Construction and analysis method of cfDNA-TCR library based on liquid phase probe and application of construction and analysis method

Abstract

The invention provides a construction and analysis method of a cfDNA-TCR library based on a liquid-phase probe and application thereof, and belongs to the field of molecular immunology, and the construction and analysis method comprises the following steps: plasma cfDNA extraction, cfDNA UMI-UDI library construction, liquid-phase probe hybridization capture, enrichment of a cfDNA library containing a TCR gene, next-generation sequencing, offline data filtration and cleaning, double-end sequencing data merging, and analysis of the cfDNA-TCR library based on the liquid-phase probe. Sequencing data is compared to a reference genome, a non-specific capture sequence is removed, PCR deduplication is carried out according to UMI molecules, the data is input into TRUST4 for contig assembly and annotation, and further immune repertoire analysis is carried out according to an annotated CDR3 result. According to the method disclosed by the invention, the TCR molecules in the cfDNA can be more accurately detected by distinguishing PCR repetition from cloning repetition.

Description

Technical Field

[0001] The present invention relates to the field of gene detection technology, and in particular to a method for constructing and analyzing a cfDNA-TCR library based on a liquid phase probe and its application. Background Art

[0002] TCRs, or T cell antigen receptors, are essential components of T cell function. During individual development, T cells generate a large number of different TCRs through random gene rearrangement to recognize a wide variety of antigens in nature. The TCR immune repertoire refers to the sum total of all TCRs in the body, reflecting the diversity of the body's T cell population and its ability to recognize antigens. Research on the TCR immune repertoire contributes to a deeper understanding of the workings of the immune system and the immunopathogenesis of diseases. Significant progress has been made in the field of oncology. For example, by investigating the peripheral-tumor infiltration and the tumor immune microenvironment, the efficacy of tumor immunotherapy can be predicted. Immune response levels have also shown potential as biomarkers for disease onset or progression in some tumors, and they hold great potential for the translation of TCR-T therapies.

[0003] The widespread application of TCR immune repertoire research has promoted the development of personalized tumor immunotherapy, but current research on it mainly focuses on peripheral blood mononuclear cells (PBMC) and tumor tissues. There are very few studies on TCRs in circulating cell-free DNA (cfDNA) in peripheral blood. This is because approximately 79% of cfDNA released from leukocytes is regarded as noise, and the large amount of immune dynamic change information contained therein is ignored; in addition, the mainstream method of TCR immune repertoire research is based on the principle of multiplex PCR. The template chain length requirements of most multiplex kits limit their application on cfDNA samples. However, studying the TCR immune repertoire solely on PBMCs and tissues presents limitations. First, studies of peripheral blood (PBMCs) often miss local immune information not present in circulating blood cells, such as tissue-resident T cells, which play a key role in anti-tumor immunity. While studying tissue TCRs provides more direct access to tumor-reactive TCRs, tissue sample collection is more challenging and may miss information about T cells that carry tumor-reactive TCRs but have lost activity. Combined PBMC and tissue studies are also widely used, but due to the high antigenic heterogeneity of tumors, the immune system must simultaneously recruit a diverse array of T cell clones to recognize and target tumor antigens, resulting in a vast and diverse PBMC-tissue shared TCR repertoire. This complicates our focus on tumor-reactive TCRs. Therefore, supplementing studies with cfDNA-TCRs can simultaneously capture circulating and tissue-derived T cells undergoing apoptotic turnover. This not only potentially addresses previously missing information but, due to the low abundance and susceptibility of cfDNA to degradation, also allows us to focus on key immune dynamics with extremely high turnover rates.

[0004] Therefore, it is necessary to develop a method for library construction data analysis for cfDNA-TCR. Summary of the Invention

[0005] To solve the above technical problems, the present invention provides a method for constructing and analyzing a cfDNA-TCR library based on a liquid phase probe and its application.

[0006] The first object of the present invention is to provide a liquid-phase probe panel for hybridization capture of four chains, namely, TCR α, β, γ, and δ, wherein:

[0007] Liquid phase probes were designed based on the V and J gene sequences of human TCR genes in the International Immunogenetics ImMunoGeneTics (IMGT) database. The IMGT data numbers for some genes are as follows:

[0008] AE000658|TRAV1, AE000660|TRAV25, X58768|TRAV30, M95394|TRAV38, AE000660|TRAV8, 1444|TRBV7, M22198|TRDV1, X15207|TRDV2, M23326|TRDV3, M13431|TRGV5, M13434|TRGV8,

[0009] Furthermore, there are 113 alleles for TRAV of the TCR α chain, 147 alleles for TRBV of the TCR β chain, 19 alleles for TRGV of the TCR γ chain, and 21 alleles for TRDV of the TCR δ chain. 100 bp probes were designed targeting the 5' and 3' ends of these TCR V genes.

[0010] Furthermore, there are 71 alleles for TRAJ of the TCR α chain, 16 alleles for TRBJ of the TCR β chain, 6 alleles for TRGJ of the TCR γ chain, and 4 alleles for TRDJ of the TCR δ chain. 100bp probes were designed for the 5' and 3' ends of these TCR J genes. Since some J genes are short (<200bp), some constant region TCRC region genes will be combined for 5' end probe design.

[0011] Furthermore, a total of 843 probes were designed, including 433 probes targeting V genes and 410 probes targeting JC combined genes. Some of their sequences are shown in the sequence listing as SEQ ID NO.1-NO.39.

[0012] The second object of the present invention is to provide a method for constructing a second-generation sequencing library based on the above-mentioned probe, comprising the following steps:

[0013] (1) Construction of UMI-UDI library for cfDNA fragments;

[0014] (2) Capture and enrich the cfDNA library using liquid phase probes for cfDNA-TCR genes;

[0015] (3) Second-generation sequencing.

[0016] Furthermore, in step (1), UMI molecules are introduced into the library construction, which allows PCR deduplication of the results in the subsequent analysis step, thereby distinguishing PCR amplification duplicates from cloned amplification duplicates. At the same time, UDI molecules are introduced to reduce read mismatches.

[0017] Furthermore, in step (2), the capture probe is modified with biotin. After the capture probe hybridizes with the target nucleic acid sequence overnight for 24 hours, it is affinity captured using magnetic beads modified with streptavidin. The streptavidin magnetic beads are then incubated with the hybridization reaction system, washed, heat eluted, and eluted at room temperature. Finally, the streptavidin magnetic beads and the captured target DNA fragments are separated by high-temperature heating in a PCR instrument, and these target DNA fragments are subjected to about 10-13 rounds of capture and amplification.

[0018] Furthermore, in step (3), the sequencing library obtained by the above enrichment is applied to the second-generation sequencing Illumina platform, and the sequencing data volume is 6G / sample.

[0019] The third object of the present invention is to provide a process for immune repertoire analysis based on the above-mentioned captured and enriched cfDNA-TCR sequencing data, comprising the following steps:

[0020] (1) Filter and clean the raw data from sequencing and merge the double ends.

[0021] (2) Align the data to the reference genome hg38 and distinguish poorly mapped reads from well-mapped reads using the MAPQ value.

[0022] (3) By comparing the annotation information, the well-aligned data (mapped reads) are divided into TCR gene mapping data (TCR-mapped reads) and non-TCR gene mapping data (non-TCR mapped reads), and the non-TCR gene mapping data are removed.

[0023] (4) Merge the TCR gene alignment data and the poorly aligned data and input them into the subsequent steps to continue UMI deduplication.

[0024] (5) The number after deduplication according to Input the TRUST4 tool for contig assembly and TCR gene annotation.

[0025] Furthermore, in step (1), the data quality control was performed by using the Trimmomatic tool, and the Fastqc tool was used to control the data quality. according to Low-quality sequences were cleaned (phred score > 20), adapter sequences were removed, UMI sequences were extracted using Picard and Figbio tools, and then the R1 and R2 reads generated by double-end sequencing were merged by overlap using the flash tool.

[0026] Furthermore, in step (2), the data were aligned to the reference genome using the BWA tool, and the alignment score (MAPQ) value was used to distinguish between well-aligned data and poorly aligned data.

[0027] Furthermore, in step (3), considering the fragmentation characteristics of cfDNA, the data that match the hg38 TCR gene (14q11.2, 7q34, 7p14) in the well-matched data will be retained as TCR gene matching data, which will be beneficial for the subsequent contig assembly process. The proportion of non-TCR gene matching data will be used as the nonspecific capture rate of the technology.

[0028] Furthermore, in step (4), the data with completely identical UMI molecules and insert sequences are deduplicated.

[0029] Furthermore, in step (5), the deduplicated data is input into the TRUST4 tool as single-ended data.

[0030] The fourth object of the present invention is to provide the application of the cfDNA-TCR library construction and analysis process in human colorectal cancer samples, wherein:

[0031] The study included 30 CRC patients who underwent neoadjuvant therapy. They first received radiotherapy on Day 8 and then immunotherapy on weeks 2-7. Tissue and whole blood samples were collected on Day 0 (-RT / -ICI), Day 8 (+RT / -ICI), and week 8 (+RT / +ICI). The tissues underwent single-cell TCR sequencing, and cfDNA samples were detected using the above-mentioned cfDNA-TCR library construction and analysis process, and were jointly analyzed with paired tissue samples.

[0032] Beneficial effects:

[0033] (1) Liquid-phase probes do not have strict requirements on template chains, making them compatible with shorter cfDNA. Liquid-phase probe hybridization capture technology supports the simultaneous construction of libraries for the four chains of TCR in the same reaction, which reduces the loss of cfDNA samples and makes up for the information that may be missed by some TCR-seq kits based on the multiplex PCR principle that only construct libraries for the β chain.

[0034] (2) The UMI molecules introduced during the library construction process support deduplication in subsequent analysis, thereby distinguishing clonal duplications from PCR amplification duplications. This is a major advantage compared to previous multiplex PCR methods: conventional bulk-TCR-seq methods based on multiplex PCR lack UMI molecules, which will lead to the inability to accurately estimate the frequency of clonal amplification due to PCR amplification preference, thereby biasing the obtained dominant clone results (Appendix Figure 8 ).

[0035] (3) In cfDNA-TCR, not only can a certain amount of overlap be generated with the paired tissue single-cell TCR sequencing results (cfDNA-tissue shared CDR3), but also some tissue-TCR paired α β chains can be captured at the same time, verifying the accuracy of the cfDNA-TCR results (Appendix Figure 6 ).

[0036] (3) By comparing with the commonly used immune repertoire analysis process (fastp+MIXCR), it was found that the experimental analysis process of this patent can enrich more highly amplified cfDNA-tissue shared CDR3 (see Appendix Figure 7 ).

[0037] (4) cfDNA-TCR molecules can respond to dynamic changes in immunity and show a significant trend of increasing cfDNA-tissue shared CDR3 diversity after immunotherapy (Appendix Figure 9 ).

[0038] (5) After cell type annotation, cfDNA-TCR has a very high rank correlation with the cell type annotation results of tissue-TCR after immunotherapy, indicating that cfDNA-TCR reflects the immune status inside the tissue well (Appendix Figure 10 ). BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 Schematic diagram of the library construction and analysis process of the present invention.

[0040] Figure 2 This is the quality inspection result of the library fragments generated using the process of the present invention.

[0041] Figure 3 This is a schematic diagram of distinguishing poorly aligned data from well aligned data by using the alignment score MAPQ value in the analysis process of the present invention.

[0042] Figure 4 This is the intersection between the poorly compared data and the IMGT database in the analysis process of the present invention.

[0043] Figure 5 These are the respective proportions of poor alignment data, TCR gene alignment data, and non-TCR gene alignment data (non-specific capture).

[0044] Figure 6 This is the sharing situation of cfDNA-TCR and tissue-TCR obtained by the process of the present invention.

[0045] Figure 7 This is the clone sorting distribution of cfDNA-tissue shared CDR3 obtained by the analysis process of the present invention and the MIXCR process.

[0046] Figure 8 The changes in clone ranking before and after UMI deduplication.

[0047] Figure 9 The changes in cfDNA-tissue shared CDR3 diversity in CRC patients at different treatment stages.

[0048] Figure 10 The cell type rank correlation between cfDNA-TCR and tissue TCR obtained by the process of the present invention. DETAILED DESCRIPTION

[0049] The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. However, the following embodiments are intended only to explain the present invention, and the scope of protection of the present invention should include the entire contents of the claims. Moreover, through the description of the following embodiments, those skilled in the art can fully implement the entire contents of the claims of the present invention.

[0050] Unless otherwise specified, the chemical reagents used in the examples are conventional commercially available reagents, and the technical means used in the examples are conventional means well known to those skilled in the art. The sequencing apparatus used in the examples of the present invention is the Illumina NovaSeq 6000, and the sequencing steps of the present invention are not limited to this sequencing apparatus.

[0051] Example 1 Design of a liquid-phase probe for detecting TCR in plasma cfDNA

[0052] The CDR3 region of the TCR is the primary site for binding to the antigenic peptide-MHC complex (pMHC). Due to the random nature of V(D)J recombination, this region exhibits a high degree of diversity in length and amino acid sequence. This diversity enables the immune system to recognize a wide variety of antigens. Consequently, the CDR3 region has become a major focus of immune repertoire research. The CDR3 region is located at the junction of the V(D) and J genes. To capture CDR3-containing fragments in cfDNA-TCRs, 100-bp liquid-phase probes were designed at either end of the CDR3 region: the 5' end of the V gene and the 3' end of the J gene. Furthermore, due to the random fragmentation of cfDNA, fragments containing portions of the V and J genes but not the CDR3 region may exist. Therefore, probes were also designed for the 3' end of the V gene and the 5' end of the J gene. Since some J genes are shorter (<200 bp), probes were designed for the 5' end of the constant region TCR C region gene, which was combined with a portion of the constant region TCR C region gene.

[0053] Probe design for the TCRV gene

[0054] There are 113 alleles for TRAV of the TCR α chain, 147 alleles for TRBV of the TCR β chain, 19 alleles for TRGV of the TCR γ chain, and 21 alleles for TRDV of the TCR δ chain. 100 bp probes were designed targeting the 5' and 3' ends of these TCR V genes.

[0055] Probe design for TCRJ genes

[0056] There are 71 alleles for TRAJ of the TCR α chain, 16 alleles for TRBJ of the TCR β chain, 6 alleles for TRGJ of the TCR γ chain, and 4 alleles for TRDJ of the TCR δ chain. 100bp probes were designed for the 5' and 3' ends of these TCR J genes. Some J genes are shorter (<200bp), so some constant region TCR C region genes will be combined for 5' end probe design.

[0057] Finally, a total of 843 probes were designed, including 433 probes targeting V genes and 410 probes targeting JC combined genes. Some of their sequences are shown in the sequence listing as SEQ ID NO.1-NO.39.

[0058] Example 2 Establishment of TCR immune library from plasma cfDNA of patients with colorectal cancer

[0059] The samples tested in the embodiment of the present invention include 30 patients with colorectal cancer who underwent neoadjuvant therapy. They first received radiotherapy on Day 8 and then immunotherapy on week 2-7. Tissue and whole blood samples were collected on Day 0 (-RT / -ICI), Day 8 (+RT / -ICI), and week 8 (+RT / +ICI). The tissues underwent single-cell TCR sequencing, and the cfDNA samples were tested using the cfDNA-TCR library construction and analysis process, and were jointly analyzed with paired tissue samples. Figure 1 shown.

[0060] 1. Plasma cfDNA extraction:

[0061] 1.1 Plasma Separation: Plasma was extracted and placed in an EDTA anticoagulant tube. Centrifuged at 1600 g for 10 minutes at 4°C. The plasma was separated into three layers after centrifugation. The supernatant (plasma) was aliquoted into 1.5 mL / 2 mL centrifuge tubes. During the aspiration process, avoid aspirating the middle layer (white blood cells and platelets) and the lower layer (red blood cells). The upper layer of plasma was centrifuged at 16,000 g for 10 minutes at 4°C. The supernatant was aspirated and the residual cells at the bottom were removed. The plasma required for separation was obtained and stored in a -80°C refrigerator to avoid repeated freezing and thawing.

[0062] 1.2 Plasma cfDNA Extraction and Quantification: Plasma cfDNA was extracted from the separated plasma according to the instructions of the Magbead Free-Circulating DNA Maxi Kit (CWBIO, Jiangsu, China). Plasma cfDNA was then quantified using a 1× dsDNA HS Assay Kit (YEASEN, Shanghai, China) on a Qubit 4.0 Fluorometer. The total amount was approximately 3 to 250 ng.

[0063] 2. cfDNA library construction:

[0064] cfDNA libraries were constructed according to the VAHTS Universal DNA Library Prep Kit (Vazyme Biotech, ND607) instructions, and adapters with UMI and UDI molecules were introduced using the VAHTS Dual UMI UDI Adapters Set 1–Set 4 for Illumina kit. The specific steps are as follows:

[0065] 2.1 First, the cfDNA sample is end-repaired and the repaired sample is connected to the adapter in a two-step method. First, the adapter sequence with the UMI molecule is connected. Different UMI adapter concentrations are added according to the initial amount of cfDNA input.

[0066] 2.2 Perform the first round of 0.6X purification using VAHTS DNA Clean Beads. The beads should be equilibrated to room temperature (30 minutes) before use and washed with freshly prepared 80% ethanol equilibrated to room temperature.

[0067] 2.3 Library amplification: In this step, the second step of the two-step adapter construction was performed, introducing amplification adapters with different indices. The PCR reaction program was as follows: pre-denaturation at 95°C for 3 min; denaturation at 98°C for 20 s, annealing at 65°C for 20 s, and extension at 72°C for 30 s. The number of cycles was adjusted based on the initial amount of cfDNA input, approximately 7-13X cycle number; final extension at 72°C for 5 min; and storage at 4°C.

[0068] 2.4 cfDNA was sorted in two rounds using magnetic beads at the recommended concentrations in the manufacturer's instructions: 0.73x / 0.25x. Finally, the library was quantified on a Qubit 4.0 Fluorometer using a 1× dsDNA HS Assay Kit (YEASEN, Shanghai, China), yielding approximately 500–1800 ng.

[0069] 3. cfDNA-TCR library hybridization capture:

[0070] Following the instructions for the TargetSeq One® Hyb & Wash Kit v2.0 condition (iGeneTech, Beijing, Shanghai), cfDNA was hybridized and captured to enrich the cfDNA library containing TCR genes. The specific steps are as follows:

[0071] 3.1 Pool the prepared libraries. Combine up to six libraries of similar total volume in a 1.5 mL centrifuge tube and vacuum concentrate to dryness. Control the drying time based on the volume, approximately 20-40 minutes. Once the tube is just dry, add the hybridization solution. Add the probe after adding RNase Block, as RNase Block protects the RNA probe from degradation. Then, incubate the probe and library for hybridization at 50°C for 24 hours overnight.

[0072] 3.2 Using Invitrogen™ 65601 Dynabeads MyOne Streptavidin T1 magnetic beads, biotinylated probes that have captured cfDNA library fragments are screened. After washing with 80% ethanol, the bead-probe-library complex is placed in a PCR instrument for capture and amplification. Since the probe binds to the pre-library for a maximum of 100 bp (i.e., the probe length), high temperatures will cause the probe to melt at the site of binding, resulting in single-stranded DNA in the pre-library, which serves as a template for enrichment. Furthermore, only captured pre-library fragments contain amplification primer binding sites, so theoretically, only captured pre-library fragments will be amplified. The number of post-capture amplification cycles is 10-13x.

[0073] 3.3 Perform 1.1X purification using VAHTS DNA Clean Beads.

[0074] 4. Library concentration determination, fragment quality inspection and next-generation sequencing

[0075] After capture, the cfDNA-TCR library was quantified using the 1× dsDNA HS Assay Kit (YEASEN, Shanghai, China) on a Qubit 4.0 Fluorometer, with a total of approximately 25–200 ng. The fragments were then quality-checked using an Agilent 4200 TapeStation instrument, as shown in the attached image. Figure 2 The following figure shows the quality control of the library fragments for an example sample. The sequencing platform was an Illumina NovaSeq 6000, and 6GB of data was measured for each sample.

[0076] 5. Immune repertoire analysis

[0077] 5.1 Filter and clean the original data and merge the two ends:

[0078] Trimmomatic (v0.39) and Fastqc (v0.12.1) were used to filter low-quality data and remove adapter sequences. UMI sequence information was extracted using Picard (v3.0) and Figbio (v2.2.1). Finally, paired-end merging was performed using Flash (v1.2.11).

[0079] 5.2 Align sequencing data to the reference genome:

[0080] 5.2.1 Use the tool BWA (v0.7.18) to align the sequencing data to the hg38 reference genome. The alignment score MAPQ value will be obtained. This value reflects the quality of the data alignment. The MAPQ median value is selected as the threshold. Data with a value greater than the threshold are considered to be well aligned, while data with a value less than the threshold are considered to be poorly aligned. Figure 3 Due to the highly random gene rearrangement process of TCR molecules, the alignment quality of their hypervariable regions (such as CDR3) with the reference genome is poor, so the poorly aligned data are used as TCR candidate data.

[0081] 5.2.2 The reliability of poorly aligned data as TCR candidate data was verified by performing an intersection analysis between the TCR sequences in the IMGT database and the sequences of poorly aligned data. It was found that the intersection between the two could reach about 80%, as shown in the attached figure. Figure 4 , proving that most of these poorly matched data do belong to TCR genes.

[0082] 5.2.3 In addition, due to the random fragmentation of cfDNA, even well-aligned data may contain short fragments containing genes in relatively conserved TCR regions, resulting in a high-quality alignment with the reference genome. Therefore, we used Bedtools (v2.31.1) to obtain alignment annotation information and distinguish between TCR gene alignment data and non-TCR gene alignment data.

[0083] 5.3 Removal of non-specific captured data:

[0084] Remove the non-TCR gene comparison data and use its proportion as the non-specific capture rate of the technology, which is about 15%. Figure 5 .

[0085] 5.4 De-duplicate UMIs from sequencing data;

[0086] Using Awk (4.0.2) and Figbio (v2.2.1), the data with the same UMI molecules and insert sequences were according to Perform deduplication.

[0087] 5.5 The data were input into TRUST4 (v1.0.10.1) for contig assembly and TCR gene annotation.

[0088] 6. Example result analysis

[0089] By performing the above-mentioned steps on 90 CRC cfDNA samples for library construction and result analysis, cfDNA-TCR results were obtained and combined with paired tissue single-cell TCR-seq results:

[0090] 6.1 Intersection analysis results of paired cfDNA-TCR and tissue-TCR: as shown in the attached Figure 6 , it was found that there were a certain number of shared CDR3s between cfDNA-TCR and tissue-TCR results (only part of them is shown in the figure). Moreover, since the cfDNA-TCR process captures the four TCR chains at the same time, it was found that some of the tissue-TCR paired α β chains were captured at the same time in the bulk cfDNA-TCR results, which proves the accuracy of the cfDNA-TCR results.

[0091] 6.2 Further verification of the analysis process: The data generated by the above library construction process were simultaneously input into our supporting analysis process and the widely used MIXCR process, and the analysis results generated by the two were compared with the supporting analysis process. Pair Weaving- The CDR3 overlap analysis was performed to obtain the cfDNA-tissue shared CDR3, and then the distribution of these CDR3 in the two results was observed, as shown in the attached figure. Figure 7 , and found significant differences in the distribution: in the cfDNA-TCR results generated by our analysis process, the cfDNA-tissue shared CDR3 distribution peak was closer and located in the ~Top50 clones, which means that they have higher clone counts and clone frequencies, while in the cfDNA-TCR results generated by the MIXCR process, the clones with the cfDNA-tissue shared CDR3 distribution peak were ranked relatively backward.

[0092] 6.3 Changes in the order of major clones before and after deduplication: Figure 8 It can be found that in the original clone ranking, the clone count is the sum of PCR duplication and clone duplication. After removing PCR duplication, the clone ranking changes significantly, indicating that the deduplication process can indeed obtain more accurate clone frequency statistics.

[0093] 6.4 Changes in diversity of CRC samples during different treatment periods: Figure 9 , showing that the diversity of cfDNA-tissue shared CDR3 increased significantly after immunotherapy, indicating that the part shared by cfDNA-TCR and tissue-TCR can show the dynamic changes of the patient's immunity as the treatment progresses.

[0094] 6.5 Rank correlation between cfDNA-TCR and tissue-TCR in different cell types at different CRC treatment stages: It can be found that whether it is pCR patients with complete remission after treatment or nonpCR patients with incomplete remission after treatment, they all show extremely high rank correlation of cell types in the +RT / +ICI stage after immunotherapy, indicating that after immune stimulation, a large number of T cells turn over to produce cfDNA-TCR, thus leading to a close connection between cfDNA-TCR and tissue-TCR.

[0095] The foregoing is merely a list of specific embodiments of the present application, intended to enable those skilled in the art to understand or implement the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the broadest scope consistent with the principles and novel features of the present application.

Claims

1. A method for constructing and analyzing a cfDNA-TCR library based on a liquid phase probe, characterized in that: The following steps are involved: (1) Extracting cfDNA molecules from plasma; (2) Construction of cfDNA libraries containing UMI and UDI labels; (3) Enrichment of cfDNA libraries containing TCR genes using liquid-phase probe hybridization capture technology; (4) Perform next-generation sequencing on the enriched library; (5) Data processing and analysis, including: a) Filtering and cleaning of raw data and merging of both ends; b) Align the sequencing data to the reference genome hg38 and distinguish between poorly aligned and well aligned data; c) Identify TCR gene alignment data from well-aligned data; d) remove non-TCR gene alignment data, i.e., non-specific capture data; e) UMI molecule deduplication; f) Contig assembly and TCR gene annotation were performed using the TRUST4 tool.

2. The method according to claim 1, characterized in that In step (2), the UMI molecule is a random oligonucleotide sequence used to distinguish PCR amplification repeats from cloned amplification repeats; the UDI molecule is a double-ended unique linker used to reduce mismatches in double-ended sequencing data.

3. The method according to claim 1, characterized in that In step (3), the liquid phase probes were designed to cover the V / J gene regions of the TCR α, β, γ, and δ chains. The probe length was 100 bp, and 843 probes were designed for the 5' and 3' ends of the TRAV (113), TRBV (147), TRGV (19), and TRDV (21) genes of the IMGT database, as well as the combined constant regions of the TRAJ (71), TRBJ (16), TRGJ (6), and TRDJ (4) genes.

4. The method according to claim 1, wherein In step (5) b), the alignment quality is divided by MAPQ value, the MAPQ median is used as the threshold, and the poorly aligned data below the threshold are regarded as TCR candidate data, and the intersection rate of more than 80% is verified by the IMGT database.

5. The method according to claim 1, wherein In step (5) c), considering the random fragmentation characteristics of cfDNA, the data of the hg38 TCR gene aligned with the well-aligned data will be retained as the TCR gene alignment data for subsequent contig splicing.

6. The method according to claim 1, characterized in that The proportion of non-TCR genes in step (5) d) is used to calculate the nonspecific capture rate, which is 15% on average.

7. The method according to claim 1, characterized in that In step (5) e), UMI deduplication is performed based on the consistency of UMI molecules and insert sequences, and identical reads are removed to eliminate PCR duplication, making the calculation of clone frequency more accurate.

8. The method according to claim 1, characterized in that In the step (5) f), the immune repertoire tools such as TRUST4 and MIXCR are supported to perform contig assembly and output CDR3 annotation results.

9. The method according to claim 1, characterized in that It is used for treatment monitoring of colorectal cancer patients, including 30 patients receiving neoadjuvant therapy (radiotherapy combined with immunotherapy). Paired whole blood and tissue samples were collected at three time points before and after treatment. The shared CDR3 analysis of cfDNA-TCR and tissue single-cell TCR sequencing results was used to track dynamic changes in the immune system.

10. An application based on any method of claims 1-9, characterized in that: Used for tumor immunotherapy monitoring or immune repertoire diversity analysis, the shared clone ratio and diversity changes between cfDNA-TCR and tissue TCR are used as biomarkers to monitor immune dynamic changes.

Images