A method for telomere-to-telomere genomic multidimensional quality assessment

Abstract

This invention belongs to the field of genomics and bioinformatics technology, specifically relating to a method for multidimensional quality assessment of telomere-to-telomere genomes. The method includes the following steps: preprocessing the genome data to be assessed to obtain preprocessed data; quantitatively assessing multiple core features of the preprocessed data, including accuracy, completeness, continuity, and region specificity; the region specificity includes telomere identification, centromere identification, and collinearity with a reference genome; calculating a comprehensive score based on the weighted results of the quantitative assessment of the core features, and using the comprehensive score to determine the quality of the genome to be assessed. This assessment method is specifically adapted to the core features of T2T genomes, taking into account continuity, completeness, accuracy, and region specificity, thus solving the problem that traditional methods cannot assess T2T-specific features.

Description

Technical Field

[0001] This invention belongs to the field of genomics and bioinformatics technology, specifically involving a high-precision and specific quality assessment process for telomere-to-telomere (T2T) complete genomes. It is applicable to the detection of T2T genome assembly results of third-generation long-read sequencing (PacBio HiFi, Oxford Nanopore Ultra-long) and multi-platform joint assembly, and can be applied to scenarios with extremely high requirements for genomes, such as model organism genome analysis and evolutionary genomics. Background Technology

[0002] With breakthroughs in ultra-long read sequencing technology and advanced assembly algorithms, telomere-to-telomere genome assembly has been achieved in multiple species, including humans, rice, and maize. T2T genome assembly is a cutting-edge direction in genomics research, aiming to obtain gap-free, complete genome sequences that cover the entire chromosome and include complex repetitive regions such as centromeres and segment repetitions. Compared to traditional fragmented genomes, T2T genomes can more accurately reflect the true structure of the genome, providing a crucial foundation for elucidating the function of repetitive regions, chromosome evolution, and the pathogenesis mechanisms of rare diseases.

[0003] The most representative advantage of the T2T genome is its gap-free integrity, with telomeres and centromeres being key targets in T2T assembly. Traditional assessment metrics such as Contig N50, BUSCO, and Merqury can only individually evaluate continuity, integrity, and accuracy. QUAST can assess multiple metrics, but its correlation with overall genome assembly quality is not high. These assessment metrics cannot detect key T2T regions such as telomeres and centromeres, and lack systematic validation methods for structural levels beyond single-base accuracy, such as large-scale structural variations and complex inversions. Therefore, traditional assessment methods and metrics are not significant in the T2T genome, necessitating a completely new assessment system for comprehensive evaluation of the T2T genome. Summary of the Invention

[0004] The purpose of this invention is to provide a multi-dimensional quality assessment method applicable to T2T genomes, which can comprehensively assess the integrity, continuity, accuracy, and regional specificity (telomere identification, centromere identification, collinearity, etc.) of T2T assembly results, and output standardized and quantifiable assessment reports, providing core technical support for the reliability verification of T2T genomes.

[0005] The technical solution provided by this invention is as follows:

[0006] A method for multidimensional quality assessment of telomere-to-telomere genomes includes the following steps:

[0007] Genomic data to be quality assessed are preprocessed to obtain preprocessed data;

[0008] The preprocessed data were quantitatively evaluated for several core features, including accuracy, completeness, continuity, and region specificity. The region specificity included telomere identification, centromere identification, and collinearity with the reference genome.

[0009] The quantitative evaluation results of the core features are weighted to calculate a comprehensive score, and the quality of the genome to be evaluated is determined by the comprehensive score.

[0010] Furthermore, the genomic data to be quality assessed includes T2T genome assembly results, raw sequencing data, and species-specific reference information.

[0011] Furthermore, the data preprocessing includes performing data statistics, filtering, and format conversion using fastp second-generation NGS data, and using seqkit to detect read length distribution and average quality value of Nanopore data and PacBio HiFi data.

[0012] Furthermore, the continuity assessment metrics include N50; NGS, HiFi, and ONT data return ratios and coverage and AQI value; the integrity assessment metrics include the proportion of complete single-copy BUSCO genes, the proportion of duplicate genes, and the proportion of deleted genes; the accuracy assessment metric is QV; the assessment method is to assess genomic QV using HiFi database construction via meryl.

[0013] Furthermore, the evaluation index for telomeres includes telomere sequence integrity; the evaluation index for centromeres is centromere sequence integrity, and the evaluation method is to identify the location of the centromere region after annotating the whole genome repetitive regions using the TRF+RepeatMasker+SRF tool.

[0014] Furthermore, the evaluation metric for collinearity with the reference genome includes the percentage of collinear region length.

[0015] Furthermore, in the comprehensive score calculated according to weights, the weights are allocated as follows: accuracy 10%; completeness 10%; continuity 35%; telomere identification 20%; centromere identification 20%; and collinearity assessment with the reference genome 5%.

[0016] The present invention also provides an apparatus for multidimensional quality assessment of telomere-to-telomere genomes, the apparatus comprising:

[0017] The data preprocessing module is used to preprocess the data of each gene in the genome to be quality assessed, and obtain the preprocessed data.

[0018] The evaluation module is used to quantitatively evaluate several core features of the preprocessed data, including accuracy, completeness, continuity, and region specificity; the region specificity includes telomere identification, centromere identification, and collinearity with the reference genome.

[0019] The comprehensive scoring module is used to calculate a comprehensive score based on the quantitative evaluation results of the core features according to their weights, and to determine the quality of the genome to be evaluated based on the comprehensive score.

[0020] Furthermore, the apparatus for telomere-to-telomere genome multidimensional quality assessment also includes a report generation module for integrating the results of the data preprocessing module, the assessment module, and the comprehensive scoring module to generate a telomere-to-telomere genome multidimensional quality assessment report.

[0021] The present invention also provides a computer-readable storage medium comprising a stored program, wherein, when the program is executed, the device on which the storage medium is located executes the method for telomere-to-telomere genome multidimensional quality assessment as described in any one of claims 1-7.

[0022] This T2T genome quality assessment workflow uses the Nextflow workflow framework to encapsulate the data preprocessing module, assessment module, comprehensive scoring module, and report generation module, forming a complete assessment workflow.

[0023] The data preprocessing module mainly includes: preprocessing of NGS, HiFi and ONT offline data, data volume statistics, and generation of read length distribution charts.

[0024] The evaluation modules mainly include: continuity evaluation; integrity evaluation; accuracy evaluation; and region-specific evaluation (telomere identification; centromere identification; collinearity analysis).

[0025] Comprehensive scoring module: Construct a T2T genome-specific scoring system, highlighting the weight of core features: accuracy (10%); completeness (10%); continuity (35%); telomere assessment (20%), centromere assessment (20%), and collinearity assessment with the reference genome (5%).

[0026] Report generation module: Automatically integrates the results from various modules to generate a T2T genome-specific assessment report.

[0027] Beneficial effects

[0028] The evaluation method described in this invention is specifically adapted to the core features of the T2T genome, taking into account continuity, integrity, accuracy and region specificity, and solving the problem that traditional methods cannot evaluate T2T-specific features;

[0029] The evaluation method of this invention automates the entire process from preprocessing to report generation, standardizes core indicators and scoring systems, avoids human error, and saves analysis time.

[0030] The evaluation method of this invention provides clear quality level classification and application scenario suggestions, and gives specific optimization directions for indicators that do not meet the standards, which helps to improve the T2T genome iteratively.

[0031] The evaluation method of this invention is compatible with the T2T assembly results of mainstream third-generation sequencing platforms such as PacBio and Nanopore, and can be adapted to a variety of plants and animals. Attached Figure Description

[0032] Figure 1 This is a flowchart of a method for multidimensional quality assessment of telomere-to-telomere genomes. Detailed Implementation

[0033] Example 1

[0034] The T2T genome quality assessment workflow includes a data preprocessing module, an assessment module, a comprehensive scoring module, and a report generation module. The specific workflow is as follows:

[0035] 1. Data Preprocessing Module

[0036] Input the T2T genome assembly results (FASTA format), raw sequencing data (second-generation NGS data from DNBSEQ-T7 sequencing in FASTQ format, third-generation HiFi data from PacBio Revio sequencing in bam format, and third-generation ONT data from Nanopore PromethION (R10.4.1) sequencing in FASTQ format), and species-specific reference information (such as telomere repeat motifs and centromere characteristic sequences of the species verified in the literature), and perform the following operations:

[0037] Raw data quality verification: Data was analyzed using FASTP 2nd generation NGS data, including data statistics, connector removal, and filtering of low-quality reads. SAMtools was used to convert the raw BAM data to FASTQ format for subsequent analysis. Seqkit was used to check read length distribution and average quality values ​​for Nanopore and PacBio HiFi data. Data depth and quality were considered to determine suitability for further evaluation. If data depth and quality were insufficient, resulting in a low continuity module score, it was not recommended to use NGS, HiFi, or ONT data for read ratios and coverage.

[0038] 2. Evaluation Module

[0039] (1) Continuity assessment

[0040] Key metrics: N50, NGS, HiFi, ONT data return ratio and coverage, AQI value;

[0041] Analysis method:

[0042] The N50 of the genome was calculated using a custom Python 3 script; the raw data was aligned to the assembled genome using the Minimap2 tool; the NGS, HiFi, and ONT data return ratios and coverage were calculated using Samtools; and the AQI value was calculated using CRAQ software.

[0043] (2) Integrity assessment

[0044] Key metrics: percentage of complete single-copy BUSCO genes, percentage of duplicate genes, and percentage of deleted genes;

[0045] Analysis method: High-resolution BUSCO gene sets corresponding to the species were selected, and gene integrity was detected by BUSCO software.

[0046] (3) Accuracy assessment

[0047] Key metric: QV;

[0048] Analysis methods: A library was constructed using HiFi data via meryl, and genomic QV was evaluated using Merqury software.

[0049] (4) Telomere assessment

[0050] Key indicator: Telomere sequence integrity (whether both ends of each chromosome contain telomere sequences);

[0051] Analysis method:

[0052] Based on species-specific telomere repeat motifs (e.g., human telomere motif is TTAGGG, which can be manually entered or automatically identified by the system), the TelomereFinder tool is used to scan both ends of the assembled sequence to detect the presence of telomere sequences (telomeric sequences at both ends are "complete telomere", those at only one end are "partially complete", and those at neither end are "missing"). A Python 3 script is used to count the proportion of the three types.

[0053] (5) Centromere assessment

[0054] Key indicator: Centromere sequence integrity (whether each chromosome contains a centromere sequence);

[0055] Analysis method:

[0056] After annotating the whole genome repetitive regions using the TRF+RepeatMasker+SRF tool, the centromere region locations were identified. Regions found were marked as "present" and those not found were marked as "deleted". The percentage of chromosomes with present and missing centromeres was calculated using a Python 3 script.

[0057] (6) Assessment of collinearity with the reference genome

[0058] Key metric: Percentage of collinear region length;

[0059] Analysis method:

[0060] The minimap2 tool was used to align the T2T genome with the genomes of homologous species. Syri was used to obtain the collinearity of regions, and the pyhton3 script was used to calculate the proportion of collinearity lengths.

[0061] Comprehensive scoring module

[0062] (1) Continuity assessment scoring rules

[0063] N50≥10M earns 10 points;

[0064] 10 points are awarded if the NGS data recovery rate is ≥90%;

[0065] NGS data coverage ≥ 90% earns 10 points;

[0066] HiFi data return rate ≥ 90% earns 10 points;

[0067] HiFi data coverage ≥ 90% earns 10 points;

[0068] 10 points are awarded if the ONT data recovery rate is ≥90%;

[0069] 10 points are awarded for ONT data coverage ≥ 90%;

[0070] AQI > 90: 30 points; 80-90: 20 points; 70-80: 10 points; below 70: 0 points.

[0071] (2) Integrity assessment scoring rules

[0072] 100 points for complete single-copy gene percentage >99%; 80 points for 95%-99%; 60 points for 90%-95%; 0 points for less than 90%.

[0073] (3) Accuracy assessment scoring rules

[0074] A score of 100 is awarded for an overall genome QV ≥ 60; 90 for 50%-60%; 80 for 40%-50%; 70 for 30%-40%; and no score is awarded for scores below 30%.

[0075] (4) Telomere assessment scoring rules

[0076] 100% of chromosomes with intact telomeres receive 100 points; 90%-100% receive 90 points; 80%-90% receive 80 points; less than 80% receive no points.

[0077] (5) Centromere evaluation scoring rules

[0078] 100% of chromosomes containing mitochondria receive 100 points; 90%-100% receive 90 points; 80%-90% receive 80 points; less than 80% receive no points.

[0079] (6) Scoring rules for assessing collinearity with the reference genome

[0080] 100 points for 90% collinearity with the reference genome; 90%-90% for 90%; 70%-80% for 80%; 60%-70% for 70%; 50%-60% for 60%; 40%-50% for 50%; 30%-40% for 40%; 20%-30% for 30%; 10%-20% for 20%; 0%-10% for 10%.

[0081] Construct a T2T genome-specific scoring system, highlighting the weight of core features:

[0082] Weighting (default, configurable):

[0083] Accuracy (10%); Completeness (10%); Continuity (35%); Telomere assessment (20%), Centromere assessment (20%), Collinearity assessment with reference genome (5%);

[0084] Scoring rules: Overall score = Σ (score of each dimension × corresponding weight);

[0085] Quality grade classification:

[0086] Excellent T2T grade (≥90 points): No gaps, complete and accurate telomeres / centromeres, all indicators meet the standards, suitable for high-precision research, such as in-depth analysis of centromere information in species; investigation of centromere methylation; investigation of transcriptionally active genes in the centromere region; obtaining more new genes and genetic variation information; clarifying the specific functions of new genes in growth, development, environmental response, etc. through gene knockout, overexpression and other experiments, combined with transcriptome, proteome and other multi-omics data; identifying previously unassembled SDs, analyzing the precise sequence, copy number, insertion position and evolution time of SDs, and combining epigenetic data to analyze the regulatory effects of SDs on the expression of neighboring genes, etc.

[0087] T2T level qualified (70-90 points): No gaps, core characteristics meet the standards, a few non-critical indicators deviate slightly, can be used for scientific research and applied research;

[0088] Quasi-T2T level (50-69 points): A small number of tiny gaps exist (total length ≤1kb). The core features meet the standards, but further optimization is needed. The small number of tiny gaps can be further improved by using TGS-GapCloser (GitHub - BGI-Qingdao / TGS-GapCloser: A gap-closing software tool that uses long reads to enhance genome assembly.) with ultra-long data. Telomere loss can be further improved by Telomere_expand (https: / / github.com / zskey-zn / Telomere_expand).

[0089] Unqualified (<50 points): There are obvious gaps or core features that do not meet the requirements of T2T genome.

[0090] 4. Report Generation Module

[0091] The system automatically integrates the results from each module to generate a T2T genome-specific assessment report, which includes:

[0092] Basic information: Sample name, sequencing platform, analysis time, species information;

[0093] T2T genome assessment details: specific indicators of accuracy, completeness, and continuity; telomere data and compliance status; centromere data and compliance status; compliance status of collinearity assessment with the reference genome;

[0094] Visualization charts: Chromosomal gap distribution diagram, BUSCO gene integrity bar chart, QV assessment error kmer distribution chart, telomere localization map, centromere localization map, and collinearity with reference genome;

[0095] Optimization suggestions: For indicators that do not meet the standards, provide specific optimization directions (such as supplementing long read length data coverage, adjusting assembly algorithm parameters, etc.).

[0096] Example 2

[0097] A method for multidimensional quality assessment of telomere-to-telomere genomes includes a data preprocessing module, an assessment module, a comprehensive scoring module, and a report generation module. The specific workflow is as follows:

[0098] 1. Data Preprocessing Module

[0099] Analysis results:

[0100] NGS Data Statistics

[0101]

[0102] HiFi Data Statistics

[0103]

[0104] ONT Data Statistics

[0105]

[0106] 2. Evaluation Module

[0107] (1) Continuity assessment:

[0108] N50 Statistics

[0109]

[0110] NGS data return ratio

[0111]

[0112] NGS data coverage

[0113]

[0114] HiFi data return ratio

[0115]

[0116] HiFi data return ratio

[0117]

[0118] ONT data return ratio

[0119]

[0120] ONT data return ratio

[0121]

[0122] AQI

[0123]

[0124] (2) Integrity assessment

[0125]

[0126] (3) Accuracy assessment

[0127]

[0128] (4) Telomere assessment

[0129]

[0130] The percentage is calculated as 11 / 12 * 100% = 91.6%.

[0131] (6) Assessment of collinearity with the reference genome

[0132] syri

[0133]

[0134] Percentage = Length_qry / Length_genome = 338754751 / 382799448 * 100% = 88.5%.

[0135] 3. Comprehensive Scoring Module

[0136] (1) Continuity assessment

[0137] Continuity score = N50 score + NGS data return ratio score + NGS data coverage score + HiFi data return ratio score + HiFi data coverage score + ONT data return ratio score + ONT data coverage score + AQI >90 score;

[0138] Sample continuity score = 10 + 10 + 10 + 10 + 10 + 10 + 30 = 100.

[0139] (2) Integrity assessment

[0140] Sample integrity assessment score = 100.

[0141] (3) Accuracy assessment

[0142] Sample accuracy score = 100.

[0143] (4) Telomere assessment

[0144] Telomere assessment score = 80.

[0145] (5) Centromere assessment

[0146] Centromere evaluation score = 90.

[0147] (6) Assessment of collinearity with the reference genome

[0148] Collinearity assessment score with reference genome = 90.

[0149] Sample overall score = 100*10% (accuracy) + 100*10% (completeness) + 100*35% (completeness) + 80*20% (telomere assessment) + 90*20% (centromere assessment) + 90*5% (collinearity assessment with reference genome) = 93.5.

[0150] Comparative Example 1

[0151] Other genomic assessment procedures

[0152] QUAST:

[0153] The QUAST software (V5.3.0) was used to assess the quality of the genome, but only statistical information such as GC and N50 could be obtained. The results are shown below:

[0154]

[0155] BUSCO

[0156] The completeness assessment results obtained using BUSCO are shown below:

[0157]

[0158] Compared to Comparative Example 1, N50 arranges all assembled contigs in descending order of length. When the cumulative length reaches 50% of the total length, the length of the current contig only focuses on the physical length of the genome. This invention not only uses the traditional N50 for continuity assessment but also uses NGS, HiFi, and ONT data return ratios, NGS, HiFi, and ONT data coverage, and AQI for genome continuity assessment. It removes regions with potential erroneous connections (verified by Hi-C or long reads) based on the original data coverage, paying more attention to biological authenticity and physical length, making continuity assessment more reliable. In addition, this invention adds accuracy assessment, telomere assessment, centromere assessment, and collinearity assessment with the reference genome, making it more suitable for T2T genome quality assessment needs.

Claims

1. A method for multidimensional quality assessment of telomere-to-telomere genome, characterized in that, Includes the following steps: Genomic data to be quality assessed are preprocessed to obtain preprocessed data; The preprocessed data were quantitatively evaluated for several core features, including accuracy, completeness, continuity, and region specificity. The region specificity included telomere identification, centromere identification, and collinearity with the reference genome. The quantitative evaluation results of the core features are weighted to calculate a comprehensive score, and the quality of the genome to be evaluated is determined by the comprehensive score.

2. The method for multidimensional quality assessment of telomere-to-telomere genomes according to claim 1, characterized in that, The genomic data to be quality assessed includes T2T genome assembly results, raw sequencing data, and species-specific reference information.

3. The method for multidimensional quality assessment of telomere-to-telomere genomes according to claim 1, characterized in that, The data preprocessing includes using fastp second-generation NGS data for data statistics, filtering, and format conversion, and using seqkit to detect read length distribution and average quality value of Nanopore data and PacBio HiFi data.

4. The method for multidimensional quality assessment of telomere-to-telomere genomes according to claim 1, characterized in that, The continuity assessment metrics include N50, NGS, HiFi, ONT data return ratio and coverage, and AQI value; the integrity assessment metrics include the proportion of complete single-copy BUSCO genes, the proportion of duplicate genes, and the proportion of deleted genes; the accuracy assessment metric is QV; the assessment method is to assess genomic QV using HiFi database construction via meryl.

5. The method for multidimensional quality assessment of telomere-to-telomere genomes according to claim 1, characterized in that, The evaluation metrics for telomeres include telomere sequence integrity; the evaluation metrics for centromeres are centromere sequence integrity, and the evaluation method is to identify the location of centromere regions after annotating whole-genome repetitive regions using TRF+RepeatMasker+SRF tools.

6. The method for multidimensional quality assessment of telomere-to-telomere genomes according to claim 1, characterized in that, The evaluation metrics for collinearity with the reference genome include the percentage of collinear region length.

7. The method for multidimensional quality assessment of telomere-to-telomere genomes according to claim 1, characterized in that, The weighted comprehensive score is calculated as follows: accuracy 10%; completeness 10%; continuity 35%; telomere identification 20%; centromere identification 20%; and collinearity assessment with the reference genome 5%.

8. A device for multidimensional quality assessment of telomere-to-telomere genome, characterized in that, The device includes: The data preprocessing module is used to preprocess the data of each gene in the genome to be quality assessed, and obtain the preprocessed data. The evaluation module is used to quantitatively evaluate several core features of the preprocessed data, including accuracy, completeness, continuity, and region specificity; the region specificity includes telomere identification, centromere identification, and collinearity with the reference genome. The comprehensive scoring module is used to calculate a comprehensive score based on the quantitative evaluation results of the core features according to their weights, and to determine the quality of the genome to be evaluated based on the comprehensive score.

9. The apparatus for telomere-to-telomere genome multidimensional quality assessment according to claim 8, characterized in that, The apparatus for telomere-to-telomere genome multidimensional quality assessment also includes a report generation module, which integrates the results of the data preprocessing module, the assessment module, and the comprehensive scoring module to generate a telomere-to-telomere genome multidimensional quality assessment report.

10. A computer-readable storage medium, characterized in that, The storage medium includes a stored program, wherein, when the program is executed, the device containing the storage medium is controlled to perform the method for telomere-to-telomere genome multidimensional quality assessment as described in any one of claims 1-7.

Images