Data processing method and system for judging parent-child relationship

Abstract

The invention relates to the technical field of nucleic acid data processing, and discloses a data processing method and system for parent-child relationship determination, and the method comprises the steps: 1, receiving off-machine original data, and carrying out the splitting and cleaning; step 2, aligning the cleaned off-machine original data with a reference genome and removing repeated read segments; step 3, performing fixed-point typing on a preset single nucleotide polymorphism panel coordinate, and screening to obtain available site data; step 4, determining a transfer consistency weight based on the sequencing depth, the equipotential balance and the unique mask; 5, calculating a transfer consistency label, a prohibitive anti-homozygous label and a parent heterozygous single allelic shared proxy label; step 6, carrying out weighted aggregation on the labels and counting hard conflict counts; and step 7, performing joint judgment according to an aggregation result, outputting a parent-child relationship conclusion and generating a one-time check abstract. According to the invention, deterministic determination and result reviewable output of the parent-child relationship are realized.

Description

Technical Field

[0001] This invention belongs to the field of nucleic acid data processing technology, specifically relating to a data processing method and system for determining parentage. Background Technology

[0002] Paternity determination is a crucial application in nucleic acid testing and analysis, widely used in forensic identification, medical testing, and population management. Current paternity determination methods typically rely on single nucleotide polymorphism (SNP) sites or short tandem repeat (STDL) sequences, comparing the genotypes of the tested individual with those of suspected individuals, and then using statistical models or empirical rules to arrive at a conclusion. With the application of high-throughput sequencing technology in paternity testing, the process is gradually shifting from traditional site detection to comprehensive analysis based on sequencing data.

[0003] However, several shortcomings remain in existing parentage determination processes based on sequencing data. Firstly, raw sequencing data is susceptible to sequencing noise, amplification bias, and variations in sample management in terms of sample identification, indexing information, and data quality. Without a unified data cleaning and consistency verification mechanism, subsequent analyses can easily be based on unreliable data. Secondly, current techniques in genetic locus analysis often only perform simple screening or statistical analysis using fixed rules, failing to adequately consider quality factors such as sequencing depth, allelic balance, and genomic uniqueness to differentiate between different loci. This results in low-quality loci disproportionately impacting the overall determination results. Furthermore, some determination methods rely on probabilistic models or complex statistical inferences, lacking interpretability and verifiability, making it difficult to meet the requirements of forensic identification and other applications demanding certainty and traceability. Summary of the Invention

[0004] This invention provides a data processing method and system for determining parentage, which solves the technical problems in related technologies when determining parentage based on nucleic acid sequencing data, such as insufficient locus quality control, lack of unified quantitative rules for genetic consistency evaluation, and poor interpretability and verifiability of determination results.

[0005] This invention provides a data processing method for determining parentage, comprising the following steps: Step 1: Receive raw data from the paternity testing equipment and identification information of the individuals to be tested and suspected individuals, generate a sample list, and split and clean the raw data to obtain cleaned raw data. Step 2: Align the cleaned raw data with the reference genome and remove duplicate reads to obtain deduplicated alignment data; Step 3: Perform site-specific typing on the deduplication alignment data on the preset single nucleotide polymorphism panel coordinates to obtain diallelic genotypes, and record sequencing depth, allelic balance and uniqueness mask. Useful site data are obtained by screening according to a fixed threshold. Step 4: Determine the transit consistency weights based on sequencing depth, allelic balance, and uniqueness mask in the available site data, forming a weight vector and total weights; Step 5: Calculate the transitive consistency label, the prohibitive anti-homozygous label, and the parent heterozygous monoalle shared surrogate label based on available site data; Step 6: Based on the weight vector, perform weighted aggregation of transitive consistency labels, prohibitive anti-homozygous labels, and parental heterozygous mono-equal shared proxy labels to obtain the transitive consistency ratio, prohibitive mode ratio, and parental heterozygous mono-equal shared proxy ratio, and count hard conflicts. Step 7: Jointly determine the proportion of transitive consistency, the proportion of prohibited patterns, the proportion of parental heterozygous mono-equivalent shared proxy, and the hard conflict count, output the parent-child relationship conclusion, and generate a one-time verification summary.

[0006] This invention provides a data processing system for determining parentage, comprising: The sample data processing module is used to receive the raw data from the paternity testing equipment and the identification information of the individuals to be tested and the suspected individuals, generate a sample list, and split and clean the raw data to obtain the cleaned raw data. The sequence alignment and deduplication module is used to align the cleaned raw data after processing with the reference genome and remove duplicate reads to obtain deduplicated alignment data. The panel typing and screening module is used to perform site-specific typing on the deduplication alignment data on the preset single nucleotide polymorphism panel coordinates to obtain diallelic genotypes, and record sequencing depth, allelic balance and uniqueness mask, and screen usable site data according to a fixed threshold. The site weight calculation module is used to determine the transit consistency weight based on sequencing depth, allelic balance, and uniqueness mask in the available site data, forming a weight vector and total weight. The genetic relationship marker module is used to calculate the transmission consistency tag, the prohibition against homozygosity tag, and the parental heterozygous monoalle sharing surrogate tag based on available locus data; The relation index aggregation module is used to perform weighted aggregation of transitive consistency labels, prohibitive anti-homozygous labels, and parental heterozygous mono-equal shared surrogate labels based on weight vectors, to obtain the transitive consistency ratio, prohibitive mode ratio, and parental heterozygous mono-equal shared surrogate ratio, and to count hard conflicts. The relationship determination and verification module is used to jointly determine the proportion of transitive consistency, the proportion of prohibited modes, the proportion of parent-child heterozygous single-equal shared proxy, and the hard conflict count, output the parent-child relationship conclusion, and generate a one-time verification summary.

[0007] The beneficial effects of this invention are as follows: The data processing method and system for determining parentage provided by this invention address the problems of large data quality differences, difficulty in uniformly assessing site reliability, and insufficient verifiability of determination results in the process of determining parentage using nucleic acid sequencing data. It establishes a systematic data processing and determination process. By performing consistency verification, splitting and cleaning, alignment and deduplication on the raw data, and site-specific typing and screening of available sites based on single nucleotide polymorphism panels, it effectively ensures that the nucleic acid sites involved in the determination have a stable data quality foundation. Based on this, a site weighting mechanism based on sequencing depth, allelic balance, and genome uniqueness is introduced to differentiate the influence of different genetic sites in parentage analysis, avoiding disproportionate interference from low-quality sites on the overall determination results. Furthermore, by weighted aggregation and joint determination of consistency tags, prohibitive anti-homozygous tags, and parental heterozygous monoallelic shared proxy tags, deterministic determination of parentage is achieved without relying on probabilistic models. By generating a one-time verification summary, the determination results and their key calculation basis are solidified and encapsulated, enhancing the traceability and verifiability of the determination process. Attached Figure Description

[0008] Figure 1 This is a flowchart of the data processing method for determining parent-child relationships according to the present invention. Detailed Implementation

[0009] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that these embodiments are discussed only to enable those skilled in the art to better understand and implement the subject matter described herein, and changes may be made to the function and arrangement of the elements discussed without departing from the scope of this specification. Various processes or components may be omitted, substituted, or added as needed in the examples. Furthermore, features described in some examples may be combined in other examples.

[0010] like Figure 1 As shown, the data processing method for determining parentage includes the following steps: Step 1: Receive raw data from the paternity testing equipment and identification information of the individuals to be tested and suspected individuals, generate a sample list, and split and clean the raw data to obtain cleaned raw data. Step 2: Align the cleaned raw data with the reference genome and remove duplicate reads to obtain deduplicated alignment data; Step 3: Perform site-specific typing on the deduplication alignment data on the preset single nucleotide polymorphism panel coordinates to obtain diallelic genotypes, and record sequencing depth, allelic balance and uniqueness mask. Useful site data are obtained by screening according to a fixed threshold. Step 4: Determine the transit consistency weights based on sequencing depth, allelic balance, and uniqueness mask in the available site data, forming a weight vector and total weights; Step 5: Calculate the transitive consistency label, the prohibitive anti-homozygous label, and the parent heterozygous monoalle shared surrogate label based on available site data; Step 6: Based on the weight vector, perform weighted aggregation of transitive consistency labels, prohibitive anti-homozygous labels, and parental heterozygous mono-equal shared proxy labels to obtain the transitive consistency ratio, prohibitive mode ratio, and parental heterozygous mono-equal shared proxy ratio, and count hard conflicts. Step 7: Jointly determine the proportion of transitive consistency, the proportion of prohibited patterns, the proportion of parental heterozygous mono-equivalent shared proxy, and the hard conflict count, output the parent-child relationship conclusion, and generate a one-time verification summary.

[0011] In one embodiment of the present invention, the system receives raw data from the machine and identification information of individuals to be inspected and suspected individuals, generates a sample list, and splits and cleans the raw data from the machine to obtain cleaned raw data from the machine, including: Step 11: Receive the raw data output from the paternity testing equipment, and the identification information of the individuals to be tested and the suspected individuals corresponding to the raw data. The raw data is the raw sequence data obtained through nucleic acid sequencing technology, usually stored in the form of a sequence file containing index information; the identification information includes at least the sample identifier, the individual to be tested identifier, and the suspected individual identifier, used to characterize the correspondence between the sample and the individual.

[0012] After receiving the raw data and identification information, a consistency check is performed on both. This consistency check determines whether the sample identifier in the identification information can form a one-to-one mapping with the index information carried in the raw data, thus ensuring that each sample identifier corresponds to a unique index and that there are no missing or duplicate sample identifiers. Simultaneously, the consistency check further confirms that the identifier of the individual to be tested and the identifier of the suspected individual are each associated with at least one sample identifier, ensuring that subsequent parentage determination has a complete nucleic acid data foundation. Through this consistency check, anomalies such as sample mismatch and identifier confusion can be eliminated in the initial stage of data processing.

[0013] Step 12: After the consistency verification passes, a sample list is generated based on the verified identification information of the individuals to be inspected and the suspected individuals. The sample list is a structured data object describing the relationship between samples and data, recording at least the correspondence between sample identifiers, index information, individual identifiers to be inspected, and suspected individual identifiers, as well as the storage location information of the original offline data. By generating the sample list, the original offline data, comparison data, classification results, and judgment results involved in subsequent steps can be accurately traced through sample identifiers and individual identifiers, thus forming a data index foundation that runs through the entire process.

[0014] Step 13: After generating the sample list, the raw data is split according to the correspondence between sample identifiers and index information in the sample list. Specifically, the raw data is divided into data unit sets organized by sample identifiers based on the index information, so that each data unit contains only the nucleic acid sequence data of the corresponding sample. After the data splitting is completed, adapter removal, low-quality base cleaning, and abnormal read removal are performed sequentially on the data unit sets. Adapter removal is used to remove adapter sequences introduced during sequencing; low-quality base cleaning is used to remove base sequences that do not meet the sequencing quality requirements; and abnormal read removal is used to exclude reads with abnormal length or quality. Through the above processing, the cleaned raw data is obtained.

[0015] Through the above steps, nucleic acid sequencing data can be standardized and quality controlled in the early stages of parentage determination, ensuring that the correspondence between samples and individuals is accurate and reducing the interference caused by sample mismatch, low-quality sequences or abnormal reads on subsequent genetic locus analysis, thereby providing a reliable data foundation for parentage determination based on nucleic acid sequences.

[0016] In one embodiment of the present invention, the cleaned raw data is aligned with a reference genome and duplicate reads are removed to obtain deduplicated alignment data, including: After obtaining the cleaned raw sequencing data, the cleaned raw data is aligned and matched with a pre-selected reference genome. The reference genome is a standard genome sequence used for nucleic acid sequence localization and variation analysis, which provides a unified coordinate reference system for sequencing reads. Through alignment and matching, a mapping relationship is established between each read in the cleaned raw sequencing data and its corresponding position in the reference genome, thereby forming alignment data.

[0017] After obtaining the alignment data, it undergoes further screening to retain only reads that clearly correspond to a specific location in the reference genome as unique comparable reads. A read that clearly corresponds to a specific location in the reference genome means that the read has only one definite alignment position on the reference genome, and there are no multiple equivalent matching positions, thus avoiding location ambiguity introduced by repetitive sequences or multiple alignments. By retaining unique comparable reads, it ensures that subsequent genetic locus analysis is based on clearly defined and reproducibly located nucleic acid sequence data.

[0018] After identifying unique comparable reads, duplicate read removal is performed on these reads. These duplicate reads primarily originate from repetitive amplification products generated during PCR amplification, exhibiting high consistency in sequence content and alignment position. Without removal, duplicate reads would artificially amplify the read count at the same genetic locus, thus affecting the accurate calculation of subsequent parameters such as sequencing depth and allelic balance. Therefore, this embodiment identifies and removes the duplicate reads generated during PCR amplification to obtain alignment data free from duplication effects.

[0019] After removing duplicate reads, the alignment data is subjected to buck storage processing. Buck storage is used to compress or re-encode the alignment data without altering the alignment results and sequence information integrity, thereby reducing data storage footprint and improving the efficiency of subsequent data reading and processing. Through these processes, the deduplicated alignment data is finally obtained.

[0020] By following the steps described above, redundant information introduced by sequencing and amplification processes can be effectively eliminated in the data processing workflow for parentage determination. This ensures that the nucleic acid sequence data used for genetic locus analysis has a clear genomic location and a true read count basis, thereby providing reliable data input for subsequent site-specific typing, locus quality assessment, and parentage determination on single nucleotide polymorphism panels.

[0021] In one embodiment of the present invention, the deduplicated alignment data is genotyped on a preset single nucleotide polymorphism panel coordinate system to obtain diallelic genotypes, and sequencing depth, allelic balance, and uniqueness mask are recorded. Usable site data are obtained by screening according to a fixed threshold, including: Step 21: Based on the deduplicated alignment data and limited to the preset single nucleotide polymorphism (SNP) panel coordinates, extract the read sets covered at each panel coordinate for both the tested individual and the suspected individual. The preset SNP panel coordinates are a pre-selected set of genomic coordinate positions, corresponding to genetic loci significant in paternity determination, used to limit subsequent analysis to the specified nucleotide loci range. The read set refers to the set of nucleotide sequence reads aligned at the corresponding panel coordinate positions, reflecting the sequencing coverage of that genetic locus. By limiting the extraction of read sets to the panel coordinate range, interference from irrelevant genomic regions can be avoided in subsequent genetic analysis.

[0022] Step 22: After obtaining the read sets at each panel coordinate, perform base calling processing on the read sets. Base calling refers to determining the base type at a given position based on the base information of the read at the panel coordinate. During base calling, the count of reference base reads and the count of variant base reads at each panel coordinate are counted. Reference bases are the base types of the reference genome at that coordinate, and variant bases are base types different from the reference bases. Based on the reference base read count and the variant base read count, the bisellic genotypes of the tested individual and the suspected individual at each panel coordinate are determined. The bisellic genotype is used to characterize the base composition corresponding to the two alleles at that genetic locus and is crucial basic data for subsequent parentage determination.

[0023] Step 23: After determining the biselenic genotype, record the parameters related to the site quality for each panel coordinate, including sequencing depth, allelic balance value, and uniqueness mask. Sequencing depth is the sum of the reference read count and the variant read count, reflecting the sequencing coverage intensity at that panel coordinate. The allelic balance value reflects the proportional relationship between the two alleles at the read level. It is calculated by using the minor allele read count as the numerator and the sum of the reference and variant read counts as the denominator, and then calculating the ratio of the numerator to the denominator. The minor allele read count refers to the smaller of the reference and variant read counts. The allelic balance value can be used to determine whether there is a significant shift in the sequencing results for that site. The uniqueness mask identifies whether the panel coordinate is located in a genomic repetitive region. It is determined according to a preset uniqueness mask rule: 0 is recorded when the panel coordinate is in a repetitive region, and 1 is recorded when it is outside a repetitive region.

[0024] Step 24: After recording the above parameters, the panel coordinates are filtered according to fixed thresholds. These fixed thresholds include: sequencing depth not lower than the minimum sequencing depth, allelic balance value within a preset reasonable allelic range, and a uniqueness mask of 1. The minimum sequencing depth is preferably 15, and the preset reasonable allelic range is 0.25 to 0.75. A panel coordinate is only identified as a usable site if it simultaneously meets all the above fixed threshold conditions. The final filtered usable site data includes panel coordinates that meet the fixed thresholds, their corresponding biselic genotypes, sequencing depths, allelic balance values, and uniqueness masks.

[0025] The panel-based site-specific genotyping and available site screening process in this embodiment can effectively exclude genetic loci with insufficient sequencing coverage, abnormal allelic ratios, or located in repetitive regions during parentage determination, retaining only reliable and discriminative nucleic acid loci for subsequent genetic consistency analysis. This single nucleotide polymorphism panel-based site-specific genotyping and site quality screening mechanism helps improve the stability and verifiability of parentage determination results.

[0026] In one embodiment of the present invention, a transitivity consistency weight is determined based on sequencing depth, allelic balance, and uniqueness mask in available site data, forming a weight vector and a total weight, including: Step 31: Based on available site data, obtain the sequencing depth, allelic balance value, and uniqueness mask for each available site for both the individual to be tested and the suspected individual, and simultaneously read the minimum sequencing depth. By aggregating the above information, site-level input data indexed by available sites is formed, thus providing a unified data structure for subsequent site-by-site weight calculations.

[0027] Step 32: Calculate the sequencing depth normalization factor and allelic balance penalty factor for each available site for both the individual to be tested and the suspected individual. The sequencing depth normalization factor is calculated by comparing the sequencing depth of the available site with the minimum sequencing depth standard, reflecting the sufficiency of sequencing coverage at that site relative to the minimum requirement. By introducing the sequencing depth normalization factor, the impact of low-coverage sites on paternity determination can be reduced in subsequent weight calculations.

[0028] The absolute value of the difference between the available site's isoplegic balance value and the preset equivalent benchmark value is taken to obtain the isoplegic balance deviation value. Then, the ratio of the isoplegic balance deviation value to the preset equivalent benchmark value is calculated to obtain the deviation ratio. Finally, the isoplegic balance penalty factor is obtained by subtracting the deviation ratio from 1. Through this calculation method, the closer the isoplegic balance value is to the preset equivalent benchmark value, the larger the isoplegic balance penalty factor, thus reflecting the reliability of the site in terms of isoplegic ratio.

[0029] After calculating the sequencing depth normalization factor and the allelic balance penalty factor, for each available locus, the sequencing depth normalization factor of the tested individual and the suspected individual are multiplied together, and the allelic balance penalty factor of the tested individual and the suspected individual is multiplied together. Then, the result of these multiplications is multiplied by the uniqueness mask corresponding to the available locus to obtain the transitivity weight of that available locus. The transitivity weight is used to characterize the reliability of each available locus in subsequent parentage determination and is an important intermediate result for introducing nucleic acid sequencing quality information into genetic consistency analysis. The uniqueness mask is used to identify whether the available locus is located in a genomic repetitive region. When the uniqueness mask indicates that the locus is unavailable, its transitivity weight is suppressed, thereby avoiding interference from repetitive regions in genetic consistency analysis.

[0030] Step 33: After obtaining the transitivity weights of each available locus, the transitivity weights are aggregated according to the order of available loci to form a weight vector. The transitivity weights in the weight vector are then summed to obtain the total weight. The weight vector and the total weight serve as the basic inputs for the subsequent weighted aggregation of transitivity labels, prohibitive anti-homozygotic labels, and parental heterozygous monoallelic surrogate labels, used to construct a locus-quality-weighted parentage determination index.

[0031] The transitivity weighting calculation process in this embodiment comprehensively considers multiple quality factors in parentage determination, such as nucleic acid sequencing coverage, allele ratio stability, and genomic region uniqueness. Differential weights are assigned to different genetic loci, thus avoiding the disproportionate impact of low-quality loci on the overall determination results. This approach, which constructs weight vectors based on available locus data and performs subsequent weighted analysis, helps improve the stability, interpretability, and verifiability of parentage determination.

[0032] In one embodiment of the present invention, calculating a transitive consistency label, a prohibitive anti-homozygous label, and a parental heterozygous monoalle shared surrogate label based on available site data includes: Step 41: Based on available locus data, obtain the genotypes of the individual to be tested and the suspected individual for each available locus. The genotype is the biselenate determined at the corresponding available locus, used to describe the base composition of the two alleles at that genetic locus. Subsequently, according to a preset allele set rule, map the genotype of the individual to be tested to the allele set of the individual to be tested, and map the genotype of the suspected individual to the allele set of the suspected individual. The allele set rule is as follows: when the genotype is AA, it corresponds to a single allele; when the genotype is AB, it corresponds to two alleles; when the genotype is BB, it also corresponds to a single allele. By converting the genotype into an allele set, subsequent genetic relationship determination can be unified into a relationship judgment between sets, thereby simplifying the determination logic and avoiding ambiguity.

[0033] Step 42: After constructing the allele sets, for each available locus, a transmissibility label is calculated based on the overlap between the allele sets of the individual to be tested and those of the suspected individual. Specifically, when there is overlap between the allele sets of the individual to be tested and those of the suspected individual, the transmissibility label for that available locus is marked as 1; when there is no overlap, the transmissibility label is marked as 0. The transmissibility label reflects whether the individual to be tested can obtain at least one allele from the suspected individual at that locus, given that the genotype of the other parent is unknown.

[0034] While calculating the transmission consistency label, a prohibitive anti-homozygotic label is determined. When the genotype of the suspected individual is AA or BB, and the genotype of the individual to be tested is also AA or BB, and their corresponding allele sets do not overlap, the prohibitive anti-homozygotic label for that available locus is marked as 1; otherwise, the prohibitive anti-homozygotic label is marked as 0. The prohibitive anti-homozygotic label is used to identify genetic incompatibility at that available locus, which has clear exclusionary significance in determining parentage.

[0035] Step 43: For each available locus, calculate the parental heterozygous monoalleic surrogate tag. When the genotype of the suspected individual is AB, and the overlap between the allele set of the individual to be tested and the allele set of the suspected individual contains only one allele, mark the parental heterozygous monoalleic surrogate tag for that available locus as 1; otherwise, mark the parental heterozygous monoalleic surrogate tag as 0. The parental heterozygous monoalleic surrogate tag is used to characterize the genetic relationship feature that, when the suspected individual has a heterozygous genotype, the individual to be tested shares only one allele at that locus. This feature can serve as auxiliary discriminant information for subsequent overall genetic relationship analysis.

[0036] Through the tag calculation process in this embodiment, the allele transmission relationship of each available genetic locus can be characterized site by site in the determination of parentage using deterministic rules, avoiding the introduction of probabilistic models or inference assumptions, thereby ensuring the interpretability and verifiability of the determination process.

[0037] In one embodiment of the present invention, the transitive consistency label, the prohibitive anti-homozygous label, and the parental heterozygous monoalle shared proxy label are weighted and aggregated according to a weight vector to obtain the transitive consistency ratio, the prohibitive mode ratio, and the parental heterozygous monoalle shared proxy ratio, and hard conflict counts are calculated, including: Step 51: Using the list of available loci as an index, establish a one-to-one correspondence between the transitivity weight vector and various tags according to the order of available loci in the genome. This correspondence ensures that the transitivity weight of each available locus can be consistently matched with the transitivity tag, the prohibition against homozygosity tag, and the parental heterozygous monoallelic surrogate tag for that locus, thereby avoiding aggregation errors caused by locus misalignment. Simultaneously, a zero-value check is performed on the total weight. When the total weight is zero, the state of zero is retained and subsequent proportion calculations are not performed; when the total weight is not zero, the relationship index calculation process begins. This zero-value check avoids meaningless ratio results in the event of missing weights.

[0038] Step 52: Assuming the total weight is not zero, for each available locus, the transmission consistency weight is multiplied by the transmission consistency label of that locus to obtain the transmission consistency weighted value for each available locus. Then, the transmission consistency weighted values ​​for each available locus are summed to obtain a transmission consistency weighted sum. This weighted sum is then compared to the total weight to calculate the transmission consistency percentage. The transmission consistency percentage reflects the overall degree of allele transmission consistency between the tested individual and the suspected individual across all available loci.

[0039] The prohibited homozygous anti-pattern labels are processed using the same weighted aggregation method. Specifically, the transmission consistency weight of each available locus is multiplied by the corresponding prohibited homozygous anti-pattern label and summed. The sum is then compared to the total weight to calculate the proportion of prohibited patterns. This proportion reflects the weighted ratio of genetically incompatible patterns appearing across all available loci. Similarly, the transmission consistency weight of each available locus is multiplied by the corresponding parental heterozygous monoallelic surrogate label and summed. The sum is then compared to the total weight to calculate the proportion of parental heterozygous monoallelic surrogate, which reflects the weighted proportion of individuals sharing a single allele when the suspected individual is heterozygous.

[0040] Step 53 involves counting the available sites in the available site list that are marked with a prohibitive anti-homozygotic tag of 1, thus obtaining a hard conflict count. This hard conflict count is determined within the site range defined by the available site list, explicitly excluding the influence of unavailable sites on the statistical results. The final output includes the proportion of transitive consistency, the proportion of prohibitive patterns, the proportion of parental heterozygous monoallelic surrogate representation, and the hard conflict count, serving as the basis for subsequent joint determination of parentage.

[0041] Through the weighted aggregation and hard conflict statistics process in this embodiment, the genetic consistency information, genetic incompatibility information, and heterozygous shared features of each locus can be structurally summarized in the parentage determination. By passing consistency weights, the quality differences of different loci can be adjusted, thereby avoiding the disproportionate impact of low-quality loci on the overall determination result.

[0042] In one embodiment of the present invention, a joint determination is made based on the proportion of transitive consistency, the proportion of prohibited modes, the proportion of parental heterozygous monoalle shared surrogate, and the hard conflict count, outputting a parentage relationship conclusion and generating a one-time verification digest, including: Step 61: The sample identifier and panel version in the sample list are used to identify the individuals to be tested and those suspected of being involved in the determination. The sample identifier is used to uniquely identify the individuals to be tested and those suspected of being involved in the determination. The panel version is used to limit the range of single nucleotide polymorphism panels used in this determination, thereby ensuring a clear correspondence between the determination result and the specific sample and panel. Subsequently, the total weight obtained in the aforementioned weighted aggregation step is zeroed. When the total weight is zero, it indicates that there are no usable sites for weighted analysis in this determination. In this case, the parentage relationship conclusion is directly output as non-compliant input to avoid giving a result without a basis for determination in the absence of effective nucleic acid site support.

[0043] Step 62: When the total weight is not zero, proceed to the joint verification process. The joint verification is performed on the proportion of transmission consistency, the proportion of prohibited patterns, the hard conflict count, and the proportion of parental heterozygous monoallelic surrogate. Specifically, it verifies whether the proportion of transmission consistency exceeds a preset consistency threshold to determine whether the overall allele transmission consistency meets the requirements; it verifies whether the proportion of prohibited patterns is zero and whether the hard conflict count is zero to determine whether there are definite genetic incompatibility sites; and it verifies whether the proportion of parental heterozygous monoallelic surrogate is within the preset surrogate proportion range to determine whether the heterozygous surrogate characteristic conforms to the overall distribution characteristics of parent-child relationships. The aforementioned preset consistency threshold and preset surrogate proportion range are pre-set fixed judgment parameters used to ensure the certainty and verifiability of the judgment rules.

[0044] Step 63: After completing the above three checks, the parentage conclusion is output as parent-child only if all three results—the transitivity consistency ratio check, the prohibition mode and hard conflict check, and the parent heterozygous monoallelic surrogate ratio check—are simultaneously true; otherwise, the parentage conclusion is output as non-parent-child. By jointly determining multiple indicators based on nucleic acid site statistics, the decisive influence of a single abnormal indicator on the overall determination result can be avoided, thereby improving the robustness of parentage determination.

[0045] Step 64: After determining the parent-child relationship conclusion, the determination result and its related intermediate indicators are solidified. Specifically, the parent-child relationship conclusion, the proportion of transitive consistency, the proportion of prohibited patterns, the proportion of heterozygous single alleles sharing surrogacy, the total weight, and the hard conflict count are recorded as solidified content. Furthermore, combined with the transitive consistency weights corresponding to available loci and various tag details, a one-time verification summary is generated through summary calculation. This one-time verification summary is used to verify the completeness and consistency of the current parent-child relationship determination result, and it supports recalculation and verification of the determination result during subsequent review processes.

[0046] Through the joint determination and one-time verification summary generation process implemented in this paper, parentage can be finally determined according to deterministic rules in nucleic acid testing and analysis applications. The determination results and their key calculation inputs are encapsulated in a summary manner, thereby ensuring the clarity of the determination conclusion while improving the traceability and verifiability of the results. This meets the technical requirements for the reliability and standardization of parentage determination results in application scenarios such as forensic identification and medical testing.

[0047] This invention provides a data processing system for determining parentage, comprising: The sample data processing module is used to receive the raw data from the paternity testing equipment and the identification information of the individuals to be tested and the suspected individuals, generate a sample list, and split and clean the raw data to obtain the cleaned raw data. The sequence alignment and deduplication module is used to align the cleaned raw data after processing with the reference genome and remove duplicate reads to obtain deduplicated alignment data. The panel typing and screening module is used to perform site-specific typing on the deduplication alignment data on the preset single nucleotide polymorphism panel coordinates to obtain diallelic genotypes, and record sequencing depth, allelic balance and uniqueness mask, and screen usable site data according to a fixed threshold. The site weight calculation module is used to determine the transit consistency weight based on sequencing depth, allelic balance, and uniqueness mask in the available site data, forming a weight vector and total weight. The genetic relationship marker module is used to calculate the transmission consistency tag, the prohibition against homozygosity tag, and the parental heterozygous monoalle sharing surrogate tag based on available locus data; The relation index aggregation module is used to perform weighted aggregation of transitive consistency labels, prohibitive anti-homozygous labels, and parental heterozygous mono-equal shared surrogate labels based on weight vectors, to obtain the transitive consistency ratio, prohibitive mode ratio, and parental heterozygous mono-equal shared surrogate ratio, and to count hard conflicts. The relationship determination and verification module is used to jointly determine the proportion of transitive consistency, the proportion of prohibited modes, the proportion of parent-child heterozygous single-equal shared proxy, and the hard conflict count, output the parent-child relationship conclusion, and generate a one-time verification summary.

[0048] It should be noted that the interval and threshold sizes are set for ease of comparison. The size of the threshold depends on the amount of sample data and the base number set by those skilled in the art for each set of sample data, as long as it does not affect the proportional relationship between the parameter and the quantized value. Furthermore, the above formulas are all dimensionless calculations, and the formulas are derived from software simulations using a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0049] The embodiments of the present invention have been described above, but the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms based on the guidance of the present embodiments, all of which are within the protection scope of the present embodiments.

Claims

1. A data processing method for parentage testing, characterized in that, The method comprises the following steps: Step 1, receiving raw data from a parentage identification detection device and identification information of a to-be-tested individual and a suspected individual, generating a sample list, and splitting and cleaning the raw data to obtain cleaned raw data; Step 2, aligning the cleaned raw data with a reference genome and removing duplicate reads to obtain de-duplicated alignment data; Step 3, performing site typing on the de-duplicated alignment data at a preset single nucleotide polymorphism panel coordinate to obtain di-allele genotypes, and recording sequencing depth, allele balance and uniqueness mask, and screening to obtain usable site data according to a fixed threshold; Step 4, determining transmission consistency weight based on the sequencing depth, allele balance and uniqueness mask in the usable site data to form a weight vector and a total weight; Step 5, calculating transmission consistency tags, forbidden anti-homozygous tags and parent heterozygous single allele shared proxy tags based on the usable site data; Step 6, weighting and aggregating the transmission consistency tags, the forbidden anti-homozygous tags and the parent heterozygous single allele shared proxy tags according to the weight vector to obtain transmission consistency proportion, forbidden mode proportion and parent heterozygous single allele shared proxy proportion, and counting hard conflict count; Step 7, jointly determining the transmission consistency proportion, the forbidden mode proportion, the parent heterozygous single allele shared proxy proportion and the hard conflict count to output a parent-child relationship conclusion and generate a one-time verification summary.

2. The data processing method for parentage testing according to claim 1, characterized in that, The method comprises the following steps: Step 11, performing consistency verification on the raw data and the identification information of the to-be-tested individual and the suspected individual to verify that the sample identifiers in the identification information and the index information in the raw data can form a one-to-one mapping, and that the to-be-tested individual identifier and the suspected individual identifier are respectively associated with at least one sample identifier; Step 12, generating a sample list based on the identification information of the to-be-tested individual and the suspected individual that passes the consistency verification, wherein the sample list records the correspondence between the sample identifiers, the index information, the to-be-tested individual identifier and the suspected individual identifier, and the storage positioning information of the raw data; Step 13, splitting the raw data according to the correspondence between the sample identifiers and the index information in the sample list to form a data unit set organized by sample identifiers, and performing joint removal, low-quality base cleaning and abnormal read removal on the data unit set to obtain cleaned raw data.

3. The data processing method for parentage testing according to claim 1, wherein, The method comprises the following steps: The method comprises the following steps: Aligning and matching the cleaned raw data with the reference genome to obtain alignment data, retaining only the reads that can be clearly corresponded to specific positions of the reference genome in the alignment data as unique comparable reads, removing duplicate reads generated in the PCR amplification process from the unique comparable reads, and performing pressure reduction storage on the removed data to obtain de-duplicated alignment data.

4. The data processing method for parentage testing according to claim 1, wherein, The de-duplicated alignment data is genotyped on the preset single nucleotide polymorphism panel coordinates to obtain di-allelic genotypes, and sequencing depth, allele balance and uniqueness mask are recorded, and the available site data is screened according to the fixed threshold, including: Step 21, based on the de-duplicated alignment data and limited to the preset single nucleotide polymorphism panel coordinates, the read set covered by the individual to be detected and the suspicious individual at each panel coordinate is extracted respectively; Step 22, base calling is performed on the read set, and the reference base read count and the variant base read count of each panel coordinate are counted respectively during base calling, and the di-allelic genotypes of the individual to be detected and the suspicious individual at each panel coordinate are determined based on the reference base read count and the variant base read count; Step 23, for each panel coordinate, the sequencing depth, allele balance value and uniqueness mask are recorded respectively, wherein the sequencing depth is the sum of the reference base read count and the variant base read count, the read count of the minor allele is the smaller one of the reference base read count and the variant base read count, the read count of the minor allele is taken as the numerator, the sum of the reference base read count and the variant base read count is taken as the denominator, and the ratio of the numerator to the denominator is calculated to obtain the allele balance value; the uniqueness mask determines the repeat region according to the preset uniqueness mask rule and records it as 0, and records the region outside the repeat region as 1; Step 24, the fixed threshold is used to screen each panel coordinate, and the fixed threshold includes that the sequencing depth is not less than the minimum sequencing depth, the allele balance value is within the preset reasonable allele range and the uniqueness mask is 1, and the available site data is screened, which contains the panel coordinates and their di-allelic genotypes, sequencing depth, allele balance value and uniqueness mask that meet the fixed threshold.

5. The data processing method for parentage testing according to claim 1, wherein, Based on the sequencing depth, allele balance and uniqueness mask in the available site data, the transmission consistency weight is determined to form a weight vector and a total weight, including: Step 31, based on the available site data, the sequencing depth, allele balance value and uniqueness mask of the individual to be detected and the suspicious individual are obtained for each available site, and the minimum sequencing depth is read to form site-level input data indexed by available sites; Step 32, for each available site, the sequencing depth normalization factor and the allele balance penalty factor of the individual to be detected and the suspicious individual are calculated respectively, wherein the sequencing depth normalization factor is obtained by the ratio of the sequencing depth to the minimum sequencing depth; the absolute value of the difference between the allele balance value of the available site and the preset equivalent reference value is taken to obtain the allele balance deviation value; the ratio of the allele balance deviation value to the preset equivalent reference value is taken as the deviation ratio; the allele balance penalty factor is obtained by subtracting 1 from the deviation ratio; Step 33, for each available site, the sequencing depth normalization factor of the individual to be detected and the suspicious individual is multiplied, the allele balance penalty factor of the individual to be detected and the suspicious individual is multiplied, and then the multiplication result is multiplied with the uniqueness mask of the available site to obtain the transmission consistency weight of each available site, and the weight vector is formed by collecting the transmission consistency weights of the available sites in order, and the sum of the transmission consistency weights in the weight vector is obtained to obtain the total weight.

6. The data processing method for parentage testing according to claim 1, wherein, The transmission consistency label, the prohibitive anti-homozygous label and the parent heterozygous single allele shared proxy label are calculated based on available locus data, including: In step 41, based on available locus data, the genotype of the individual to be tested and the genotype of the suspicious individual are obtained for each available locus, and the genotype of the individual to be tested is mapped to the allele set of the individual to be tested and the genotype of the suspicious individual is mapped to the allele set of the suspicious individual according to the allele set rule, wherein the allele set rule is that AA corresponds to a single allele, AB corresponds to two alleles, and BB corresponds to a single allele; In step 42, for each available locus, the transmission consistency label is calculated based on the overlapping relationship between the allele set of the individual to be tested and the allele set of the suspicious individual, and when the allele set of the individual to be tested and the allele set of the suspicious individual overlap, the transmission consistency label is marked as 1, and when there is no overlap, the transmission consistency label is marked as 0, and when the genotype of the suspicious individual is AA or BB and the genotype of the individual to be tested is AA or BB and the allele sets of the two do not overlap, the prohibitive anti-homozygous label is marked as 1, and the prohibitive anti-homozygous label is marked as 0 except for the above cases; In step 43, for each available locus, when the genotype of the suspicious individual is AB and the overlapping of the allele set of the individual to be tested and the allele set of the suspicious individual contains only one allele, the parent heterozygous single allele shared proxy label is marked as 1, and the parent heterozygous single allele shared proxy label is marked as 0 except for the above cases.

7. The data processing method for parentage testing according to claim 1, wherein, The transmission consistency label, the prohibitive anti-homozygous label and the parent heterozygous single allele shared proxy label are aggregated according to the weight vector, and the transmission consistency proportion, the prohibitive mode proportion and the parent heterozygous single allele shared proxy proportion are obtained, and the hard conflict count is counted, including: In step 51, the transmission consistency weight vector and the one-to-one correspondence of each label are established in the order of the available loci with the available locus list as the index; the total weight is checked for zero value, when the total weight is 0, the state of the total weight is kept as 0 and the proportion calculation is not performed, when the total weight is not 0, the proportion calculation is entered; In step 52, the transmission consistency weight of each available locus is multiplied by the transmission consistency label to obtain the transmission consistency weighted value, the sum of the transmission consistency weighted values of each available locus is obtained to obtain the transmission consistency weighted sum, and the transmission consistency weighted sum is divided by the total weight to obtain the transmission consistency proportion; the transmission consistency weight of each available locus is multiplied by the prohibitive anti-homozygous label and the sum is divided by the total weight to obtain the prohibitive mode proportion; the transmission consistency weight of each available locus is multiplied by the parent heterozygous single allele shared proxy label and the sum is divided by the total weight to obtain the parent heterozygous single allele shared proxy proportion; In step 53, the available loci with the prohibitive anti-homozygous label marked as 1 are counted to obtain the hard conflict count, and the transmission consistency proportion, the prohibitive mode proportion, the parent heterozygous single allele shared proxy proportion and the hard conflict count are outputted.

8. The data processing method for parentage testing according to claim 7, characterized in that, The hard conflict count is determined by the number of available loci with the prohibitive anti-homozygous label marked as 1 in the available loci included in the available locus list.

9. The data processing method for parentage testing according to claim 1, wherein, The transmission consistency proportion, the forbidden mode proportion, the parent heterozygous single allele shared agent proportion and the hard conflict count are jointly judged, and the parent-child relationship conclusion is output, and a one-time verification digest is generated, including: Step 61, the sample identification in the sample list is locked with the panel version to determine the object of this judgment, and the total weight is zero value determination, when the total weight is 0, the output of the parent-child relationship conclusion is input illegal; Step 62, when the total weight is not 0, the check is carried out, respectively checking the transmission consistency proportion exceeding the preset consistency proportion threshold, the forbidden mode proportion being 0 and the hard conflict count being 0, and the parent heterozygous single allele shared agent proportion being in the preset shared agent proportion range; Step 63, only when the three checks of step 62 are simultaneously established, the output of the parent-child relationship conclusion is parent-child, and when the three checks are not simultaneously established, the output of the parent-child relationship conclusion is non-parent-child; Step 64, the parent-child relationship conclusion, the transmission consistency proportion, the forbidden mode proportion, the parent heterozygous single allele shared agent proportion, the total weight and the hard conflict count are solidified, and a one-time verification digest is generated based on the solidified content and the weight and label details corresponding to the available site.

10. A data processing system for parentage testing, characterized in that, The data processing method for parent-child relationship determination according to any one of claims 1-9, comprising: a sample data processing module for receiving raw data from a parent identification detection device and identification information of a to-be-tested individual and a suspected individual, generating a sample list, and splitting and cleaning the raw data to obtain cleaned raw data; a sequence alignment and deduplication module for aligning the cleaned raw data with a reference genome and removing duplicate reads to obtain deduplicated alignment data; a panel typing screening module for performing fixed-point typing on the deduplicated alignment data at a preset single nucleotide polymorphism panel coordinate to obtain di-allele genotypes, and recording sequencing depth, allele balance and uniqueness mask, and screening to obtain available site data according to a fixed threshold; a site weight calculation module for determining transmission consistency weight based on the sequencing depth, allele balance and uniqueness mask in the available site data, forming a weight vector and a total weight; a genetic relationship marking module for calculating transmission consistency labels, forbidden anti-homozygous labels and parent heterozygous single allele shared agent labels based on the available site data; a relationship index aggregation module for weighting and aggregating the transmission consistency labels, the forbidden anti-homozygous labels and the parent heterozygous single allele shared agent labels according to the weight vector to obtain the transmission consistency proportion, the forbidden mode proportion and the parent heterozygous single allele shared agent proportion, and counting the hard conflict count; a relationship determination and verification module for jointly judging the transmission consistency proportion, the forbidden mode proportion, the parent heterozygous single allele shared agent proportion and the hard conflict count, outputting the parent-child relationship conclusion, and generating a one-time verification digest.

Images