Quality control method and device for clinical test sample, electronic equipment and storage medium

Abstract

The application discloses a quality control method and device for clinical detection samples, electronic equipment and a storage medium, and belongs to the technical field of high-throughput sequencing technology. The quality control method for the clinical detection samples comprises the following steps: performing quality control filtering on an original FASTQ data file to obtain a target FASTQ data file; performing alignment on the target FASTQ data file and a reference genome sequence file to obtain a BAM file; determining an alignment rate, rRNA content and globin RNA content according to the BAM file; performing sequencing depth detection by using the BAM file to obtain gene 3' end sequencing depth and gene 5' end sequencing depth; performing consistency checking on multiple sequencing data obtained by using multiple sequencing strategies on the clinical detection samples to obtain a consistency checking result; and comprehensively judging whether the clinical detection samples are qualified. The application can accurately detect the quality of the clinical detection samples and improve the effectiveness of the data.

Description

Technical Field

[0001] This application relates to the field of high-throughput sequencing technology, and in particular to a quality control method, apparatus, electronic device and storage medium for clinical test samples. Background Technology

[0002] RNA sequencing of blood samples can help analyze human physiology and health status. However, in practical applications, the quality of clinical test samples is affected by a variety of factors. When clinical test samples are contaminated or the procedures for extracting clinical test samples are not standardized, the quality of clinical test samples is poor.

[0003] Therefore, how to accurately detect the quality of clinical test samples and improve the validity of the data is a technical problem that needs to be solved by those skilled in the art. Summary of the Invention

[0004] The purpose of this application is to provide a quality control method for clinical test samples, a quality control device for clinical test samples, an electronic device, and a storage medium, which can accurately detect the quality of clinical test samples and improve the validity of the data.

[0005] To address the aforementioned technical problems, this application provides a quality control method for clinical test samples, which includes:

[0006] RNA sequencing was performed on clinical test samples to obtain raw FASTQ data files, and the raw FASTQ data files were then subjected to quality control filtering to obtain target FASTQ data files;

[0007] The target FASTQ data file is compared with the reference genome sequence file to obtain the BAM file;

[0008] The alignment rate, rRNA (ribosomal RNA) content, and globin RNA (globin RNA) content were determined based on the BAM file.

[0009] The sequencing depth was determined using the BAM file to obtain the sequencing depth at the 3' end and the sequencing depth at the 5' end of the gene.

[0010] The consistency test results were obtained by performing multiple sequencing data from the clinical test samples using various sequencing strategies.

[0011] The quality control results of the clinical test samples are generated based on the alignment rate, rRNA content, globin RNA content, gene 3' end sequencing depth, gene 5' end sequencing depth, and the consistency test results.

[0012] Optionally, the original FASTQ data file is subjected to quality control filtering to obtain the target FASTQ data file, including:

[0013] Remove the target connector sequence and low-quality sequence from the original FASTQ data file to obtain the target FASTQ data file;

[0014] The low-quality sequences include sequences with quality values ​​lower than a quality value threshold and sequences with lengths less than a length threshold.

[0015] Optionally, the target FASTQ data file is aligned with a reference genome sequence file to obtain a BAM file, including:

[0016] Obtain the reference genome sequence file in FASTA format, create an index sequence for the reference genome sequence file, and align the target FASTQ data file with the index sequence to obtain the BAM file.

[0017] Optionally, determining the alignment rate based on the BAM file includes:

[0018] The target sequence aligned to the reference genome sequence file is determined based on the BAM file;

[0019] The proportion of the target sequence to all sequences in the target FASTQ data file is set as the alignment rate.

[0020] Optionally, after determining the alignment rate based on the BAM file, the method further includes:

[0021] If the comparison rate is less than the first threshold, a quality control result is generated determining that the clinical test sample is unqualified.

[0022] Optionally, before determining the rRNA content and globin RNA content based on the BAM file, the method further includes:

[0023] Determine the rRNA sequence coordinates and globin RNA sequence coordinates in the reference genome sequence file;

[0024] The levels of control rRNA and control globin RNA in N control blood samples were statistically analyzed based on the rRNA sequence coordinates and the globin RNA sequence coordinates.

[0025] The first standard content range was determined based on the control rRNA content of N control blood samples.

[0026] The second standard content range was determined based on the control globin RNA content of N control blood samples.

[0027] Accordingly, after determining the rRNA content and globin RNA content based on the BAM file, the process also includes:

[0028] If the rRNA content is not within the first standard content range, a quality control result is generated to determine that the clinical test sample is unqualified;

[0029] If the globin RNA content is not within the second standard content range, a quality control result is generated indicating that the clinical test sample is unqualified.

[0030] Optionally, before performing sequencing depth alignment of the BAM file using control blood samples and genome annotation files, the method further includes:

[0031] Genomic annotation files were used to obtain gene 3' end position information and gene 5' end position information;

[0032] The average sequencing depth of the gene 3' end in M ​​control blood samples was determined based on the gene 3' end position information, and the average sequencing depth of the gene 5' end in M ​​control blood samples was determined based on the gene 5' end position information.

[0033] The ratio of the average sequencing depth of the 3' end of the gene to the average sequencing depth of the 5' end of the gene corresponding to the same gene is set as the feature value;

[0034] The KMeans algorithm is used to model the feature values ​​to obtain a clustering model, and the number of clusters in the clustering model is determined according to the elbow method.

[0035] The distance threshold is determined based on the clustering model;

[0036] Accordingly, after using the BAM file to perform sequencing depth detection and obtain the sequencing depth of the 3' end and 5' end of the gene, the process also includes:

[0037] The ratio of the sequencing depth at the 3' end of the gene to the sequencing depth at the 5' end of the gene is set as the sample feature value;

[0038] Calculate the Euclidean distance between the sample feature values ​​and the cluster centers of the clustering model;

[0039] If the Euclidean distance is greater than the distance threshold, a quality control result is generated indicating that the clinical test sample is unqualified.

[0040] Optionally, the quality control results of the clinical test samples are generated based on the alignment rate, the rRNA content, the globin RNA content, the sequencing depth of the gene 3' end, the sequencing depth of the gene 5' end, and the consistency test results, including:

[0041] Calculate the first score corresponding to the comparison rate;

[0042] Calculate the second score corresponding to the rRNA content;

[0043] Calculate the third score corresponding to the globin RNA content;

[0044] Calculate the fourth score corresponding to the sequencing depth of the 3' end and the sequencing depth of the 5' end of the gene;

[0045] Calculate the fifth score corresponding to the consistency test result;

[0046] The sum of the first score, the second score, the third score, the fourth score, and the fifth score is set as the total sample score, and the quality control results of the clinical test sample are generated based on the total sample score.

[0047] This application also provides a quality control device for clinical test samples, the device comprising:

[0048] The raw data quality control module is used to perform RNA sequencing on clinical test samples to obtain raw FASTQ data files, and to perform quality control filtering on the raw FASTQ data files to obtain target FASTQ data files;

[0049] The mapping file construction module is used to align the target FASTQ data file with the reference genome sequence file to obtain a BAM file;

[0050] The alignment rate detection module is used to determine the alignment rate based on the BAM file;

[0051] The target RNA content statistics module is used to determine the rRNA content and globin RNA content based on the BAM file.

[0052] The sequencing depth statistics module is used to perform sequencing depth detection using the BAM file to obtain the sequencing depth at the 3' end and the sequencing depth at the 5' end of the gene.

[0053] The consistency verification module is used to perform consistency verification on the various sequencing data obtained by multiple sequencing strategies of the clinical test sample, and obtain the consistency verification results.

[0054] The judgment module is used to generate quality control results for the clinical test samples based on the alignment rate, rRNA content, globin RNA content, gene 3' end sequencing depth, gene 5' end sequencing depth, and consistency test results.

[0055] This application also provides a storage medium storing a computer program thereon, which, when executed, implements the steps of the above-described quality control method for clinical test samples.

[0056] This application also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor invokes the computer program in the memory to implement the steps of the above-described quality control method for clinical test samples.

[0057] This application provides a quality control method for clinical test samples, comprising: performing RNA sequencing on the clinical test samples to obtain a raw FASTQ data file; performing quality control filtering on the raw FASTQ data file to obtain a target FASTQ data file; aligning the target FASTQ data file with a reference genome sequence file to obtain a BAM file; determining the alignment rate, rRNA content, and globin RNA content based on the BAM file; performing sequencing depth detection using the BAM file to obtain the 3' end sequencing depth and 5' end sequencing depth of the genes; performing consistency checks on multiple sequencing data obtained from the clinical test samples using multiple sequencing strategies to obtain consistency check results; and generating quality control results for the clinical test samples based on the alignment rate, rRNA content, globin RNA content, 3' end sequencing depth, 5' end sequencing depth, and consistency check results.

[0058] This application performs quality control filtering on the original FASTQ data file corresponding to the clinical test sample and compares it with the reference genome sequence file to obtain the BAM file. Based on the BAM file, the alignment rate, rRNA content, and globinRNA content can be determined. This application also uses a control blood sample and a genome annotation file to compare the sequencing depth of the BAM file, obtaining the 3' end sequencing depth and 5' end sequencing depth of the clinical test sample. This application further performs consistency checks on various sequencing data obtained from different sequencing strategies for the clinical test sample, obtaining consistency check results. This application uses the alignment rate, rRNA content, globin RNA content, 3' end sequencing depth, 5' end sequencing depth, and consistency check results as statistical indicators to comprehensively determine whether the clinical test sample is qualified. The above process detects the quality of the clinical test sample from multiple aspects and combines multiple statistical indicators to comprehensively determine whether the clinical test sample is qualified. Therefore, this application can accurately detect the quality of clinical test samples and improve the validity of the data. This application also provides a quality control device for clinical test samples, a storage medium, and an electronic device, which have the above-mentioned beneficial effects, and will not be elaborated further here. Attached Figure Description

[0059] To more clearly illustrate the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0060] Figure 1 A flowchart illustrating a quality control method for clinical test samples provided in this application embodiment;

[0061] Figure 2 A box plot of rRNA content statistics provided in an embodiment of this application;

[0062] Figure 3 A box plot of globin RNA content statistics provided in an embodiment of this application;

[0063] Figure 4 A schematic diagram illustrating an elbow method for determining the optimal number of clusters, provided in an embodiment of this application;

[0064] Figure 5 This is a schematic diagram of an abnormal sample provided in an embodiment of this application;

[0065] Figure 6 A flowchart for quality control of RNAseq in blood samples provided in this application embodiment;

[0066] Figure 7 This is a schematic diagram of the structure of a quality control device for clinical test samples provided in an embodiment of this application. Detailed Implementation

[0067] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0068] Please see below. Figure 1 , Figure 1 This is a flowchart of a quality control method for clinical test samples provided in an embodiment of this application.

[0069] Specific steps may include:

[0070] S101: RNA sequencing is performed on clinical test samples to obtain raw FASTQ data files, and the raw FASTQ data files are filtered for quality control to obtain target FASTQ data files;

[0071] This embodiment can be applied to servers capable of performing gene processing tasks. The aforementioned clinical test sample refers to a blood sample used for RNA sequencing. Prior to this step, there may be operations that use the aforementioned clinical test sample to perform RNA sequencing according to various sequencing strategies.

[0072] The raw FASTQ data file corresponding to the clinical test samples obtained in this step is the sequence information stored in FASTQ format. Because some low-quality sequences and adapter sequences exist in the raw FASTQ data file, this step performs quality control filtering on the raw FASTQ data file to obtain the target FASTQ data file.

[0073] S102: Align the target FASTQ data file with the reference genome sequence file to obtain the BAM file;

[0074] Prior to this step, a reference genome sequence file in FASTA format is downloaded from a public gene database. The target FASTQ data file is then aligned with the reference genome sequence file to obtain a BAM file, which is a mapping file.

[0075] S103: Determine the alignment rate, rRNA content, and globin RNA content based on the BAM file;

[0076] Based on the obtained BAM file, this step determines the alignment rate, rRNA content, and globin RNA content. The alignment rate describes the percentage of sequences in the target FASTQ data file that align to the reference genome sequence file out of the total sequence.

[0077] S104: Use the BAM file to detect sequencing depth and obtain the sequencing depth of the 3' end and the sequencing depth of the 5' end of the gene.

[0078] The reference genome contains multiple genes. By using BAM files to detect sequencing depth, the sequencing depth of the 3' end and the sequencing depth of the 5' end of each gene can be obtained.

[0079] S105: Perform consistency testing on the various sequencing data obtained from the clinical test samples using multiple sequencing strategies to obtain consistency testing results;

[0080] Prior to this step, sequencing operations can be performed using the aforementioned clinical test samples according to multiple sequencing strategies. This embodiment can obtain sequencing data corresponding to each sequencing strategy and perform a consistency check on all sequencing data. If the consistency of the sequencing data obtained from all sequencing strategies is greater than a preset value (e.g., 95%), the consistency check is considered passed; otherwise, the consistency check is considered failed.

[0081] S106: Generate the quality control results of the clinical test samples based on the alignment rate, the rRNA content, the globin RNA content, the sequencing depth of the gene 3' end, the sequencing depth of the gene 5' end, and the consistency test results.

[0082] In this embodiment, alignment rate, rRNA content, globin RNA content, gene 3' end sequencing depth, gene 5' end sequencing depth, and consistency test results can be used as statistical indicators. A comprehensive judgment is made by combining all statistical indicators. This embodiment can also upload these statistical indicators to a human-machine interface for user decision-making, and output a quality control result indicating whether the clinical test sample is qualified based on the decision. The quality control result can indicate that the clinical test sample is qualified or unqualified. As a feasible implementation method, if the clinical test sample is unqualified, it can be determined that the sample is contaminated or the sample extraction operation is not standardized.

[0083] This embodiment performs quality control filtering on the original FASTQ data file corresponding to the clinical test sample and compares it with the reference genome sequence file to obtain the BAM file. Based on the BAM file, the alignment rate, rRNA content, and globinRNA content can be determined. This embodiment also uses a control blood sample and a genome annotation file to align the sequencing depth of the BAM file, obtaining the 3' end sequencing depth and 5' end sequencing depth of the clinical test sample. This embodiment further performs consistency checks on various sequencing data obtained from different sequencing strategies for the clinical test sample, obtaining the consistency check results. This embodiment uses the alignment rate, rRNA content, globinRNA content, 3' end sequencing depth, 5' end sequencing depth, and the consistency check results as statistical indicators to comprehensively determine whether the clinical test sample is qualified. The above process detects the quality of the clinical test sample from multiple aspects and combines multiple statistical indicators to comprehensively determine whether the clinical test sample is qualified. Therefore, this embodiment can accurately detect the quality of clinical test samples and improve the effectiveness of the data.

[0084] As for Figure 1 In a further description of the corresponding embodiment, this embodiment can perform quality control filtering on the original FASTQ data file in the following way: remove the target connector sequence and low-quality sequences from the original FASTQ data file to obtain the target FASTQ data file; wherein, the low-quality sequences include sequences with quality values ​​lower than a quality value threshold and sequences with lengths less than a length threshold.

[0085] Some data in the original FASTQ data file are of low quality, which can affect the accuracy of subsequent data analysis and therefore needs to be removed. Additionally, the original FASTQ data file contains adapter sequences, which are not target sequences and also need to be removed.

[0086] The quality control operation for the raw FASTQ data file specifically includes the following steps A1 and A2:

[0087] Step A1: Obtain the raw FASTQ data file.

[0088] The raw FASTQ data file stores sequence information in FASTQ format, with each read's information divided into 4 lines; below is a real sequencing data from the Illumina platform, containing information for one read:

[0089] Line 1: @ST-E00126:128:HJFLHCCXX:2:1101:7405:1133;

[0090] Line 2: TTGCAAAAAATTTCTCTCATTNTGTAGGTTGCCTGTTCA CTCTGATGATAGTTTGNTTTGG;

[0091] Line 3: +

[0092] Line 4: FFKKKFKKFKF <KK<F,AFKKKKK7FFK77<FKK,<F7K,,7AF<FF7FKK7AA,7<FA;

[0093] The first line primarily stores information such as coordinates during sequencing. The second line contains the sequence information, typically represented by ATCGN, where N indicates fluorescence interference that makes it impossible to determine which base it is. The third line begins with a "+" sign and can store additional information; it is usually empty. The fourth line stores quality information, corresponding one-to-one with the base sequence in the second line. The ASCII value corresponding to each symbol is called the phred value, which can be understood as the quality value of the base at that position; a higher value indicates better sequencing quality.

[0094] The following explains the phred value: When the sequencer performs sequencing, it automatically provides a reference sequencing error probability (P) based on the intensity of the fluorescence signal. In this embodiment, P is taken as log10 and then multiplied by -10, resulting in Q. For example, if P = 1%, then the corresponding Q = -10 * log10(0.01) = 20. This Q is converted to a new value by adding 33 or 64, called phred. Finally, the ASCII character corresponding to phred is mapped to this base. For example, if Q = 20, phred = 20 + 33 = 53, and the corresponding symbol is "5". This embodiment uses the sequencing error probability to represent the sequencing quality, and the ASCII value of each symbol represents the quality of each base.

[0095] Step A2: Use the software cutadapt to remove the connector sequences from the original FASTQ data file based on the provided connector sequences, and remove low-quality sequences based on the set quality threshold.

[0096] The command line is as follows:

[0097] cutadapt-a adapter1-A adapter2-e 0.1-O 1-m 36--max-n 5-q 30,30-j nt-oSample.trim.R1.fastq.gz-p Sample.trim.R2.fastq.gz R1_raw R2_raw;

[0098] The parameters in the command line above are explained as follows:

[0099] -a: Connector 1 sequence;

[0100] -A: Connector 2 sequence;

[0101] -e: Maximum allowable mismatch rate for connector sequence matching;

[0102] -m: Select which sequences to keep based on the shortest length;

[0103] -O: The adapter and reads must have minimal overlap for successful recognition;

[0104] --max-n: The number of N values ​​in a sequence. If the value is greater than this, the sequence will be discarded.

[0105] -q: Threshold for pruning low-quality bases. If there is only one value, only the 3' end is pruned. If there are two values ​​separated by a comma, both the 5' and 3' ends are pruned.

[0106] -o: First output file;

[0107] -p: Output of the second reads;

[0108] R1_raw: The input file for the first read;

[0109] R2_raw: The input file for the second read;

[0110] The input file for the above process is the original FASTQ data file; the software used is cutadapt sequence quality control software; and the output file is the quality-controlled file (i.e., the target FASTQ data file).

[0111] The two ends of a sequence are called the 5' end and the 3' end. By default, low-quality bases at both ends are pruned. In some cases, only one end is pruned. If only one threshold is provided for the -q parameter, only the 3' end will be pruned. If two thresholds are provided, both ends of the sequence will be pruned. The purpose of pruning the 3' and 5' ends is to remove low-quality sequences and obtain high-quality sequences.

[0112] As for Figure 1 A further description of the corresponding embodiment is that the BAM file can be constructed in the following manner: obtain the reference genome sequence file in FASTA format, create an index sequence for the reference genome sequence file, align the target FASTQ data file with the index sequence, and obtain the BAM file (i.e., the mapping file).

[0113] Specifically, the above process can use the reference genome sequence file and the target FASTQ data file to perform whole-genome alignment to generate a BAM file, and then obtain the alignment rate and count the rRNA content and globin RNA content through the BAM file.

[0114] The process of building the BAM file mentioned above includes the following steps B1 to B3:

[0115] Step B1: Download the FASTA format reference genome sequence file (hereinafter referred to as ref.fa) from public gene databases (NCBI, UCSC, Ensembl, etc.);

[0116] Step B2: Input the reference genome sequence file ref.fa, and use sequence alignment software to create an alignment index for the reference genome. The generated related files include ref.fa.fai, ref.fa.amb, ref.fa.ann, ref.fa.bwt, ref.fa.pac, and ref.fa.sa.

[0117] Step B3: Using an alignment tool, the FASTQ data of the clinical sample and the index sequence obtained in step B are used to perform sequence alignment with the reference genome to obtain the original alignment BAM file of the sample.

[0118] The input files for the above process include a reference genome sequence file and a target FASTQ data file. The pipeline software used includes sequence indexing and alignment software. The output files include a reference genome sequence file, an index file, and a BAM file.

[0119] As for Figure 1 A further description of the corresponding embodiment suggests that the alignment rate can be determined as follows: The target sequence aligned to the reference genome sequence file is determined based on the BAM file; the proportion of the target sequence to all sequences in the target FASTQ data file is set as the alignment rate. After determining the alignment rate based on the BAM file, if the alignment rate is less than a first threshold, a quality control result indicating that the clinical test sample is unqualified can be generated.

[0120] This embodiment can detect abnormal samples based on the alignment rate. For example, the alignment rate of a sample can be used to determine whether a sample is contaminated during library construction and sequencing. Specifically, this embodiment can obtain the BAM file of the sample to be tested and use publicly available software to calculate the percentage of sequences aligned to the human reference genome out of the total sequence. Based on experience, if the alignment rate of the sample is less than 90%, the sample is considered contaminated and needs to be reconstructed and sequenced.

[0121] As for Figure 1 In a further description of the corresponding embodiment, before determining the rRNA content and globin RNA content according to the BAM file, the following operations are performed: determining the rRNA sequence coordinates and globin RNA sequence coordinates in the reference genome sequence file; calculating the control rRNA content and control globin RNA content in N control blood samples according to the rRNA sequence coordinates and globin RNA sequence coordinates; determining a first standard content range based on the control rRNA content of the N control blood samples; and determining a second standard content range based on the control globin RNA content of the N control blood samples.

[0122] Accordingly, after determining the rRNA content and globin RNA content according to the BAM file, if the rRNA content is not within the first standard content range, a quality control result is generated indicating that the clinical test sample is unqualified; if the globin RNA content is not within the second standard content range, a quality control result is generated indicating that the clinical test sample is unqualified.

[0123] For blood sample RNA sequencing data, rRNA and globin RNA can be manually removed during library construction. The adequacy of rRNA and globin RNA removal can be determined based on their levels. Abnormal sample detection based on rRNA and globin RNA levels includes the following steps C1 to C4:

[0124] Step C1: Download the reference genome sequence in FASTA format (hereinafter referred to as ref.fa) from public gene databases (NCBI, UCSC, Ensembl, etc.);

[0125] Step C2: Obtain the corresponding rRNA and globin RNA sequences from the downloaded human reference genome file, and generate a location information file by obtaining the coordinates of these sequences. Location information format: chromosome number: gene start coordinates - gene end coordinates;

[0126] Step C3: Obtain the reference confidence interval for rRNA & globin RNA content: Based on the location information file obtained in Step C2, use publicly available software to statistically analyze the expression percentages of rRNA and globin genes in 80 locally sourced blood RNA-seq control data, establishing a local rRNA & globin RNA expression database. Obtain the mean and confidence interval [mean ± 3 × std] of rRNA & globin RNA in the local control. Define an rRNA content percentage greater than 0.133 or a globin RNA content percentage greater than 0.132 as abnormal samples for rRNA & globin RNA experiments.

[0127] Step C4: Statistically analyze the rRNA and globin RNA content of clinical diagnostic test samples to determine whether the test samples meet the experimental removal requirements.

[0128] Please see Figure 2 and Figure 3 , Figure 2 This is a box plot of rRNA content statistics provided in an embodiment of this application. Figure 3 This is a box plot of globin RNA content statistics provided in an embodiment of this application. Figure 2 and Figure 3 The vertical axis represents the sample name, the horizontal axis rRNA_freq represents the rRNA content, and the horizontal axis globin_freq represents the globin RNA content.

[0129] The input files in the above embodiments include BAM files, rRNA and globin RNA location information files, and the software used is Reads software, which extracts mapping files based on location information. The output information includes statistical results of rRNA and globin RNA content.

[0130] As a feasible implementation, before performing sequencing depth alignment of the BAM file using control blood samples and a genome annotation file, the following operations are performed: obtaining gene 3' end position information and gene 5' end position information using the genome annotation file; determining the average sequencing depth of the gene 3' end in M ​​control blood samples based on the gene 3' end position information, and determining the average sequencing depth of the gene 5' end in M ​​control blood samples based on the gene 5' end position information; setting the ratio of the average sequencing depth of the gene 3' end to the average sequencing depth of the gene 5' end corresponding to the same gene as a feature value; using the KMeans algorithm to model the feature value to obtain a clustering model, and determining the number of clusters in the clustering model using the elbow method; and determining a distance threshold based on the clustering model.

[0131] Accordingly, after using the BAM file to perform sequencing depth detection and obtain the sequencing depth of the 3' end and the sequencing depth of the 5' end of the gene, the ratio of the sequencing depth of the 3' end to the sequencing depth of the 5' end of the gene can be set as a sample feature value; the Euclidean distance between the sample feature value and the cluster center of the clustering model is calculated; if the Euclidean distance is greater than the distance threshold, a quality control result is generated to determine that the clinical test sample is unqualified.

[0132] Specifically, by analyzing the sequencing depth differences between the 3' and 5' exons of disease-associated genes and the consistency analysis of experimental RNA integrity, possible heterogeneity caused by the library construction process is identified, and false negatives in diagnosis caused by the experiment are minimized as much as possible. Therefore, the abnormal sample detection based on the sequencing depth difference analysis of the 3' and 5' ends of genes includes the following steps D1 to D7:

[0133] Step D1: Download genome annotation files from public databases, collect disease-related gene files, and obtain the first 50 bp of the first exon of each gene as the 5' end position information and the last 50 bp of the last exon as the 3' end position information;

[0134] Step D2: Using the BAM files of 80 local control blood samples and the location information files obtained in step D1, the average sequencing depth of the 3' and 5' ends of different genes in each sample was calculated, and the ratio of the average sequencing depth of the 3' and 5' ends was used as a standard to measure the difference. The larger the ratio, the greater the difference in sequencing depth between the 3' and 5' ends. A total of 4658 feature values ​​were obtained.

[0135] Step D3: Feature selection and model training. The 4658 feature values ​​were filtered out, removing features with very low sequencing depth at the 3' and 5' ends, resulting in 1933 features. The KMeans algorithm module in Python was then used to model the data.

[0136] Step D4: Select the optimal number of clusters based on the elbow plot method. KMeans uses minimizing the squared error between samples and data points as its objective function. The sum of the squared distance errors between data points in each cluster and the samples within that cluster is called the distortion. For a cluster, lower distortion indicates a more compact cluster structure, while higher distortion indicates a looser cluster structure. Distortion decreases with increasing number of clusters, but for data with a certain degree of discriminative power, distortion improves significantly at a certain critical point before slowly decreasing. This critical point can be considered a point with good clustering performance. Statistically, a 2-cluster ratio is appropriate. Please refer to [link to relevant documentation]. Figure 4 , Figure 4 This is a schematic diagram illustrating an elbow method for determining the optimal number of clusters, provided in an embodiment of this application. Figure 4 The vertical axis, score, represents the degree of distortion; the horizontal axis, number clusters, represents the number of clusters; and 'e' indicates scientific notation. Figure 4 This represents the distortion value calculated when different numbers of clusters are selected.

[0137] Step D5: Threshold determination;

[0138] Based on the model obtained in step D4, calculate the Euclidean distance from each sample to the cluster center based on the feature values. Set the proportion of outliers to 1%, because in the case of a standard normal distribution (N(0,1)), we generally consider data outside 3 standard deviations as outliers, and data within 3 standard deviations contain more than 99% of the data in the dataset, so the remaining 1% of data can be considered outliers; based on the Euclidean distances calculated for all samples and the proportion of outliers, determine the threshold to be 81.18.

[0139] Step D6: Outlier detection, mainly through the following steps;

[0140] Calculate the feature values ​​of the clinical test samples, use the model obtained in step D to predict the cluster centers of the clinical test samples, calculate the Euclidean distance from the sample to the cluster center, and if the Euclidean distance of the sample is greater than 81.18, the clinical sample is considered abnormal.

[0141] Step D7: Visualization of abnormal samples.

[0142] Please see Figure 5 , Figure 5This is a schematic diagram illustrating an abnormal sample provided in an embodiment of this application. Figure 5 The ordinate 'distance' represents the Euclidean distance, and 'sample index' represents the sample index, such as... Figure 5 The sample indices MS22050748P1 and MS20041621P1 shown are abnormal samples. Anomaly1 is used to mark abnormal samples, with 0 indicating normal and 1 indicating abnormal.

[0143] The input files for the above embodiments include BAM files for control samples and test samples, and genome annotation files; the software used includes reads software for extracting mapping files based on location information, and clustering software; the output information includes: 5' and 3' end location information files for different genes, 5' and 3' sequencing depth files for different genes, and Euclidean distance files for each sample.

[0144] When a sample is sequenced using multiple strategies, consistency analysis is required. Subsequent analysis can only proceed after sample consistency is confirmed. As a feasible implementation method, the following approach can be used for data consistency analysis: Obtain RNA FASTQ and DNA FASTQ data of the sample to be tested; calculate the consistency between different data using publicly available software, as follows:

[0145] python ncm_fastq.py-l sample.list-O fastq_check-N Matched_list-p 4-ptSNP.pt;

[0146] The parameters are explained below:

[0147] -l indicates the input file for sample data, as shown in the example below:

[0148] / data / LSJ_R1.fastq

[0149] / data / LSJ_R2.fastq LSJ

[0150] / data / LSH_R1.fastq

[0151] / data / LSH_R2.fastq LSH

[0152] -O indicates the selected data type. In this example, FASTQ data is used, so fastq_check is selected.

[0153] -N indicates the prefix for the output file;

[0154] -p indicates the number of threads used by the software;

[0155] -pt indicates the database file included with the software;

[0156] Determine whether the samples are consistent based on the final calculation result file. If it shows "match", it means that the two data are the same sample.

[0157] The aforementioned different data refers to the different sequencing data obtained when a sample is subjected to multiple sequencing strategies, such as sequencing DNA and RNA data for the same sample.

[0158] After obtaining the alignment rate, rRNA content, globin RNA content, gene 3' end sequencing depth, gene 5' end sequencing depth, and consistency test results, this implementation can determine whether the sample is qualified in the following ways:

[0159] If the sample comparison rate is less than 90%, it is preliminarily determined that the sample may be contaminated by other species, and the sample is judged as unqualified.

[0160] If the rRNA content of a sample is greater than 0.133 or the globin RNA content is greater than 0.132, it indicates an abnormality in the rRNA and / or globin RNA content. This may result in the abnormal signal of the gene that is not expressed at low levels being undetectable due to the low amount of effective data. If the final amount of effective data is less than 12G, the library needs to be reconstructed.

[0161] Based on the anomaly detection of 5' and 3' sequencing depth, the Euclidean distance from the clinical test sample features to the cluster center is calculated. If it is greater than 81.1, it indicates that the proportion of 5' and 3' sequencing is significantly higher than that of the control group, which may be due to abnormal RNA quality or library construction. It is recommended to resample or construct a new library.

[0162] If the consistency test shows that multiple sequencing data sources are consistent, subsequent data analysis can be carried out. If they are inconsistent, the cause needs to be further investigated and a remedial plan needs to be determined.

[0163] Furthermore, in this embodiment, the quality control results of the clinical test samples can also be generated in the following ways: calculating the first score corresponding to the alignment rate; calculating the second score corresponding to the rRNA content; calculating the third score corresponding to the globin RNA content; calculating the fourth score corresponding to the sequencing depth at the 3' end of the gene and the sequencing depth at the 5' end of the gene; calculating the fifth score corresponding to the consistency test result; setting the sum of the first score, the second score, the third score, the fourth score, and the fifth score as the total sample score, and generating the quality control results of the clinical test samples based on the total sample score.

[0164] The first score is determined based on the deviation of the alignment rate from the standard alignment rate; the second score is determined based on the deviation of the rRNA content from the standard rRNA content; the third score is determined based on the deviation of the globin RNA content from the standard globin RNA content; the fourth score is determined based on the deviation of the gene 3' end sequencing depth from the standard gene 3' end sequencing depth and the deviation of the gene 5' end sequencing depth from the standard gene 5' end sequencing depth; and the fifth score is determined based on the deviation of the consistency test result from the standard consistency test result. The greater the deviation, the lower the score. When the total sample score is greater than the preset total score, it is considered qualified; when the total sample score is less than or equal to the preset total score, it is considered unqualified.

[0165] The process described in the above embodiments is illustrated below through examples in practical applications.

[0166] RNA-seq has become a routine method for transcriptome analysis relying on high-throughput sequencing (HTS) technology, which provides deeper and more precise measurements at the transcriptional level than microarrays and other traditional gene expression analysis methods. It can be used to identify new transcripts, alternative splicing and gene fusion events, analyze allele-specific expression, and perform differential gene expression analysis. However, due to the inherent limitations of RNA sequencing technology, quality issues are very common in raw RNA sequence data. Besides the common quality problems found in various high-throughput sequencing data, such as sequencing data quality (read count, length distribution, base quality, sequence complexity, GC content, etc.) and contamination from other species, RNA-seq also suffers from some "RNA sequence-specific" quality issues, such as ribosomal RNA (rRNA) residues, RNA degradation, and differences in coverage between different sequences. Therefore, RNA sequencing data must undergo rigorous quality control (QC) according to the experimental objectives before downstream analysis to ensure the reliability of the results.

[0167] The widespread application of high-throughput genome sequencing in the diagnosis of genetic diseases has led to a continuous increase in variants of unknown clinical significance (VUS). According to the ACMG guidelines, VUS cannot be used as the basis for clinical decision-making, and more evidence is needed to investigate the functional impact of these variants. RNA sequencing (RNA-seq), as an effective complementary method to genome sequencing, can be used to interpret the significance of VUS variants and uncover possible transcriptional abnormalities. It has been widely applied to the molecular diagnosis of various Mendelian diseases (improving diagnostic rates by 7.5-36%). When performing RNA data analysis, it is necessary to evaluate all RNA-mediated events (alternative splicing, gene expression, gene fusion, ASE, etc.) to obtain comprehensive genetic information mechanisms and further obtain genetic diagnostic results. RNA-seq technology has gradually shown diagnostic value in specific tissues and diseases, such as for muscle biopsy tissue samples from patients with undiagnosed neuromuscular diseases and for the detection of fibroblasts from patients with mitochondrial diseases. However, both disease-related tissues and cultured fibroblasts are difficult to obtain. Peripheral blood, as a more readily available clinical sample, has shown varying degrees of improvement in molecular diagnostic rates when using blood tissue RNA-seq combined with WES / WGS analysis.

[0168] Peripheral blood is a readily available biological fluid, providing a rich source of information about human physiology and health. Eukaryotic cells contain several high-abundance RNAs, with ribosomal RNA (rRNA) being the most abundant, while globin RNA is the second most abundant RNA in blood. These two types of RNA account for more than 85% of the total RNA in blood samples. However, for the diagnosis of rare genetic diseases using RNA sequencing (RNA-Seq) technology, if sequencing is performed without pretreatment, the high-abundance RNA occupies a large portion of the data, wasting sequencing costs and obscuring the target RNA data. This is especially true for low-abundance but crucial target genes related to patient diseases and rare splicing abnormalities, making it difficult to detect significant abnormalities and thus affecting the sensitivity of clinical diagnosis. For blood sample RNA sequencing data, rRNA and globin RNA can be removed experimentally during library construction. The effectiveness of this removal can be evaluated through a specially designed quality control system.

[0169] RNA quality is influenced by multiple factors, and its integrity is crucial for the success of globin removal and downstream applications. Even moderate levels of RNA degradation can lead to insufficient removal of globin mRNA by capture reagents. Blood sample preservation, RNA purification methods, and residual RNase activity all affect RNA quality. The RIN value is commonly used for assessment (RIN is a parameter of RNA integrity, more reliable than 18S:28S, unaffected by sample concentration, indicating RNA quality, and providing predictions for sequencing data). Generally, sequencing requires an RNA RIN greater than 8, but some special samples and experiments require RINs greater than 7, or even lower values, for downstream experiments. Therefore, understanding RNA experiments and data is crucial for subsequent result interpretation, especially for analyzing the causes of abnormal results. Conventional RNA-seq sequencing often uses polyA for mRNA enrichment, randomly fragments the mRNA, and uses fragmented mRNA as templates and random oligonucleotides as primers for cDNA synthesis. The purified cDNA undergoes end repair, A-tailing, and sequencing adapter ligation. Amplification of appropriate fragment sizes yields sequencing libraries. This method leads to 3' end bias (the worse the RNA integrity, the worse the detection capability for sequences farther from the 3' end). Differential analysis of sequencing depth at the 3' and 5' exons of disease-associated genes can assess RNA integrity, the stability of RNA-seq library preparation and sequencing, and minimize false negatives in diagnosis caused by experiments. Furthermore, RNA-seq is often used as a supplement to DNA testing. When a sample undergoes multiple tests, it is necessary to ensure sample consistency and guarantee that the DNA and RNA data originate from the same individual.

[0170] Currently, many tools are available for QC of high-throughput sequencing data, such as FastQC, FASTX-Toolkit, QC-Chain, and NGS QC Toolkit. However, most of these focus primarily on adapting general high-throughput sequencing data rather than addressing specific RNA sequence QC issues. While some tools are specifically designed for RNA sequence data, they cannot perform quality control on hemoglobin mRNA, rRNA, disease-associated gene end coverage, and DNA and RNA sample consistency in blood tissues. Therefore, analysts urgently need a new technological approach that, based on high-throughput sequencing quality control, can further control the characteristics of blood samples to improve data quality, determine data validity, and increase the sensitivity of data analysis.

[0171] This embodiment proposes a novel technical solution for quality control of RNAseq in blood samples, mainly through the following five aspects: (1) Basic quality control of high-throughput sequencing (QC, alignment rate, data volume, etc.): obtaining high-quality target sequences by removing low-quality data and adapter sequences; (2) Quality control of rRNA and globin RNA content, judging whether the removal of rRNA and globin RNA during library construction and sequencing is qualified; (3) Genome alignment rate analysis: judging whether the data is contaminated; (4) Quality control of sequencing depth at the 5' and 3' ends of genes: judging the integrity of sample mRNA and the stability of library construction experiments; (5) Sample consistency test: judging whether there is sample confusion. Through a series of data quality controls, the robustness of experiments and data is fully understood, and the reliability of analysis results is determined.

[0172] The overall process of the technical solution in this embodiment includes the following steps: raw data quality control; mapping file construction; abnormal sample detection based on alignment rate; rRNA and globin RNA coordinate file construction; abnormal sample detection based on rRNA and globin RNA content; abnormal sample detection based on alignment rate; abnormal sample detection based on sequencing depth difference analysis at the 3' and 5' ends of the gene; data consistency analysis; and determining whether the sample quality control is qualified based on statistical results.

[0173] Please see Figure 6 , Figure 6 A flowchart for quality control of blood sample RNAseq provided in this application embodiment is shown in the figure, illustrating the following process:

[0174] (1) Perform data quality control on the original RNAseq FASTQ data (i.e., the original FASTQ data file) to obtain the quality-controlled FASTQ data (i.e., the target FASTQ data file);

[0175] (2) Generate an index file using the genome sequence file, and generate a BAM file by sequence alignment based on the index file and the quality-controlled FASTQ data;

[0176] (3) Calculate the comparison rate using BAM files;

[0177] (4) Use gene sequence files to construct rRNA and globin RNA coordinate files, and use rRNA and globin RNA coordinate files and BAM files to perform rRNA content statistics and globin RNA content statistics.

[0178] (5) Determine whether the FASTQ data after quality control contains both DNA and RNA data in order to perform data consistency verification;

[0179] (6) Gene 3' and 5' coordinate files are generated using the genome annotation file, and the sequencing depth of the 3' and 5' ends of the gene is statistically analyzed using the BAM file;

[0180] (7) The data quality control is judged comprehensively based on the alignment rate, rRNA content statistics, globin RNA content statistics, consistency test results and gene 3' and 5' sequencing depth statistics.

[0181] Conventional RNA-seq sequencing only performs quality control on the data, lacking a quality control process specific to different sample types. High levels of globin RNA and rRNA in blood can occupy valuable sequencing space, affecting the detection and quantification of normal RNA-seq analysis, thus impacting the sensitivity of clinical diagnosis. This embodiment uses statistical analysis of globin RNA and rRNA levels to determine whether the removal of globin RNA and rRNA during library preparation and sequencing is adequate.

[0182] Abnormal RNA quality or library construction can lead to uneven sequencing at the 5' and 3' ends during RNA-seq sequencing, resulting in false negatives. This embodiment uses abnormal sequencing depth at the 5' and 3' ends of genes to determine if there are abnormalities in RNA quality or library construction. Data contamination often occurs during RNA-seq sequencing due to various reasons; this embodiment uses statistical alignment rates to determine sample contamination. Sample confusion is a frequent problem during sequencing; this invention uses data consistency checks to determine sample consistency, thereby reducing unnecessary losses. The processing flow designed in this embodiment has low requirements for server computing resources; a typical 8-core, 64GB memory server can allow simultaneous processing of dozens of target genes. The workflow of this embodiment is simple to deploy and easy to use; only the relevant computing nodes need to be deployed to complete the entire analysis process.

[0183] Please see Figure 7 , Figure 7 This is a schematic diagram of the structure of a quality control device for clinical test samples provided in an embodiment of this application;

[0184] The device may include:

[0185] The raw data quality control module 701 is used to perform RNA sequencing on clinical test samples to obtain raw FASTQ data files, and to perform quality control filtering on the raw FASTQ data files to obtain target FASTQ data files.

[0186] The mapping file construction module 702 is used to align the target FASTQ data file with the reference genome sequence file to obtain a BAM file;

[0187] The alignment rate detection module 703 is used to determine the alignment rate based on the BAM file;

[0188] The target RNA content statistics module 704 is used to determine the rRNA content and globin RNA content based on the BAM file.

[0189] The sequencing depth statistics module 705 is used to perform sequencing depth detection using the BAM file to obtain the sequencing depth at the 3' end of the gene and the sequencing depth at the 5' end of the gene.

[0190] The consistency verification module 706 is used to perform consistency verification on the various sequencing data obtained by multiple sequencing strategies of the clinical test sample, and obtain the consistency verification result.

[0191] The judgment module 707 is used to generate the quality control result of the clinical test sample based on the alignment rate, the rRNA content, the globin RNA content, the sequencing depth of the gene 3' end, the sequencing depth of the gene 5' end, and the consistency test result.

[0192] This embodiment performs quality control filtering on the original FASTQ data file corresponding to the clinical test sample and compares it with the reference genome sequence file to obtain the BAM file. Based on the BAM file, the alignment rate, rRNA content, and globinRNA content can be determined. This embodiment also uses a control blood sample and a genome annotation file to align the sequencing depth of the BAM file, obtaining the 3' end sequencing depth and 5' end sequencing depth of the clinical test sample. This embodiment further performs consistency checks on various sequencing data obtained from different sequencing strategies for the clinical test sample, obtaining the consistency check results. This embodiment uses the alignment rate, rRNA content, globinRNA content, 3' end sequencing depth, 5' end sequencing depth, and the consistency check results as statistical indicators to comprehensively determine whether the clinical test sample is qualified. The above process detects the quality of the clinical test sample from multiple aspects and combines multiple statistical indicators to comprehensively determine whether the clinical test sample is qualified. Therefore, this embodiment can accurately detect the quality of clinical test samples and improve the effectiveness of the data.

[0193] Furthermore, the process by which the raw data quality control module 701 performs quality control filtering on the raw FASTQ data file to obtain the target FASTQ data file includes: removing the target connector sequence and low-quality sequences from the raw FASTQ data file to obtain the target FASTQ data file; wherein, the low-quality sequences include sequences with quality values ​​lower than a quality value threshold and sequences with lengths less than a length threshold.

[0194] Furthermore, the process by which the mapping file construction module 702 aligns the target FASTQ data file with the reference genome sequence file to obtain the BAM file includes: acquiring the reference genome sequence file in FASTA format, creating an index sequence for the reference genome sequence file, and aligning the target FASTQ data file with the index sequence to obtain the BAM file.

[0195] Furthermore, the alignment rate detection module 703 includes: determining the target sequence aligned to the reference genome sequence file based on the BAM file; and setting the proportion of the target sequence to all sequences in the target FASTQ data file as the alignment rate.

[0196] Furthermore, it also includes:

[0197] The comparison rate judgment module is used to generate a quality control result that determines the clinical test sample is unqualified if the comparison rate is less than a first threshold after the comparison rate is determined according to the BAM file.

[0198] Furthermore, it also includes:

[0199] Before determining the rRNA and globin RNA content based on the BAM file, the target RNA content statistics module 704 further performs the following operations: determining the rRNA sequence coordinates and globin RNA sequence coordinates in the reference genome sequence file; calculating the control rRNA and control globin RNA content in N control blood samples according to the rRNA sequence coordinates and the globin RNA sequence coordinates; determining a first standard content range based on the control rRNA content of the N control blood samples; and determining a second standard content range based on the control globin RNA content of the N control blood samples.

[0200] Accordingly, after determining the rRNA content and globin RNA content based on the BAM file, the operation performed by the target RNA content statistics module 704 further includes: if the rRNA content is not within the first standard content range, generating a quality control result indicating that the clinical test sample is unqualified; if the globin RNA content is not within the second standard content range, generating a quality control result indicating that the clinical test sample is unqualified.

[0201] Furthermore, before performing sequencing depth alignment of the BAM file using control blood samples and genome annotation files, the sequencing depth statistics module 705 further includes the following operations: obtaining gene 3' end position information and gene 5' end position information using the genome annotation file; determining the average sequencing depth of the gene 3' end in M ​​control blood samples based on the gene 3' end position information, and determining the average sequencing depth of the gene 5' end in M ​​control blood samples based on the gene 5' end position information; setting the ratio of the average sequencing depth of the gene 3' end to the average sequencing depth of the gene 5' end corresponding to the same gene as a feature value; using the KMeans algorithm to model the feature value to obtain a clustering model, and determining the number of clusters in the clustering model using the elbow method; and determining a distance threshold based on the clustering model.

[0202] Accordingly, after using the BAM file to perform sequencing depth detection and obtain the sequencing depth at the 3' end and 5' end of the gene, the sequencing depth statistics module 705 further performs the following operations: setting the ratio of the sequencing depth at the 3' end of the gene to the sequencing depth at the 5' end of the gene as a sample feature value; calculating the Euclidean distance between the sample feature value and the cluster center of the clustering model; and generating a quality control result that determines the clinical test sample to be unqualified if the Euclidean distance is greater than the distance threshold.

[0203] Furthermore, the process by which the judgment module 707 generates the quality control result of the clinical test sample based on the alignment rate, the rRNA content, the globin RNA content, the gene 3' end sequencing depth, the gene 5' end sequencing depth, and the consistency test result includes: calculating a first score corresponding to the alignment rate; calculating a second score corresponding to the rRNA content; calculating a third score corresponding to the globin RNA content; calculating a fourth score corresponding to the gene 3' end sequencing depth and the gene 5' end sequencing depth; calculating a fifth score corresponding to the consistency test result; setting the sum of the first score, the second score, the third score, the fourth score, and the fifth score as the total sample score; and generating the quality control result of the clinical test sample based on the total sample score.

[0204] Since the embodiments of the apparatus and the embodiments of the method correspond to each other, please refer to the description of the embodiments of the method for the embodiments of the apparatus, which will not be repeated here.

[0205] This application also provides a storage medium on which a computer program is stored, which, when executed, can perform the steps provided in the above embodiments. The storage medium may include various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0206] This application also provides an electronic device that may include a memory and a processor. The memory stores a computer program, and when the processor calls the computer program in the memory, it can implement the steps provided in the above embodiments. Of course, the electronic device may also include various network interfaces, power supplies, and other components.

[0207] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section. It should be noted that those skilled in the art can make various improvements and modifications to this application without departing from the principles of this application, and these improvements and modifications also fall within the protection scope of the claims of this application.

[0208] It should also be noted that, in this specification, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

Claims

1. A quality control method for clinical test samples, characterized in that, include: RNA sequencing was performed on clinical test samples to obtain raw FASTQ data files, and the raw FASTQ data files were then subjected to quality control filtering to obtain target FASTQ data files; The target FASTQ data file is compared with the reference genome sequence file to obtain the BAM file; The alignment rate, rRNA content, and globin RNA content were determined based on the BAM file. The sequencing depth was determined using the BAM file to obtain the sequencing depth at the 3' end and the sequencing depth at the 5' end of the gene. The consistency test results were obtained by performing multiple sequencing data from the clinical test samples using various sequencing strategies. The quality control results of the clinical test samples are generated based on the alignment rate, rRNA content, globin RNA content, gene 3' end sequencing depth, gene 5' end sequencing depth, and the consistency test results. The process of determining the rRNA content and the globin RNA content includes: using the rRNA and globin RNA coordinate files and the BAM file to perform rRNA content statistics and globin RNA content statistics to obtain the rRNA content and the globin RNA content.

2. The quality control method for clinical test samples according to claim 1, characterized in that, The original FASTQ data file is subjected to quality control filtering to obtain the target FASTQ data file, including: Remove the target connector sequence and low-quality sequence from the original FASTQ data file to obtain the target FASTQ data file; The low-quality sequences include sequences with quality values ​​lower than a quality value threshold and sequences with lengths less than a length threshold.

3. The quality control method for clinical test samples according to claim 1, characterized in that, The target FASTQ data file is aligned with the reference genome sequence file to obtain a BAM file, which includes: Obtain the reference genome sequence file in FASTA format, create an index sequence for the reference genome sequence file, and align the target FASTQ data file with the index sequence to obtain the BAM file.

4. The quality control method for clinical test samples according to claim 1, characterized in that, Determining the alignment rate based on the BAM file includes: The target sequence aligned to the reference genome sequence file is determined based on the BAM file; The proportion of the target sequence to all sequences in the target FASTQ data file is set as the alignment rate.

5. The quality control method for clinical test samples according to claim 1, characterized in that, After determining the alignment rate based on the BAM file, the process also includes: If the comparison rate is less than the first threshold, a quality control result is generated determining that the clinical test sample is unqualified.

6. The quality control method for clinical test samples according to claim 1, characterized in that, Before determining the rRNA and globin RNA content based on the BAM file, the following steps are also included: Determine the rRNA sequence coordinates and globin RNA sequence coordinates in the reference genome sequence file; The levels of control rRNA and control globin RNA in N control blood samples were statistically analyzed based on the rRNA sequence coordinates and the globin RNA sequence coordinates. The first standard content range was determined based on the control rRNA content of N control blood samples. The second standard content range was determined based on the control globin RNA content of N control blood samples. Accordingly, after determining the rRNA content and globin RNA content based on the BAM file, the process also includes: If the rRNA content is not within the first standard content range, a quality control result is generated to determine that the clinical test sample is unqualified; If the globin RNA content is not within the second standard content range, a quality control result is generated indicating that the clinical test sample is unqualified.

7. The quality control method for clinical test samples according to claim 1, characterized in that, Before performing sequencing depth alignment of the BAM file using control blood samples and genome annotation files, the following steps are also included: Genomic annotation files were used to obtain gene 3' end position information and gene 5' end position information; The average sequencing depth of the gene 3' end in M ​​control blood samples was determined based on the gene 3' end position information, and the average sequencing depth of the gene 5' end in M ​​control blood samples was determined based on the gene 5' end position information. The ratio of the average sequencing depth of the 3' end of the gene to the average sequencing depth of the 5' end of the gene corresponding to the same gene is set as the feature value; The KMeans algorithm is used to model the feature values ​​to obtain a clustering model, and the number of clusters in the clustering model is determined according to the elbow method. The distance threshold is determined based on the clustering model; Accordingly, after using the BAM file to perform sequencing depth detection and obtain the sequencing depth of the 3' end and 5' end of the gene, the process also includes: The ratio of the sequencing depth at the 3' end of the gene to the sequencing depth at the 5' end of the gene is set as the sample feature value; Calculate the Euclidean distance between the sample feature values ​​and the cluster centers of the clustering model; If the Euclidean distance is greater than the distance threshold, a quality control result is generated indicating that the clinical test sample is unqualified.

8. The quality control method for clinical test samples according to any one of claims 1 to 7, characterized in that, The quality control results of the clinical test samples are generated based on the alignment rate, rRNA content, globin RNA content, gene 3' end sequencing depth, gene 5' end sequencing depth, and the consistency test results, including: Calculate the first score corresponding to the comparison rate; Calculate the second score corresponding to the rRNA content; Calculate the third score corresponding to the globin RNA content; Calculate the fourth score corresponding to the sequencing depth of the 3' end and the sequencing depth of the 5' end of the gene; Calculate the fifth score corresponding to the consistency test result; The sum of the first score, the second score, the third score, the fourth score, and the fifth score is set as the total sample score, and the quality control results of the clinical test sample are generated based on the total sample score.

9. A quality control device for clinical test samples, characterized in that, include: The raw data quality control module is used to perform RNA sequencing on clinical test samples to obtain raw FASTQ data files, and to perform quality control filtering on the raw FASTQ data files to obtain target FASTQ data files; The mapping file construction module is used to align the target FASTQ data file with the reference genome sequence file to obtain a BAM file; The alignment rate detection module is used to determine the alignment rate based on the BAM file; The target RNA content statistics module is used to determine the rRNA content and globin RNA content based on the BAM file. The sequencing depth statistics module is used to perform sequencing depth detection using the BAM file to obtain the sequencing depth at the 3' end and the sequencing depth at the 5' end of the gene. The consistency verification module is used to perform consistency verification on the various sequencing data obtained by multiple sequencing strategies of the clinical test sample, and obtain the consistency verification results. The judgment module is used to generate the quality control results of the clinical test samples based on the alignment rate, the rRNA content, the globin RNA content, the sequencing depth of the gene 3' end, the sequencing depth of the gene 5' end, and the consistency test results. The process of determining the rRNA content and the globin RNA content includes: using the rRNA and globin RNA coordinate files and the BAM file to perform rRNA content statistics and globin RNA content statistics to obtain the rRNA content and the globin RNA content.

10. An electronic device, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program, and the processor, when calling the computer program in the memory, implements the steps of the quality control method for clinical test samples as described in any one of claims 1 to 8.

11. A storage medium, characterized in that, The storage medium stores computer-executable instructions, which, when loaded and executed by a processor, implement the steps of the quality control method for clinical test samples as described in any one of claims 1 to 8.

Images