gray sites represent the SNP sites selected in Example 1. The horizontal axis of Figure A represents the reference genome location, and the vertical axis represents the log2 value of all bin partitions corresponding to the MTAP gene. The red line in the figure represents the calculated LogR value. In Figure B, the horizontal axis represents the reference genome location information, and the vertical axis represents the VAF value. The blue line represents the calculated BAF_mean and (1-BAF_mean) values.

[0032] Figure 2The diagram shows the distribution of heterozygous deletion logR and BAF_mean values. In Figure A, the gray sites represent the SNP sites selected in Example 1. The horizontal axis of Figure A represents the reference genome location, and the vertical axis represents the log2 value of all bin partitions corresponding to the MTAP gene. The red line in the figure represents the calculated LogR value. In Figure B, the horizontal axis represents the reference genome location information, and the vertical axis represents the VAF value. The blue line represents the calculated BAF_mean and (1-BAF_mean) values. Detailed Implementation

[0033] Unless otherwise defined, all technical and scientific terms used in this invention have the same meaning as commonly used in the field to which this invention pertains. For the purposes of interpreting this specification, the following definitions will apply, and where appropriate, terms used in the singular will also include the plural forms, and vice versa.

[0034] definition The term "about" as used herein is as understood by one of ordinary skill in the art and varies within a certain range depending on the context in which it is used. If one of ordinary skill in the art is unfamiliar with the use of this term in the context in which it is used, "about" will mean a particular value plus or minus 10%.

[0035] The present invention will be further described in detail below through specific embodiments.

[0036] It should be understood that these embodiments are for illustrative purposes only and are not intended to limit the scope of the invention.

[0037] Unless otherwise specified, experimental methods in the following examples are generally performed under standard conditions or as recommended by the manufacturer. Unless otherwise specified, the materials and reagents used in these examples are commercially available. For reagents whose manufacturers are listed, similar products from other manufacturers are substituted.

[0038] Example Example 1: SNP site selection SNP loci with frequencies between 0.05 and 0.95 in Asian populations were selected from the gnomAD database. Loci in simple repeat regions and centromere regions were filtered out using the annotated rmsk database. Loci that could not be accurately mapped to human genome regions during sequencing were filtered out using the wgEncodeDukeMapabilityRegionsExcludable database. This resulted in the SNP locus set for this embodiment.

[0039] In this application, the gene used for threshold fitting is MTAP, and the gene used for verification is PTEN. Therefore, SNP sites of two genes were selected from the SNP site set obtained in the previous step by means of the gene start and end positions. The start and end positions of the MTAP gene are Chr9: 21795134-21865970, and the start and end positions of PTEN are Chr10: 89615694-89728532. Since there are relatively few PTEN gene sites in the site set, the range of expanded sites is Chr10: 89605694-89738532. The location information of the selected SNP sites is shown in Tables 1 and 2.

[0040] Table 1: Location information of selected PTEN gene SNP sites

[0041] Table 2: Location information of selected MTAP gene SNP sites

[0042] Example 2: Acquisition and Analysis of Sequencing Data 100 clinical tissue samples with heterozygous deletion of the MTAP gene verified by FISH, 100 clinical tissue samples with homozygous deletion of the MTAP gene verified by FISH, 100 clinical tissue samples with wild-type MTAP gene verified by FISH, and 30 plasma samples from healthy individuals were collected. DNA was extracted, and hybridization capture was performed using a gene mutation detection kit (GenePlus 1021 gene mutation detection kit) with probes targeting the SNPs in Table 2. NGS sequencing was then performed to obtain sequencing data.

[0043] The sequencing data were analyzed according to the following steps: 1) Use the mem2 algorithm of BWA software to map the sequencing data to the human reference genome (hg19) and remove redundant sequences; 2) Based on the hybridization capture region, multiple bin partitions are obtained; 3) Use the cnvkit software to calculate the log2 value in each bin partition: log2 = log2depth, where depth is the coverage of the target bin partition; 4) Use GPTK software to perform mutation detection, obtain the mutation frequency VAF of the SNP sites corresponding to the positions shown in Table 2, and calculate the minor allele mutation frequency BAF of the SNP sites based on the VAF value, i.e. BAF=min(VAF,1-VAF). 5) For the SNP sites in step 4, use the K-means clustering method to classify the sites into two categories, somatic mutations and embryonic mutations, based on the BAF value of the sites, and calculate the average BAF value BAF_mean of all SNP sites in the embryonic mutations. 6) Based on the log2 values ​​of each bin partition calculated in step 3), use the Cumulative Distribution Function (CDF) of the Kernel Density Estimation (KDE) in the R package stats to calculate the median of the log2 values ​​of all bin partitions as the median value. 7) Based on the log2 value obtained in step 3), take the median of the log2 values ​​of all bin partitions corresponding to the MTAP gene as the logR value.

[0044] Figure 1 and Figure 2 The example shows the results of statistical analysis of the log2 and VAF values ​​of all bin partitions corresponding to the MTAP gene in a single sample with heterozygous deletion of the MTAP gene (from the above 100 MTAP heterozygous deletion samples) and a single sample with homozygous deletion of the MTAP gene (from the above 100 MTAP homozygous deletion samples), respectively, using the ggplot2 toolkit in the R language.

[0045] Example 3: Threshold Fitting for Homozygous and Heterozygous Deletions The BAF_mean, logR, and median values ​​obtained in Example 2 were subjected to threshold fitting using the following method.

[0046] To reduce the impact of overall log2 value shift, we use the logR-median value to determine whether the MTAP gene is deleted. To fit a threshold, the logR-median value is divided into threshold intervals based on a gradient of 0.1 within the range [-1, 1]. The concordance rate between the interpretation results for wild-type and deleted (including homozygous or heterozygous deletions) MTAP genes and the FISH validation results is calculated when each value within the interval is used as the interpretation threshold. The logR-median value is used to determine whether a sample is deleted or wild-type, but it cannot distinguish between homozygous and heterozygous deletions. When the logR-median threshold is -0.2, the maximum concordance rate of 100% for wild-type and deleted genes is achieved; therefore, a logR-median threshold of -0.2 is selected.

[0047] The concordance rate is calculated as follows: Concordance Rate = (TP + TN) / (TP + TN + FP + FN). Where TP represents the number of true positive samples, TN represents the number of true negative samples, FP represents the number of false positive samples, and FN represents the number of false negative samples.

[0048] The BAF_mean value was divided into threshold intervals based on a gradient of 0.01 within the range [0, 0.5]. The consistency rate between the interpretation results of homozygous and heterozygous deletions of the MTAP gene and the FISH validation results was calculated when each value within the interval was used as the interpretation threshold. BAF_mean is used to determine whether a sample is homozygous or heterozygous. When the BAF_mean threshold is 0.35, the maximum consistency rate of 90% for homozygous and heterozygous deletion samples can be achieved; therefore, a BAF_mean threshold of 0.35 was chosen.

[0049] Furthermore, to avoid incorrect interpretation of homozygous deletions due to insufficient germline mutation sites, the logR value was used as one of the parameters for identifying homozygous deletions. The logR value was divided into threshold intervals of [-3, 0] with a gradient of 0.1. The consistency rate between the interpretation results of homozygous and heterozygous deletions of the MTAP gene and the FISH validation results was calculated for each value within the interval as the interpretation threshold. When the logR threshold was -1.6, the maximum consistency rate of 100% for heterozygous and homozygous deletions was achieved; therefore, the logR threshold was set to -1.6.

[0050] The final rules for determining homozygous and heterozygous deletions of genes based on a combination of logR, median, and BAF_mean are as follows: 1) If logR <median - 0.2; i. If logR is less than -1.6, it indicates homozygous missing data; ii. BAF_mean in the range [0.35, 0.5] indicates homozygous deletion; iii. BAF_mean is not in the range [0.35, 0.5], indicating heterozygous missing values; 2) If median - 0.2 ≤ logR ≤ median + 0.2, it is wild type.

[0051] Example 4: Validation based on clinical samples Fourteen clinical tissue samples with heterozygous deletion of the PTEN gene verified by FISH, one clinical tissue sample with homozygous deletion of the PTEN gene verified by FISH, and five clinical tissue samples with normal PTEN gene verified by FISH were selected. DNA extraction, hybridization capture, and NGS sequencing were performed to obtain sequencing data for the SNP sites listed in Table 1. BAF value, logR value, and median value were obtained according to the analysis method in Example 2. The deletion status of the PTEN gene in the samples was judged using the judgment criteria in Example 3. The results are shown in Table 3. Among them, the interpretation results of 20 samples were consistent with the FISH verification results, with an accuracy rate of 100%.

[0052] Table 3: Detection results of homozygous and heterozygous deletions of PTEN

[0053] The applicant declares that the above description is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Those skilled in the art should understand that any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention fall within the protection and disclosure scope of the present invention.

Claims

1. A method for detecting gene deletion states, characterized in that, The method includes the following steps: S1, Obtain sequencing data: Extract DNA from the sample to be tested, perform hybridization capture and sequencing to obtain sequencing data; S2, the region captured by hybridization is divided into partitions, and the log2 value of the sequencing coverage in each bin partition is obtained; S3, calculate the median of the log2 values ​​of all bin partitions obtained in step S2 as the median value; S4, based on the log2 value obtained in step S2, calculate the median of the log2 values ​​of all bin partitions corresponding to the gene to be tested as the logR value; and S5. Based on the difference between logR and median (logR - median) and the threshold of logR, determine the deletion status of the gene to be tested in the sample to be tested.

2. The method according to claim 1, characterized in that, The hybridization capture includes hybridization capture using probes targeting multiple SNP sites of the gene to be tested, as well as NGS detection reagents or kits; and / or The multiple SNP sites of the gene to be tested can be obtained through the following steps: (a1) Loci with frequencies between 0.05 and 0.95 in Asian populations were selected from the database; (a2) Filter out sites in simple repeat regions, centromere regions, and / or sites that cannot be accurately mapped to human genome regions during sequencing; and / or (a3) Select the SNP sites based on the start and end positions of the gene to be tested.

3. The method according to claim 1 or 2, characterized in that, In step S1, the sequencing data is sequencing data of multiple genes including the gene to be tested, or sequencing data of the whole genome.

4. The method according to any one of claims 1-3, characterized in that, In step S3, the cumulative distribution function of kernel density estimation in the R language package stats is used to obtain the median value of the log2 value of each bin partition obtained in step S2. and / or In step S5, determining the deletion status of the gene to be tested in the sample to be tested includes: a) if logR < median -0.2, then the gene to be tested is determined to be deleted; b) if median - 0.2 ≤ logR ≤ median + 0.2, then the gene to be tested is determined to be wild-type.

5. The method according to any one of claims 1-4, characterized in that, The method further includes the following steps: S6. Based on the sequencing data obtained in step S1, the minor allele mutation frequency (BAF) value of the multiple SNP sites is obtained. S7, calculate the mean BAF value (BAF_mean) for all SNP sites in embryonic cell mutations; and S8. Based on BAF_mean and logR, determine whether the gene to be tested in the sample is a heterozygous deletion or a homozygous deletion. Preferably, step S6 includes: obtaining the mutation frequency VAF of the SNP site, and calculating the minor allele mutation frequency BAF of the SNP site based on the VAF value, where BAF = min(VAF, 1-VAF). Step S7 includes: classifying the BAF values ​​obtained in step S6 into two categories: somatic cell mutations and embryonic cell mutations, and calculating the average BAF value BAF_mean of all SNP sites in embryonic cell mutations.

6. The method according to any one of claims 1-5, characterized in that, If logR < median - 0.2, then in step S8, determining whether the gene to be tested in the sample is heterozygous or homozygous deletion includes: i. If logR is less than -1.6, the gene to be tested is determined to be a homozygous deletion; ii. If 0.35 ≤ BAF_mean ≤ 0.5, then the gene to be tested is determined to be homozygous deletion; iii. If BAF_mean < 0.35 or BAF_mean > 0.5, the gene to be tested is determined to be heterozygous deletion.

7. A device for detecting gene deletion states, characterized in that, The device includes: The sequencing data acquisition module is set to extract DNA from the sample to be tested, perform hybridization capture and sequencing, and obtain sequencing data; The log2 value determination module is configured to partition the region captured by hybridization and obtain the log2 value of sequencing coverage in each bin partition. The median value determination module is set to use the median value as the median value of all bin partitions obtained in step S2. The logR value determination module is set to calculate the median of the log2 values ​​of all bin partitions corresponding to the gene to be tested as the logR value based on the log2 value obtained in step S2. The BAF_mean determination module is configured to, based on the sequencing data, obtain the minor allele mutation frequency (BAF) values ​​for the multiple SNP loci; calculate the mean BAF_mean value for all SNP loci in the embryonic cell mutation; and... The judgment module is configured to determine the deletion status of the test gene in the test sample based on the difference between logR and median (logR - median) and a threshold value of logR.

8. The apparatus according to claim 7, characterized in that, In the sequencing data acquisition module, multiple SNP sites of the gene to be tested can be obtained through the following steps: (a1) Loci with frequencies between 0.05 and 0.95 in Asian populations were selected from the database; (a2) Filter out sites in simple repeat regions, centromere regions, and / or sites that cannot be accurately mapped to human genome regions during sequencing; and (a3) Select the SNP sites based on the start and end positions of the gene to be tested; and / or In the median value determination module, the cumulative distribution function (CDF) of the kernel density estimation (KDE) in the R language package stats is used to obtain the median of the log2 value of each bin partition obtained in step S2 as the median value; and / or The BAF_mean determination module includes: using K-means clustering to classify the obtained BAF values ​​into two categories: somatic mutations and embryonic mutations; calculating the average BAF value BAF_mean for all SNP sites in embryonic mutations; and / or In the judgment module, the deletion status of the test gene in the test sample is determined according to the following rules: 1) If logR < median - 0.2; i. If logR < -1.6, then the gene to be tested is determined to be a homozygous deletion; ii. If BAF_mean is in the range of [0.35, 0.5], then the gene to be tested is determined to be homozygous deletion; iii. If BAF_mean is not in the range of [0.35, 0.5], then the gene to be tested is determined to be heterozygous deletion; 2) If median - 0.2 ≤ logR ≤ median + 0.2, then the gene to be tested is determined to be wild-type.

9. A system for detecting homozygous and heterozygous deletions of genes, the system comprising: Memory and processor; The memory is used to store program instructions; The processor is used to invoke program instructions, which, when executed, are used to perform the method for detecting gene deletions as described in any one of claims 1-6.

10. A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for detecting gene deletions as described in any one of claims 1-6.

Images