A biological identity key generation method and system based on multi-source gene data and application thereof

Abstract

The application discloses a biological identity key generation method and system based on multi-source gene data and application thereof, relates to the cross field of bioinformatics, cryptography and computer technology, and aims to solve the problems of non-endogenous, easy forgery, privacy leakage and poor compatibility of traditional biological identity recognition technology. The method specifically comprises the following steps: constructing an identity molecule marker set (IMMS) and adapting different scenes through multi-source gene data acquisition technology; carrying out digital coding, fault-tolerant error correction coding and sequence standardization splicing on the genotype data, and combining an encryption hash algorithm to generate a GBBK; designing a privacy protection verification mechanism based on error correction decoding, hash verification and zero-knowledge proof, and integrating Internet of Things and block chain technology to realize whole life cycle management of biological asset identity. The application can be applied to the fields of biological asset right confirmation, traceability, genetic resource management and financialization, and has the advantages of uniqueness, stability, security and universality, thereby providing a core trust basis for related industries.

Description

Technical Field

[0001] This invention belongs to the interdisciplinary field of bioinformatics, cryptography and computer technology, specifically a method, system and application for generating biological identity keys based on multi-source gene data. Background Technology

[0002] With the growing global demand for food safety, breed protection, biodiversity monitoring, and the valuation of biological assets, assigning a unique, tamper-proof identity to biological assets (such as a livestock, a plant, a plant variety, an animal population, an asexual reproduction line, a strain of microorganisms, or a batch of cell lines) throughout its life cycle has become a key technological bottleneck for the development of modern agriculture, the food industry, the bio-industry, and regulatory technology.

[0003] 1. Fundamental flaws in traditional biological asset identification methods

[0004] Traditional methods of biometric identification, such as physical ear tags, RFID chips, tattoos, and branding, are essentially externally attached identifiers. These methods have fundamental flaws:

[0005] Non-endogeneity: The identifier is separate from the organism itself, making it impossible to fundamentally prove the authenticity and uniqueness of the asset.

[0006] Easily damaged and lost: Physical tags are extremely easy to be damaged or fall off during the growth, transportation and processing of biological assets. For example, in large-scale breeding scenarios, the annual damage / fall-off rate of RFID ear tags can reach 10%-20%, which directly leads to the permanent loss of identity information and the cost of replacement tags is high.

[0007] Easily counterfeited and tampered with: Physical identifiers can be easily copied, replaced, or maliciously altered. QR code labels pose a risk of "code-object separation." The counterfeiting rate of QR codes for some high-end agricultural products exceeds 25%, providing opportunities for asset fraud, brand counterfeiting, and agricultural insurance fraud. They also fail to meet the stringent requirements for the confirmation of ownership of high-value assets and financial collateral.

[0008] Physiological rejection and health risks: Implantable chips have biocompatibility issues, which can easily cause local inflammation, tissue fibrosis, and even chip displacement or potential association with tumors, which limits the acceptance of them by farmers.

[0009] Data silos and poor interoperability: Different manufacturers and regions have different data formats and communication protocols for identification systems, which can easily lead to the loss of identity information when the system is transferred across the industry chain, making it difficult to achieve collaborative supervision.

[0010] 2. Limitations of Phenotypic Biometric Technology

[0011] To overcome the limitations of physical identifiers, the industry has explored phenotypic biometric technologies such as bovine nose prints, pig facial features, and animal irises, but large-scale application remains a challenge.

[0012] Insufficient feature stability: Phenotypic features will undergo "template aging" due to factors such as age, nutritional status, and disease, affecting the recognition accuracy; and in large-scale farming scenarios, long-distance high-precision feature collection requires specialized equipment, making deployment infeasible.

[0013] Privacy and security risks: Biometric templates are irreversible, and database leaks can easily lead to "template reverse engineering attacks" and "replay attacks"; liveness detection requires the fusion of multimodal sensors, which further increases system complexity and cost.

[0014] 3. Limitations of existing gene recognition technologies

[0015] To overcome the aforementioned shortcomings, the industry has begun exploring gene identification technologies based on endogenous biological information. Existing gene technology solutions and their limitations are as follows:

[0016] 1) DNA Barcoding: This technology identifies species at the level by analyzing a few standard gene fragments (such as the COI gene in animals). Its main limitation is insufficient resolution, making it unable to effectively distinguish different varieties or families within the same species, let alone individuals, and therefore cannot meet the needs of personalized asset management.

[0017] 2) Traditional molecular marker technology (SSR / SNP microarray): These technologies rely on a pre-screened, limited set of polymorphic genetic markers (such as microsatellite repeat sequences (SSRs) or single nucleotide polymorphism sites (SNPs)) for individual identification. While more accurate than DNA barcoding, their limitations include:

[0018] High development costs: Developing dedicated SSR marker combinations or SNP chips for specific species or varieties requires significant R&D investment and time.

[0019] Poor flexibility and scalability: Preset marker combinations may not be able to cover newly discovered loci with higher discriminative power, and are inadequate when faced with species diversity and population genetic variation.

[0020] Data standardization is difficult: SSR or SNP typing results generated by different laboratories and different technology platforms may present challenges in standardization and comparison.

[0021] 3) Whole Genome Sequencing (WGS): WGS can provide the most comprehensive genetic information about an individual and theoretically has the highest identification accuracy. However, its high sequencing costs, massive data storage requirements, and complex bioinformatics analysis processes make it completely unsuitable for large-scale, commercial bio-asset individual identification at the current stage, resulting in extremely low economic benefits.

[0022] In recent years, high-throughput genotyping technologies, represented by Genotyping-by-Sequencing (GBS), have achieved breakthroughs. GBS uses restriction endonucleases to digest the genome and selectively sequences portions of the genome, thereby efficiently discovering and genotyping tens of thousands of SNP sites across the entire genome at a relatively low cost. This provides an important technological foundation for low-cost, high-throughput individual genetic identity profiling.

[0023] However, even with advanced data acquisition methods, existing technologies still have the following key gaps:

[0024] Rigid data acquisition methods and scenario mismatch: Existing solutions often rely on a single gene data acquisition technology (such as solely depending on GBS), neglecting the diverse needs of real-world applications. For example, for paternity testing scenarios requiring verification of only a few key loci, qPCR may be the most cost-effective option; while for rapid screening of large populations, customized gene chips may be more advantageous. Current technologies lack a unified framework that can flexibly adapt to multiple data input sources.

[0025] There is a lack of methods for transforming massive heterogeneous genetic data into standardized digital identities: How can we screen out stable, highly discriminative core marker combinations from noisy data generated by different technologies and in various formats (such as raw sequencing data, microarray fluorescence signals, PCR amplification curves) containing multiple marker types (SNP, SSR, Indel, etc.)? How can we transform this complex biological data into a concise, fixed-length digital key that can be efficiently transferred and verified in information systems? Existing technologies mostly remain at the laboratory level for genetic distance comparisons, failing to achieve a leap to industrial-grade "digital identities".

[0026] The lack of a universal fault-tolerance mechanism and privacy-preserving design for data errors is a significant problem: any genetic testing technology inevitably contains a small number of errors (such as sequencing errors, genotyping errors, missing data, allele decoupling, etc.). A robust identity generation algorithm must be able to tolerate these "noises," ensuring that the same individual generates a stable and unique identity key when sampled at different times and under different conditions, or even when tested using different but compatible technology platforms. Meanwhile, raw genetic data is highly sensitive biological privacy information, and direct storage and comparison pose significant security and ethical risks. How to achieve identity verification while protecting genetic privacy remains a challenging problem that current technologies have failed to effectively solve.

[0027] Lack of system architecture to support large-scale commercial applications: The market lacks a comprehensive application system that can integrate sample management, heterogeneous detection task scheduling, multi-source bioinformatics analysis, key generation, secure storage, encrypted verification, and provide standardized API services, making it impossible to seamlessly connect gene identity technology with downstream application scenarios such as blockchain evidence storage, financial risk control, and consumer anti-counterfeiting.

[0028] In summary, there is an urgent need for an innovative technical solution that can not only flexibly integrate multiple gene data acquisition technologies to adapt to different application scenarios, but also transform multi-source, heterogeneous raw gene data into a "biological identity gene key" (GBBK) that is unique, stable, secure, tamper-proof, and easy to use through sophisticated algorithm design and systems engineering. Furthermore, it can build a supporting application system to create a closed-loop technology from gene ownership confirmation to the value transfer of the entire industry chain. Summary of the Invention

[0029] 1. Technical problems to be solved

[0030] This invention aims to systematically address the core pain points of existing biometric asset identification technologies in terms of accuracy, cost, security, standardization, flexibility, and application integration. Specifically, this invention is committed to solving the following technical problems:

[0031] 1) The issue of flexibility and universality of data sources: break away from the dependence on single gene data acquisition technology, and establish a gene data framework that can be compatible with and process multiple technology sources including but not limited to PCR, qPCR, gene sequencing, and gene chips, so as to adapt to the differentiated requirements of sample type, quantity, accuracy and cost in different application scenarios.

[0032] 2) Molecular marker diversity compatibility issues: Enabling identity generation algorithms to handle and utilize various types of molecular markers, including but not limited to known or newly discovered genetic variations such as single nucleotide polymorphisms (SNPs), insertions and deletions (Indels), multiple nucleotide polymorphisms (MNPs), and microsatellites (SSRs).

[0033] 3) Uniqueness and accuracy issues: A method is provided to generate unique identification marks at the individual level (including asexual reproduction lines) for biological assets such as plants and animals. Its accuracy can be flexibly adjusted according to the selected mark set. At the same time, it has multi-dimensional traceability capabilities such as individual, kinship, production area, species, variety, pathogen, etc., far exceeding traditional physical identification and conventional molecular marker technology.

[0034] 4) Data standardization, keying, and visualization: Create a deterministic algorithm that can transform high-dimensional, complex, and heterogeneous raw genotype data into a fixed-length, uniformly formatted, and highly information-entropy encrypted digital key (GBBK) through core tag set filtering, encoding, and cryptographic hashing, while generating an intuitive genetic fingerprint nesting code to visualize identity information.

[0035] 5) Robustness and fault tolerance issues: Fault tolerance mechanisms such as fuzzy extractor are used to address the interference of minor errors and omissions in genotyping data from different technical sources on identity consistency, ensuring that the same individual can stably generate the same GBBK.

[0036] 6) Data security and privacy protection issues: Design an irreversible key generation mechanism to prevent GBBK from being reverse-engineered to obtain the original genetic privacy information, and use advanced cryptographic techniques such as zero-knowledge proof or homomorphic encryption in the verification process to achieve secure verification that "data is usable but not visible".

[0037] 7) System integration and application expansion issues: Build a modular and scalable application system to realize the full lifecycle management of GBBK (generation, storage, verification, and cancellation), integrate decentralized identity (DID) and blockchain evidence storage capabilities, and support its in-depth application in diversified scenarios such as traceability, rights confirmation, finance, supervision and DSI benefit sharing through standard API interfaces.

[0038] This invention provides a method for extracting, generating, and verifying digital identity keys from the genetic data of biological samples. This method is particularly suitable for the confirmation, traceability, valuation, and financialization of biological assets, and also encompasses the software systems, hardware devices, and overall business ecosystem supporting this method.

[0039] Specifically, this invention relates to a technical solution for creating unique, stable, and verifiable digital identities for biological assets such as plants and animals. The core of this solution lies in a Genetic Biological Biometric Key (GBBK) generation algorithm, an application system supporting multi-source heterogeneous gene data input and full lifecycle key management, and integrated capabilities for gene fingerprint visualization, multi-dimensional gene traceability analysis, and decentralized identity verification. This technology can be widely applied to fields such as biological asset ownership verification, full-chain traceability, genetic resource management, biological asset financialization, food authenticity verification, and digital sequence information (DSI) benefit sharing, providing a core trust infrastructure for modern agriculture, the food industry, the biotechnology industry, and regulatory technology.

[0040] To achieve the above objectives, this invention discloses a method for generating a biometric identity key based on multi-source gene data, a verification system, and its application.

[0041] The technical solution adopted in this invention is:

[0042] In a first aspect, the present invention provides a method for generating a biometric identity key based on multi-source gene data, comprising the following steps:

[0043] S1: Screening and Construction of the Identity Molecular Marker Set IMMS: Preprocessing and quality control of raw gene data, converting them into a sample-marker genotype matrix; initial screening of markers based on MAF > 0.05, call rate > 95%, HWE and LD pruning; iterative optimization of the marker set using a heuristic search algorithm, with the marker with the highest information entropy as the core, adding markers that maximize cumulative discriminative ability until a plateau is reached to form the identity molecular marker set IMMS;

[0044] S2: Multi-source gene data acquisition: Select technologies according to the application scenario, including GBS / RAD-seq or WGS for high-throughput exploration, SNPArray or GBTS for medium-to-high-throughput validation, PCR / qPCR / Sanger for low-throughput high-precision validation, dedicated PCR or targeted capture for trace / degraded samples, and uniformly convert them to GFIS format;

[0045] S3: GBBK Generation and Encoding: Digitally encode IMMS marker genotypes, construct virtual tags by clustering species without reference genomes; introduce BCH codes or fuzzy extractors to tolerate ≤2% errors; splice site information to form genotype fingerprints; generate GBBKs through SHA-256 or SHA-3 hashing;

[0046] S4: GBBK verification: Obtain the gene data of the verification sample to generate a temporary G-String'; correct and decode to recover the random string C, and compare Hash(C)' with the database Hash(C); verify through ZKP to achieve privacy protection through Pedersen commitment.

[0047] Furthermore, it includes the following steps:

[0048] In step 1, the screening and construction of the identity molecular marker set IMMS is performed.

[0049] 1) Preprocess and quality control the raw gene data, and convert gene data of different formats into a sample-marker genotype matrix;

[0050] 2) Candidate markers were initially screened based on minimum allele frequency > 0.05, call rate > 95%, Hardy-Weinberg equilibrium test (HWE), and linkage disequilibrium (LD) pruning criteria;

[0051] 3) Employ a heuristic search algorithm to iteratively optimize the candidate tag set. Use the tag with the highest information entropy as the initial core set, and iteratively add tags that maximize the group's "cumulative distinguishing ability" until the distinguishing ability reaches a plateau, thus forming IMMS and solidifying it.

[0052] The acquisition of multi-source gene data in step S2 includes:

[0053] 1) Select the appropriate technology to acquire gene data according to the application scenario requirements: for high-throughput exploratory scenarios, use simplified genome sequencing (GBS / RAD-seq) or whole genome resequencing (WGS); for medium-to-high-throughput validation scenarios, use gene chip SNPArray or targeted sequencing (GBTS); for low-throughput high-precision validation scenarios, use PCR, qPCR, or Sanger sequencing; for sample trace / degradation scenarios, use dedicated PCR or targeted capture sequencing; and convert the raw data in multiple formats into the standard GFIS format.

[0054] In step S3, GBBK is generated and encoded.

[0055] 1) Genotype digital encoding: The genotypes of SNP and SSR markers in IMMS are converted into digital / binary codes according to preset rules. Virtual tags are constructed by clustering for species without reference genomes.

[0056] 2) Error-tolerant and error-correcting coding: Introduce BCH codes or fuzzy extractors to tolerate ≤2% of marker typing errors and restore the normal genotype sequence;

[0057] 3) Sequence assembly and normalized sorting: Assemble error-corrected site information according to the physical location or logical order of the genome to form an individual's genotype fingerprint;

[0058] 4) Generate GBBK using cryptographic hash: Input the genotype fingerprint into the SHA-256 or SHA-3 hash function, and output a fixed-length hash value as GBBK;

[0059] The GBBK verification in step S4 is as follows:

[0060] 1) Obtain gene data from the sample to be validated and generate a temporary G-String;

[0061] 2) Recover the original random string C based on error correction decoding, calculate Hash(C)' and compare it with Hash(C) in the database. If they match, the verification is successful.

[0062] 3) Zero-Knowledge Proof (ZKP) Verification: Proof circuits are generated through Pedersen commitments to achieve privacy-preserving verification where data is available but not visible.

[0063] Preferably, the heuristic search algorithm in step S1 is the CoreHunter or PowerCore algorithm, and the "cumulative distinguishing ability" is measured by the incremental number of uniquely distinguishable individuals.

[0064] Preferably, the BCH code parameters are n=4000, k=2000, t=80, correcting 80-bit random errors. The fuzzy extractor includes a generation algorithm Gen and a regeneration algorithm Rep. The Gen algorithm binds the genotype to the key, and the Rep algorithm recovers the key from noisy data.

[0065] Preferably, the zero-knowledge proof adopts the zk-SNARKs protocol, which verifies the consistency of G-String' with the original commitment within the fault tolerance range and hash verification logic by constructing an arithmetic circuit.

[0066] Secondly, this invention provides a GBBK validation system based on multi-source gene data, comprising the following modules:

[0067] The data access and preprocessing module provides a standardized API interface, supports uploading gene data in multiple formats such as FASTQ, BAM, VCF, and CEL, has a built-in automated bioinformatics analysis workflow, and allows configuration of analysis parameters;

[0068] The core algorithm engine module includes an IMMS management library, a GBBK generator, a GBBK validator, a gene tracing analysis engine, and a gene nesting code generator, enabling IMMS management, GBBK generation and verification, multi-dimensional tracing analysis, and gene fingerprint visualization.

[0069] The secure storage and management module is used to store sample information and GBBK using a distributed database, protect the original gene data with strong encryption, integrate the W3C standard DID to achieve decentralized identity management, and configure the DSI management module to record gene information access logs.

[0070] The API service and application integration module provides GBBK registration / verification APIs, blockchain interfaces, and third-party integration interfaces. It supports 1:1 comparison, 1:N search, and kinship identification multi-mode verification, anchoring GBBK to the blockchain to achieve evidence storage and rights confirmation.

[0071] Preferably, the gene tracing analysis engine supports individual origin tracing with IBS similarity ≥95% to determine homology, parentage identification with PP ≥99.99% and no non-exclusionary loci to determine parentage, group origin tracing using NJ tree and 1000 bootstrap tests, species / variety tracing with ≥5 specific SNPs and consistency ≥90%, and pathogen screening marker matching rate ≥80% and ≥5 loci.

[0072] Preferably, the gene nesting code generator is used to convert IMMS marker genotypes into 3-7 nested color nesting codes to achieve gene fingerprint visualization and intuitive comparison.

[0073] Thirdly, the present invention provides an application of the method described in the first aspect and the system described in the second aspect, applicable to the following scenarios:

[0074] GBBK, generated from breeding livestock and precious trees, serves as the sole identity credential for credit collateral / insurance verification.

[0075] Establish a "gene-product" traceability link for high-end foods and breeding varieties;

[0076] Establish GBBK archives for germplasm resource banks and endangered wild animals;

[0077] Based on GBBK, electronic pedigrees are constructed to guide breeding, and production performance data is linked to achieve early prediction. Disease tracing is completed by combining pathogen-specific markers.

[0078] Compared with the prior art, the present invention has the following significant advantages:

[0079] (I) Root-cause binding and ultra-high anti-counterfeiting: Identity information originates from endogenous genes, coexists with the organism, and cannot be physically separated or tampered with. Counterfeiting GBBK is equivalent to cloning an organism at the molecular level, which is not feasible both technically and economically, fundamentally solving the fraud problems of "labeling" and "cloning".

[0080] (ii) High universality and flexibility: It innovatively adopts a multi-source data access framework, which is compatible with a variety of technologies from PCR to high-throughput sequencing. The optimal solution can be flexibly selected according to cost, accuracy and throughput requirements, and it supports a variety of molecular markers such as SNP, SSR, and Indel, which greatly expands the application scope and economic feasibility.

[0081] (III) High uniqueness and accuracy: Through a scientific core marker set screening strategy, GBBK is ensured to have ultra-high resolution at the individual level, which is far superior to traditional identification methods.

[0082] (iv) Good stability and robustness: The innovative fault-tolerant and error-correcting coding mechanism effectively overcomes random errors in the gene detection process, ensuring that the same individual can generate stable and consistent GBBK under different conditions, achieving applicability throughout the entire life cycle.

[0083] (v) Strong security and privacy protection: Based on the one-way transformation of cryptographic hash functions and optional advanced cryptographic protocols such as zero-knowledge proofs, the irreversibility of identity keys and the "data usable but not visible" verification process are realized, which completely solves the risk of gene privacy leakage.

[0084] (vi) Standardization and easy integration: It transforms complex biological genetic information into standardized digital keys (GBBK) and provides services to the outside world through standard APIs. It is easy to integrate with existing information systems such as blockchain, Internet of Things, and big data platforms, empowering a wide range of downstream applications.

[0085] (vii) Immense Commercial Application Value: By constructing a complete technological closed loop from data to keys to application services, it solves the core trust issues in the process of digitizing and valuing biological assets, providing a powerful technological engine for industrial upgrading in multiple fields such as agriculture, food, finance, and environmental protection. It provides core trust infrastructure for many high-value fields such as agricultural insurance, supply chain finance, brand protection, and germplasm resource management, with broad market prospects. Attached Figure Description

[0086] Figure 1 This is a diagram illustrating the overall logical architecture of the GBBK technical solution of the present invention.

[0087] Figure 2 This is a flowchart of the GBBK generation and verification algorithm of the present invention;

[0088] Figure 3 This is a schematic diagram of the GBBK application system module architecture of the present invention;

[0089] Figure 4 This is a schematic diagram of the gene nesting code of the beef sample of the present invention. Detailed Implementation

[0090] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention. The technical solutions of the embodiments of this application consist of four parts: GBBK generation and verification algorithm, multi-dimensional gene tracing analysis, GBBK application system and application scenarios, realizing full-link coverage from biological samples to digital identity and then to value transfer. The overall technical solution logical architecture is as follows: Figure 1 As shown.

[0091] Part 1: GBBK Generation and Verification Algorithm (Core Original Innovation)

[0092] This algorithm is the core technology of this invention, aiming to convert genetic data from different sources and formats (such as whole-genome sequencing (WGS), simplified genome sequencing (RAD / GBS), SNP microarray genotyping, STR genotyping, etc.) into standardized GBBK. The logical architecture of the GBBK generation and validation algorithm is as follows: Figure 2 As shown.

[0093] GBBK generation is a multi-stage conversion process from biological samples to digital keys. The core of this process lies in balancing recognition accuracy, cost, and algorithm robustness. Specifically, it consists of the following stages:

[0094] Phase 1: Screening and Construction of the Identity Molecular Marker Set (IMMS)

[0095] IMMS is a set of molecular markers selected from a vast pool of potential molecular markers in the genome of a species or variety. These markers possess high individual distinguishing ability, strong genetic stability, and excellent technical detectability. It forms the biological genetic basis for the generation of GBBK. The screening process follows these standards and steps:

[0096] Data preprocessing and quality control: Standardized bioinformatics data quality control and analysis are performed on raw genetic data from different sources. For example, sequencing data is subjected to adapter removal, quality filtering, alignment with a reference genome (if available), or de novo assembly; VCF, PLINKped / map, and other formats are converted into matrices (rows = samples, columns = SNP sites, cells = genotypes).

[0097] Preliminary screening of candidate markers: Filtering out marker sites with poor quality; screening criteria include:

[0098] Minimum allele frequency (MAF): Markers with sufficient polymorphism in the population (e.g., MAF>0.05) are removed after rare variants are eliminated.

[0099] Call Rate: Removes markers with severe data gaps in a large number of samples (e.g., Call Rate > 95%).

[0100] Hardy-Weinberg equilibrium test (HWE): Remove marker sites that may be misclassified or significantly deviate from the HWE.

[0101] Linkage disequilibrium (LD) pruning: Removes markers that are too close in genetic distance and have highly redundant information to ensure the independence between markers.

[0102] IMMS Iterative Optimization (Heuristic Search): It adopts an algorithmic approach similar to Core Hunter or PowerCore to construct the optimal tag set through iteration.

[0103] Initialization: Select the tag with the highest information entropy (e.g., the highest heterozygosity) as the initial core set.

[0104] Iterative addition: In each iteration, a label is selected from the candidate labels that maximizes the "cumulative discriminative power" of the current core set against the population samples and added. The evaluation metric can be the increment in the number of uniquely distinguishable pairwise individuals.

[0105] Convergence Criterion: The iteration terminates when the size of the core set reaches a preset upper limit, or when its discriminative power growth curve plateaus (i.e., adding new markers no longer significantly improves the ability to distinguish all individuals). The final marker combination is the molecular identity marker set (IMMS) for individual identification and verification of this species / variety. This IMMS will be fixed and used for the generation of GBBKs for all subsequent individuals of this species / variety. See Appendix 1 for the Molecular Identity Marker Information Table (TIMM).

[0106] Comparison of CoreHunter and PowerCore algorithms:

[0107]

[0108] Phase Two: Gene Data Acquisition (Flexible Adaptation Layer)

[0109] This invention is not limited to any specific gene data acquisition technology. Based on the needs of the actual application scenario, such as sample type (blood, tissue, hair, leaves, seeds, etc.), sample quantity, required identification accuracy, and cost budget, the most suitable technical means can be flexibly selected to acquire gene data from biological samples.

[0110] High-throughput, exploratory scenarios: For species or large-scale population surveys that require the construction of core marker sets from scratch, simplified genome sequencing (GBS / RAD-seq) or whole genome resequencing (WGS) technologies can be used to efficiently discover massive molecular markers (SNPs, Indels, etc.) across the entire genome.

[0111] High-throughput, validation scenarios: For scenarios where a large number of samples need to be rapidly genotyped using an existing reference marker set, gene chips (SNP Array) and targeted sequencing (GBTS) can be used to detect thousands to hundreds of thousands of known loci at once.

[0112] Low-throughput, high-precision targeted validation scenarios: For scenarios that only require validation of a few key sites (such as species identification, variety identification, specific microbial identification, and specific trait-related site validation), polymerase chain reaction (PCR), quantitative PCR (qPCR), or Sanger sequencing can be used to achieve low-cost, rapid, and accurate detection.

[0113] For samples with trace amounts or severe degradation (such as environmental DNA or processed food), specialized PCR or targeted capture sequencing protocols can be designed.

[0114] A comprehensive comparison of gene data acquisition technologies (categorized by applicable scenarios)

[0115]

[0116] The system can receive and parse raw data generated by these different technologies, such as Ct values ​​from FASTQ (sequencing), CEL (array), ABI (Sanger sequencing), or qPCR, and can ultimately convert gene data in different formats into standard GFIS format data.

[0117] Phase 3: GBBK Generation and Encoding (Cryptographic Conversion Layer)

[0118] This stage converts an individual's biological information (genotype) at the IMMS locus into a standardized digital key. It has the following characteristics:

[0119] Irreversibility: The one-way nature of the hash function ensures that no original genotype information can be deduced from GBBK, effectively protecting genetic privacy.

[0120] Collision resistance: ensures that it is extremely difficult for different individuals to generate the same GBBK.

[0121] Avalanche effect: Even if there are slight differences in genotype fingerprints (exceeding the fault tolerance threshold), the generated GBBK will be completely different, ensuring strict differentiation of identity.

[0122] Genotype digital encoding: For each marker locus in IMMS, genetic variations (SNPs, Indels, MNPs, SSRs, etc.) from different data sources are uniformly mapped onto a reference genome coordinate system, forming a standardized set of variant loci. For species without a reference genome, a virtual shared label is constructed using a clustering algorithm. The individual's genotype (e.g., the A / T / C / G combination of SNP loci) is converted into numerical or binary codes according to predefined rules. For example, a biseleural SNP locus (e.g., A / G) can be encoded as: AA=00, AG=01, GG=10, deletion (N / N)=11. For SSR markers, their repetition count can be directly used as the numerical code.

[0123] Error-tolerant and error-correcting coding: This is a crucial step in ensuring the stability of GBBK. Based on digital coding, an error-correcting code mechanism (such as BCH code or a fuzzy extractor) is introduced. This mechanism allows for the recovery of a unique and correct "canonical genotype sequence" even when the input genotype data contains a small number of errors (e.g., 2% marker genotyping errors in IMMS). This significantly enhances GBBK's robustness to minor differences between different batches of experiments and different detection platforms.

[0124] Sequence splicing and normalized sorting: All error-corrected coded site information is spliced ​​together according to their physical location on the genome or a preset logical order to form a long string, namely the "individual genotype fingerprint".

[0125] Cryptographic hashing to generate GBBK: The above "individual genotype fingerprint" string is used as input, and a cryptographically secure hash function (such as SHA-256 or SHA-3) is applied for calculation. The output fixed-length (e.g., 256 bits) hash value is the final biometric genetic key (GBBK).

[0126] Phase Four: GBBK Verification System and Security Mechanism

[0127] The verification system is used to confirm whether a biological sample taken in the field matches a registered GBBK, while protecting the genetic privacy of the sample.

[0128] The verification process is as follows:

[0129] Sampling and Data Acquisition: Samples (such as blood, hair, and tissue) are taken from the biological individuals to be validated, and their genotype data is acquired using any compatible technology platform (not necessarily the same as that used during registration).

[0130] Generate a temporary G-String': Following the exact same stages one through three as during registration, the newly acquired genotype data is converted into a temporary, normalized binary string G-String'. Due to detection errors, G-String' may have slight differences from the original G-String.

[0131] Key recovery (error correction decoding):

[0132] The verifier retrieves the public information corresponding to the target GBBK from the database: Hash(C) and ECC(δ).

[0133] The validator provides ECC(δ) to the sample holder (or its proxy device).

[0134] The holder uses G-String' and ECC(δ) for RS decoding. Specifically, it calculates δ' = G-String' ⊕ C, and then tries to use the redundant information in ECC(δ) to correct the errors in δ', thereby recovering the original δ.

[0135] If the difference between G-String' and the original G-String is within the error correction capability of the RS code, decoding will succeed and recover δ. Then, the original random string C can be recovered by calculating C = G-String'⊕δ. If the difference is too large, decoding will fail.

[0136] Hash verification:

[0137] The holder calculates the hash value of the recovered C, i.e., Hash(C)'.

[0138] Compare Hash(C)' with Hash(C) obtained from the database.

[0139] If Hash(C)' == Hash(C), then the current sample and the registered sample are the same entity, and the verification passes. Otherwise, the verification fails.

[0140] Enhanced Privacy Protection: Verification Scheme Based on Zero-Knowledge Proof (ZKP)

[0141] To achieve ultimate privacy protection during the verification process, the verifier must not have access to any intermediate data (such as G-String', C).

[0142] The present invention further designs a verification protocol based on zk-SNARKs (Zero-Knowledge Succinct Non-Interactive Argument of Knowledge).

[0143] Application of the Commitment Scheme: During the registration phase, individuals not only store GBBK, but also generate commitments for each bit or block of the G-String using the Pedersen Commitment Scheme and publish these commitments. The Commitment Scheme allows individuals to "commit" to a value without revealing it, but cannot deny it afterward.

[0144] Construct a proof circuit: Design an arithmetic circuit whose inputs are private information (the G-String' of the sample to be verified) and public information (target GBBK, original commitment, etc.). This circuit executes the following logic:

[0145] i. Verify that the G-String' matches the original commitment (within the tolerance range).

[0146] ii. Simulate the RS decoding and hash verification process described above.

[0147] iii. If all steps are successful, the circuit output is true.

[0148] ZKP verification process:

[0149] i. The prover (the sample holder) uses its private G-String as input to generate a short zero-knowledge proof for the circuit described above.

[0150] ii. The prover sends this proof to the verifier.

[0151] iii. Verifiers can quickly verify whether the circuit logic is valid using only publicly available information and the Proof.

[0152] iv. Advantages: Throughout the process, G-String' and any related genetic information are not disclosed to the verifier, achieving "zero-knowledge" verification and fundamentally eliminating the risk of genetic privacy leakage. This is especially important for processing human biological information or high-value breeding core materials.

[0153] The following section uses beef samples as an example to analyze the process according to the GBBK generation and verification workflow in Part 1, and details the specific operating procedures:

[0154] Part Two: Multidimensional Gene Origin Analysis (Commercial Adaptation Development)

[0155] Based on GBBK and IMMS tag data, it integrates multiple traceability and analysis capabilities to meet the identity verification needs of different scenarios:

[0156] Individual origin traceability: Allele sharing (IBS) analysis is used to calculate the IBS similarity between samples. If ≥95% is determined to be from the same source, the genetic fingerprint nesting code of the submitted sample and the earliest sample in the database is displayed to achieve intuitive comparison.

[0157] Paternity testing: Based on the SNIPAR model and the paternity index (PI) model, the genetic consistency of loci is first tested using Mendel's laws, and then the cumulative paternity index (CPI) and paternity probability (PP) are calculated. When PP ≥ 99.99% and there are no exclusionary loci, a parent-child relationship is determined.

[0158] Population / Origin Tracing: A genetic phylogenetic tree is constructed using IBS / Nei's genetic distance calculation. The reliability of branches is evaluated using the neighbor-joining (NJ) method combined with 1000 bootstrap tests. The evolutionary position of the submitted samples is visualized through a circle diagram to infer their geographical origin.

[0159] Species / Variety Origin Tracing: Construct a species / variety-specific SNP reference library, confirm the species / variety type through site matching and consistency threshold screening (e.g., species need ≥5 specific SNPs, consistency ≥90%), and quantify the relative content of each component based on sequencing depth.

[0160] Pathogen screening: Based on the pathogen-specific SNP database, the pathogen species are determined by the marker matching rate (≥80%) and the number of sites (≥5). The relative content of pathogens is calculated based on the average coverage depth, and the reliability of the results is ensured by repeat sequencing and Sanger validation.

[0161] Gene fingerprint nesting code generation: IMMS marker genotypes are converted into colored graphic units and arranged according to rules to form nesting codes, such as... Figure 4 As shown, this enables the visualization of gene fingerprints and supports intuitive comparison of nested codes from different samples.

[0162] Part Three: GBBK Application System (Industry Collaborative Original Innovation)

[0163] This is a comprehensive platform that integrates the GBBK algorithm and gene tracing analysis software, and combines Internet of Things (IoT) and blockchain technology, aiming to connect data flow, trust chain and value chain.

[0164] Multi-role collaborative mechanism for the credible value system of biological assets:

[0165] GBBK application system architecture design, such as Figure 3 As shown.

[0166] Data access and preprocessing module:

[0167] It provides a standardized API interface that supports uploading various gene data formats (FASTQ, BAM, VCF, CEL, etc.).

[0168] It has built-in automated bioinformatics analysis pipelines for different data types, performing tasks such as quality control, comparison, and variant detection.

[0169] Analysis parameters can be flexibly configured according to customer needs.

[0170] Core algorithm engine module:

[0171] IMMS Management Library: Stores and manages optimized core genotype sets (IMMS) for different species and application scenarios.

[0172] GBBK Generator: Implements the above GBBK generation algorithm, including encoding, fault tolerance, and hash calculation.

[0173] GBBK Validator: Receives the genetic data of the sample to be validated, generates a "real-time GBBK" through the same process, and compares it with the GBBK already registered in the database.

[0174] Gene tracing analysis engine: integrates multi-dimensional tracing capabilities such as IBS calculation, parentage index analysis, and phylogenetic tree construction;

[0175] Gene nesting code generator: Enables the visualization, conversion, and comparative display of gene fingerprints.

[0176] Secure storage and management module:

[0177] Distributed database: Stores sample information, raw data (optional, according to privacy policy), analysis results, and generated GBBKs. GBBKs are associated with the metadata of the biological assets (such as date of birth, owner, geolocation, etc.).

[0178] Privacy Protection Mechanism: Sensitive information such as raw genetic data is stored using strong encryption. GBBK's verification process can integrate zero-knowledge proof (ZKP) protocols. The verifier only needs to submit the genetic data of the sample to be verified, and the system can return a "match" or "no match" result without exposing the raw data of either party, achieving the highest level of privacy protection.

[0179] Decentralized identity management: Following the W3C DID standard, GBBK is used as the master key to create a DID for biometric assets, anchored to the blockchain, and to achieve immutable identity.

[0180] DSI Management Module: Records access and usage logs for gene information, supporting transparent auditing of DSI benefit sharing.

[0181] API service and application integration module:

[0182] GBBK Registration API: Allows authorized users to register GBBK for new biological assets.

[0183] GBBK Verification API: This API is used by third-party applications (such as traceability platforms, financial risk control systems, and insurance companies) to verify the authenticity of biological assets. Multiple verification modes can be designed for the API, such as: Individual Origin Tracing: "1:1 Comparison" (verifying if A is A) and "1:N Search" (searching for the existence of A in the database); Kinship Identification: "1:1 Parentage Identification" (verifying the kinship between A and B) and "1:N Screening" (searching for individuals related to A in the database); Population Origin Tracing: Utilizing methods such as Allele Sharing (IBS), genetic distance, principal component analysis (PCA), or phylogenetic tree construction to infer the most probable origin region of an individual.

[0184] Blockchain interface: The hash values ​​of GBBK and its key lifecycle events (generation, transfer, cancellation) can be anchored on the blockchain, and its decentralized and tamper-proof characteristics can be used to provide the highest level of evidence preservation and ownership protection for biological assets.

[0185] Third-party integration interface: Supports seamless integration with traceability platforms, financial risk control systems, and insurance companies, and reserves module nesting interfaces, which can be embedded as independent components into other systems.

[0186] Part Four: Application Scenarios

[0187] The ultimate goal of this invention is to construct a closed-loop "credible value system for biological assets." This system, with its core architecture of "top-level coordination of a gene credibility index matrix + full-chain traceability using GBBK technology + insurance risk mitigation + bank financial empowerment," connects ten key stakeholders to achieve a complete closed loop of "genetic ownership confirmation - process traceability - index endorsement - financial support - regulatory protection." The credible value system for biological assets constructed in this application can be widely applied in the following fields:

[0188] Confirmation and financialization of high-value biological assets:

[0189] Live animal asset collateral: Genetic biometric data (GBBK) is generated for high-value live assets such as breeding livestock and valuable trees, serving as the sole identity credential for financial institutions to use for loans and collateral. In the event of asset disposal or disputes, ownership can be clearly established through genetic verification, solving the problems of difficult supervision and verification of traditional collateral.

[0190] Biological asset insurance: Insurance companies can use GBBK to accurately identify insured livestock, crops, etc., to prevent insurance fraud and achieve accurate claims settlement.

[0191] Full supply chain traceability and brand protection:

[0192] High-end food traceability: Establishing a unique genetic profile for branded beef cattle, premium fruits, etc., from farm to table, with a unique code for each item. Consumers can scan the QR code on the product packaging to access a verification API and verify whether the purchased product matches the genetic information of the brand's origin, thus preventing counterfeit and substandard goods.

[0193] Variety Rights Protection: Authoritative GBBK (Guidelines for Plant and Animal Varieties) are generated for new plant and animal varieties with independent intellectual property rights. During market circulation, suspected infringing products can be sampled for genetic testing at any time, and infringement can be quickly and accurately determined through GBBK comparison, effectively protecting the legitimate rights and interests of breeders.

[0194] Genetic resource management and biodiversity conservation:

[0195] Germplasm Resource Bank Management: Generate GBBK for each material in the germplasm resource bank to achieve digital and precise management, prevent sample confusion and loss, and efficiently assess genetic diversity and population structure.

[0196] Digital Sequence Information (DSI) Management: By combining blockchain, a transparent and auditable log is established for the access and use of digital sequence information of genetic resources, providing a technological tool for global benefit sharing.

[0197] Wildlife Conservation: GBBK files are established for endangered wildlife individuals for individual tracking, population monitoring, and combating illegal poaching and trade in animal products. Law enforcement agencies can trace the origin of seized animal products by verifying their GBBKs.

[0198] Precision agriculture / livestock management:

[0199] Genealogy and Breeding Management: Accurately record the GBBK of breeding animals and their parentage, construct reliable electronic pedigrees, guide precision breeding, avoid inbreeding, and accelerate the selection process for superior traits.

[0200] Production performance prediction: By performing correlation analysis between GBBK and individual production performance data (such as milk yield and growth rate), we can develop an early performance prediction model based on genetic identity.

[0201] Pathogen screening and disease tracing: By using pathogen-specific SNP markers and combining them with the GBBK system, the types and amounts of pathogens in samples can be quickly screened, and the source of the disease can be accurately traced through genomic tracing.

[0202] Example 1

[0203] Genetic Biometric Bio-Key (GBBK) is an innovative biometric identification system that integrates molecular genetics, error-correcting coding theory, and cryptography. This system aims to transform an individual's genomic variation information (such as SNPs and Indels) into an irreversible, collision-resistant, and fault-tolerant standardized digital key, which can be applied to fields such as livestock product traceability, individual identity authentication, and genetic privacy protection.

[0204] GBBK's core technology stack comprises four layers: genotype digitization encoding, error-tolerant and error-correcting encoding, sequence splicing and normalized sorting, and cryptographic hash generation.

[0205] This application uses 2000 SNP molecular marker genotype data from three different batches (batch A, B, and C) of the same beef sample as a specific case to deeply analyze the entire process of GBBK generation and verification, and focuses on explaining the mathematical implementation details of BCH code and fuzzy extractor in the fault tolerance mechanism.

[0206] In practical applications of livestock products, cross-batch genotype data faces three core challenges:

[0207] 1. Experimental platform heterogeneity: Differences in DNA extraction, microarray hybridization, and sequencing depth between different batches lead to fluctuations in genotyping error rates.

[0208] 2. Accumulated data noise: base interpretation errors and missing data contamination during SNP calling.

[0209] 3. Species without reference genomes: Some local cattle breeds lack high-quality reference genomes, necessitating the construction of virtual shared tags.

[0210] The GBBK system, by introducing the algebraic error correction capability of BCH codes and the information-theoretic security framework of the fuzzy extractor, can theoretically tolerate a maximum of 2% tag classification error, ensuring consistency in key generation across batches.

[0211] 2. GBBK Generation and Encoding: A Detailed Explanation of the Cryptographic Transformation Layer

[0212] 2.1 Genotype Digital Encoding: From ATGC to Binary Vectors

[0213] 2.1.1 SNP site normalization mapping

[0214] For 2000 SNP markers, the genotypes must first be uniformly mapped to reference genome coordinates (e.g., the ARS-UCD1.2 bovine reference genome). The genotype coding rules for each biselequential SNP locus are as follows:

[0215] Mathematical expression: For the i-th SNP locus, let the allele set be {A, G}, then the encoding function E(g_i) is:

[0216]

[0217] 2.1.2 Example of multiple batches of data

[0218] Suppose that among 2000 SNP loci, the data from the first 10 loci in 3 batches are as follows:

[0219] 2.2 Fault-Tolerant and Error-Correcting Coding: Implementation by Fusion of BCH Code and Fuzzy Extractor

[0220] 2.2.1 BCH Code Parameter Selection and Construction

[0221] BCH codes are powerful linear block error-correcting codes. Their error-correcting capability is determined by parameters (n, k, t), where n is the codeword length, k is the information bit length, and t is the number of correctable errors. In the GBBK system, BCH codes need to be designed based on the SNP error rate.

[0222] Parameter design principles:

[0223] Code length n: Determined by the number of SNP sites. 2000 SNPs × 2 bits = 4000 bits, which can be designed as a (4095, 2050, 255) BCH code.

[0224] Error correction capability t: set at 2% of the total number of bits, i.e., 4000 × 2% = 80 bits of error.

[0225] The generating polynomial g(x) must satisfy the condition that g(x) has 2t consecutive roots in the field GF(2^12).

[0226] Specific parameter calculations: For a BCH code (n=4095, k=2050, t=255), its generator polynomial degree is nk=2045. This code can correct any 255 random bit errors, far exceeding the 2% fault tolerance requirement [^9][^10]. In practical applications, the BCH code can be shortened (n'=4000, k'=2000, t'=80) to precisely match the SNP data length.

[0227] Generating polynomial construction: On GF(2^12), let α be a primitive element, the generating polynomial is:

[0228]

[0229] Where M_i(x) is the minimal polynomial of α^i. This polynomial ensures that the Hamming distance d_min≥2t+1=161[^11][^12].

[0230] 2.2.2 Implementation Framework of Fuzzy Extractor

[0231] The fuzz extractor consists of two core algorithms: the Gen algorithm and the Rep algorithm, and its security is based on information theory and computational complexity theory.

[0232] Gen Algorithm (Registration Phase): Input: Genotype binary string X∈{0,1}^n, security parameter λ Output: Key R∈{0,1}^λ, auxiliary information P

[0233] 1. Random Key Generation: Randomly select a key R from a uniform distribution, with a length λ = 256 bits.

[0234] 2. BCH encoding: Encode R into codeword C = BCH_Encode(R) to obtain a codeword of length n.

[0235] 3. XOR binding: Calculate the difference vector δ = X⊕C

[0236] 4. Hash Commitment: Calculate the commitment value Com = Hash(C) (SHA-256)

[0237] 5. Auxiliary information encapsulation: P=(δ,Com)

[0238] Mathematical expression:

[0239] \text{Gen}(X)\rightarrow(R,P)\quad\text{where}P=(X\oplus\text{BCH_Encode}(R),\text{Hash}(\text{BCH_Encode}(R)))

[0240] Rep Algorithm (Verification Phase): Input: Noisy genotype X', auxiliary information P=(δ,Com) Output: Recovered key R' or failure symbol ⊥

[0241] 1. XOR recovery: Calculate C' = X' ⊕ δ

[0242] 2. BCH Decoding: Try R'=BCH_Decode(C')

[0243] 3. Hash Verification: Check if Hash(BCH_Encode(R')) == Com

[0244] 4. Output: Return R' if validation passes, otherwise return ⊥.

[0245] Correctness condition: When Hamming distance HD(X,X')≤t, Pr[R'=R]≥1-ε[^15][^16].

[0246] 2.2.3 Analysis of Cross-Batch Fault Tolerance

[0247] For the data from 3 batches of beef samples, assume the inter-batch genotyping error rate p_e = 0.5% (i.e., 2000 × 0.5% = 10 SNP site errors):

[0248] BCH(4000,2000,80) code: can correct 80 bit errors, with a fault tolerance margin of 80 / 10 = 8 times.

[0249] Decoding success rate: According to BCH code theory, the decoding success rate is 100% when the number of errors is ≤ t, and the theoretical decoding failure probability P_fail≈2^{-λ} (negligible) [^17][^18]

[0250] 2.3 Sequence Concatenation and Normalized Sorting

[0251] The error-corrected binary strings of 2000 SNP loci are sorted and concatenated according to their genomic physical location (chromosome:position):

[0252] Concatenation function:

[0253] \text{FINGERPRINT}=\text{Concat}_{i=1}^{2000}\text{BCH_Encode}_i(R)\quad\text{st}\text{Pos}(SNP_i)<\text{Pos}(SNP_{i+1})

[0254] For species without a reference genome, a virtual coordinate system needs to be constructed:

[0255] 1. Clustering to construct shared tags: Using the USTACKS algorithm in the Stacks software package, reads are clustered according to 99% similarity to form "tags".

[0256] 2. Virtual coordinate assignment: Assign virtual chromosome IDs and positions based on label length and abundance.

[0257] 3. Sorting rules: First, arrange in ascending order by chromosome ID, then assemble in ascending order by position.

[0258] 2.4 Generating GBBK using cryptographic hashing

[0259] The final key is generated using the SHA-256 cryptographic hash function:

[0260]

[0261] Cryptographic properties:

[0262] Irreversibility: The computational complexity of recovering FINGERPRINT from GBBK is 2. 256 This operation;

[0263] Collision resistance: The probability of finding a collision under a birthday attack is approximately 2. -128 ;

[0264] Avalanche effect: A 1-bit change in the input results in an average 128-bit flip in the output;

[0265] 3. GBBK Verification System and Security Mechanism

[0266] 3.1 Cryptographic Implications and Generation Algorithms of δ and C

[0267] 3.1.1 The Cryptographic Role of Delta

[0268] δ is the Difference Binding Vector, whose core function is to bind biometric features to a key without exposing the original genotype X. In fuzzy commitment schemes, δ is mathematically defined as:

[0269]

[0270] in:

[0271] X: Original genotype binary string (n=4000 bits)

[0272] C: BCH encoded codeword (n=4000 bits)

[0273] ⊕: Bitwise XOR operation on the GF(2) field

[0274] Security attributes:

[0275] δ itself does not reveal information about X or C, because for any X, there exists C = X ⊕ δ, and C is uniformly distributed in the key space.

[0276] δ is stored in a public database, so even if an attacker obtains it, it is impossible to deduce X or C.

[0277] 3.1.2 The cryptographic meaning of C

[0278] C is the commitment codeword, whose construction follows the algebraic structure of error-correcting codes:

[0279] C=\text{BCH_Encode}(R)=R\cdotG

[0280] Where G is the generator matrix of the BCH code (dimension k×n).

[0281] On GF(2), the weight distribution of Chinese characters in codeword C satisfies:

[0282]

[0283] Key properties:

[0284] C must belong to a linear subspace of the BCH code, i.e., satisfy C·H^T=0, where H is an (nk)×n parity check matrix.

[0285] The entropy of C is at least λ = 256 bits, ensuring that the key space is large enough.

[0286] 3.1.3 Generation mechanism of ECC(δ)

[0287] ECC(δ) is not an independent error-correcting code, but rather a helper data wrapper for the δ vector. Its generation process is as follows:

[0288] \text{ECC}(\delta)=(\delta,\text{Hash}(C),\text{ECC_Params})

[0289] ECC_Params contains:

[0290] The (n,k,t) parameters of the BCH code

[0291] coefficients of the generator polynomial g(x)

[0292] Hash algorithm identifier (e.g., SHA-256)

[0293] Storage optimization: To reduce storage overhead, Syndrome Binding technology can be used to store only the (nk) bit parity of δ instead of the complete n-bit vector [^28][^29].

[0294] 3.2 Mathematical Implementation of the Verification Process

[0295] 3.2.1 Temporary G-String Generation

[0296] The 2000 SNP sites for the validation samples followed the exact same coding process as in the registration phase:

[0297]

[0298] Noise model: Assuming the error probability per bit is p_e=0.005, the expected number of errors E[Δ]=n·p_e=4000×0.005=20 bits, which is much smaller than t=80.

[0299] 3.2.2 Key Recovery (Error Correction Decoding)

[0300] After the verifier retrieves ECC(δ) from the database, the holder executes:

[0301] 1. XOR recovery: Calculate the noisy codeword

[0302] 2. Syndrome Calculation: Calculate the syndrome

[0303]

[0304] 3. Berlekamp-Massey Decoding: Solving for the error location polynomial σ(x), whose roots correspond to the error locations.

[0305] 4. Chien search: Exhaustively search all elements in GF(2^m) and locate t incorrect positions.

[0306] 5. Error correction: Flip the bits at the corresponding position in C' to obtain C*.

[0307] Decoding complexity: O(n·t) GF(2^m) operations, which is approximately 3.2×10^5 operations for the (4000,2000,80) code, and is completed in milliseconds.

[0308] 3.2.3 Hash Verification and Decision

[0309] Verify the equation:

[0310] \text{Accept iff}\text{SHA-256}(\text{BCH_Encode}(\text{RS_Decode}(X'\oplus\delta)))==\text{Hash}(C)

[0311] Probability analysis:

[0312] False positive rate (FPR): The probability that an attacker's random guess will pass is ≤ 2^{-256}

[0313] False Negative Rate (FNR): When HD(X,X')>t, the probability of decoding failure is approximately 1.

[0314] True Positive Rate (TPR): The success rate of validating correct samples ≈ 1 - ε, where ε is the probability of decoding failure (< 10^{-6}).

[0315] 4. Cross-batch data consistency guarantee mechanism

[0316] 4.1 Sources and Assessment of Systematic Errors

[0317] According to systematic error studies in the field of food testing, the systematic error in beef sample genotype data mainly originates from:

[0318] 1. Sample preparation error: Batch-to-batch variation in DNA extraction efficiency (CV = 5-8%)

[0319] 2. Platform systematic errors: Differences in probe hybridization efficiency between different gene chips (e.g., Illumina GGP Bovine 150K vs. Affymetrix Axiom).

[0320] 3. Bioinformatics error: Differences in threshold settings of the SNP calling algorithm

[0321] Evaluation method: The standard substance calibration method was used, with bovine cell line standards of known genotypes (such as ATCC B-350) as quality control samples, and the batch-to-batch consistency rate was calculated.

[0322]

[0323] 4.2 Clustering Algorithm for Constructing Virtual Shared Labels

[0324] For local cattle breeds without a reference genome (such as the Chinese Yellow Cattle), de novo clustering should be used to construct virtual labels.

[0325] 4.2.1 USTACKS Clustering Algorithm

[0326] When using the USTACKS module from the Stacks package, the parameters are set as follows:

[0327] • Similarity threshold: -M 3 (allows 3 mismatches)

[0328] Minimum coverage depth: -m 3

[0329] ·Maximum number of gaps: --gap 2

[0330] Algorithm flow:

[0331] 1. k-mer decomposition: dividing reads into k-mer values ​​of k=31.

[0332] 2. Hash table construction: Use rolling hashing to map k-mer to buckets.

[0333] 3. Greedy clustering: Clustering is performed based on k-mer overlap > 97%, forming a "Stack".

[0334] 4. Common sequence generation: Perform multiple sequence alignment on each stack to generate a common tag sequence.

[0335] 4.2.2 Standardized Sorting Rules

[0336] Virtual chromosome construction follows:

[0337] Chromosome IDs: Sorted in descending order of tag length, assigned Chr1, Chr2,...

[0338] Position coordinates: sorted lexicographically by k-mer within the chromosome.

[0339] Assemble order: Assemble all sites in ascending order of (ChrID, Position).

[0340] 4.3 Robustness Enhancement Strategies

[0341] 4.3.1 Dynamic Threshold Adjustment

[0342] Automatically adjust SNP calling thresholds (QD, MQ, DP) based on batch quality:

[0343] 4.3.2 Concatenation of Multiple Error Correction Codes

[0344] The robustness is further improved by using BCH+RS concatenated codes:

[0345] Outer BCH code: Corrects random errors

[0346] Inner RS ​​code: Corrects burst errors (such as consecutive missing values).

[0347] The concatenated code has a total error correction capability of t_total=t_BCH×t_RS=80×8=640 bits, which can cope with extreme batch differences.

[0348] 5. Security and Privacy Protection Analysis

[0349] 5.1 Information Theory Security Boundaries

[0350] According to the security definition of the fuzzy extractor, the amount of information about X leaked by δ is constrained by the upper bound of mutual information:

[0351] ,

[0352] For 256-bit security parameters, leakage of δ is negligible.

[0353] 5.2 Collision Resistance Attack Analysis

[0354] SHA-256's collision resistance safety is based on the birthday paradox:

[0355]

[0356] When an attacker attempts q = 2^64 GBBK, the collision probability is only 2^{-129}.

[0357] 5.3 Side-channel attack protection

[0358] Constant-time decoding algorithm is used to prevent timing attacks:

[0359] Use a lookup table to replace conditional branches.

[0360] Perform virtual error correction on all locations, regardless of whether they contain errors.

[0361] Appendix 1: Identification Molecular Marker Information Table

[0362] The Information Table of Identity Molecular Marker (TIMM) is a template file specifically designed to identify the molecular markers of a particular species or variety, tailored to specific application scenarios. The file name is in English according to its intended use. For example, a TIMM containing 2050 SNPs for tea origin traceability could be named "TIMM-TEA2050," while a TIMM containing 100 SNPs for meat species composition traceability could be named "TIMM-MEAT100." To prevent discrepancies between identical species and marker counts with different actual loci, a two-digit serial number is added after the name for differentiation, such as "TIMM-TEA2050-01" or "TIMM-MEAT100-01." The specific format is as follows:

[0363]

[0364] The first two rows contain information on the species, number of markers, and serial number.

[0365] The third row is the header row. Cells 1-5 of the third row contain the SN, ID, chrom number, Position value, and Ref value, respectively. Cell 6 and onwards contain the sample number. Columns 6 and onwards, from the fourth to the fifth row, contain molecular marker information.

[0366] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this invention.

Claims

1. A method for generating a biometric key based on multi-source gene data, characterized in that, The following steps are required: S1: Screening and Construction of the Identity Molecular Marker Set IMMS: Preprocessing and quality control of raw gene data, converting them into a sample-marker genotype matrix; initial screening of markers based on MAF > 0.05, call rate > 95%, HWE and LD pruning; iterative optimization of the marker set using a heuristic search algorithm, with the marker with the highest information entropy as the core, adding markers that maximize cumulative discriminative ability until a plateau is reached to form the identity molecular marker set IMMS; S2: Multi-source gene data acquisition: Select technologies according to the application scenario, including GBS / RAD-seq or WGS for high-throughput exploration, SNPArray or GBTS for medium-to-high-throughput validation, PCR / qPCR / Sanger for low-throughput high-precision validation, dedicated PCR or targeted capture for micro / degraded samples, and uniformly convert them to GFIS format; S3: GBBK Generation and Encoding: Digitally encode the genotype of the IMMS molecular marker set for identity, construct virtual tags by clustering species without reference genomes; introduce BCH codes or fuzzy extractors to tolerate ≤2% errors; splice site information to form genotype fingerprints; generate GBBKs through SHA-256 or SHA-3 hashing; S4: GBBK verification: Obtain the gene data of the verification sample to generate a temporary G-String'; correct and decode to recover the random string C, and compare Hash(C)' with the database Hash(C); verify through ZKP to achieve privacy protection through Pedersen commitment.

2. The method for generating a biometric key based on multi-source gene data according to claim 1, characterized in that, Includes the following steps: In step 1, the screening and construction of the identity molecular marker set IMMS is performed. 1) Preprocess and quality control the raw gene data, and convert gene data of different formats into a sample-marker genotype matrix; 2) Candidate markers were initially screened based on minimum allele frequency > 0.05, call rate > 95%, Hardy-Weinberg equilibrium test (HWE), and linkage disequilibrium (LD) pruning criteria; 3) Employ a heuristic search algorithm to iteratively optimize the candidate tag set. Use the tag with the highest information entropy as the initial core set, and iteratively add tags that maximize the group's "cumulative distinguishing ability" until the distinguishing ability reaches a plateau, thus forming IMMS and solidifying it. The acquisition of multi-source gene data in step S2 includes: 1) Select the appropriate technology to acquire gene data according to the application scenario requirements: for high-throughput exploratory scenarios, use simplified genome sequencing (GBS / RAD-seq) or whole genome resequencing (WGS); for medium-to-high-throughput validation scenarios, use gene chip SNPArray or targeted sequencing (GBTS); for low-throughput high-precision validation scenarios, use PCR, qPCR, or Sanger sequencing; for sample trace / degradation scenarios, use dedicated PCR or targeted capture sequencing; and convert the raw data in multiple formats into the standard GFIS format. In step S3, GBBK is generated and encoded. 1) Genotype digital encoding: The genotypes of SNP and SSR markers in IMMS are converted into digital / binary codes according to preset rules. Virtual tags are constructed by clustering for species without reference genomes. 2) Error-tolerant and error-correcting coding: Introduce BCH codes or fuzzy extractors to tolerate ≤2% of marker typing errors and restore the normal genotype sequence; 3) Sequence assembly and normalized sorting: Assemble error-corrected site information according to the physical location or logical order of the genome to form an individual's genotype fingerprint; 4) Generate GBBK using cryptographic hash: Input the genotype fingerprint into the SHA-256 or SHA-3 hash function, and output a fixed-length hash value as GBBK; The GBBK verification in step S4 is as follows: 1) Obtain gene data from the sample to be validated and generate a temporary G-String; 2) Recover the original random string C based on error correction decoding, calculate Hash(C)' and compare it with Hash(C) in the database. If they match, the verification is successful. 3) Zero-Knowledge Proof (ZKP) Verification: Proof circuits are generated through Pedersen commitments to achieve privacy-preserving verification where data is available but not visible.

3. The method for generating a biometric key based on multi-source gene data according to claim 2, characterized in that, The heuristic search algorithm mentioned in step S1 is either the CoreHunter algorithm or the PowerCore algorithm, and the "cumulative distinguishing ability" is measured by the incremental number of uniquely distinguishable individuals.

4. The method for generating a biometric key based on multi-source gene data according to claim 2, characterized in that, The BCH code parameters are n=4000, k=2000, t=80, correcting 80-bit random errors. The fuzzy extractor includes a generation algorithm Gen and a regeneration algorithm Rep. The Gen algorithm binds the genotype to the key, and the Rep algorithm recovers the key from noisy data.

5. The method for generating a biometric key based on multi-source gene data according to claim 2, characterized in that, The zero-knowledge proof adopts the zk-SNARKs protocol, which verifies the consistency of G-String' with the original commitment within the fault tolerance range and uses hash verification logic by constructing an arithmetic circuit.

6. A GBBK validation system based on multi-source gene data, characterized in that, Includes the following modules: The data access and preprocessing module provides a standardized API interface, supports uploading gene data in multiple formats such as FASTQ, BAM, VCF, and CEL, has a built-in automated bioinformatics analysis workflow, and allows configuration of analysis parameters; The core algorithm engine module includes an IMMS management library, a GBBK generator, a GBBK validator, a gene tracing analysis engine, and a gene nesting code generator, enabling IMMS management, GBBK generation and verification, multi-dimensional tracing analysis, and gene fingerprint visualization. The secure storage and management module is used to store sample information and GBBK using a distributed database, protect the original gene data with strong encryption, integrate the W3C standard DID to achieve decentralized identity management, and configure the DSI management module to record gene information access logs. The API service and application integration module provides GBBK registration / verification APIs, blockchain interfaces, and third-party integration interfaces. It supports 1:1 comparison, 1:N search, and kinship identification multi-mode verification, anchoring GBBK to the blockchain to achieve evidence storage and rights confirmation.

7. The system according to claim 5, characterized in that, The gene tracing analysis engine supports individual origin tracing with IBS similarity ≥95% to determine homology, parentage identification with PP ≥99.99% and no non-exclusionary loci to determine parentage, group origin tracing using NJ tree and 1000 bootstrap tests, species / variety tracing with ≥5 specific SNPs and consistency ≥90%, and pathogen screening marker matching rate ≥80% with ≥5 loci.

8. The system according to claim 5, characterized in that, The gene nesting code generator is used to convert IMMS marker genotypes into 3-7 nested color nesting codes, enabling gene fingerprint visualization and intuitive comparison.

9. An application of the method according to claims 1-5 and the system according to claims 6-8, characterized in that, Applicable to the following scenarios: GBBK, generated from breeding livestock and precious trees, serves as the sole identity credential for credit collateral / insurance verification. Establish a "gene-product" traceability link for high-end foods and breeding varieties; Establish GBBK archives for germplasm resource banks and endangered wild animals; Based on GBBK, electronic pedigrees are constructed to guide breeding, and production performance data is linked to achieve early prediction. Disease tracing is completed by combining pathogen-specific markers.

Images