Sample cross contamination evaluation method, apparatus, device, and medium

Abstract

The present application belongs to the technical field of sample pollution evaluation, and discloses a sample cross-contamination evaluation method, device, equipment and medium, comprising: performing quality control on batch sequencing data, and performing mutation site detection and genetic marker typing according to a human reference genome sequence and a site interval file; according to the genetic marker typing result and the site interval file, the number of heterozygous genotype sites is counted and the sample heterozygosity is calculated, according to the sample heterozygosity and a preset threshold, the sample contamination state is determined; according to the copy number estimation of the sample, the allele frequency is diploid standardized correction; based on the allele frequency in the genetic marker typing result and the mutation site detection result, the cross-contamination relationship between each two samples is analyzed by using identification rules to determine the pollution source; the site for calculating the pollution proportion is selected, and the weighted average algorithm is used to calculate the pollution proportion. The sample pollution degree can be accurately estimated and the pollution source sample can be identified.

Description

Technical Field

[0001] This invention belongs to the field of sample contamination assessment technology, specifically relating to a method, apparatus, equipment, and medium for assessing sample cross-contamination. Background Technology

[0002] In the process of genome sequencing and genetic marker analysis, cross-contamination of samples can seriously affect the accuracy of test results. Therefore, the assessment of the degree of sample contamination is a key step in genome testing.

[0003] Currently, the commonly used methods for assessing sample contamination in the industry mainly rely on various specialized software tools, including VerifyBamID, ContEst, and Conpair.

[0004] VerifyBamID2, based on population allele frequencies and Haven's law, uses maximum likelihood estimation to determine contamination levels. It is highly dependent on allele frequencies from specific populations (e.g., the 1000 Genomes Study) and is inaccurate for rare variants or samples from non-reference populations. While it can identify potential contamination sources under certain patterns, its core strength lies in accurately estimating the contamination rate, making computation relatively time-consuming. Conpair operates on the same principle as VerifyBamID2; however, it is only designed for paired samples and cannot analyze contamination in unpaired samples, nor can it identify contamination sources. It can only accurately estimate severe cross-contamination levels exceeding 50%. ContEst, based on Bayesian methods, combines genotype information and sequencing data to calculate the maximum a posteriori probability estimate of cross-contamination levels, also making computation relatively time-consuming.

[0005] However, these software programs are all based on diploid models, which will severely distort the estimation of sample contamination levels in samples with copy number variations (CNV) or loss of heterozygosity (LOH), such as tumors. Summary of the Invention

[0006] The purpose of this invention is to provide a method, apparatus, device and storage medium for assessing cross-contamination of samples, which can accurately estimate the degree of sample contamination and identify the source of contamination.

[0007] The first aspect of this invention discloses a method for assessing cross-contamination of samples, comprising:

[0008] Quality control is performed on batch sequencing data, and mutation site detection and genetic marker typing are performed based on the human reference genome sequence and pre-defined site interval files;

[0009] Based on the genetic marker typing results and the locus interval file, the number of heterozygous genotype loci is counted and the sample heterozygosity rate is calculated. Based on the sample heterozygosity rate and a preset threshold, the sample contamination status is determined.

[0010] For genomic regions with copy number variations or loss of heterozygosity, diploid-normalized corrections are applied to allele frequencies in mutation site detection results based on copy number estimates of the samples.

[0011] Based on the allele frequencies in the genetic marker typing results and mutation site detection results, the cross-contamination relationship between each pair of samples is analyzed using preset identification rules to determine the source of contamination. The identification rules are used to find the best matching source of contamination sample in the same batch of samples based on the genotype and gene frequency of the preset site of the contamination receptor sample.

[0012] Loci for calculating contamination ratios were selected, and a weighted average algorithm was used to calculate the contamination ratios based on the corrected allele frequencies of the contaminated samples and the sequencing depth of the loci.

[0013] In some embodiments, identifying genomic regions with copy number variations and loss of heterozygosity includes:

[0014] The copy number of the sample is estimated, and the genomic regions with copy number variations are identified based on the copy number estimate; the genomic purity, ploidy and copy number of the sample are estimated, the heterozygosity loss status is determined, and the genomic regions with heterozygosity loss are screened.

[0015] In some embodiments, selecting the sites for calculating the contamination ratio includes:

[0016] Sites that meet the following conditions are used to calculate the pollution ratio: the genotype in the pollution source sample is 1 / 1 and the mutation frequency is 0.9, the genotype in the cross-contaminated sample is 0 / 0, and the ratio of the mutation frequency of the cross-contaminated sample to the mutation frequency of the pollution source sample is less than or equal to the empirical value.

[0017] In some embodiments, the heterozygosity of each sample is calculated based on the mutation results of SNP genetic markers in a preset number of healthy, uncontaminated population samples. The preset threshold is determined based on the heterozygosity using the third quartile and interquartile range statistical method. The preset threshold includes an uncontaminated threshold, a suspected contaminated threshold, and a confirmed contaminated threshold.

[0018] In some embodiments, based on the allele frequencies in the genetic marker typing results and mutation site detection results, a preset identification rule is used to analyze the cross-contamination relationship between each pair of samples to determine the source of contamination, including:

[0019] Combining the location of the mutated chromosome, allele information, genetic marker typing results, and the allele frequency, the samples are verified according to the identification rules. If the first sample meets all the verification rules, and all specific mutations in the first sample can be matched and detected in the second sample, then the first sample is determined to be the source of contamination for the second sample.

[0020] In some embodiments, after identifying the source of pollution, the method further includes:

[0021] Two samples were selected from all samples. Based on the allele frequencies in the genetic marker typing results and mutation site detection results of the two samples, the identification rules were used to filter out false positive mutation sites and count the proportion of effective contamination mutation sites until the proportion of effective contamination mutation sites for each pair of samples was counted.

[0022] A cross-contamination matrix between samples is constructed based on the proportion of all the effective contamination mutation sites.

[0023] A second aspect of the present invention discloses a sample cross-contamination assessment device, comprising:

[0024] The genetic marker genotyping module is used for quality control of sequencing data and for mutation site detection and genetic marker genotyping based on the human reference genome sequence and preset locus interval files.

[0025] The cross-contamination identification module is used to count the number of heterozygous genotype loci and calculate the sample heterozygosity rate based on the genetic marker typing results and the locus interval file, and to determine the sample contamination status based on the sample heterozygosity rate and a preset threshold.

[0026] The copy number correction module is used to perform diploid normalization correction on allele frequencies in mutation site detection results for genomic regions with copy number variations or loss of heterozygosity, based on the copy number estimation results of the samples.

[0027] The pollution source determination module is used to analyze the cross-contamination relationship between each two samples based on the allele frequency in the genetic marker typing results and mutation site detection results, and to determine the pollution source. The identification rules are used to find the best matching pollution source sample in the same batch of samples according to the genotype and gene frequency of the preset site of the pollution receptor sample.

[0028] The contamination ratio calculation module is used to select the loci for calculating the contamination ratio. It uses a weighted average algorithm to calculate the contamination ratio based on the corrected allele frequency of the contaminated sample and the sequencing depth of the loci.

[0029] A third aspect of the present invention discloses an electronic device, including a memory storing executable program code and a processor coupled to the memory; the processor calls the executable program code stored in the memory to execute the sample cross-contamination assessment method disclosed in the first aspect.

[0030] A fourth aspect of the present invention discloses a computer-readable storage medium storing a computer program, wherein the computer program causes a computer to perform the sample cross-contamination assessment method disclosed in the first aspect.

[0031] The beneficial effects of this invention are as follows: by automatically calculating the heterozygosity of batch samples and determining the sample contamination status based on a preset threshold, cross-contamination between samples from the same batch can be detected in batches; copy number estimation is corrected to avoid reporting low-frequency mutations introduced by contamination in cross-contaminated clinical samples, thus obtaining an accurate contamination ratio; based on genetic marker genotyping results and allele frequencies, the cross-contamination relationship between each pair of samples is analyzed using identification rules to automatically determine the source sample of contamination; and the contamination ratio is calculated based on the corrected allele frequencies and sequencing depth of the contaminated samples. This allows for accurate estimation of the degree of sample contamination and identification of the source sample. Attached Figure Description

[0032] The accompanying drawings illustrate specific examples of the technical solutions described in this invention and, together with the detailed embodiments, form part of the specification, serving to explain the technical solutions, principles, and effects of this invention.

[0033] Unless otherwise specified or defined, the same reference numerals in different figures represent the same or similar technical features, and different reference numerals may be used to represent the same or similar technical features.

[0034] Figure 1 This is a flowchart of a sample cross-contamination assessment method disclosed in an embodiment of the present invention;

[0035] Figure 2 A schematic diagram of the sample cross-contamination assessment device according to an embodiment of the present invention;

[0036] Figure 3 A schematic diagram of the structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0037] Unless otherwise specified or defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art. When combined with the technical solutions of the invention in a real-world scenario, all technical and scientific terms used herein may also have meanings corresponding to the purpose of achieving the technical solutions of the invention. The terms "first," "second," etc., used herein are merely for distinguishing names and do not represent a specific number or order. The term "and / or," as used herein, includes any and all combinations of one or more of the associated listed items.

[0038] It should be noted that when a component is considered "fixed" to another component, it can be directly fixed to the other component or there can be an intervening component; when a component is considered "connected" to another component, it can be directly connected to the other component or there can be an intervening component; when a component is considered "mounted" on another component, it can be directly mounted on the other component or there can be an intervening component; when a component is considered "placed" on another component, it can be directly placed on the other component or there can be an intervening component.

[0039] Unless otherwise specified or defined, the terms "described" or "the" as used herein refer to the technical features or technical content mentioned or described prior to the relevant section, which may be the same as or similar to the technical features or technical content mentioned herein. Furthermore, the terms "comprising" and "having," and any variations thereof, as used herein, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the steps or units listed, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such processes, methods, products, or apparatus.

[0040] This invention discloses a method for assessing sample cross-contamination, which can batch detect whether cross-contamination exists among samples from the same batch, automatically and accurately deduce specific contamination sources from mixed samples, and calculate the contamination ratio of cross-contaminated samples. This method can be implemented through computer programming. The executing entity of this method can be an electronic device such as a computer, laptop, or tablet, or a control chip embedded in an electronic device; this invention does not limit this. To facilitate understanding of this invention, specific embodiments will be described in more detail below with reference to the accompanying drawings.

[0041] like Figure 1 As shown, the method includes the following steps:

[0042] Step S100: Perform quality control on the batch sequencing data, and perform mutation site detection and genetic marker typing based on the human reference genome sequence and the preset site interval file;

[0043] The sequencing process is monitored in real time. After sequencing is completed and the sample is removed from the sequencing machine, the raw sequencing data file is automatically collected. The raw sequencing data file is parsed, split, and integrated to obtain batch sequencing data corresponding to the samples from the same batch. Then, quality control is performed on the batch sequencing data to comprehensively evaluate the quality of the raw sequencing data, the integrity of the sample DNA, and the degree of degradation damage to FFPE tissue samples. Qualified and valid sequencing data are screened, and low-quality, contaminated, and severely degraded unqualified data are removed to ensure the accuracy and reliability of subsequent sequence alignment, variant detection, and bioinformatics analysis results. Specifically: the BCL file of the sequencing data is split to obtain the raw fastq.gz file; data quality control is performed on FFPE samples (formalin-fixed paraffin-embedded tissue samples) in the sequencing data.

[0044] Then, the quality-controlled FASTQ data, based on the human reference genome sequence and a pre-defined BED (site interval file), was used with Sentieon TNscope software for mutation site detection and genetic marker genotyping. Specifically, 500 SNP sites with an allele frequency (AF) between 0.4 and 0.6, evenly distributed on the hg19 reference genome, were selected from the gnomAD v2.1.1 (Genome Aggregation Database) as known SNP genetic markers. hg19 was used as the human reference genome sequence, and the 500 SNP genetic markers were used as the BED. After quality control, the FASTQ data was aligned to the reference sequence using the Sentieon BWA MEM module. The Sentieon Dedup module was used to remove PCR duplicates, avoiding false positives due to amplification bias, resulting in a deduplicated BAM file. The Sentieon Realigner module was used to locally re-align reads in the obtained BAM file around indel regions, reducing false positive SNPs caused by mismatches, resulting in a corrected BAM file. The Sentieon QualCal module was used to systematically correct the base quality values ​​of the obtained BAM file, correcting any quality score deviations reported by the sequencer. Based on the obtained BAM file, the human reference genome sequence, and the pre-defined BED file, mutation site detection was performed using Sentieon TNscope. The detection results were then compiled for each SNP genetic marker within the pre-defined site intervals.

[0045] To verify the effect of cross-contamination, this embodiment also simulated sequencing of cross-contaminated samples with different contamination ratios to generate test data: Four uncontaminated peripheral blood samples A, B, C, D, E, and F from normal individuals were selected; 1% of the total input sample A was mixed with 99% of the input sample B to obtain sample C with 1% cross-contamination; 1% of the total input sample D was mixed with 99% of the input sample E to obtain sample F with 1% cross-contamination; similarly, samples with 2%, 3%, 5%, 10%, and 20% cross-contamination were obtained by mixing according to different ratios and then sequenced.

[0046] Step S200: Based on the genetic marker typing results and locus interval files, count the number of heterozygous genotype loci and calculate the sample heterozygosity rate. Based on the sample heterozygosity rate and a preset threshold, determine the sample contamination status.

[0047] Based on the genetic marker typing results and locus interval files, the number of valid loci with heterozygous genotype (0 / 1) in each sample was counted. The number of valid loci was divided by the total number of valid detection loci in the sample to calculate the sample heterozygosity rate.

[0048] The preset thresholds include an uncontaminated threshold, a suspected contamination threshold, and a confirmed contamination threshold. Based on the mutation results of SNP genetic markers from a preset number of healthy, uncontaminated human samples, the heterozygosity of each sample is calculated. The preset thresholds are determined using the third quartile and interquartile range (IQR) statistical methods based on the heterozygosity. For example, if 200 uncontaminated peripheral blood samples from healthy individuals are collected, the mutation results of 500 SNP genetic markers in these 200 samples are calculated, and the heterozygosity of each sample is calculated; based on the third quartile (Q3) and IQR of the heterozygosity data from these 200 samples, the uncontaminated threshold is calculated to be less than 0.4, the suspected contamination threshold is 0.4–0.6, and the confirmed contamination threshold is greater than 0.6.

[0049] Based on the simulation of cross-contaminated samples with different contamination ratios, the heterozygosity of the simulated cross-contaminated samples with different contamination ratios and the heterozygosity of actual clinical contaminated samples can be used to verify the rationality of the preset threshold.

[0050] The sample heterozygosity of each sample is compared with the uncontaminated threshold, the suspected contaminated threshold, and the confirmed contaminated threshold. If the sample heterozygosity is higher than the confirmed contaminated threshold, the sample is determined to be a contaminated sample; if the sample heterozygosity is between the suspected contaminated threshold and the confirmed contaminated threshold, the sample is determined to be a suspected contaminated sample; if the sample heterozygosity is lower than the uncontaminated threshold, the sample is determined to be an uncontaminated sample.

[0051] Step S300: For genomic regions with copy number variations or loss of heterozygosity, diploid-normalized correction is performed on the allele frequencies in the mutation site detection results based on the copy number estimation of the sample.

[0052] Copy number estimation for samples: For unpaired samples, CNVkit software is used to construct a "reference copy number baseline" for a group of normal human peripheral blood samples or normal tissue samples (normal / pooled control). CNVkit software divides the BED region into "target bins" and counts the number of reads (i.e., read depth) of each clinical sample falling into each target bin and antitarget bin; GC content correction and mappability correction are performed; for each bin, the ratio of its depth to the reference depth is calculated, and log2(ratio) is taken to obtain the log2 copy ratio; CNVkit uses CBS (Circular Binary Segmentation) to merge adjacent bins into segments with the same copy number state; each segment outputs a smoothed log2 ratio and copy number estimate. For paired samples, CNVkit software can be used directly to calculate the copy number estimate for the tumor sample BAM and the paired peripheral blood sample BAM.

[0053] The study identifies and segments genomic regions with copy number variations based on copy number estimation. Simultaneously, it assesses sample genomic purity, ploidy, and whole-genome copy number characteristics to determine loss of heterozygosity, further screening for regions with this feature. Specifically, for paired samples, ASCAT software is used to extract BAF (B allele frequency) and Log-R ratio (LRR) of SNP sites from tumor sample BAM and paired peripheral blood sample BAM, estimating sample purity, ploidy, copy number, and LOH. For unpaired samples, ASCAT software is used to extract BAF (B allele frequency) and Log-Log-R ratio (LRR) of SNP sites from tumor sample BAM, and common heterozygous SNPs are obtained from the 1000 Genomes public database to estimate sample purity, ploidy, copy number, and LOH. This integrated analysis of sample purity, ploidy, copy number, and loss of heterozygosity allows for precise screening of target genomic regions with copy number variations and loss of heterozygosity.

[0054] For the target genomic region, allele frequencies are diploid-normalized based on copy number estimation. This corrects for mutation frequency distortion caused by copy number abnormalities, restores the true mutation frequency of the locus under normal diploid conditions, eliminates interference from copy number changes in sample contamination determination, and ensures accurate contamination detection results.

[0055] Step S400: Based on the allele frequencies in the genetic marker typing results and mutation site detection results, the cross-contamination relationship between each pair of samples is analyzed using preset identification rules to determine the source of contamination;

[0056] Theoretically, real somatic mutations present in contaminant sample A may cross-contaminate sample B at a certain proportion (1%, 2%, 3%...20%), resulting in cross-contamination C. If contamination exists and the sequencing depth and mutation detection sensitivity are sufficiently high, then the mutations in the contaminant (sample A) can be observed in the cross-contaminated sample C. However, if the mutation frequency of a certain mutation in the contaminant (sample A) is less than 0.1, and it then contaminates sample B at an extremely low frequency (e.g., 1%), it may not be observable; this is normal. The higher the proportion of contamination, the greater the likelihood of observing it in the cross-contaminated sample C. When the contamination proportion reaches 1% and the sequencing depth is sufficiently high, almost all real somatic mutations present in contaminant sample A can be observed in the cross-contaminated sample C.

[0057] Therefore, this embodiment combines the location of the mutated chromosome, allele information, genetic marker genotyping results, and allele frequencies from mutation site detection results to verify the samples according to identification rules. If the first sample meets all verification rules, and all specific mutations in the first sample can be detected in the second sample, then the first sample is determined to be the source of contamination for the second sample. The identification rules are used to find the best matching source of contamination sample in the same batch of samples based on the genotype and gene frequency of the pre-set sites in the contamination receptor sample.

[0058] The identification rules in this embodiment include:

[0059] 1. For the same mutation, when the genotype is "1 / 1" in the source sample, the genotype of the cross-contaminated sample cannot be "0 / 0" or the genotype with a frequency greater than or equal to 0.9 is "0 / 1" mutation. This mutation is marked as FALSE.

[0060] 2. For the same mutation, if the genotype in the source sample is "0 / 0", the genotype in the cross-contaminated sample cannot be "1 / 1" or have a frequency of less than or equal to 0.1. This mutation is marked as FALSE.

[0061] 3. If the same mutation has a genotype of "0 / 1" and a frequency of less than 0.9 in the source sample, the cross-contaminated sample cannot have a mutation with a genotype of "1 / 1". This mutation is marked as FALSE.

[0062] 4. If the same mutation has a genotype of "0 / 1" and a frequency greater than 0.1 in the source sample, the cross-contaminated sample cannot have a mutation with a genotype of "0 / 0". This mutation is marked as FALSE.

[0063] 5. If the same mutation has a genotype of "0 / 1" and a frequency of less than or equal to 0.1 in the source sample, the cross-contaminated sample cannot have the same mutation with a frequency of less than 0.1 and a genotype of "0 / 1". This mutation is marked as FALSE.

[0064] 6. Mutations in other cases are marked as TRUE.

[0065] Combining the location information of the mutation site on the chromosome, the reference allele, the mutated allele, the sample genotype, and the mutation frequency, all mutation sites in the two groups of test samples are compared one by one. If all the mutation sites detected in sample A can be matched to the crossover sample B in terms of chromosomal location, allele type, and genotype characteristics, and all indicators of sample A meet the identification rules, then sample A is determined to be the source of contamination and sample B is the contaminated sample.

[0066] In some implementations, an inter-sample contamination matrix is ​​also generated to centrally display the cross-contamination associations between pairs of samples from the same batch (e.g., cross-contamination can be verified by the proportion of mutations with TRUE accounting for more than 70% of the total mutations). This visually identifies the correspondence between contamination source samples and contaminated samples, and presents the contamination incorporation ratio of each paired sample, enabling batch visualization monitoring of the contamination status of the entire batch of samples and rapid source tracing. The specific process for generating the inter-sample contamination matrix is ​​as follows: Two samples are selected from all samples. Based on the allele frequencies in the genetic marker genotyping results and mutation site detection results of the two samples, false positive mutation sites are filtered out using identification rules. The proportion of effective contamination mutation sites is counted until the proportion of effective contamination mutation sites for each pair of samples is counted. Then, an inter-sample cross-contamination matrix is ​​constructed based on the proportion of all effective contamination mutation sites.

[0067] Step S500: Select the loci for calculating the contamination ratio, and calculate the contamination ratio using a weighted average algorithm based on the corrected allele frequencies of the contaminated samples and the sequencing depth of the loci.

[0068] First, sites meeting the following criteria are selected for calculating the contamination ratio: genotype 1 / 1 and mutation frequency of 0.9 in the source sample; genotype 0 / 0 in the cross-contaminated sample; and the ratio of the mutation frequency of the cross-contaminated sample to the mutation frequency of the source sample is less than an empirical value. After selecting the sites for calculating the contamination ratio, allele frequencies corrected for copy number and loss of heterozygosity regions are combined with the effective sequencing depth data of the target site, introducing sequencing depth as a weighting factor. A weighted average algorithm is used to fuse and calculate the characteristic parameters of multiple effective genetic marker sites, weakening the interference of random sequencing errors at low-depth sites, improving the stability and accuracy of the calculation, and finally accurately calculating the actual cross-contamination ratio of the sample to be tested.

[0069] By performing copy number correction, the degree of sample contamination can be accurately estimated even when large-scale copy number changes such as CNV (copy number variation) and LOH (loss of heterozygosity) occur in the sample.

[0070] Finally, based on the above results—sample contamination status, contamination sources, cross-contamination matrix, contamination ratio, etc.—a relevant report is issued, and the test results are copied to the designated results directory and notified to relevant personnel via email.

[0071] In summary, this embodiment can automatically calculate the sample heterozygosity rate, the contamination matrix between batches of samples, and the cross-contamination ratio of samples in a batch. It features fast detection speed, low resource requirements, and high stability, greatly improving work efficiency. Moreover, it can batch detect whether there is cross-contamination between samples in the same batch. For cross-contaminated samples, it identifies the source of contamination and calculates the contamination ratio of cross-contaminated samples, avoiding the reporting of low-frequency mutations introduced by contamination in cross-contaminated clinical samples, which could lead to misdiagnosis.

[0072] like Figure 2 As shown, based on the above-mentioned sample cross-contamination assessment method, this embodiment of the invention discloses a sample cross-contamination assessment device, comprising:

[0073] The genetic marker typing module 600 is used for quality control of sequencing data and for mutation site detection and genetic marker typing based on the human reference genome sequence and preset site interval files;

[0074] The cross-contamination identification module 610 is used to count the number of heterozygous genotype loci and calculate the sample heterozygosity rate based on the genetic marker typing results and the locus interval file, and to determine the sample contamination status based on the sample heterozygosity rate and a preset threshold.

[0075] The copy number correction module 620 is used to perform diploid normalization correction on the allele frequencies in the mutation site detection results based on the copy number estimation results of the sample for genomic regions with copy number variations or loss of heterozygosity.

[0076] The pollution source identification module 630 is used to analyze the cross-contamination relationship between each two samples based on the allele frequency in the genetic marker typing results and mutation site detection results, and to identify the pollution source by using preset identification rules;

[0077] The contamination ratio calculation module 640 is used to select the loci for calculating the contamination ratio. The contamination ratio is calculated using a weighted average algorithm based on the corrected allele frequency of the contaminated sample and the sequencing depth of the loci.

[0078] In some embodiments, a cross-contamination matrix construction module is further included, which is used to select two samples from all samples, filter false positive mutation sites based on the allele frequencies in the genetic marker typing results and mutation site detection results of the two samples, and count the proportion of effective contamination mutation sites until the proportion of effective contamination mutation sites for each pair of samples has been counted; and construct a cross-contamination matrix between samples based on the proportion of all effective contamination mutation sites.

[0079] like Figure 3 As shown in the present invention, an electronic device is disclosed, including a memory 401 storing executable program code and a processor 402 coupled to the memory 401;

[0080] The processor 402 calls the executable program code stored in the memory 401 to execute the sample cross-contamination assessment method described in the above embodiments.

[0081] This invention also discloses a computer-readable storage medium storing a computer program that causes a computer to execute the sample cross-contamination assessment method described in the above embodiments.

[0082] The purpose of the above embodiments is to reproduce and derive the technical solution of the present invention by way of example, and to fully describe the technical solution, purpose and effect of the present invention. The purpose is to enable the public to have a more thorough and comprehensive understanding of the disclosure of the present invention, and not to limit the scope of protection of the present invention.

[0083] The above embodiments are not an exhaustive list based on the present invention, and there may be many other embodiments not listed. Any substitutions and improvements made without departing from the concept of the present invention are within the protection scope of the present invention.

Claims

1. A method for assessing cross-contamination of samples, characterized in that, include: Quality control is performed on batch sequencing data, and mutation site detection and genetic marker typing are performed based on the human reference genome sequence and pre-defined site interval files; Based on the genetic marker typing results and the locus interval file, the number of heterozygous genotype loci is counted and the sample heterozygosity rate is calculated. Based on the sample heterozygosity rate and a preset threshold, the sample contamination status is determined. For genomic regions with copy number variations or loss of heterozygosity, diploid-normalized corrections are applied to allele frequencies in mutation site detection results based on copy number estimates of the samples. Based on the allele frequencies in the genetic marker typing results and mutation site detection results, the cross-contamination relationship between each pair of samples is analyzed using preset identification rules to determine the source of contamination. The identification rules are used to find the best matching source of contamination sample in the same batch of samples based on the genotype and gene frequency of the preset site of the contamination receptor sample. Loci for calculating contamination ratios were selected, and a weighted average algorithm was used to calculate the contamination ratios based on the corrected allele frequencies of the contaminated samples and the sequencing depth of the loci.

2. The sample cross-contamination assessment method as described in claim 1, characterized in that, Genomic regions with copy number variations and loss of heterozygosity were identified, including: The copy number of the sample is estimated, and the genomic regions with copy number variations are identified based on the copy number estimate; the genomic purity, ploidy and copy number of the sample are estimated, the heterozygosity loss status is determined, and the genomic regions with heterozygosity loss are screened.

3. The sample cross-contamination assessment method as described in claim 1, characterized in that, The selection of sites for calculating the contamination ratio includes: Sites that meet the following conditions are used to calculate the pollution ratio: the genotype in the pollution source sample is 1 / 1 and the mutation frequency is 0.9, the genotype in the cross-contaminated sample is 0 / 0, and the ratio of the mutation frequency of the cross-contaminated sample to the mutation frequency of the pollution source sample is less than or equal to the empirical value.

4. The sample cross-contamination assessment method as described in claim 1, characterized in that, The heterozygosity of each sample is calculated based on the mutation results of SNP genetic markers in a preset number of healthy, uncontaminated human samples. The preset threshold is determined based on the heterozygosity using the third quartile and interquartile range statistical methods. The preset threshold includes an uncontaminated threshold, a suspected contaminated threshold, and a confirmed contaminated threshold.

5. The sample cross-contamination assessment method as described in claim 1, characterized in that, Based on allele frequencies from genetic marker typing and mutation site detection results, pre-defined identification rules were used to analyze cross-contamination relationships between every two samples to determine contamination sources, including: Combining the location of the mutated chromosome, allele information, genetic marker typing results, and the allele frequency, the samples are verified according to the identification rules. If the first sample meets all the verification rules, and all specific mutations in the first sample can be matched and detected in the second sample, then the first sample is determined to be the source of contamination for the second sample.

6. The sample cross-contamination assessment method as described in claim 5, characterized in that, After identifying the source of pollution, the following also includes: Two samples were selected from all samples. Based on the allele frequencies in the genetic marker typing results and mutation site detection results of the two samples, the identification rules were used to filter out false positive mutation sites and count the proportion of effective contamination mutation sites until the proportion of effective contamination mutation sites for each pair of samples was counted. A cross-contamination matrix between samples is constructed based on the proportion of all the effective contamination mutation sites.

7. A sample cross-contamination assessment device, characterized in that, include: The genetic marker genotyping module is used for quality control of sequencing data and for mutation site detection and genetic marker genotyping based on the human reference genome sequence and preset locus interval files. The cross-contamination identification module is used to count the number of heterozygous genotype loci and calculate the sample heterozygosity rate based on the genetic marker typing results and the locus interval file, and to determine the sample contamination status based on the sample heterozygosity rate and a preset threshold. The copy number correction module is used to perform diploid normalization correction on allele frequencies in mutation site detection results for genomic regions with copy number variations or loss of heterozygosity, based on the copy number estimation results of the samples. The pollution source determination module is used to analyze the cross-contamination relationship between each two samples based on the allele frequency in the genetic marker typing results and mutation site detection results, and to determine the pollution source. The identification rules are used to find the best matching pollution source sample in the same batch of samples according to the genotype and gene frequency of the preset site of the pollution receptor sample. The contamination ratio calculation module is used to select the loci for calculating the contamination ratio. It uses a weighted average algorithm to calculate the contamination ratio based on the corrected allele frequency of the contaminated sample and the sequencing depth of the loci.

8. The sample cross-contamination assessment device as described in claim 7, characterized in that, It also includes a cross-contamination matrix construction module, which is used to select two samples from all samples, filter false positive mutation sites based on the allele frequencies in the genetic marker typing results and mutation site detection results of the two samples, and count the proportion of effective contamination mutation sites until the proportion of effective contamination mutation sites for each pair of samples has been counted; and construct a cross-contamination matrix between samples based on the proportion of all the effective contamination mutation sites.

9. An electronic device, characterized in that, It includes a memory storing executable program code and a processor coupled to the memory; the processor calls the executable program code stored in the memory to execute the sample cross-contamination assessment method according to any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein the computer program causes a computer to perform the sample cross-contamination assessment method according to any one of claims 1 to 6.

Images