Amplification method and device of DNA (Deoxyribose Nucleic Acid) molecule for three-generation sequencing in single cell chromosome conformation capture

Abstract

The invention discloses a method and a device for amplifying DNA (deoxyribonucleic acid) molecules for three-generation sequencing in single cell chromosome conformation capture. According to the method disclosed by the invention, when a sequencing library for single cell three-generation sequencing is prepared, a barcode sequence is added into a linker sequence of Tn5 transposase and a barcode sequence is added into a PCR (Polymerase Chain Reaction) amplification primer sequence, and different single cells are distinguished according to the combined barcode sequences; dNA amplification products from different sources are mixed, so that the requirement of a third-generation sequencing platform on the template quantity is met, single-cell three-generation Hi-C sequencing can be carried out according to the requirements of single cells with different fluxes, and a matched bioinformatics analysis method is combined; the method can realize single cell three-dimensional genome modeling, single cell dimensionality reduction and clustering, single cell copy number variation, single cell structure variation identification, single cell assembly assistance and / or single cell chromatin high-order interaction identification, and chromatin and / or chromosome conformation capture.

Description

Technical Field

[0001] The present invention relates to a method and a device for amplifying DNA molecules for third-generation sequencing in capturing single-cell chromosome conformation. Background Art

[0002] The development of chromatin conformation capture technology (C-technology) and its derivative technologies 3C, 4C, 5C, and Hi-C has greatly promoted the study of three-dimensional genomes. Studies have shown that the mammalian genome performs its normal functions through a highly organized gene regulatory network. However, due to the limitation that C-technology can capture basically paired regulation, it is difficult to explore how multiple regulatory elements interact with each other to regulate genes and thus exert their biological functions. In recent years, many studies have been devoted to breaking through this bottleneck, including technologies based on proximity ligation, such as TM3C, c-walk, COLA, MC-4C, and Tri-C, and non-ligation methods such as SPRITE, ChIA-drop, and GAM. Thanks to the development of third-generation long-read sequencing, the recently developed Pore-C method based on Oxford Nanopore Technology (ONT) has shown excellent performance in detecting high-order interactions, and can effectively analyze the synergy between multi-directional chromatin interactions across the genome by applying new statistical methods. Although these population-based methods have shown their ability to explore precise and complete interactions in 3D genomes, these methods cannot solve the problem of cell-to-cell heterogeneity. More importantly, how individual cells organize multiple functional regulatory elements dispersed in the genome through genome structural conformation to ensure precise spatiotemporal expression of genetic programs within the cell can only be studied using single-cell methods. However, the single-cell 3D genomics methods developed in recent years have the following limitations for detecting multidirectional interactions in single cells: (1) Due to the limited read length of Illumina next-generation sequencing, the single-cell Hi-C (scHi-C) method based on proximity ligation cannot currently be used to capture multidirectional interactions; (2) It is difficult to balance the high throughput of capturing the heterogeneity of large cell populations and the high sequencing depth of constructing high-resolution 3D genome structures.

[0003] Hi-C (High-throughput / resolution chromosome conformation capture) is a chromatin conformation capture technology based on 3C technology and high-throughput sequencing. Summary of the invention

[0004] The technical problem to be solved by the present invention is how to reduce the starting amount of cells for detecting whole-genome high-order interactions at the single-cell level, and / or how to improve the signal-to-noise ratio for detecting whole-genome high-order interactions at the single-cell level, and / or how to improve the throughput for detecting whole-genome high-order interactions at the single-cell level and / or how to capture single-cell chromatin and / or chromosome conformation.

[0005] In order to solve the above technical problems, the present invention first provides a method for amplifying DNA molecules for third-generation sequencing in single-cell chromatin and / or chromosome conformation capture, the method comprising the following steps:

[0006] S1. performing in vitro cross-linking and fixation of target cells to obtain fixed target cells, collecting the fixed target cells and lysing them with a lysate to obtain a lysate;

[0007] S2, after removing the protein in the lysate, digesting with restriction endonucleases and then ligating with T4 DNA ligase to obtain target cells containing proximity-ligated DNA; separating the target cells containing proximity-ligated DNA into single cells, decrosslinking the single cells and lysing the cell genome to obtain a single cell genome lysate;

[0008] S3. The genome lysate of each single cell is subjected to genome fragmentation and labeling using one Tn5 transposase complex, and the genome lysates of n1 single cells are subjected to genome fragmentation and labeling using n1 Tn5 transposase complexes to obtain fragmented genome products of n1 single cells, wherein the labels in the fragmented genome products of each single cell in the n1 fragmented genome products of the single cells are different, and the labels are single-stranded DNA used to distinguish different single cells; in the n1 Tn5 transposase complexes, each Tn5 transposase complex is a substance formed by the combination of a Tn5 linker and a Tn5 transposase, and the Tn5 linker can be A double-stranded DNA fragment formed by annealing two single-stranded primers, one strand of which can be connected by a fixed sequence, a barcode sequence and a reverse complementary sequence of an ME sequence, the barcode sequence can be a single-stranded DNA used to distinguish different individual cells, and the ME sequence can be a chimeric end sequence of a Tn5 transposon; the n1 can be a natural number greater than or equal to 1; the barcode sequences of the n1 Tn5 connectors in the n1 Tn5 transposase complexes are all different, and the reverse complementary sequences of the fixed sequence and the ME sequence are all the same; the nucleotide sequence of the other strand of the Tn5 connector can be a single-stranded DNA fragment of the ME sequence, and the 5' terminal nucleotide of the other strand of the Tn5 connector can be phosphorylated (phos). The n1 can be a natural number between 1 and 96 (such as 24). The fixed sequence can be a single-stranded DNA used as a protective base and / or a part of the primer sequence.

[0009] S4, the fragmented genomic products of each n1 single cell are mixed to obtain a group of fragmented genomic products, each group of the fragmented genomic products in the B group of fragmented genomic products is amplified using one barcode primer and a common B barcode primers to obtain B PCR products, and the B PCR products are mixed to obtain PCR products for third-generation sequencing in single-cell chromatin and / or chromosome conformation capture;

[0010] The sequences of the B amplification primers are all different, and each of the barcode primers in the B amplification primers is a single-stranded DNA fragment connected by an anchor sequence, a barcode sequence, and a fixed sequence X, and the anchor sequence is used to protect the single-stranded DNA. One amplification primer contains a barcode sequence, and the barcode sequence is a single-stranded DNA used to distinguish different groups of fragmented genomic products; the barcode sequences in the B amplification primers are all different, and the anchor sequence and the fixed sequence X are the same; the fixed sequence X is a single-stranded DNA used as a protective base and / or a part of the primer sequence, and B is a natural number greater than or equal to 1. The anchor sequence is used to determine the position of amplification.

[0011] In S3 of the above amplification method, the length of the tag may be m1, and m1 may be a natural number between 16 and 35. The length of the fixed sequence may be 5 nt to 20 nt. The length of the barcode sequence may be m1 nt. The length of n1 may be 1 to 4 m1 A natural number in , such as 24. The nucleotide sequence of the fixed sequence may be 5'-TCGTCGGCAGCGTC-3' (nucleotides 4-14 of sequence 1). The barcode sequences of the n1 types of Tn5 connectors are all different, and the n1 types of Tn5 connectors may specifically be 24 types of Tn5 connectors, and the structures of the 24 types of Tn5 connectors may be as shown in Formula 1:

[0012]

[0013] 3'-NH3-TCTACACATATTCTCTGTC-phos-5' (sequence 2) Formula 1.

[0014] The barcode sequences of the 24 Tn5 adapters are named Tn5 barcode1-24, and the sequences are shown in Table 1. The reverse complementary sequence of the ME sequence can be 5'-AGATGTGTATAAGAGACAG-3' (nucleotides 39-57 of sequence 1). The barcode sequences of the 24 Tn5 adapters can be applied to both the Nanopore platform and the PacBio platform.

[0015] In S3 of the above amplification method, the length of the genomic lysate of the single cell can be controlled by adjusting the concentration of the Tn5 transposase complex. The length of the genomic lysate of the single cell can be 3-6 kb.

[0016] In S4 of the above amplification method, the nucleotide sequence of the anchor sequence may be 5'-AATGATACGGCGACCACCGAGATCT-3' (nucleotides 1-25 of sequence 3). The length of the barcode sequence may be m1. B may be 1 to 4 m1 / n1, such as a natural number from 1 to 96. The fixed sequence X can be selected from the full length or a fragment of the fixed sequence in S3. The length of the fixed sequence X can be 5 nt to 20 nt. The nucleotide sequence of the fixed sequence X can be specifically 5'-TCGTCGGCAGCGTC-3' (nucleotides 50 to 63 of sequence 3). The B amplification primers can be 96 amplification primers named I5 PCR primer 1-96, and the general formula of the 96 barcode primers can be Formula 3:

[0017]

[0018] In S1 of the above amplification method, the in vitro cross-linking and fixation may be performed by treating the target cells with formaldehyde and then with disuccinimidyl glutarate.

[0019] In S1 of the above amplification method, the target cells may be 0.25×10 6 Cells-1.00×10 6 cells.

[0020] In S1 of the above amplification method, in order to reduce the noise of random connection, the in vitro cross-linking fixation can be to treat the target cells with formaldehyde and then treat them with disuccinimidyl glutarate. The formaldehyde treatment can be to treat with a 0.8%-2.0% formaldehyde solution, such as a 1% formaldehyde solution. The subsequent treatment with disuccinimidyl glutarate can be to treat with a 0.6mM-3.0mM disuccinimidyl glutarate (DSG) solution, such as a 1.5mM disuccinimidyl glutarate solution.

[0021] In S4 of the above amplification method, the B PCR products can be mixed and purified using 0.8×, 0.7×, 0.6×, 0.5× or 0.4× XP magnetic beads, and can be purified twice or three times, or other purification methods can be used for purification. The DNA polymerase used in the amplification can be a DNA polymerase that can amplify longer fragments and has high fidelity, such as TksGflex DNA polymerase.

[0022] The above amplification method can flexibly select the throughput according to the experimental requirements, and can amplify at least one cell or 24×96 cells at a time.

[0023] In order to solve the above technical problems, the present invention also provides a method for capturing chromatin and / or chromosome conformation, which is named scNanoHi-C.

[0024] The method for capturing chromatin and / or chromosome conformation provided by the present invention comprises preparing a sequencing library for third-generation sequencing in capturing single-cell chromatin and / or chromosome conformation according to the above method, performing third-generation sequencing on the sequencing library to obtain sequencing data, and performing data analysis on the sequencing data to complete the capture of chromatin and / or chromosome conformation.

[0025] In the above method, the data analysis may include the following steps:

[0026] A1) preprocessing and filtering the sequencing data to obtain valid sequencing read data of a single cell, including splitting the sequencing data into single cells according to the barcode sequence on the Tn5 adapter and the barcode sequence on the amplification primer to obtain sequencing data of a single cell; removing the sequencing adapter of the single cell sequencing data and performing preliminary quality control to obtain valid sequencing data of a single cell;

[0027] A2) performing Pore-C Snakemake process analysis on the single-cell effective sequencing data to obtain the original single-cell concatemer interaction file;

[0028] A3) filtering the single-cell concatemers in the original single-cell concatemer interaction file to remove the single-cell concatemers containing abnormal chromatin interaction information to obtain a filtered single-cell concatemer interaction file;

[0029] A2) The Pore-C Snakemake process analysis may include the following steps:

[0030] A2-1) Genome alignment: Align the single-cell effective sequencing data to the reference genome to obtain an alignment result file of the sequencing reads; the alignment result file may include alignment fragments and genomic position information of the alignment fragments. The sequencing reads may include more than or equal to 1 alignment fragment;

[0031] A2-2) Haplotype marking: performing haplotype marking on the alignment fragments in the alignment result file according to the known single nucleotide polymorphism site data to obtain an alignment result file of marked haplotypes;

[0032] A2-3) Positioning the restriction fragments: In the comparison result file of the marker haplotype, the restriction fragment information (the restriction fragment in the fragmented genome product) corresponding to the comparison fragment is determined according to the genomic position information of the midpoint of the comparison fragment, and the genomic position information of the restriction fragment is determined as the chromatin interaction position information of the sequencing read segment,

[0033] A2-4) defining the sequencing reads containing the chromatin interaction information as single-cell concatemers, extracting the set of single-cell concatemers in a single cell, and obtaining an original single-cell concatemer interaction file;

[0034] A2-5) constructing a directed acyclic graph (DAG) based on the chromatin interaction information of the sequencing reads and performing optimal traversal identification to obtain high-order concatemers of single cells.

[0035] In the above method, the abnormal chromatin interaction information (i.e., abnormal chromatin interaction) described in A3) may include:

[0036] A3-1) Chromatin interactions originate from two enzyme-cut fragments at adjacent positions on the genome;

[0037] A3-2) Close-range chromatin interaction: The distance between the two enzyme-cut fragments derived from the chromatin interaction on the genome is less than or equal to 1000 nucleotide pairs;

[0038] A3-3) Repeated chromatin interactions: chromatin interactions at the same chromatin interaction position pair in the same single cell;

[0039] A3-4) Chaotic chromatin interactions: chromatin interactions containing more than 10 different chromatin interactions at the chromatin interaction site;

[0040] A3-5) Isolated chromatin interactions: Chromatin interactions in which there are no other chromatin interactions within 10 million nucleotide pairs upstream and downstream of the chromatin interaction site.

[0041] In the above method, the data analysis may also include the step of constructing a single-cell three-dimensional chromatin structure model for the sequencing data, including using the hickit and dip-C algorithms to construct a single-cell three-dimensional chromatin structure model using the filtered single-cell concatemer interaction file at a resolution of 20 kilonucleotide pairs.

[0042] In the above method, the data analysis may also include the steps of performing single-cell dimensionality reduction and clustering of sequencing data, identifying single-cell copy number variations and structural variations, assisting single-cell assembly, identifying single-cell chromatin high-order interactions and / or identifying single-cell high-order collaborative genomic window sets.

[0043] The dimensionality reduction and clustering of single cells may include using the dip-C algorithm to calculate the single cell compartmentalization value at a resolution of one million nucleotide pairs using the filtered single cell concatemer interaction file (chromatin interaction map) to perform dimensionality reduction and clustering of single cells.

[0044] The single-cell copy number variation may include the following steps: using the calculate-cnv algorithm, the circular binary segmentation algorithm and the filtered single-cell concatemer interaction file in the NeoLoopFinder software to predict the copy number variation of the single cell; using the hic_breakfinder algorithm and the filtered single-cell concatemer interaction file to identify the structural variation of the genome.

[0045] The auxiliary single-cell assembly may include the following steps: using the SALSA2 process to assist in assembling the single-cell genome assembly draft using the filtered single-cell concatemer interaction file; using the QUAST algorithm to calculate NG50, NGA50 and using the BUSCO algorithm to calculate the completeness of the assembled genome.

[0046] The identifying of high-order chromatin interactions of single cells may include the steps of calculating the proportion of high-order concatemers in the genome and the proportion of high-order cells containing high-order concatemers.

[0047] The high-order concatemer may be a concatemer with an order greater than or equal to 3. The order may be the number of monomers contained in a single-cell concatemer. The monomer may be each aligned fragment in a sequencing read containing chromatin interaction information. The concatemer may be the entire sequencing read (containing greater than or equal to 2 monomers).

[0048] The method for identifying a single-cell high-order cooperative genomic window set may include the following steps: selecting a target region in the genome and dividing the window; grouping the concatemers according to the coverage of the single-cell concatemers between different cells in the region and generating a preliminary common window set for each group; randomly sampling the preliminary window set to generate a background prediction model; according to the background prediction model and actual observations, counting the observed expected ratio of the number of single cells and the number of concatemers contained in each subset of the window set to be detected; dividing each subset of the detection window set into a high-order interaction subset and a pairwise interaction subset, respectively fitting a generalized linear model of whether the high order is related to the number of single cells or the number of concatemers, and testing its significance to obtain a genomic window set enriched in high-order interactions.

[0049] In order to solve the above technical problems, the present invention also provides a kit for implementing the above-mentioned amplification method, wherein the kit may include the above-mentioned n1 Tn5 transposase complex, the B barcode primer and a cross-linking agent. The cross-linking agent may be formaldehyde and disuccinimidyl glutarate. The cross-linking agent may also be formaldehyde.

[0050] The kit may also contain other reagents required for chromatin and / or chromosome conformation capture, such as all solutions for Hi-C treatment (such as cell membrane lysis solution), restriction endonucleases, DNA polymerases, amplification buffers, DNA purification reagents and one or more of water.

[0051] In order to solve the above technical problems, the present invention also provides a device for capturing single cell chromatin and / or chromosome conformation, characterized in that the device may include the following modules:

[0052] B1) Sequencing data acquisition module: used to prepare a sequencing library for third-generation sequencing of single-cell chromatin and / or chromosome conformation capture using the amplification method of DNA molecules for third-generation sequencing in single-cell chromatin and / or chromosome conformation capture described above, and perform third-generation sequencing on the sequencing library to obtain sequencing data of a single cell;

[0053] B2) Sequencing data analysis module: used to analyze the sequencing data to obtain the results of chromatin and / or chromosome conformation capture of single cells.

[0054] B2) The sequencing data analysis module may include the following modules:

[0055] B2-1) Single-cell data splitting and quality control module: used to split the sequencing data into single cells according to the barcode sequence on the Tn5 connector and the barcode sequence on the amplification primer of the sequencing library to obtain single-cell sequencing data; perform quality control on the single-cell sequencing data to obtain single-cell valid sequencing data;

[0056] B2-2) Pore-C Snakemake process analysis module: used to perform Pore-C Snakemake process analysis on the single-cell effective sequencing data to obtain the original single-cell concatemer interaction file;

[0057] B2-3) Abnormal chromatin interaction information filtering module: used to filter the single-cell concatemers in the original single-cell concatemer interaction file to remove the single-cell concatemers containing abnormal chromatin interaction information to obtain a filtered single-cell concatemer interaction file;

[0058] B2-4) Chromatin and / or chromosome conformation analysis module: used to construct a single-cell three-dimensional chromatin structure model for the filtered single-cell concatemer interaction file, perform single-cell dimensionality reduction and clustering, identify single-cell copy number variation and structural variation, assist single-cell assembly, identify single-cell chromatin high-order interactions and / or identify single-cell high-order collaborative genome window sets, and obtain single-cell chromatin and / or chromosome conformation analysis results.

[0059] In the above device, the Pore-C Snakemake process analysis module described in B2-2) can be established by a method comprising the following steps:

[0060] B2-2-1) Genome alignment: aligning the single-cell effective sequencing data to the reference genome to obtain an alignment result file of sequencing reads;

[0061] B2-2-2) Haplotype labeling: haplotype labeling is performed on the alignment fragments in the alignment result file according to the known single nucleotide polymorphism site data to obtain a haplotype-labeled alignment result file; the alignment result file may include alignment fragments and genomic position information of the alignment fragments. The sequencing read segment may include more than or equal to 1 alignment fragment;

[0062] B2-2-3) Positioning the restriction fragments: In the comparison result file of the labeled haplotypes, the restriction fragments (the restriction fragments in the fragmented genome products) corresponding to the comparison fragments are determined according to the genomic position information of the midpoints of the comparison fragments, and the genomic position information of the restriction fragments is determined as the chromatin interaction position information of the sequencing reads.

[0063] B2-2-4) defining the sequencing reads containing the chromatin interaction information as single-cell concatemers, extracting the set of single-cell concatemers in a single cell, and obtaining an original single-cell concatemer interaction file;

[0064] B2-2-5) Based on the chromatin interaction information of the sequencing reads, a directed acyclic graph (DAG) is constructed and optimal traversal identification is performed to obtain high-order concatemers of single cells.

[0065] In the above device, the abnormal chromatin interaction information (i.e., abnormal chromatin interaction) described in B2-3) may include:

[0066] B2-3-1) Chromatin interaction originates from two enzyme-cut fragments at adjacent positions on the genome;

[0067] B2-3-2) Close-range chromatin interaction: the distance between the two enzyme-cut fragments derived from the chromatin interaction on the genome is less than or equal to 1000 nucleotide pairs;

[0068] B2-3-3) Repeated chromatin interactions: chromatin interactions at the same chromatin interaction position pair in the same single cell;

[0069] B2-3-4) Chaotic chromatin interactions: chromatin interactions containing more than 10 different chromatin interactions at the chromatin interaction site;

[0070] B2-3-5) Isolated chromatin interactions: Chromatin interactions in which there are no other chromatin interactions within 10 million nucleotide pairs upstream and downstream of the chromatin interaction site.

[0071] In the above-mentioned device, the dimensionality reduction and clustering of single cells may include the steps of using the dip-C algorithm to calculate the single cell compartmentalization value at a resolution of one million nucleotide pairs using the filtered single cell concatemer interaction file (chromatin interaction map) to perform dimensionality reduction and clustering of single cells.

[0072] The single cell copy number variation may include the following steps:

[0073] The calculate-cnv algorithm, the circular binary segmentation algorithm and the filtered single-cell concatemer interaction file in the NeoLoopFinder software were used to predict the copy number variation of single cells; the hic_breakfinder algorithm and the filtered single-cell concatemer interaction file were used to identify the structural variation of the genome.

[0074] The auxiliary single-cell assembly may include the following steps: using the SALSA2 process to assist in assembling the single-cell genome assembly draft using the filtered single-cell concatemer interaction file; using the QUAST algorithm to calculate NG50, NGA50 and using the BUSCO algorithm to calculate the completeness of the assembled genome.

[0075] The identification of single-cell chromatin high-order interactions may include the steps of calculating the proportion of high-order concatemers in the genome and the proportion of high-order cells containing high-order concatemers. The identification of single-cell high-order collaborative genome window sets may include the following steps: selecting a target region in the genome and dividing the window; grouping the concatemers according to the coverage of the single-cell concatemers in the region between different cells and generating a preliminary common window set for each group; randomly sampling the preliminary window set to generate a background prediction model; according to the background prediction model and the actual observation value, the observation expectation ratio of the number of single cells and the number of concatemers contained in each subset of the window set to be detected is counted; each subset of the detection window set is divided into a high-order interaction subset and a paired interaction subset, and a generalized linear model of whether the high order is related to the number of single cells or the number of concatemers is fitted respectively, and its significance is tested to obtain a genome window set enriched in high-order interactions.

[0076] In order to solve the above technical problem, the present invention also provides a computer-readable storage medium, which can enable a computer to execute the steps of any of the above methods.

[0077] The sequencing described above can be performed based on a third-generation sequencing platform, such as a PacBio sequencing platform or a Nanopore platform.

[0078] In order to solve the above technical problems, the present invention also provides a cell processing method for capturing single cell chromatin and / or chromosome conformation.

[0079] The cell processing method for capturing single cell chromatin and / or chromosome conformation provided by the present invention comprises the following steps:

[0080] S1. performing in vitro cross-linking and fixation of target cells to obtain fixed target cells, collecting the fixed target cells and lysing them with a lysate to obtain a lysate;

[0081] S2. After removing the protein in the lysate, the lysate is digested with a restriction endonuclease and then ligated with T4 DNA ligase to obtain target cells containing adjacently ligated DNA; the target cells containing adjacently ligated DNA are separated into single cells, and the single cells are decrosslinked and the cell genome is lysed to obtain a genomic lysate of the single cell, thereby completing the cell treatment.

[0082] In the above S1, the in vitro cross-linking and fixation may be performed by treating the target cells with formaldehyde and then with disuccinimidyl glutarate.

[0083] In the above S1, the target cells may be 0.25×10 6 Cells-1.00×10 6 cells.

[0084] In the above S1, in order to reduce the noise of random connection, the in vitro cross-linking fixation can be to treat the target cells with formaldehyde and then treat them with disuccinimidyl glutarate. The formaldehyde treatment can be to treat with a 0.8%-2.0% formaldehyde solution, such as a 1% formaldehyde solution. The subsequent treatment with disuccinimidyl glutarate can be to treat with a 0.6mM-3.0mM disuccinimidyl glutarate (DSG) solution, such as a 1.5mM disuccinimidyl glutarate solution.

[0085] The cells in the present invention may be mammalian cells, which is a broad term and may refer to cell lines or cells after tissue digestion, including but not limited to cells isolated from bone marrow, blood, serum, platelets, plasma, interstitial fluid, urine, and cerebrospinal fluid.

[0086] The high-order concatemer described above may be a concatemer with an order greater than or equal to 3. The order may be the number of monomers contained in a single-cell concatemer. The monomer may be each aligned fragment in a sequencing read containing chromatin interaction information. The concatemer may be the entire sequencing read (containing greater than or equal to 2 monomers).

[0087] In view of the shortcomings of the prior art and actual needs, the present invention provides a single-cell chromatin conformation capture (Hi-C) method based on proximity ligation of third-generation sequencing. Single-cell Hi-C is combined with third-generation sequencing technology. The starting amount of cells can be reduced by optimizing the reaction system, and the signal-to-noise ratio can be improved by combining cross-linking agents. The flexible amplification method invented can appropriately select the flux, thereby providing sufficient single-cell-level interaction information for detecting whole-genome higher-order chromatin, and can be expanded to meet the high-throughput requirements of distinguishing cell populations through alternative sequence strategies.

[0088] In the present invention, the third generation sequencing includes single molecule real-time DNA sequencing (SMRT) and nanopore single molecule sequencing.

[0089] The present invention combines the advantages of high signal-to-noise ratio single-cell Hi-C and long read length of third-generation sequencing, providing a method for detecting comprehensive and accurate high-order interactions at the single-cell level.

[0090] The present invention adds a barcode sequence to the linker sequence of the Tn5 transposase and to the PCR amplification primer sequence, and flexibly combines the two to achieve single-cell third-generation Hi-C sequencing for the needs of single cells with different fluxes, and mixes DNA amplification products from different sources to meet the requirements of the third-generation sequencing platform for the amount of templates to achieve accurate sequencing of single-cell third-generation Hi-C. Combined with supporting bioinformatics analysis methods, this method can achieve single-cell three-dimensional genome modeling, single-cell dimensionality reduction and clustering, single-cell copy number variation, single-cell structural variation identification, auxiliary single-cell assembly and / or identification of single-cell chromatin high-order interactions. BRIEF DESCRIPTION OF THE DRAWINGS

[0091] Figure 1 This is a basic flowchart of the construction of the scNanoHi-C library based on nanopore single-cell sequencing in the present invention.

[0092] Figure 2 This is an analysis diagram of the genome length after the restriction endonuclease cuts the cross-linked cells in Example 1 of the present invention, and after T4 DNA ligase ligation and amplification.

[0093] Figure 3Comparison of chromatin domains, chromatin A / B compartments, topologically associated domains (TADs), and chromatin loops captured by the sequencing method of the present invention (scNanoHi-C) and second-generation Hi-C sequencing (Bulk Hi-C). A is a heat map of the chromatin interaction matrix of scNanoHi-C and Bulk Hi-C at the whole genome level; B is a scatter plot of the correlation between the original interaction number of scNanoHi-C and Bulk Hi-C, with the horizontal axis being the original interaction number of Bulk Hi-C and the vertical axis being the original interaction number of scNanoHi-C; C is a heat map of representative chromatin compartments in scNanoHi-C and Bulk Hi-C; D is a scatter plot of the correlation between the chromosome compartment score of representative chromosomes in scNanoHi-C and Bulk Hi-C, with the horizontal axis being the chromatin compartment score of scNanoHi-C and the vertical axis being the chromatin compartment score of Bulk Hi-C; E is a heat map of representative topologically related domains in scNanoHi-C and Bulk Hi-C; F is a scatter plot of the correlation between the insulation score of scNanoHi-C and Bulk Hi-C, with the horizontal axis being the insulation score of scNanoHi-C and the vertical axis being the insulation score of Bulk Hi-C.

[0094] Figure 4 Flowchart for scNanoHi-C sequencing data preprocessing.

[0095] Figure 5 Flowchart for filtering single-cell interaction pairs.

[0096] Figure 6 Schematic diagram for calculating the proportion of high-order concatemers. A is an analysis summary of the proportion of high-order interactions in single cells; B is a schematic diagram for calculating the proportion of high-order interactions in different genomic windows. The upper part of the figure shows the distribution of chromatin compartment scores in the selected region, the middle part of the figure shows the distribution of single-cell concatemers in the region, and the lower part of the figure shows the total number of single-cell concatemers in each genomic window in the region (curve graph) and the proportion of high-order concatemers (bar graph). The horizontal axis is the genomic window, and the vertical axis is the number of single-cell concatemers (right, curve graph) and the proportion of high-order concatemers (left, bar graph).

[0097] Figure 7The figure is a schematic diagram of the single-cell synergy analysis process. i is to select the target region within the genome and divide it into genomic windows of the same size (usually 20 kilonucleotide pairs), and at the same time, construct a nearest neighbor graph for the concatemers according to the distribution of concatemers in the region of different single cells; ii is to divide the concatemer groups according to the nearest neighbor graph of single-cell concatemers, and only the groups containing at least three concatemers from three cells are retained for subsequent analysis; iii is to identify the common window set in each concatemer group; iv is to use all the preliminarily identified window sets as training sets, and randomly sample to generate a random background model. The random background model can predict the number of single cells and concatemers contained in each subset of the window set under the condition of given window set properties. The window set properties include the average distance between windows in the window set, the number of windows, the GC base content of the window, the total length of the window, etc.; v is to evaluate the significance of the enrichment of high-order interactions in the preliminarily identified window set. First, according to the above random background model, the expected number of single cells and the expected number of concatemers for each subset of the window set to be tested were generated, and compared with the actual number of single cells and the actual number of concatemers observed for each subset of the window set, and the observed expectation ratio was calculated. Then, according to whether it belongs to a high-order interaction, the subset of the window set is divided into a high-order interaction subset and a pairwise interaction subset, which are used to fit the gamma-Poisson generalized linear model of the subset order and the number of single cells or concatemers, respectively, to determine whether the number of single cells or concatemers in the high-order interaction subset is significantly higher than that in the pairwise interaction. For this model, the significant P value is calculated using the Wald test and the Benjamini–Hochberg method is used for multiple testing correction to obtain the false discovery rate (FDR). When the FDR of the single cell number and concatemer number models of the window set are both less than 0.05, the window set is identified as a window set enriched for high-order synergy. DETAILED DESCRIPTION

[0098] In order to better explain the present invention, reference to embodiments of the present invention will now be provided in detail, one or more examples of which are described below, but the content of the present invention is not limited to the following examples. In fact, it is obvious to those skilled in the art that various modifications and changes can be made to the present invention without departing from the scope or spirit of the present invention. For example, a feature described or described as part of one embodiment can be used in another embodiment to produce a further embodiment.

[0099] Therefore, it is intended that the present invention covers such modifications and variations that fall within the scope of the appended claims and their equivalents. Other objects, features and aspects of the present invention are disclosed in or are apparent from the following detailed description. It should be understood by those of ordinary skill in the art that this discussion is only a description of exemplary embodiments and is not intended to limit the broader aspects of the present invention. If no specific techniques or conditions are specified in the examples, the techniques or conditions described in the literature in the art or in accordance with the product instructions are used. If the manufacturer of the reagents or instruments used is not specified, they are all conventional products that can be purchased through regular channels.

[0100] Example 1: Single-cell high-throughput chromatin conformation capture (Hi-C) third-generation sequencing method - scNanoHi-C

[0101] 1. Cell processing methods for single cell Hi-C

[0102] The present invention uses the B cell-derived cell line GM12878 (Coriell Institute) for experiments. The cell treatment method of scNanoHi-C is to use formaldehyde and disuccinimidyl glutarate (DSG) to perform in vitro double crosslinking fixation on the target cell line; in the crosslinked DNA proximity ligation step, the crosslinked DNA after enzyme digestion is directly adjacently ligated using a ligase, omitting the biotin enrichment step to reduce losses; and finally sorting to obtain single cells containing DNA interaction information.

[0103] 1.1 Cross-linking and fixation of cells

[0104] The target cells are fixed by double cross-linking in vitro using formaldehyde and disuccinimidyl glutarate (DSG) successively to obtain fixed target cells. The specific method is as follows:

[0105] Collect 0.5×10 6 GM12878 cells were centrifuged at 300×g for 5 minutes, resuspended in HBSS and centrifuged again, and a formaldehyde solution with a final concentration of 1% was added (the solvent of the formaldehyde solution was PBS with a pH of 7.4, and the solute was formaldehyde. The composition of PBS is (1mM KH2PO4, 155mM NaCl, 3mM Na2HPO4-7H2O, and the rest is water), and the cross-linking was fixed and rotated on a rotator at room temperature for 10 minutes. Subsequently, 1% BSA solution was added to terminate the cross-linking reaction, and the reaction was terminated at room temperature for 5 minutes and then on ice for 15 minutes to obtain GM12878 cells fixed with formaldehyde.

[0106] The GM12878 cells fixed with formaldehyde were centrifuged at 1000×g for 10 minutes and then fixed with 1.5mM disuccinimidyl glutarate (DSG) (Thermo Scientific, Fisher cat#20593) liquid on a rotator at room temperature for 40 minutes. After that, 1% BSA solution was added to terminate the reaction, thereby obtaining GM12878 cells fixed with formaldehyde and DSG. The 1.5mM DSG liquid is a liquid obtained by dissolving DSG in DMSO and then diluting it with water to a DSG content of 1.5mM.

[0107] The GM12878 cells double-fixed with formaldehyde and DSG were centrifuged at 2000×g for 10 minutes, then washed once with 0.1% BSA solution, and the fixed cell pellets were collected after centrifugation at 2000×g for 10 minutes to obtain the fixed GM12878 cells, which were stored at -80°C for subsequent experiments.

[0108] 1.2 Cell lysis

[0109] Add cell membrane lysis buffer to the fixed GM12878 cell pellet obtained in step 1.1 and lyse on ice for 20 minutes. Centrifuge at 2500×g for 5 minutes, resuspend with 50 μL 0.5% SDS solution, react at 62°C for 10 minutes to remove certain proteins, and obtain a lysate solution, DNA in the lysate, and DNA cross-links.

[0110] The composition of the cell membrane lysis solution is: pH 8.0, 10 mM Tris-HCl, 10 mM NaCl, 0.2% Igepal CA630 (Sigma, Catalog No.: I8896-50ML), 2.5 μL 100×complete proteinase inhibitor (Roche, Catalog No.: 11836145001), and the rest is water.

[0111] 1.3 Endonuclease digestion

[0112] A mixture for neutralizing SDS (145 μL ddH2O and 25 μL 10% (v / v) Triton X-100) was directly added to the cell lysate solution obtained in step 1.2, and the mixture was reacted at 37°C for 15 minutes to neutralize the SDS and obtain a cross-linked solution containing only DNA and DNA.

[0113] Then add 25 μL of 10× rCutSmart TM Buffer and 20 μL of 25,000 units / mL 4-base restriction endonuclease MboI (NEB, cat#R0147M) were added and rotated at 37°C for 12 hours to obtain a cross-linked product solution after enzyme digestion.

[0114] 1.4 Proximity Connection

[0115] T4 DNA ligase was directly added to the cross-linked product solution after enzyme digestion for ligation, and the reaction was carried out at 16°C for 4 hours to obtain target cells containing adjacent ligated DNA.

[0116] The ligation reaction system was as follows: 100 μL 10×T4 DNA ligase buffer, 5 μL BSA, 865 μL H2O, and 5 μL T4 DNA ligase.

[0117] 1.5 Single cell sorting

[0118] The target cells containing the adjacent ligated DNA were centrifuged at 2500×g for 5 minutes, washed once with 0.1% BSA, and then the treated cells were separated by single cells using flow cytometry. The separated 96 single cells (single cells) were placed in a 96-well plate, with one single cell in one well.

[0119] 1.6 Cross-linking and cell genome lysis

[0120] Add 2 μL of lysis buffer to each well containing single cells to reverse cross-linking and lyse the cell genome: react at 50°C for 3 hours, 70°C for 20 minutes, and 68°C for 45 minutes to obtain genomic lysates of 96 single cells.

[0121] The composition of the lysis buffer is as follows: 0.04 μL 1M Tris-HCl (pH 8.0), 0.08 μL 500 mM NaCl, 0.03 μL 10% Triton X-100 (Sigma), 0.0004 μL 0.5M EDTA, 0.05 μL 20 mg / mL QIAGEN protease, and the rest is water.

[0122] The present invention uses a third-generation sequencing method based on nanopore single-cell sequencing. The basic process of constructing a third-generation sequencing library of scNanoHi-C is as follows: Figure 1 As shown, the specific steps are as follows:

[0123] 2.1 Preparation of fragmented genomic products from single cells

[0124] The adjacent connected genome fragments after cleavage were fragmented using Tn5 transposase. In order to amplify longer fragments, the same adapters (containing barcode sequences) were added to both ends of the Tn5 transposon sequence. In order to improve the subsequent amplification throughput, 24 Tn5 adapters with different barcodes of 24nt (Tn5bar1-Tn5bar24) were designed, which are suitable for both Nanopore and PacBio platforms. Each Tn5 adapter sequence contains an anchor sequence, a barcode sequence, and an ME sequence. The barcode sequence in each Tn5 adapter sequence is different, which is used to distinguish the DNA of different individual cells.

[0125] In a 96-well plate, the genome lysate of each single cell is subjected to genome fragmentation and labeling using one Tn5 transposase complex. The genome lysate of each 24 single cells of the 96 single cells is subjected to genome fragmentation and labeling using 24 Tn5 transposase complexes, respectively, to obtain fragmented genome products of 24 single cells, wherein the label (barcode sequence of the Tn5 adapter) in the fragmented genome product of each single cell in the fragmented genome products of the 24 single cells is different, and the label is a DNA for distinguishing different single cells, and the label is a DNA with a length of 24 nt.

[0126] Among the 24 Tn5 transposase complexes, each of the Tn5 transposase complexes is a substance formed by combining a Tn5 adapter and a Tn5 transposase, and the same Tn5 adapter is added to both ends of the Tn5 transposase. The 24 Tn5 adapters are double-stranded DNA fragments, one chain (Adapter barcode primer, the nucleotide sequence is sequence 1) is connected by a fixed sequence, a barcode sequence and a reverse complementary sequence of an ME sequence, the fixed sequence is a single-stranded DNA used as a protective base and / or a part of the primer sequence, the barcode sequence is a single-stranded DNA used to distinguish different individual cells, and the ME sequence is a mosaic end (Mosaic end, ME) sequence of the Tn5 transposon. The length of the fixed sequence is 14 nt, and the sequence is 5'-TCGTCGGCAGCGTC-3' (nucleotides 1-14 of sequence 1); the length of the barcode sequence is 24 nt (Formula 1, i.e., 24 N in sequence 1, N is A, T, C or G), and the reverse complementary sequence of the ME sequence is 5'-AGATGTGTATAAGAGACAG-3' (nucleotides 39-57 of sequence 1). The barcode sequences of the Tn5 adapters in the 24 Tn5 transposase complexes are all different, and the fixed sequence and the ME sequence are all the same; the other chain of the Tn5 adapter (Adapter MEprimer) is a single-stranded DNA fragment with a nucleotide sequence of the ME sequence, and the sequence is 3'-TCTACACATATTCTCTGTC-5' (nucleotides 1-19 of sequence 2), and the 5' end is phosphorylated (phos). The specific method is as follows.

[0127] 2.2 Preparation of Tn5 transposase complex

[0128] 24 kinds of Tn5 linkers were prepared, and the structures of the 24 kinds of Tn5 linkers are as shown in Formula 1:

[0129]

[0130] 3'-NH3-TCTACACATATTCTCTGTC-phos-5' (sequence 2) Formula 1.

[0131] The barcode sequences of the Tn5 adapters in the 24 Tn5 transposase complexes are all different, and the fixed sequence and the ME sequence are all the same. The barcode sequences of the Tn5 adapters in the 24 Tn5 transposase complexes are named Tn5 barcode1-24, and the sequences are shown in Table 1.

[0132] Table 1. Barcode sequences of Tn5 adapters in 24 Tn5 transposase complexes

[0133]

[0134]

[0135] Note: The bold marked sequences are barcode sequences.

[0136] Each of the 24 Tn5 linkers was prepared as follows:

[0137] The adapter barcode primers diluted to 200 μM were annealed with Adapter ME at a ratio of 1:1 to obtain a Tn5 adapter mix. The gradient annealing reaction was incubated at 95°C for 5 minutes in a PCR instrument, then cooled to 75°C at a rate of 0.1°C / s, and finally the mixture was transferred to 75°C hot water and naturally cooled to room temperature to obtain Tn5 Adapter Mix (containing Tn5 adapters). 24 types of Tn5 adapters were prepared according to the above method.

[0138] Each of the 24 Tn5 transposase complexes was prepared as follows:

[0139] 1 μL of 2 μg / μL Tn5 transposase solution (Nanjing Novozyme Co., Ltd., Catalog No. TD601), 0.7 μL of 100 μM Tn5AdapterMix, 3 μL Coupling Buffer (Nanjing Novozyme Co., Ltd., Catalog No. TD601), fully pipetted and mixed, incubated at 30°C for 1 hour, and obtained a Tn5 transposase complex with different barcodes (TTE Mix, Tn5 transposase mixture), the stock concentration was 400 ng / μL, and stored at -20°C. According to the above method, 24 Tn5 transposase complexes with different barcodes (TTEMix, Tn5 transposase mixture) were prepared.

[0140] 2.3 Genome fragmentation and labeling

[0141] The genomic lysate of each single cell was fragmented and labeled using a Tn5 transposase complex, and each well of a 96-well plate was added

[0142] 2μL 5×TAPS_PEG8K (50mM TAPS-NaOH Sigma, Catalog No. T0647-100G, pH 8.3, 25mMMgCl2, 40% PEG8K (Sigma, Catalog No. P2139-500G), the rest is water), 1μL 0.2ng / μL Tn5 transposase mixture, 4.5μL H2O. React at 37℃ for 2h to obtain fragmented genome products of 96 single cells. Among them, the barcodes of the Tn5 adapters in the Tn5 transposase mixture are any one of Tn5 barcode1-24, and the specific addition method of 96 wells is as follows Figure 1As shown. Eight Tn5 transposase mixtures with barcodes of Tn5 adapters Tn5barcode1-8 were added to the 8 wells A1, B1, C1, D1, E1, F1, G1 and H1, eight Tn5 transposase mixtures with barcodes of Tn5 adapters Tn5barcode1-8 were added to the 8 wells A4, B4, C4, D4, E4, F4, G4 and H4, eight Tn5 transposase mixtures with barcodes of Tn5 adapters Tn5barcode1-8 were added to the 8 wells A7, B7, C7, D7, E7, F7, G7 and H7, eight Tn5 transposase mixtures with barcodes of Tn5 adapters Tn5barcode1-8 were added to the 8 wells A10, B10, C10, D10, E10, F10, G10 and H10, eight Tn5 transposase mixtures with barcodes of Tn5 adapters Tn5barcode1-8 were added to the 8 wells.

[0143] Eight kinds of Tn5 transposase mixtures with barcodes of Tn5 adapters Tn5barcode9-16 were added to the 8 wells A2, B2, C2, D2, E2, F2, G2 and H1, eight kinds of Tn5 transposase mixtures with barcodes of Tn5 adapters Tn5barcode9-16 were added to the 8 wells A5, B5, C5, D5, E5, F5, G5 and H5, eight kinds of Tn5 transposase mixtures with barcodes of Tn5 adapters Tn5barcode9-16 were added to the 8 wells A8, B8, C8, D8, E8, F8, G8 and H8, eight kinds of Tn5 transposase mixtures with barcodes of Tn5 adapters Tn5barcode9-16 were added to the 8 wells A11, B11, C11, D11, E11, F11, G11 and H11, eight kinds of Tn5 transposase mixtures with barcodes of Tn5 adapters Tn5barcode9-16 were added to the 8 wells

[0144] The barcodes of Tn5 adapters are Tn5barcode17-24, and the barcodes of Tn5 adapters are 8 kinds of Tn5 transposase mixtures added to the 8 wells A3, B3, C3, D3, E3, F3, G3 and H1. The barcodes of Tn5 adapters are Tn5barcode17-24, and the ... The barcodes of the 8 Tn5 transposase mixtures of barcode17-24 are respectively added to the 8 wells A12, B12, C12, D12, E12, F12, G12 and H12, and the barcodes of the 8 Tn5 transposase mixtures of barcode17-24 are respectively added to the Tn5 adapters.

[0145] The 24 wells in columns 1-3 are the first group, only one Tn5 transposase mixture is added to each well, and the Tn5 transposase mixture added to each well of the first group is different; the 24 wells in columns 4-6 are the second group, only one Tn5 transposase mixture is added to each well, and the Tn5 transposase mixture added to each well of the second group is different; the 24 wells in columns 7-9 are the third group, only one Tn5 transposase mixture is added to each well, and the Tn5 transposase mixture added to each well of the third group is different; the 24 wells in columns 10-12 are the fourth group, only one Tn5 transposase mixture is added to each well, and the Tn5 transposase mixture added to each well of the fourth group is different.

[0146] The average length of the fragmented genomic products from 96 individual cells was between 3 and 6 kb ( Figure 2 ).

[0147] 2.4 Preparation of PCR products for next-generation sequencing in single-cell chromatin and / or chromosome conformation capture

[0148] After mixing the fragmented genome products of every 24 single cells with different barcode sequence tags, a group of fragmented genome products was obtained, and PCR amplification was performed using the same amplification primer. After mixing the fragmented genome products of 96 single cells, 4 groups of fragmented genome PCR products were obtained. Each group of the fragmented genome products in the 4 groups of fragmented genome products was amplified using 1 amplification primer and 4 amplification primers were used to amplify to obtain 4 PCR products. The PCR products of each single cell in each PCR product have different barcode label combinations, and each barcode label combination is a combination of two barcode labels (the barcode sequence of the Tn5 adapter and the barcode sequence of the amplification primer). Subsequently, the different barcode label combinations carried by the PCR products of each single cell are used to split the third-generation sequencing data of the PCR products of each single cell.

[0149] The four PCR products are mixed to obtain PCR products for third-generation sequencing in single-cell chromatin and / or chromosome conformation capture.

[0150] The sequences of the four barcode primers are all different. The 96 amplification primers are all single-stranded DNA fragments formed by connecting an anchor sequence, a barcode sequence and the fixed sequence X. The anchor sequence is 5'-AATGATACGGCGACCACCGAGATCT-3' (the double-underlined sequence in Formula 3, i.e., nucleotides 1-25 of Sequence 3). One amplification primer contains a barcode sequence, which is a single-stranded DNA used to distinguish different groups of fragmented genomic products. The barcode sequences in the 96 amplification primers (24 Ns in Formula 3, N is A, T, C or G) are all different. The anchor sequence and the fixed sequence X are all the same. The sequence of the fixed sequence X is 5'-TCGTCGGCAGCGTC-3' (the single-underlined sequence in Formula 3, i.e., nucleotides 50-63 of Sequence 3).

[0151] The general formula of the four amplification primers (named I5 PCR primer1-4) is Formula 3:

[0152]

[0153] The barcode sequences in the four amplification primers are all different. The barcode sequences in the four amplification primers are named PCRbar1-4, and the sequences are shown in Table 2.

[0154] Table 2. Barcode sequences of the barcodes in the four amplification primers

[0155]

[0156] The barcode sequences in the 96 amplification primers are all different. The names of the barcode sequences in the 96 amplification primers are PCR bar1-96, and the sequences of the 96 amplification primers are shown in Table 3.

[0157] Table 3. Sequences of 96 amplification primers containing 96 barcode sequences

[0158]

[0159]

[0160]

[0161]

[0162] Note: The bold marked sequences are barcode sequences.

[0163] In this example, an amplification experiment of genomic lysates of 96 single cells was performed, in which each 24 cells were fragmented and labeled with TTE Mix with different barcodes to obtain 24 fragmented genomic products of single cells (i.e., 24 products after Tn5 transposase fragmentation with different barcodes). The fragmented genomic products of the 24 single cells were then mixed together, purified once with 0.8×XP beads, and eluted with 25-50μL H2O. The fragmented genomic products of 96 single cells were purified in four tubes to obtain 4 groups of fragmented genomic products. Each of the four groups of fragmented genome products was amplified using one barcode primer, and the barcode primer used for each of the four groups of fragmented genome products was different. The PCR product obtained by amplifying the first group of fragmented genome products (corresponding to the 24 wells in columns 1-3) using I5 PCR primer1 was named PCR product 1; the PCR product obtained by amplifying the second group of fragmented genome products (corresponding to the 24 wells in columns 4-6) using I5 PCR primer2 was named PCR product 2, the PCR product obtained by amplifying the third group of fragmented genome products (corresponding to the 24 wells in columns 7-9) using I5 PCR primer3 was named PCR product 3, and the PCR product obtained by amplifying the fourth group of fragmented genome products (corresponding to the 24 wells in columns 10-12) using I5 PCR primer4 was named PCR product 4.

[0164] Each tube of purified fragmented genomic products (each group of fragmented genomic products) was amplified with PCR primers with different barcodes. The reaction system was as follows: 2μL 1.25U / μL Tks Gflex DNA polymerase (TAKARA, Cat#R060B), 50μL 2×Gflex PCR buffer, 13μL H2O and 5μL 10uM I5 PCR primers. Among them, I5PCR primers were I5 ​​PCR primer1, I5 PCR primer2, I5 PCR primer3 or I5PCR primer4.

[0165] The amplification reactions were as follows:

[0166] The gaps were filled at 72°C for 3 minutes, and then amplified at 98°C for 1 minute for 18-24 cycles: 98°C for 15 s, 60°C for 30 s, and 68°C for 5 minutes.

[0167] PCR product 1, PCR product 2, PCR product 3 and PCR product 4 were mixed together, purified once with 0.4-0.8×XP beads (Beckman, catalog number 19644800), and purified a second time with 0.6-0.8×XP beads according to the situation to obtain a PCR product for third-generation sequencing. The PCR product for third-generation sequencing is a DNA molecule for third-generation sequencing used in single-cell chromatin and / or chromosome conformation capture.

[0168] 2.5. Construction and sequencing of Nanopore libraries

[0169] According to the requirements of the Oxford Nanopore PromethION sequencing platform, the PCR products used for the third-generation sequencing were subjected to sequencing library construction and single-cell sequencing. The nanopore sequencing library was constructed according to the instructions using the Ligation Sequencing Kit 1D (ONT, Cat. SQK-LSK109), and connected to the 1D connector using a quick connection module (NEB, Cat. E6056); thereafter, each library was loaded into a FLOPRO002 channel and sequenced on the PromethION to obtain sequencing data, which was referred to as scNanoHi-C sequencing data.

[0170] 3. scNanoHi-C sequencing data analysis

[0171] 3.1 scNanoHi-C sequencing data preprocessing and filtering

[0172] The present invention uses Snakemake software (https: / / snakemake.github.io) to develop a reproducible processing flow for scNanoHi-C sequencing data ( Figure 4 ).

[0173] 3.1.1 Single-cell data splitting

[0174] The sequencing data was firstly split into single cells according to the barcode sequence (including the barcode sequence on the Tn5 adapter and the barcode sequence on the barcode primer) using the Nanoplexer algorithm (https: / / github.com / hanyue36 / nanoplexer) to obtain the scNanoHi-C sequencing data of each single cell.

[0175] 3.1.2 Sequencing data quality control

[0176] For the scNanoHi-C sequencing data of each single cell, Cutadapt (https: / / github.com / marcelm / cutadapt) was used to remove the sequencing adapter sequence; at the same time, the sequencing reads with a length of less than 500 nucleotide pairs were filtered out to obtain the valid sequencing read data of each single cell.

[0177] 3.1.3 Pore-C Snakemake Process Analysis

[0178] The valid sequencing read data of each single cell were preliminarily processed using the following Pore-C snakemake pipeline (https: / / github.com / nanoporetech / Pore-C-Snakemake / ).

[0179] (1) Genome alignment

[0180] The bwasw algorithm (https: / / github.com / lh3 / bwa) was used to align the valid sequencing read data of each single cell from the GM12878 cell line with the GRCh38 human reference genome to obtain the alignment result file of the sequencing reads. The alignment result file of the sequencing reads contains the alignment fragments (multiple genomic fragments contained in each sequencing read) and the genomic position information of the alignment fragments (each sequencing read contains multiple genomic fragments, and each genomic fragment is aligned to the corresponding genomic position).

[0181] (2) Haplotype markers

[0182] Use the whatshap algorithm (https: / / whatshap.readthedocs.io / ) to perform haplotype tagging on each alignment fragment contained in each sequencing read in the alignment result file according to the known GM12878 single nucleotide polymorphism sites (from https: / / github.com / Illumina / PlatinumGenomes) to obtain the alignment result file of the marked haplotype;

[0183] (3) Positioning of enzyme-cut fragments

[0184] In the comparison result file of the marker haplotype obtained, according to the position coordinates of the base at the midpoint of each comparison fragment sequence on the reference genome, the comparison fragment (the comparison fragment is converted into its corresponding enzyme fragment according to the midpoint of the comparison) is assigned to the restriction endonuclease digested fragment containing the position coordinates, that is, the restriction endonuclease digested fragment where each comparison fragment is located on the sequencing read is determined, and the restriction endonuclease digested fragments where multiple comparison fragments are located on the same sequencing read are considered to have DNA interactions, and the restriction endonuclease digested fragments contained in the same sequencing read are called interaction pairs, that is, the chromatin interaction information contained in the sequencing read. In the sequencing results of scNanoHi-C, the present invention refers to each comparison fragment in the sequencing read containing chromatin interaction information as a monomer, and the entire sequencing read (containing more than or equal to 2 monomers) is called a single-cell concatemer. The number of monomers contained in a single-cell concatemer is called the order of the concatemer, and a concatemer with an order of 2 is called a paired concatemer, and a concatemer with an order of more than or equal to 3 is called a high-order concatemer. Extract the single-cell concatemer collection and obtain the original single-cell concatemer interaction file.

[0185] (4) Preliminary Filtration of Single-Cell Concatemers

[0186] According to the chromatin interaction information contained in each read segment, a directed acyclic graph (DAG) was constructed and optimally traversed using a greedy algorithm to remove redundant fragments in the original single-cell concatemers, identify and obtain valid single-cell concatemers, and obtain valid single-cell concatemer interaction files.

[0187] 3.1.4 Filtering of single-cell interaction pairs to remove abnormal chromatin interactions in single cells

[0188] Next, the present invention further filters the obtained valid single-cell concatemer interaction file containing the single-cell concatemer set according to the data characteristics of single-cell Hi-C, and removes abnormal chromatin interactions on the basis of retaining the high-order concatemer structure to the maximum extent. The final single-cell concatemer set (filtered single-cell concatemer interaction file) can be further decomposed into virtual paired chromatin interactions, and assigned to genomic windows at different resolutions according to their genomic coordinates to generate an interaction map.

[0189] Specifically, in order to remove abnormal chromatin interactions in single cells, the present invention first decomposes the effective single-cell concatemer interaction file single-cell concatemer into virtual paired chromatin interactions, that is, the restriction endonuclease cut fragments corresponding to each two monomers in the single-cell concatemer are combined to form paired interactions (referred to as interaction pairs).

[0190] The following aberrant chromatin interactions were removed in turn:

[0191] (1) Adjacent chromatin interactions: i.e., chromatin interactions located on adjacent restriction endonuclease fragments (i.e., the two restriction endonuclease fragments have a shared restriction endonuclease site, and the genomic locations of the two aligned fragments are adjacent (directly adjacent));

[0192] (2) close-range chromatin interactions, i.e., the distance between the two restriction endonuclease-cut fragments contained in each of the two interaction pairs on the genome is less than 1000 nucleotide pairs;

[0193] (3) Repeated chromatin interactions, that is, chromatin interactions between different concatemers in the same single cell that come from the same pair of restriction endonuclease fragments.

[0194] In order to retain the high-order structure of the concatemer while removing repeated chromatin interactions, for each set of repeated chromatin interactions in the same single cell, the present invention only retains a single pair of chromatin interactions from the concatemer with the highest order (containing the most monomers);

[0195] (4) promiscuous chromatin interactions, i.e., at least one end of the interaction pair is located on a restriction enzyme fragment containing more than 10 unique (different) pairs of chromatin interactions;

[0196] (5) Isolated chromatin interactions, that is, there are no other chromatin interactions within 10 million nucleotide pairs upstream and downstream of the site where the chromatin interaction occurs.

[0197] 3.1.5 Comparative test of scNanoHi-C and bulk Hi-C

[0198] In order to evaluate the performance of the present invention in actually characterizing the three-dimensional genome structure, this example merged 456 GM12878 single-cell scNanoHi-C results with low sequencing depth (total base amount after quality control was 81.6 Gb, containing a total of 34.1 million pairs of chromatin interaction pairs) and compared them with GM12878 bulk Hi-C data (total base amount was about 826 Gb, containing a total of 4 billion pairs of chromatin interaction pairs, downloaded from https: / / www.4dnucleome.org, accession code 4DNFIXP4QG5B) at multiple levels.

[0199] First, this example compares the merged scNanoHi-C and bulk Hi-C chromatin interaction maps at different resolutions (whole genome, 1 million base pairs, and 50 kbp), and the two show high similarity ( Figure 3 A, C and E).

[0200] Furthermore, this example calculates the raw interaction logarithms of the combined scNanoHi-C and bulk Hi-C ( Figure 3 Middle B), chromatin compartmentalization fraction ( Figure 3 D) and chromatin sequestration fraction ( Figure 3 F), and the Pearson correlation coefficient was used to evaluate their similarity, and the correlation coefficients obtained were 0.95, 0.94 and 0.88, respectively, all showing strong correlation. Specifically, this example uses the cooler software package (https: / / cooler.readthedocs.io / ) to calculate the chromatin compartment score at a resolution of one million base pairs, and the chromatin isolation score calculated under the conditions of 50 kilobase pair resolution and 250 kilobase pair window.

[0201] 3.2 Construction of single-cell three-dimensional chromatin structure model

[0202] The single-cell concatemers in the filtered single-cell concatemer interaction file obtained in step 3.1.4 were first decomposed into virtual pairwise chromatin interactions to obtain a chromatin interaction map, and the hickit algorithm (https: / / github.com / lh3 / hickit) was used to estimate the haplotypes of the remaining chromatin interaction pairs without labeled haplotype information based on the identified haplotype information.

[0203] Next, for each single cell, the present invention uses hickit and dip-C algorithms (https: / / github.com / tanlongzhi / dip-c) to perform three-dimensional genome modeling at a resolution of 20 kilonucleotide pairs (20kb), and compares the three modeling results. In order to measure the reliability of the modeling, for each 20 kilonucleotide pair modeling unit, the present invention calculates the root mean square distance (RMSD) between models, and uses the root mean square value of the RMSD of all modeling units (RMSRMSD) as a measure of modeling repeatability. The present invention only retains cells with an RMS RMSD less than 1.5 for subsequent analysis.

[0204] 3.3 Dimensionality reduction and clustering of single cells using scNanoHi-C data

[0205] For each single cell, the present invention first uses the dip-C algorithm (https: / / github.com / tanlongzhi / dip-c) to calculate the single-cell A / B compartments values ​​(scAB value) according to the chromatin interaction map at a resolution of one million nucleotide pairs, and sorts the scAB values ​​in each single cell and normalizes them to values ​​between 0 and 1. Next, the present invention selects cells containing more than 50,000 chromosome interaction pairs and less than 30% of the chromosome-to-chromosome interaction ratio, and uses the normalized scAB values ​​of the autosomal genome window covered by all these cells to perform principal component analysis (Principal Component Analysis, PCA) to reduce the dimension of the cells. The principal components of the first five dimensions obtained are further reduced using the Uniform Manifold Approximation and Projection (UMAP) (https: / / doi.org / 10.48550 / arXiv.1802.03426) algorithm to obtain the final single cell dimensionality reduction and clustering results.

[0206] 3.4 Identification of copy number variation and structural variation

[0207] 3.4.1 Copy number variation analysis based on scNanoHi-C

[0208] The present invention uses the calculate-cnv algorithm in the NeoLoopFinder software (https: / / github.com / XiaoTaoWang / NeoLoopFinder) to infer the copy number variation of single cells according to the chromosome interaction map and excludes the genomic regions located in the blacklist and the difficult-to-match region (mainly including the constitutive heterochromatin region, the blacklist list is from https: / / github.com / Boyle-Lab / Blacklist / tree / master / lists). The present invention uses the circular binary segmentation algorithm (Circular binary segmentation, CBS) (https: / / github.com / veseshan / DNAcopy) contained in the R software package DNAcopy to segment and estimate the copy number. In order to evaluate the noise size of the copy number variation analysis, the present invention calculates the coefficient of variation of the copy number variation inferred by the single cell based on the whole genome sequencing results.

[0209] 3.4.2 Structural variation analysis based on scNanoHi-C

[0210] The present invention uses the hic_breakfinder algorithm (https: / / github.com / dixonlab / hic_breakfinder) to identify genomic structural variations in the merged scNanoHi-C data and uses the visualization module of NeoLoopFinder (https: / / github.com / XiaoTaoWang / NeoLoopFinder) for visualization.

[0211] 3.5 Using scNanoHi-C to assist single-cell de novo genome assembly

[0212] The present invention uses the SALSA2 process (https: / / github.com / marbl / SALSA) to assist in the assembly of the single-cell genome draft (the single-cell genome draft used in this embodiment is from https: / / doi.org / 10.1093 / nar / gkac586) using the scNanoHi-C data of single cells. In order to evaluate the degree to which the auxiliary assembly improves the continuity and integrity of the genome, the present invention evaluates different indicators, including the calculation of NG50 and NGA50 using the QUAST algorithm and the calculation of genome integrity using the BUSCO algorithm.

[0213] 3.6 Identification of higher-order chromatin interactions

[0214] 3.6.1 Calculation of the ratio of higher-order concatemers

[0215] In order to evaluate the probability of a given genomic region participating in the formation of a higher-order structure, for each 50 kilonucleotide (50 kb) region on the genome, the present invention calculates the proportion of high-order concatemer monomers and the proportion of cells containing high-order concatemers in the cell population ( Figure 6 ). Specifically, the present invention first divides the whole genome into continuous windows of 50 kilonucleotide pairs, and removes windows with more than 5% extreme values ​​according to the distribution of the number of single cells and the number of concatemers in the studied cell population in these windows, and the remaining windows are used to calculate the high-order proportion. The high-order concatemer ratio is defined as the ratio of monomers from high-order concatemers in the region (50 kilonucleotide pair window) to all monomers in the region; the high-order cell ratio is defined as the ratio of cells containing high-order concatemers in the region to all cells covering the region.

[0216] 3.6.2 Analysis of high-order enhancer-promoter interactions based on scNanoHi-C

[0217] scNanoHi-C high-order concatemer data can be used to detect multiplex interactions between enhancers and promoters. Specifically, the present invention first uses the activity-by-contacts model (ABC model, https: / / github.com / broadinstitute / ABC-Enhancer-Gene-Prediction) to predict enhancer-promoter interaction pairs in a given cell type, and obtains paired enhancer-promoter interaction pairs contained in the corresponding cell type (GM12878 cell line in this embodiment) according to default parameters.

[0218] Next, in order to identify one-to-many interactions, the present invention divides the genome into non-repeating windows with a width of 20 kilonucleotide pairs (the resolution is determined according to the coverage of the data set in this embodiment). For each gene, the present invention takes the window where its transcription start site and the enhancer that regulates it as a set, and when there is at least one high-order concatemer that simultaneously covers the window where the transcription start site is located and at least two discontinuous enhancer windows, the gene is identified as a multi-enhancer regulated gene.

[0219] Similarly, the present invention also identifies enhancers that can simultaneously regulate multiple promoters, which requires at least one high-order concatemer that simultaneously covers the window where the enhancer is located and two other discontinuous transcription start site windows.

[0220] 3.6.3. Single-cell synergy model

[0221] In order to identify the cooperative interaction relationship between multiple genomic regions in a single cell, the present invention optimizes the algorithm originally used in Pore-C for identifying chromatin cooperative interaction in a large number of cells according to the characteristics of scNanoHi-C.

[0222] Specifically, the present invention first selects the target genomic region to be detected (in this embodiment, in order to study the synergistic interaction between enhancers and promoters in the GM12878 cell line, the active promoter and enhancer regions in GM12878 are selected according to the chromosome state annotation of the GM12878 cell line), and then divides the selected region into windows of fixed length (a length of 20 kilonucleotide pairs is used in this embodiment). Next, for a given cell population, the present invention extracts high-order concatemers that cover at least three different windows for subsequent analysis. According to the distribution of concatemers in different single cells in the region, and the windows commonly covered between concatemers, the present invention constructs a weighted graph that characterizes the Jaccard similarity between concatemers, while retaining only the relationship between concatemers that cover more than three windows together.

[0223] Furthermore, the present invention constructs a k-nearest neighbor graph based on this, and classifies the concatemers. For each concatemer group, the present invention only retains the windows covered by at least three concatemers from three different cells in the group and merges the adjacent windows to obtain the common window set of the concatemer group. Finally, only the common window set containing at least three windows is retained as the candidate window set for the next step of testing.

[0224] In order to use the synergy model for testing, the present invention first uses a window random sampling algorithm to generate a random background window set for a candidate window set, and uses these background window sets to train a generalized linear model for predicting the number of concatemers and single cells contained in each subset of the candidate window set under a given covariate. The covariates considered in the model include the maximum distance and minimum distance between windows, the window set order, the average GC content, the total number of concatemers or single cells covering the window set, and the total width of the window set. Using this background prediction model, the present invention predicts the number of concatemers and single cells covered by all subsets of each window set, and compares them with the actual observed values ​​to obtain the observed expected ratio. The present invention further establishes two other new generalized linear models for fitting the relationship between the observed expected ratio of the number of concatemers and single cells in each subset and whether the subset is a high-order subset, and then uses the Wald test to calculate the significance and P value of the model, and performs multiple test corrections by the Benjamini-Hochberg method, and converts the P value into a false discovery rate (False Discovery Rate, FDR). When the FDR of the two models is less than 0.05, the window set is identified as a synergistic window set.

[0225] In summary, the present invention adds a barcode sequence to the linker sequence of the Tn5 transposase and to the PCR amplification primer sequence, and flexibly combines the two to achieve single-cell third-generation Hi-C sequencing for the needs of single cells with different fluxes, and mixes DNA amplification products from different sources to meet the requirements of the third-generation sequencing platform for the amount of templates, thereby achieving accurate sequencing of single-cell third-generation Hi-C. Combined with supporting bioinformatics analysis methods, this method can achieve single-cell three-dimensional genome modeling, single-cell dimensionality reduction and clustering, copy number variation and structural variation identification, auxiliary single-cell assembly, and identification of chromatin higher-order interactions.

[0226] The present invention has been described in detail above. For those skilled in the art, without departing from the purpose and scope of the present invention, and without the need to carry out unnecessary experimental conditions, the present invention can be implemented in a wide range under equivalent parameters, concentrations and conditions. Although the present invention provides specific embodiments, it should be understood that the present invention can be further improved. In a word, according to the principles of the present invention, the application is intended to include any changes, uses or improvements to the present invention, including departure from the disclosed scope in the application, and changes made with conventional techniques known in the art.

Claims

1. A method for amplifying DNA molecules for third-generation sequencing in single-cell chromatin and / or chromosome conformation capture, characterized in that: The method comprises the following steps: S1, cross-linking and fixing the target cells in vitro to obtain fixed target cells, collecting the fixed target cells and lysing them with a lysate to obtain a lysate; S2, after removing the protein in the lysate, digesting with restriction endonucleases and then ligating with T4 DNA ligase to obtain target cells containing proximity-ligated DNA; separating the target cells containing proximity-ligated DNA into single cells, decrosslinking the single cells and lysing the cell genome to obtain a single cell genome lysate; S3, using one Tn5 transposase complex to fragment the genome of each single cell and adding a label to obtain a genome fragment containing one barcode sequence label; using n1 single cell genome lysates to fragment the genome of n1 single cells and adding a label to obtain n1 single cell fragmented genome products, wherein the labels in the fragmented genome products of each single cell in the n1 single cell fragmented genome products are different, and the labels are single-stranded DNA used to distinguish different single cells; each of the n1 Tn5 transposase complexes is composed of one Tn5 transposase complex. The substance is formed by combining a head with a Tn5 transposase, the Tn5 connector is a double-stranded DNA fragment, one chain is formed by connecting a fixed sequence, a barcode sequence and a reverse complementary sequence of an ME sequence, the barcode sequence is a single-stranded DNA used to distinguish different single cells, and the genome fragment of each single cell contains the same barcode sequence tag, and the ME sequence is a chimeric end sequence of the Tn5 transposon; the n1 is a natural number greater than or equal to 1; the barcode sequences of the n1 Tn5 connectors in the n1 Tn5 transposase complexes are all different, and the fixed sequence and the ME sequence are all the same; the other chain of the Tn5 connector is a single-stranded DNA fragment with a nucleotide sequence of the ME sequence; S4, the fragmented genomic products of each n1 single cell are mixed to obtain a group of fragmented genomic products, each group of the fragmented genomic products in the group B is amplified using one barcode primer and a common B amplification primer to obtain B PCR products, and the B PCR products are mixed to obtain PCR products for third-generation sequencing in single-cell chromatin and / or chromosome conformation capture; The sequences of the B amplification primers are all different, and each of the B amplification primers is a single-stranded DNA fragment connected by an anchor sequence, a barcode sequence and a fixed sequence X, the anchor sequence is used to protect the single-stranded DNA, and one amplification primer contains a barcode sequence, and the barcode sequence is a single-stranded DNA used to distinguish different groups of fragmented genomic products; the barcode sequences in the B amplification primers are all different, and the anchor sequence and the fixed sequence X are the same; the fixed sequence X is used to amplify the single-stranded DNA, and B is a natural number greater than or equal to 1.

2. The amplification method according to claim 1, characterized in that: The in vitro cross-linking fixation is to treat the target cells with formaldehyde and then with disuccinimidyl glutarate.

3. A method for capturing chromatin and / or chromosome conformation, characterized in that: The method comprises preparing a sequencing library for third-generation sequencing in single-cell chromatin and / or chromosome conformation capture according to the method described in claim 1 or 2, sequencing the sequencing library to obtain sequencing data, performing data analysis on the sequencing data, and completing chromatin and / or chromosome conformation capture.

4. The method according to claim 3, characterized in that: The data analysis comprises the following steps: A1) preprocessing and filtering the sequencing data to obtain valid sequencing read data of a single cell, including splitting the sequencing data into single cells according to the barcode sequence on the Tn5 adapter and the barcode sequence on the PCR amplification primer to obtain sequencing data of a single cell; performing quality control on the single cell sequencing data to obtain valid sequencing data of a single cell; A2) performing Pore-C Snakemake process analysis on the single-cell effective sequencing data to obtain an effective single-cell concatemer interaction file; A3) filtering the single-cell concatemers in the effective single-cell concatemer interaction file to remove the single-cell concatemers containing abnormal chromatin interaction information to obtain a filtered single-cell concatemer interaction file.

5. The method according to claim 3 or 4, characterized in that: The sequencing analysis also includes the step of constructing a single-cell three-dimensional chromatin structure model for the sequencing data, including using the hickit and dip-C algorithms to construct a single-cell three-dimensional chromatin structure model using the filtered single-cell concatemer interaction file at a resolution of 20 kilonucleotide pairs.

6. The method according to claim 3, 4 or 5, characterized in that: The sequencing analysis also includes the steps of performing single-cell dimensionality reduction and clustering of sequencing data, identifying single-cell copy number variations and structural variations, assisting single-cell assembly and / or identifying single-cell chromatin high-order interactions and / or identifying single-cell high-order collaborative genome window sets.

7. A kit for implementing the amplification method according to claim 1 or 2, the kit comprising the n1 Tn5 transposase complex according to claim 1 or 2, the B amplification primer and a cross-linking agent, wherein the cross-linking agent is formaldehyde and disuccinimidyl glutarate, or the cross-linking agent is formaldehyde.

8. A device for capturing single cell chromatin and / or chromosome conformation, characterized in that: The device has the following modules: B1) a sequencing data acquisition module: used to prepare a sequencing library for third-generation sequencing for capturing single-cell chromatin and / or chromosome conformation using the method described in claim 1 or 2, and perform third-generation sequencing on the sequencing library to obtain sequencing data of a single cell; B2) Sequencing data analysis module: used to analyze the sequencing data to obtain the results of chromatin and / or chromosome conformation capture of single cells; B2) The sequencing data analysis module includes the following modules: B2-1) Single-cell data splitting and quality control module: used to split the sequencing data into single cells according to the barcode sequence on the Tn5 connector and the barcode sequence on the sequencing library amplification primer to obtain single-cell sequencing data; perform quality control on the single-cell sequencing data to obtain single-cell valid sequencing data; B2-2) Pore-C snakemake process analysis module: used to perform Pore-C snakemake process analysis on the single-cell effective sequencing data to obtain effective single-cell concatemer interaction files; B2-3) Abnormal chromatin interaction information filtering module: used to filter the single-cell concatemers in the effective single-cell concatemer interaction file to remove the single-cell concatemers containing abnormal chromatin interaction information to obtain a filtered single-cell concatemer interaction file; B2-4) Chromatin and / or chromosome conformation analysis module: used to construct a single-cell three-dimensional chromatin structure model for the filtered single-cell concatemer interaction file, perform single-cell dimensionality reduction and clustering, identify single-cell copy number variation and structural variation, assist single-cell assembly, identify single-cell chromatin high-order interactions and / or identify single-cell high-order collaborative genome window sets, and obtain single-cell chromatin and / or chromosome conformation analysis results.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium enables a computer to execute the steps of any one of claims 4-6.

10. A cell processing method for capturing single-cell chromatin and / or chromosome conformation, comprising S1 and S2 as described in claim 1 or 2.

Images