A method for identifying genomic interactions

Abstract

The present invention relates to an in vitro method for analyzing DNA interactions and kits for implementing said method. The method of the invention is more particularly useful for analyzing interactions between i) single-strand DNA (ssDNA) and ii) double-strand or single- strand DNA (dsDNA or ssDNA).

Description

[0001] A METHOD FOR IDENTIFYING GENOMIC INTERACTIONS

[0002] The present invention relates to the field of molecular biology. It provides an in vitro method for analyzing DNA interactions and kits for implementing said method.

[0003] BACKGROUND

[0004] Spatial DNA interactions in a cell may have an impact on gene regulation, chromosome morphogenesis and genome stability and transmission. That is to say, while a gene is located at a particular position on a particular chromosome, the elements that regulate its activity can lie far away. DNA organization in eukaryotic cells is not random. Linear DNA strands are firstly packed into chromatin, which is a complex of DNA and protein. Chromatin is further folded into loops, which organize domains of preferential interactions and condense the chromosome. This dynamic organization can position two distal genomic regions in close spatial proximity.

[0005] Methods have been developed to analyze such chromatin interactions at the genome-wide scale and investigate the biological roles of the spatial organization of chromatin in the nucleus.

[0006] Molecular biology techniques such as “Chromosome Conformation Capture” coupled with high-throughput sequencing makes it possible to analyze the spatial organization of chromatin in a cell or a cell population. These methods are based on the fixation of the organization of chromatin with a crosslinking agent such as formaldehyde, followed by its partial digestion, and the re-ligation of DNA fragments in such a way that only those that are covalently linked together form chimeric ligation products. In particular, the Hi-C method, which incorporates a biotin marker into chimeric ligation products, enables selecting DNA molecules bearing an informative ligation event for sequencing. Hi-C enables determining the frequencies of contact between pairs of DNA segments of a genome and to infer their three-dimensional organization (Belton et al., 2012). However, one drawback of these methods that involve DNA fragmentation using nucleases is that only the canonical double-stranded DNA form (dsDNA) can be detected. Contacts between two single-stranded DNA (ssDNA) or between a ssDNA and a dsDNA molecule are not detected by these techniques, because the structure recognized by these nucleases is absent. It represents a limitation for these methods to study global DNA interactions in many biological processes such as DNA replication, DNA repair (in somatic cells and in meiosis), reverse-transcription of retrovirus and retrotransposons, etc, because DNA interactions may involve not only dsDNA but also ssDNA. So far, there is no technique able to directly and globally analyze interactions between i) singlestrand DNA (ssDNA) and ii) double-strand or single-strand DNA (dsDNA or ssDNA) in genomic DNA. There is therefore a need to provide a method which can remedy this drawback.

[0007] SUMMARY OF THE INVENTION

[0008] The inventors have now successfully developed a method which addresses the aforementioned drawback. The invention lies in the design of a set of specific oligonucleotides which anneal with ssDNA at target genomic sites such that, upon annealing, dsDNA is restored that encompasses a restriction site at the target genomic site of interest. Said restored dsDNA can then be submitted to restriction enzyme digestion, and then be further labeled and ligated to provide ligation junction fragments. Ligation products thus obtained can be distinguished from ligation products obtained from the genomic region involving interactions between two dsDNA.

[0009] The method of the present invention provides means for analyzing spatial proximity relationships between ssDNA and DNA that can be either ssDNA or dsDNA, in a sample. Indeed, the method of the present invention allows to measure the frequency or an absolute number of specific interactions between ssDNA and ssDNA or dsDNA in a sample. More generally, the method of the present invention makes it possible to analyze global spatial proximity relationships, i.e. interactions between ssDNA and ssDNA or dsDNA, in addition to interactions between two dsDNA, in a sample.

[0010] The method of the invention may be implemented for analyzing DNA interactions in prokaryotes, eukaryotes, archaea or virus. The method of the invention is particularly suitable for analyzing DNA interactions in chromatin.

[0011] Particularly, the present invention provides a method for analyzing interactions between i) single-strand DNA (ssDNA) and ii) double-strand or single-strand DNA (dsDNA or ssDNA), also designated as ssDNA contacts, by using a set of annealing oligonucleotides, wherein each annealing oligonucleotide is designed to anneal to a ssDNA at a target genomic site, while differing from an endogenous sequence that is complementary to said target genomic site, whereby the annealing of the oligonucleotide with the target genomic site restores a first doublestranded restriction site.

[0012] Particularly, the method of the invention comprises following steps: a) mixing genomic DNA with a set of annealing oligonucleotides as defined in the present invention, b) digesting the restored genomic DNA, by means of an enzyme that specifically recognizes said first double- stranded restriction site, to produce digested genomic DNA; c) labelling the ends of the digested genomic DNA and ligating the spatially proximal labeled digested ends, whereby labeled ligation junction fragments are provided; d) purifying the labeled ligation junction fragments, and preparing a library from the purified fragments; e) sequencing the library; wherein sequences that differ from said target genomic sites are identified and marked as a proximal interaction between ssDNA and dsDNA or ssDNA in the genomic DNA.

[0013] In a particular embodiment, the set of oligonucleotides consists of identical oligonucleotides.

[0014] In another particular and preferred embodiment, the set of annealing oligonucleotides includes oligonucleotides of different sequences. More particularly, the set of annealing oligonucleotides comprises from 2 to 1000, in particular from 5 to 500, from 5 to 400, from 5 to 300, from 5 to 200, from 5 to 100, more in particular from 10 to 40, more particularly from 15 to 30 different sequences.

[0015] In preferred embodiments, the set of annealing oligonucleotides comprise oligonucleotides that anneal at least over 40 nucleotides, and at a density of one oligonucleotide every 100 to 50000 nucleotides on the genomic DNA.

[0016] Advantageously, each annealing oligonucleotide may consist of a sequence of between 40 to 120 nucleotides, preferably between 60 and 90 nucleotides in length.

[0017] In a still preferred embodiment, the set of annealing oligonucleotides is designed so that all oligonucleotides anneal to the target genomic sites under the same hybridization conditions, in particular the same temperature and salt concentration.

[0018] The set of annealing oligonucleotides is preferably designed so that the digested annealing oligonucleotides remain annealed to the target genomic site at the denaturation temperature of the enzyme that specifically recognizes the first double-stranded restriction site. In a particular embodiment, each annealing oligonucleotide restores the first double-stranded restriction site located at about 1-15 nucleotides, preferably 4-10 nucleotides from the 3’ or the 5’ end of the oligonucleotide.

[0019] In a particular embodiment, said first double- stranded restriction site is DpnII.

[0020] In some preferred embodiments, each annealing oligonucleotide comprises at least one, preferably at least 2 or 3, mutations compared to the endogenous sequence that is complementary to said target genomic site, optionally wherein the annealing oligonucleotide comprises a single point mutation every 10 to 30 nucleotides.

[0021] In a particular embodiment, at least one of said mutations inactivates a second double-stranded restriction site.

[0022] In a more particular embodiment, said second double-stranded restriction site is Mfel or Sspl.

[0023] In a particular embodiment, the method of the invention further comprises at least a step d’ ) of submitting the library to a further enrichment of the labeled ligation junction fragments obtained from ssDNA contacts, comprising at least one cycle of denaturation and annealing with capture oligonucleotides.

[0024] In a particular embodiment, the capture oligonucleotides are biotinylated.

[0025] In another particular embodiment, each annealing oligonucleotide comprises at least a mutation inactivating a second double-stranded restriction site and, before sequencing, the method further comprises submitting the fragments enriched by step d’ ) to a digestion with an enzyme that specifically recognizes said second double- stranded restriction site.

[0026] In another particular embodiment, sequencing is high-throughput sequencing.

[0027] Another aspect of the present invention relates to a kit for conducting the method of the invention. Said kit comprises a set of annealing oligonucleotides as defined herein and optionally capture oligonucleotides that are specific to ssDNA at target genomic sites.

[0028] In a particular embodiment, said kit further comprises capture oligonucleotides that are specific to at least one reference double- stranded target genomic site. BRIEF DESCRIPTION OF THE FIGURES

[0029] Figure 1: Detection of ssDNA-dsDNA contacts with ssHi-C

[0030] (A) Experimental system in haploid S. cerevisiae. The DNA double-strand break (DSB) site is located 35 kb away from the centromere. (B) Cell synchronization, DSB induction and cell fixation protocol. >90% of cells have arrested in G2 / M at 2 and 4h post-DSB induction. (C) Hi- C coverage following DSB induction, normalized on coverage in the absence of DSB. Data is mean+SEM (n=2 and 3 biological replicates at 2 and 4h, respectively). (D) A simplified schema of an embodiment of the method of the invention (designed herein ssHi-C) for analyzing ssDNA contacts in a genomic DNA. A more detailed schema is illustrated in Figure 4A-B. (E) Annealing and capture oligonucleotides surrounding the DSB site. The capture oligonucleotides panel also target 7 control dsDNA sites on other chromosomes. (F) Left: Representative Hi-C and ssHi-C results in WT cells 4h post-DSB induction, or in cells lacking the DSB-inducible construct. Black arrows indicate the detection of ssDNA sites. PhiX represents a unique heterologous sequence, LY represents a 2 kb-long region of homology to another genomic site present on chr. II, HOcs represents the 117 bp recognition sequence for the HO endonuclease, i.e. where the DNA break will be inflicted upon induction of HO expression. Right: A representative control dsDNA site on chr. XV is efficiently captured with and without DSB. (G) DSB-dependent ssDNA sites enrichment with or without the use of Capture oligo. (H) Hi-C and ssHi-C contact maps of the whole genome (left) and of chr. V (right) 4h post-DSB induction in WT cells (APY266) showing classical features of yeast chromosome organization (e.g. centromere clustering and cohesin-mediated loop folding). Insets: stripe of inter-chromosomal contacts emanating from the DSB region on chr. V (black arrow), enriched at the donor site on chr. II (asterisk) and on chr. III. Absolute and relative coverage vs. undamaged cells show contact enrichment at dsDNA and ssDNA sites in ssHi-C. Maps are a merge of 3 biological replicates subsampled at 19 million contacts. (I) The extent of resection partly accounts for the heterogeneity of ssDNA enrichment. (J) Summary of 3 drop-out experiments with annealing oligonucleotides. (K) ssHi-C retrieves yeast chromosome organization features in addition to ssDNA-specific contacts.

[0031] Figure 2: Non-uniform distribution of genome-wide contacts along ssDNA and over time. (A) Average genome-wide contact distribution of the left and right DSB ends determined with Hi-C (left; dsDNA only) and ssHi-C (right; ssDNA only) 4h post-DSB induction in WT cells (APY266). A control dsDNA site on chr. VIII (dashed line), located at the same distance to a centromere as the DSB, is shown for comparison. Dotted lines: centromeres. Anchors: viewpoints. Contact frequency is shown per kb. Data smoothed over 30 kb. (B) Trans contacts for dsDNA and ssDNA sites determined in the DSB region (right) and at non-damaged sites genome-wide (left) with Hi-C and ssHi-C at 2 and 4h post-DSB induction in WT cells. Data points: individual captured fragment (ssHi-C) or 1 kb bin (Hi-C). (C) ssDNA trans contacts as a function of the distance to the DSB and to the resection fronts. (D) Hi-C and ssHi-C distribution of cis contacts in the DSB region (left DSB ends, right DSB ends), and of a control dsDNA site on chr. VIII. (E) Distribution of cis contacts for the most enriched ssDNA fragments. (D-E) Contact frequency is shown per kb. Data smoothed over 4 kb. (F) Distribution of ssDNA cis contacts as a function of genomic distance, compared to the genome- averaged Pc(s). (G) Aggregated trans contacts in the centromere- surrounding regions. (H) Aggregated trans contacts with the centromere bin (1 kb) for each ssDNA fragment, as a function of the distance to the DSB. (I) Trans contacts with telomeres as a function of chromosome arm length. Each data point represents a ssHi-C biological replicate. (G-I) Contact frequency is shown per kb. Average of n=3 and 4 biological replicates shown for 2 and 4h post-DSB induction, respectively (J) Schematic representation of the genomic reach of dsDNA and ssDNA contacts. (K) ssDNA sites preferentially engage with chromatin in trans, contrary to dsDNA sites. The chromatin regions contacted depend on the position along ssDNA: DSB proximal ssDNA regions engage chromatin genome-wide with little apparent spatial constraints, while DSB-distal ssDNA regions engage cis chromatin and are subjected to spatial constraints imposed by nearby anchors that direct trans contacts to specific chromatin regions (i.e. centromeres).

[0032] Figure 3: Rad51 promotes genome- wide homology search by stiffening ssDNA.

[0033] (A) Top: absolute decrease of ssDNA contacts and Hi-C coverage in a rad51A (APY679) versus a WT (APY266) strain. Bottom: Fraction of trans contacts made by dsDNA and ssDNA sites.

[0034] (B) Genome-wide contacts distribution made by the left and right ssDNA ends and of a control dsDNA site 4h post-DSB induction in WT and rad51A strains. Legend as in Figure 2A. (C) Distribution of cis contacts of all ssDNA fragments on the left and right DSB ends 4h post-DSB induction. (D) Distribution of cis contacts of individual ssDNA fragments. (C-D) Other legends as in Figure 2D-E. (E) Comparison of the genome-wide Pc(s) and the contact distribution of the average of DSB-proximal left ssDNA sites (top) and individual left ssDNA sites in a rad51A strain (bottom). (F-H) Same legend as Figure 2G-I, in a rad51D strain. (I) Model: extension of ssDNA by Rad51 promotes genome-wide contacts by DSB-proximal ssDNA sites. RPA coils ssDNA, leading to preferential contact with flanking dsDNA by all ssDNA sites. Figure 4 :

[0035] (A) Detailed schematic of an embodiment of the method of the invention for analyzing ssDNA contacts in a genomic DNA: A set of annealing oligonucleotides of 80 nucleotides are designed to anneal with ssDNA at a target genomic site to restore a double-stranded DNA structure. Each annealing oligonucleotide comprises at 5’ end 5-methyl deoxycytosine (5Me-dC) and at 3’ end a C3 spacer (3Sp-C3). Each annealing oligonucleotide comprises a sequence of nucleotides to restore the restriction site of DpnII and two nucleotide substitutions which inactivate the restriction site of Mfel or Sspl. After annealing, the genomic DNA is submitted to restrictions digestion by DpnII. The ends of the digested genomic DNA are submitted to labeling with biotinylated nucleotides and ligation to provide labeled ligation junction fragments. After crosslink reversal and DNA shearing, labeled ligation junction fragments are purified by biotin pulldown. The purified ligation junction fragments are further flanked with sequencing adapter by ligation and submitted to PCR amplification to generate a Hi-C library. Said library is submitted to a further enrichment which consists of denaturing ligation junction fragments and annealing with a set of capture oligonucleotides. Each capture oligonucleotide matches the sequence polymorphism included in the annealing oligonucleotide and is labeled at its 5’ end by biotin. Captured single- stranded ligation junction fragments are purified by biotin pulldown and PCR amplification. The purified ligation junction fragments are submitted to restriction digestion by Mfel or Sspl to eliminate ligation junction fragments obtained from dsDNA interactions, so as to retain solely ligation junction fragments obtained from ssDNA contacts.

[0036] (B) Example of annealing and capture oligonucleotides design at the site L2599 (corresponding to the restriction fragments 18613 (SEQ ID NO: 71) and 18614 (SEQ ID NO: 72)) on the leftend side of the DSB. The original targeted sequence (SEQ ID NO: 73) in the site and the targeted sequence with introduced SNPs (SEQ ID NO: 74), the designed annealing oligo (SEQ ID NO: 10) and capture oligo (SEQ ID NO: 30) are also illustrated. (C) Location of ssDNA and dsDNA capture probes relative to centromeres and telomeres.

[0037] Figure 5:

[0038] (A) ssHi-C read coverage genome-wide (top) and around the DSB region (bottom) prior to contacts filtering. (B) ssDNA fragment contacts upon various annealing oligonucleotides dropout combinations, normalized on the presence of all annealing oligonucleotides. (C) Genomewide Pc(s) and its derivative and (D) Proportion of trans contacts per chromosome in matched Hi-C and ssHi-C experiments in a WT strain 4 hours post-DSB induction (n=3 biological replicates). (E) Comparison of the proportion of trans contacts determined genome- wide on a Hi-C library and at the 7 selected dsDNA sites enriched in ssHi-C at 2 and 4 hours post-DSB induction.

[0039] Figure 6:

[0040] (A-B) Rationale for the coverage-based over-estimation of trans contacts in the resected tract by Hi-C. (A) Genome- wide proportion of trans contacts genome- wide (left), along chr. V (middle), and centered in the resected DSB region (right). WT and rad51D strains underwent DSB induction for 4 hours. Gl-arrested and metaphase-arrested cells are undamaged. Damaged cells exhibit trans contact profiles identical to metaphase-arrested cells, except for the resected DSB region. (B) The proportion of trans contacts resembles that of a Gl-arrested cells, in a Rad51 -independent manner. This effect is exacerbated close to the DSB site and the most resected surrounding region, which expands from 2 to 4 hours. These contacts thus reflect the minority of unresected G1 cells present in the population of cells (<10%), and not the contacts made by the majority of resected cells. (C) Pile-up plots at the 16 centromeres (top) and the inter-centromere interactions of e.g. chr. V (middle) and chr. VII (bottom) in untreated (left) and nocodazole-treated (right) WT cells 4 hours post-DSB induction. Data represent observed contacts normalized over matched genome-averaged contacts. Nocodazole causes a loss of centromere interaction and fails to form the hairpin signature of spindle-mediated tensions at the centromeres. (D) Aggregated trans contacts with the centromere bin (1 kb) for each ssDNA fragment, as a function of the distance to the DSB in untreated and nocodazole-treated WT cells.

[0041] Figure 7:

[0042] (A) Representative Hi-C contact maps of chr. V in WT and rad51A strains following DSB formation, and ratio map binned with Serpentine. (B) Ratio map of the DSB region in rad51A over WT conditions. Bin = 2 kb. (C) FACS profiles of representative experiments in WT and rad51A strains. (D) Genome-wide Pc(s) and its derivative and (E) proportion of trans contact for each chromosome in WT and rad51A strains in following DSB induction. (F) Genome- wide ssHi-C contact by the average of the DSB-proximal left and right ssDNA in WT and radSID strains 2h post-DSB induction. (G) ssHi-C contact distribution of cis contacts in the DSB region, and of a control dsDNA site on chr. VIII. (F-G) Other legends as in Figure 2A and D, respectively. DETAILED DESCRIPTION OF THE INVENTION

[0043] Definitions

[0044] The term “ssDNA contacts” as used herein, refers to the interactions between ssDNA and ssDNA or between ssDNA and dsDNA.

[0045] The term “interaction” as used herein, refers to the spatial proximity of two different genomic sites that may or may not involve physical contact. The spatial proximity can result from the presence of a protein, a protein complex, or an ensemble of protein complexes crosslinked together and to DNA, which bridge the two DNA molecules in a covalent manner. The spatial proximity can also result from direct DNA-DNA contacts mediated by Watson-Crick basepairing. An interaction may involve two ssDNA, two dsDNA or a ssDNA and a dsDNA.

[0046] The term "interaction frequency" as used herein, refers to measuring the spatial proximity probability of two different genomic sites. As the probability increases that the two genomic sites are physically proximal to one another, so does their measured interaction frequency. Conversely, as the interaction frequency decreases, the probability decreases that the two genomic sites are physically proximal to one another. More specifically, two genomic regions with a high interaction frequency most likely share a short range or close neighbor spatial proximity. Alternatively, two genomic sites with a low interaction frequency most likely do not share short-range or close neighbor spatial proximity.

[0047] The terms “genomic site” and “genomic region”, as used herein, refer to any defined length of a genome and / or chromosome or fragments thereof. A genomic site may be in form of doublestranded DNA or single- stranded DNA. Preferably, a genomic site may refer to a complete chromosome or a partial chromosome. Further, a genomic site may refer to a specific nucleic acid sequence on a chromosome (i.e., for example, an open reading frame and / or a regulatory gene). The “genomic site” may be a locus of interest. The term “locus”, as used herein, refers to a specific site or interval on a chromosome.

[0048] The term “genomic DNA” as used herein may comprise the entirety or a part of a genome, one or more chromosomes, fragments of chromosome, fragments of DNA packed in chromatin or nucleosome. The term “target genomic site” as used herein refers to a genomic site of interest in a genomic DNA. A target genomic site is generally a ssDNA contact or spatially close to a ssDNA contact. For example, a target genomic site may be a homology recombination site, a transcription regulation site, etc. Particularly, a target genomic site may be a DNA double-strand break (DSB) site.

[0049] The term “nucleic acids” as used herein refers to polynucleotides or oligonucleotides, and fragments or portions thereof.

[0050] The term “oligonucleotide” as used herein refers to oligodeoxyribonucleotide, which is solely composed of deoxyribonucleotides, or to oligoribonucleotide, which is solely composed of ribonucleotides, or to a nucleic acid comprising both deoxyribonucleotides and ribonucleotides. An oligonucleotide may comprise natural nucleotides and / or non-natural nucleotides. Natural nucleotides are, for examples, deoxyadenosine, deoxycytidine, deoxyguanosine, or deoxythymidine for DNA or their ribose counterparts for RNA. Non-natural nucleotide that can be cited include nucleotides with modified bases, sugars, or intemucleosidic linkages. Examples of non-natural nucleotides are, without being limited to, Peptide Nucleic Acid (PNA), locked nucleic acid (LNA), phosphorothioate intemucleosidic linkages, bases containing linking groups permitting the attachment of labels, such as fluorophores, haptens, biotinyl group, non-ligatable chemical modifications of 5’ and 3’ nucleotides such as C3 spacer and 5-Methyl groups, and the like.

[0051] The term “polynucleotide” as used herein refers to a DNA or RNA polymer.

[0052] The term “junction”, as used herein, refers to a site where two nucleic acid fragments from different genomic sites are joined, for example by a ligase. A junction encodes information about the proximity of the nucleic acid fragments that participate in formation of the junction. For example, junction formation between two nucleic acid fragments indicates that these two nucleic acid sequences were in close proximity when the junction was formed, although they may not be in proximity in linear nucleic acid sequence space. Thus, a junction can define long range interactions. In some embodiments, a junction is labeled, for example with a labeled nucleotide, for example to facilitate isolation of the nucleic acid molecule that includes the junction.

[0053] Annealing oligonucleotides The method of the present invention is based on the development of a specific set of annealing oligonucleotides. Each of said oligonucleotides is designed to simultaneously respond to three conditions: (i) each oligonucleotide anneals to a ssDNA at a target genomic site; (ii) each oligonucleotide has a sequence different from an endogenous sequence that is complementary to that of said target genomic site; and (iii) each oligonucleotide comprises a specific sequence of nucleotides which forms a first restriction site with its complementary strand.

[0054] According to the invention, each annealing oligonucleotide anneals to a ssDNA at a target genomic site to form a double- stranded DNA structure comprising a first restriction site. Such dsDNA structures can be digested by an enzyme that specifically recognizes said restriction site and is then ligated to form ligation junction products. In addition, since the annealing oligonucleotides differ from the endogenous complementary strand of the target genomic sites, the ligation junction products obtained from ssDNA contacts can be distinguished from the ligation junction products obtained from DNA interactions in their canonical double-stranded form.

[0055] In a particular embodiment, each annealing oligonucleotide is designed on the basis of the sequence of the endogenous complementary strand of a ssDNA at a target genomic site, in which one or more mutations are introduced. Said annealing oligonucleotide is long enough and shows sufficient complementarity with the ssDNA to ensure that the annealing oligonucleotide keeps hybridizing with the ssDNA at the denaturation temperature of an enzyme which recognizes the first restriction site. In other words, the annealing oligonucleotides are designed so that their annealing temperatures are higher than the denaturation temperature of the said enzyme.

[0056] The term “annealing” in the context of nucleic acids as used herein refers to the hybridization of two single- stranded nucleic acids to form a double- stranded nucleic acid. The annealing may be obtained from two nucleic acids which have complementary sequences or which have substantially complementary sequences.

[0057] The term “annealing temperature” refers to the temperature at which a single-stranded nucleic acid hybridizes with another single-stranded nucleic acid. The annealing temperature depends on the length of the nucleic acids forming a double- stranded structure and the GC / AT ratio within the nucleic acids as well as the buffer composition (salt composition and concentration, pH). The annealing temperature of a single- stranded nucleic acid with its complementary strand may be calculated according to any conventional method.

[0058] As used herein, the term "hybridization" refers to a process by which a strand of nucleic acid associates with a complementary or substantially complementary strand through Watson & Crick base pairing to form a hybridization complex. Hybridization and the stability of the paired nucleic acids product are impacted by such factors as the degree of complementarity between the nucleic acids, stringency of the conditions involved, and the GC / AT ratio within the nucleic acids.

[0059] The term “complementary” in the context of nucleic acids as used herein refers to two nucleic acid strands wherein each nucleotide of one strand hybridizes through base pairing rules with a complementary nucleotide in another strand. The term “complementary” in the context of the nucleotides refers to Watson & Crick base pairing between nucleotides. Complementary nucleotides are, generally, A and T (or A and U), or C and G.

[0060] The term “substantially complementary” in the context of nucleic acids as used herein refers to two nucleic acid strands wherein not all nucleotides on one strand hybridize to nucleotides on the other strand by base pairing rules, but having a complementarity sufficient to form a doublestranded structure under desired hybridization condition.

[0061] Two single- stranded DNA molecules are substantially complementary when the nucleotides of one strand, optimally aligned and compared, pair with at least about 80% of the nucleotides of the other strand, preferably at least about 90%, 95%, 96%, and more preferably at least 98% or at least 99% of the nucleotides of the other strand.

[0062] Within the scope the present invention, each annealing oligonucleotide is substantially complementary to a ssDNA at a target genomic site. Each annealing oligonucleotide may comprise one or more mutations compared to the endogenous sequence that is complementary to said target genomic site. The number of mutations in an annealing oligonucleotide may be determined by the length and the desired annealing temperature of said oligonucleotide. Generally speaking, the longer an annealing oligonucleotide or the higher the annealing temperature of said oligonucleotide, the greater the number of mutations that the annealing oligonucleotide can contain. The mutations in an annealing oligonucleotide may be substitutions, deletions or insertions of nucleotides compared to the endogenous sequence that is complementary to said target genomic site.

[0063] The term “substitution”, as used herein, refers to the replacement of one or more nucleotides by different nucleotides.

[0064] The term “deletion”, as used herein, refers to a change in a nucleotide sequence in which one or more nucleotides are absent.

[0065] The term “insertion” or “addition” as used herein, refers to a change in a nucleotide sequence in which one or more nucleotides are added.

[0066] In a preferable embodiment, each annealing oligonucleotide comprises at least one nucleotide substitution compared to the endogenous sequence that is complementary to said target genomic site.

[0067] In a particular embodiment, each annealing oligonucleotide comprises at least one mutation compared to the endogenous sequence that is complementary to said target genomic site.

[0068] In a more particular embodiment, each annealing oligonucleotide comprises at least 2 or 3 mutations compared to the endogenous sequence that is complementary to said target genomic site.

[0069] In a particular embodiment, each annealing oligonucleotide comprises a single point mutation every 10 to 30 nucleotides. In a still more particular embodiment, an annealing oligonucleotide of 80 nucleotide may comprise from 4 to 6 mutations, which are introduced as evenly as possible within the part which remains annealed to the target genomic site at the denaturation temperature of the enzyme that specifically recognizes the first double- stranded site.

[0070] In another more particular embodiment, each annealing oligonucleotide comprises a mutation which inactivates a second double- stranded restrictions site.

[0071] The term “a mutation which inactivates a second double-stranded restriction site” as used herein is meant that this mutation modifies the sequence of a second restriction site, so that the sequence comprising said mutation is no more recognized by an enzyme which recognizes the second restriction site. In a preferable embodiment, each annealing oligonucleotide comprises a mutation which inactivates the restrictions site of Mfel or Sspl.

[0072] In another more particular embodiment, each annealing oligonucleotide comprises one or more mutations which inactivate any cryptic restriction site with respect to the first double- stranded restriction site.

[0073] The term “cryptic restriction site” refers to a sequence that differs by one nucleotide substitution from a canonical restriction site. The mutation(s) at a cryptic restriction site allow to further improve the efficiency of the method of the present invention.

[0074] In a particular preferable embodiment, each annealing oligonucleotide comprises one or more mutations which inactivate the restrictions site of Mfel or Sspl and one or more mutations which inactivate any cryptic restriction site with respect to the restriction site Mfel or Sspl.

[0075] The length of an annealing oligonucleotide may be adapted according to the length of the ssDNA at a targeted genomic site, the number of mutations that it comprises, the desired annealing temperature, the specific sequence of nucleotides which will form the first restriction site, its desired location in said annealing oligonucleotide, and the denaturing temperature of said enzyme, etc..

[0076] In some embodiments, each annealing oligonucleotide comprises at least 40 nucleotides.

[0077] In a particular embodiment, each annealing oligonucleotide consists of a sequence of between 40 and 120 nucleotides in length.

[0078] In a more particular embodiment, each annealing oligonucleotide consists of a sequence of between 40 and 110, 40 and 100, 50 and 100, 50 and 90 or 60 and 90 nucleotides in length.

[0079] In a more particular embodiment, each annealing oligonucleotide consists of a sequence of between 65 and 85 nucleotides.

[0080] In a still more particular embodiment, each annealing oligonucleotide consists of a sequence of 80 nucleotides.

[0081] In some embodiments, the set of annealing oligonucleotides is designed so that all oligonucleotides anneal to the target genomic sites under the same hybridization condition. In a particular embodiment, the set of annealing oligonucleotides is designed so that all oligonucleotides anneal to the target genomic sites at the same temperature and same salt concentration. In a more particular embodiment, the set of annealing oligonucleotides is designed so that the digested annealing oligonucleotides remain annealed to the target genomic site at a temperature from 65-70 °C.

[0082] According to the invention, the set of annealing oligonucleotides may have identical sequence or different sequences.

[0083] In a particular embodiment, the set of annealing oligonucleotides consists of identical sequence. Such a set of annealing oligonucleotides targets the genomic sites having the same sequence.

[0084] In another particular embodiment, the set of annealing oligonucleotides includes oligonucleotides of different sequences. Such a set of annealing oligonucleotides may target genomic sites having different sequences.

[0085] In a more particular embodiment, the set includes oligonucleotides that have from 2 to 1000, in particular from 5 to 500, from 5 to 400, from 5 to 300, from 5 to 200, from 5 to 100, from 10 to 100, more in particular from 10 to 40, more particularly from 15 to 30 different sequences. All annealing oligonucleotides having the same sequence within the set target the genomic sites that have an identical sequence.

[0086] For example, in the Experimental Section below, a set of annealing oligonucleotides that includes oligonucleotides of 21 different sequences was used .

[0087] According to the invention, a set of annealing oligonucleotides has enough quantity of oligonucleotides to cover all, or substantially all, ssDNA contacts, within the genomic DNA. Suitable quantities of annealing oligonucleotides may be determined by the size of the target genomic sites. For example, between one and five annealing oligonucleotides may be used for analyzing the contacts made by 1 kilobase of ssDNA in the human genome.

[0088] In a particular embodiment, the set of annealing oligonucleotides comprises oligonucleotides that anneal at least every 40 nucleotides. In a more particular embodiment, depending on the frequency of the restriction site in the genome, the set of annealing oligonucleotides may comprise oligonucleotides that anneal at a density of one oligonucleotide every 100 to 50000 nucleotides on the genomic DNA. According to the invention, each annealing oligonucleotide comprises a specific sequence of nucleotides to restore a first double- stranded restriction site.

[0089] The term “restriction site”, as used herein, refers to a double- stranded DNA structure having a specific sequence, which is recognized by a restriction enzyme.

[0090] The term “restriction enzyme”, as used herein, refers to any protein that recognizes a specific restriction site within a dsDNA and cuts said dsDNA at or near its restriction site. Typically, a restriction enzyme can cut a dsDNA and result in the formation of 5’- or 3’- overhangs or of blunt ends. The term “5’ overhang” refers to 5’ unpaired DNA strand, which is generated by so- called 5’ overhang restriction enzymes that cut dsDNA asymmetrically and leave one or several single stranded bases end with a 5’ phosphate. The term “3’ overhang” refers to 3’ unpaired DNA strand which is generated by so-called 3’ overhang restriction enzymes that cut dsDNA asymmetrically and leave one or several single stranded bases end with a 3’ hydroxyl. The term “blunt end” refers to an end of dsDNA without unpaired DNA strand.

[0091] According to the invention, the first restriction site restored by the annealing oligonucleotides can be a restriction site of any restriction enzyme conventionally used in the art. Appropriate restriction sites to be restored may be determined according to several parameters, such as the frequency of said restriction site in a genomic DNA, the estimated frequencies of ssDNA contacts, the size of the genomic DNA to be digested, the length of annealing oligonucleotides, etc. For example, the smaller the size of a restriction site, the higher its frequency in a genomic DNA. For example, if it is desirable to decrease the fragment sizes after digestion, it is then preferable to use a shorter restriction site.

[0092] In general, the cleavage site of a restriction enzyme may be within or outside its restriction site, or even at distance from its restriction site. In a preferred embodiment, the cleavage site is inside or immediately close to the restriction site of the selected restriction enzyme. In certain embodiments, the restriction site is a 4 to 8 base-pair restriction site, preferably a 4 base-pair restriction site. In some embodiments, the cut of the restriction enzyme creates a 5’overhang end. In a more particular embodiment, said restriction site is a 4 or 5 base-pair restriction site recognized by a restriction enzyme that creates a 5’ overhang end, e.g. DpnII, Hinfl, Hpall, MluCI, or Ddel.

[0093] In a preferable embodiment, said first restriction site is the restriction site of DpnII. Generally speaking, a specific sequence of nucleotides which will form the first restriction site may be located at any part of an annealing oligonucleotide. Preferably, said specific sequence of nucleotides is located close to the 3’ or the 5’ end of the annealing oligonucleotides to ensure that the digested annealing oligonucleotides remain annealed to the ssDNA of the target genomic sites at the denaturation temperature of the enzyme that specifically recognizes the first double- stranded restriction site.

[0094] The term “digested annealing oligonucleotide” refers to the part of an annealing oligonucleotide which remains annealed to the target genomic sites at the denaturation temperature of the first restriction enzyme. In general, the digested annealing oligonucleotide may be the longer part compared to the other part of the annealing oligonucleotide.

[0095] In a particular embodiment, said specific sequence of nucleotides is located at about from 1 to 15 nucleotides from the 3’ or the 5’ end of the annealing oligonucleotide, i.e. said specific sequence of nucleotides is located from 1 to 15 nucleotides downstream of the 5’ end of an annealing oligonucleotide or from 1 to 15 nucleotides upstream of the 3’ of said oligonucleotide.

[0096] In a more particular embodiment, said specific sequence of nucleotides is located at about 4 to 10 nucleotides from the 3’ or the 5’ end of the annealing oligonucleotide, i.e. said specific sequence of nucleotides is located from 4 to 10 nucleotides downstream of the 5’ end of an annealing oligonucleotide or from 4 to 10 nucleotides upstream of the 3’ of said oligonucleotide.

[0097] In a particular embodiment, the set of annealing oligonucleotides is designed so that the digested annealing oligonucleotides remain annealed to the digested genomic DNA at the denaturation temperature of the enzyme that specifically recognizes the first double-stranded restriction site.

[0098] The annealing oligonucleotide may be an oligodeoxyribonucleotide or a nucleic acid comprising both deoxyribonucleotides and ribonucleotides.

[0099] The annealing oligonucleotides may comprise one or more non-natural nucleotides, which may be nucleotides with modified bases, modified sugar moieties, or modified phosphate backbone.

[0100] Examples of modified base that may be cited include, but are not limited to: 5 -fluorouracil, 5- bromouracil, 5-chlorouracil, 5-iodouracil, hypoxanthine, xanthine, acetylcytosine, 5- (carboxyhydroxylmethyl) uracil, 5-carboxymethylaminomethyl-2-thiouridine, 5- carboxymethylaminomethyluracil, dihydrouracil, beta-D-galactosylqueosine, inosine, N~6- sopentenyladenine, 1 -methylguanine, 1 -methylinosine, 2,2-dimethylguanine, 2-methyladenine, 2-methylguanine, 3-methylcytosine, 5-methyl cytosine, N6-adenine, 7-methylguanine, 5- methylaminomethyluracil, methoxyarninomethyl-2-thiouracil, beta-D-mannosylqueosine, 5'- methoxycarboxymethyluracil, 5-methoxyuracil, 2-methylthio-N6-isopentenyladenine, uracil- 5-oxyacetic acid, pseudouracil, queosine, 2-thiocytosine, 5-methyl-2-thiouracil, 2-thiouracil, 4- thiouracil, 5-methyluracil, uracil-5-oxyacetic acid methylester, uracil- S-oxy acetic acid, 5- methyl-2-thiouracil, 3-(3-amino-3-N-2-carboxypropyl) uracil, 2,6-diaminopurine and biotinylated analogs, amongst others.

[0101] Examples of modified sugar moieties that may be cited include, but are not limited to arabinose, 2-fluoroarabinose, xylose, hexose, sugar moieties modified with an extra bridge connecting the 2’ -oxgen and 4’ carbon, or sugar moieties comprising a modified component of the phosphate backbone, such as phosphorothioate, a phosphorodithioate, a phosphoramido thioate, a phosphoramidate, a phosphordiamidate, a methylphosphonate, an alkyl phosphotriester, or a formacetal or analog thereof.

[0102] Example of modified phosphate backbone that may be cited, but are not limited to, is phosphate backbone modified by a phosphoryl guanidine group, a triethylene glycol spacer or a phosphorothioate group.

[0103] Preferably, the non-natural nucleotides in an annealing oligonucleotide are substrates of a DNA polymerase.

[0104] In a particular embodiment, each annealing oligonucleotide comprises one or more 5-methyl deoxycytosines.

[0105] In a more particular embodiment, each annealing oligonucleotide comprises 5-methyl deoxycytosine at the 5’ end of the oligonucleotides.

[0106] In some embodiments, each annealing oligonucleotide comprises a chemical linker at 3’ end. Examples of chemical linkers that may be cited include, but are not limited to, a phosphoramidite spacer or a triethylene glycol spacer.

[0107] In a particular embodiment, each annealing oligonucleotide comprises a C3 -phosphoramidite spacer at 3’ ends of the oligonucleotides. In some embodiments, each annealing oligonucleotide is further flanked by a sequencing adapter at 5’ or 3’ end. Said adapter, which is a sequence of nucleotides, may comprise a sequence required to bind to a support surface for sequencing, the index sequence, and the sequencing primer binding site.

[0108] Step (a) of the method of the invention comprises mixing genomic DNA with a set of annealing oligonucleotides as described above.

[0109] Step (a) allows to obtain a mixture, preferably an intimate mixture, comprising both genomic DNA and a set of annealing oligonucleotides.

[0110] According to the invention, step (a) is conducted under conditions which favor hybridization of the annealing oligonucleotides with genomic DNA, e.g. in a reaction buffer comprising suitable salt concentration and at an appropriate temperature. In particular, step (a) is conducted at a temperature lower than the annealing temperature of the annealing oligonucleotides. Appropriate reactional conditions for step (a) may be determined according to the length and GC content of the annealing oligonucleotides by conventional methods described in the art. For example, step (a) may be conducted in a reaction buffer comprising 100 mM sodium chloride, 50 mM Bis-Tris-HCl pH 6, 10 mM magnesium chloride, 1 mM DTT, 0.5% sodium dodecyl sulfate and at a temperature from 50 to 70 °C.

[0111] Said genomic DNA may be DNA comprised in a sample of one or more cells, nuclear extract, cellular milieu or system of nucleic acids of interest that include nucleic acid, or a sample of virus. Appropriate samples for use in the method of the invention include any conventional biological sample obtained from an organism or a part thereof, such as plants, animals, homogeneous or heterogeneous population of prokaryotes, eukaryotes and archaea, viruses, e.g. obtained from controlled source or natural habitats. In particular embodiments, the biological sample is obtained from an animal subject, such as a human subject, including samples from a healthy or apparently healthy human subject or a human patient affected by a condition or disease to be diagnosed or investigated, such as cancer. A biological sample may be any solid or fluid sample. For example, a biological sample can be a biological fluid obtained from, for example, blood, plasma, serum, urine, bile, saliva, cerebrospinal fluid (CSF) or from any organ or tissue. In particular examples, samples are used directly (e.g., fresh or frozen), or can be manipulated prior to use, for example, by fixation , as described e.g. in Rao et al., 2014. Any method of obtaining tissue from a subject can be utilized. The selection of the method used will depend upon various factors such as the type of tissue, age of the subject, or procedures available to the practitioner. Standard techniques for acquisition of such samples are available.

[0112] The genomic DNA used in step (a) may be packaged in chromatin.

[0113] The term “chromatin” refers to a macromolecular complex of DNA and proteins. Examples of proteins comprised in chromatin are histones, e.g. and Hl, H2A, H2B, H3 and H4. These proteins package and arrange the DNA and create DNA interactions between two different genomic sites.

[0114] It is to be understood that the genomic DNA is not treated in a manner that could disrupt or affect the spatial organization of the DNA, and the DNA interactions to analyze.

[0115] In some embodiments, the genomic DNA used in step (a) is held in a fixed spatial position. The genomic DNA may be fixed by any conventional method, as described e.g. in Rao et al., 2014. for fixing protein-protein, protein-DNA, and / or DNA-DNA together. For example, the fixation of the genomic DNA may be obtained by contacting the cells containing the genomic DNA with one or more crosslinking agent. This treatment locks in the spatial relationships between different genomic sites of the genomic DNA in its natural or biologically functional spatial position.

[0116] The term “crosslinking agent” refers to a chemical agent or even light, which facilitates the covalent attachment of one molecule to another molecule. Crosslinking agents can be protein- nucleic acid crosslinking agents, nucleic acid-nucleic acid crosslinking agents, and proteinprotein crosslinking agents. Examples of such agents are known in the art. Examples of crosslinking agent include, without being limited to, an aldehyde, e.g. formaldehyde or glutaraldehyde, disuccinimidyl glutarate, UV light, psoralens and their derivatives such as aminomethyltrioxsalen, ethylene glycol bisfsuccinimidylsuccinate], bissulfosuccinimidyl suberate, l-Ethyl-3-[3-dimethylaminopropyl]carbodiimide (EDC) bisfsulfosuccinimidyl] suberate (BS3) and other compounds known to those skilled in the art.

[0117] In some embodiments, the genomic DNA is fixed with a crosslinking agent, such as an aldehyde, for example formaldehyde or glutaraldehyde. In other embodiments, the genomic DNA is fixed without using crosslinking agents. For example, the genomic DNA can be stabilized using a DNA stabilizer, e.g. spermine or spermidine (see Cullen et al, Science 261, 203 (1993)). In some embodiments, the genomic DNA is stabilized by embedding in a polymer such as agarose. Other methods of maintaining the positional relationships of genomic DNA are also known in the art.

[0118] In some embodiments, a crosslinking agent is a non-reversible crosslinking agent. The term “non-reversible crosslinking agent” refers to agents which create irreversible crosslinking bonds between two molecules.

[0119] In some embodiments, the crosslinking agent is a reversible cross-linking agent. The term “reversible crosslinking agent” refers to agents which create reversible crosslinking bonds between two molecules. In some embodiments, the crosslinking agent is reversed, for example after the digested genomic DNA are ligated. In a particular embodiment, the ligated genomic DNA is released from the cross-linked three-dimensional matrix by treatment with an agent, such as a protease, that degrade the proteinaceous material form the sample, thereby releasing the end ligated genomic DNA for further analysis, such as determination of the nucleic acid sequence. In some embodiments, a crosslinking agent is light, such as UV light. In some embodiments, a crosslinking agent is light activated. In some embodiments, a decrosslinking agent is used. A decrosslinking agent may be light, such as UV light, or high pH and heat.

[0120] In some embodiments, the method further comprises, before step a), a preliminary step a’) of providing a sample of one or more cells, cellular nuclei, or an acellular nucleic acid system comprising DNA, wherein the DNA strands are held in a fixed position with respect to each other.

[0121] In a particular embodiment, step a’) consists of contacting a sample of one or more cells, cellular nuclei, or an acellular nucleic acid system comprising genomic DNA with one or more crosslinking agents as described above, or a DNA stabilizer as described above.

[0122] In a more particular embodiment, step a’) consists of contacting a sample as described above with an aldehyde, e.g. formaldehyde or glutaraldehyde.

[0123] In a preferable embodiment, said crosslinking agent is formaldehyde.

[0124] In some embodiments, the method of the invention further comprises a preliminary step of lysing cells to release fixed genomic DNA. In general, the cell lysis method used to release fixed genomic DNA is a gentle method to avoid breaking the genomic DNA, including DNA in a single-stranded form, or destroying the three-dimensional structure of said genomic DNA. In some examples the nuclei are lysed as well, while in other examples, the nuclei are maintained intact, which can then be isolated and optionally lysed, for example using a reagent that selectively targets the nuclei or other separation technique known in the art. In some examples, the sample is a sample of permeabilized nuclei, multiple nuclei, isolated nuclei, synchronized cells (such at various points in the cell cycle, for example metaphase) or acellular.

[0125] Examples of suitable methods for lysing cells that can be cited include detergent-based lysis, freeze-thaw lysis, mechanical lysis, osmotic shock, enzymatic digestion.

[0126] Fixed genomic DNA released in this way can be therefore mixed with a set of annealing oligonucleotides for conducting step (a).

[0127] Step (a) leads to the restoration of genomic DNA, wherein ssDNA involved in ssDNA contacts are restored in double- stranded DNA structures which comprise a first double-stranded restriction site as described above.

[0128] Step (b)

[0129] Restored genomic DNA is then submitted to the digestion of an enzyme that specifically recognizes said first double-stranded restriction site.

[0130] According to the invention, step (b) leads to the digestion of restored genomic DNA. In particular, each double-stranded DNA structure formed by a ssDNA and an annealing oligonucleotide is cut at a particular position and separated into two fragments, wherein the fragments comprising the target genomic sites remain fixed together. All of the fragments comprising the target genomic sites and generated after the enzymatic digestion is designed as “digested genomic DNA”.

[0131] According to the enzyme used in step (b), the digested genomic DNA may have multiple 3’ overhanging ends, 5 ’overhanging ends or blunt ends.

[0132] Preferably, the digested genomic DNA has 5 ’overhanging ends.

[0133] The appropriate enzyme for digesting the restored genomic DNA is determined by the restriction site that is restored by annealing oligonucleotides. In a particular embodiment, step (b) comprises digesting the restored genomic DNA by DpnII, Hinfl, Hpall, MluCI, or Ddel. More particularly, step (b) comprises digesting the restored genomic DNA by DpnII.

[0134] Step (b) is conducted under conditions which are suitable for enzymatic activity of the restriction enzyme used in said step. A skilled person can choose suitable conditions for step (b) in consistency with the restriction enzyme used in said step.

[0135] For example, the digestion of restored genomic DNA with DpnII may be conducted at a temperature from 25 to 40 °C, in particular at 37°C, in a buffer comprising 100 mM sodium chloride, 50 mM Bis-Tris-HCl pH 6, 10 mM magnesium chloride, 1 mM DTT, 0.1% sodium dodecyl sulfate, 1% Triton X100.

[0136] In some embodiments, after step (b), the method of the invention further comprises a step of denaturing the enzyme used in step (b) before conducting step (c).

[0137] Enzyme denaturation may be performed according to any conventional method, e.g. by heating. The temperature for denaturing a conventional restriction enzyme known in the art may be varied from 50 to 80 °C or more. For example, DpnII may be denatured at 62 °C or more.

[0138] Digested genomic DNA obtained in step (b) is then submitted to labelling and ligation step to label the ends of the digested genomic DNA and to ligate the spatially proximal ends, whereby labeled ligation junction fragments are provided.

[0139] The term “ligation junction fragments”, as used herein, refers to chimeric fragments generated after the join of two spatially proximal ends of the digested genomic DNA, i.e. fragments containing ligation junction. Ligation junction fragments encode the information about the proximity of two target genomic sites. Ligation junction fragments comprise both the fragments derived from ssDNA contacts and the fragments derived from dsDNA interactions.

[0140] In some embodiments, the ends of digested genomic DNA are firstly labeled with a junction marker.

[0141] The term “junction marker”, as used herein, refers to any compound or chemical moiety that provide a basis for selective purification and that is capable of being incorporated within a nucleic acid and preferably used as a template by a DNA polymerase or compatible with other sequencing technologies. Non-limiting examples of labels may be, e.g. biotin.

[0142] The incorporation of a junction marker in a ligation junction fragment provides a basis for selective purification of said fragment.

[0143] A junction marker may be a labeled nucleotide or a labeled nucleotide linker.

[0144] The term “labeled nucleotide”, as used herein, refers to a nucleotide with a base attached to a detectable label as described above. Examples of labeled nucleotides include, without being limited to, biotinylated nucleotides, e.g. biotin-dATP, biotin-dCTP, allyl- or azide- containing nucleotides.

[0145] The term “labeled nucleotide linker” as used herein, refers to any nucleic acid linker comprising a detectable label as described above. Said labeled nucleotide linker may label the ends of the digested genomic DNA and serve as a linker for joining two spatially proximal ends. Said labeled nucleotide linker may be a short single- or double-stranded nucleic acid, e.g. from 5 to 20 nucleotides, comprising a labeled nucleotide as described above. When said labeled nucleotide linker is a double- stranded nucleic acid, it may have 5’ or 3’ overhang or blunt ends. A labeled nucleotide linker may comprise a predetermined universal sequence of nucleotides, which may be used as a barcode, such that the ligation junction fragments comprising said universal sequence may be extracted and amplified by polymerase chain reaction (PCR).

[0146] A junction marker may be incorporated into the ends of digested genomic DNA with a DNA polymerase when it is a labeled nucleotide, or with a ligase when it is a labeled nucleotide linker.

[0147] In one embodiment, the junction marker comprises a labeled nucleotide. In a particular embodiment, the junction marker comprises a biotinylated nucleotide.

[0148] A digested genomic DNA may undergo overhanging ends filling in with a labeled nucleotide followed by blunt-end ligation.

[0149] The term “filling in” as used herein refers to adding nucleotides on the complementary strand using the overhang as a template. Blunt ends are generated after complete fulling of the overhanging ends. The filling of overhanging ends may be implemented by any conventional chemical or enzymatic method. For example. 5 ’-overhanging ends may be filled by a DNA polymerase, e.g. T4 DNA polymerase or Klenow fragment of DNA polymerase I.

[0150] In a particular embodiment, step (c) comprises labeling the ends of the digested genomic DNA by filling in the overhanging ends with at least one labeled nucleotide.

[0151] In a more particular embodiment, step (c) comprises labeling the ends of the digested genomic DNA by filling in the 5 ’-overhanging ends with at least one labeled nucleotide.

[0152] In another more particular embodiment, step (c) comprises labeling the ends of the digested genomic DNA by filling in the overhanging ends with at least one biotinylated nucleotide.

[0153] In a still more particular embodiment, step (c) comprises labeling the ends of the digested genomic DNA by filling in the 5 ’-overhanging ends with at least one biotinylated nucleotide.

[0154] After the labeling with a junction marker, each two labeled spatial proximal ends of the digested genomic DNA are ligated together and form labeled ligation junction fragments.

[0155] The term “ligated” as used herein, refers to any linkage of two nucleic acid sequences usually comprising the formation of phosphodiester bonds between the 3 ’-hydroxyl of one nucleic acid terminus with the 5’-phosphoryl of another.

[0156] Ligation may be conducted by any conventional method, for example by a catalytic enzyme, in particular a ligase.

[0157] Any conventional ligase may be used. Non-exhau stive examples of ligases which may be used include T4 DNA ligase, its derivatives, and single-strand DNA ligase (Adaptase).

[0158] In some embodiments, step (c) is conducted in reactional conditions which are suitable both for labeling the ends of the digested genomic DNA and for ligating the labeled ends.

[0159] In some other embodiments, step (c) is conducted firstly under reactional conditions which are suitable for labelling the ends of the digested genomic DNA and then under other reactional conditions which are suitable for ligating the labeled digested ends.

[0160] Step ( d)

[0161] Labeled and ligated genomic DNA may then be typically fragmented and purified. Fragmentation results in random double strand breaks within labeled and ligated genomic DNA obtained in step (c) and thus produces a collection of DNA fragments. Said DNA fragments collection comprises both non-labeled fragments and the labeled fragments comprising ligation junction fragments.

[0162] Fragmentation of nucleic acid can be processed by several methods known in the art, including mechanical fragmentation, e.g. hydrodynamic shearing, or sonication, enzymatic digestion and chemical fragmentation, e.g. by Iron and EDTA (Gyarmati et al., 2013).

[0163] In some embodiments, step (d) comprises shearing the labeled and ligated genomic DNA and purifying the labeled ligation junction fragments.

[0164] In a particular embodiment, the labeled and ligated genomic DNA is sheared by sonication.

[0165] After fragmentation, labeled fragments are selectively purified, for example par affinity purification with regard to the label. In some embodiment, the labeled fragments are purified by immobilizing the fragments on a substrate, such as a bead, a membrane, particles, or surface (e.g,. a well or tube) that is coated with an affinity molecule suitable for immobilizing the labeled fragments.

[0166] In some embodiments, when the label is biotin, the affinity molecule is a streptavidin.

[0167] The biotin-labeled ligation junction fragments can then be purified by streptavidin pulldown.

[0168] In some embodiments, a library is created by shearing the DNA and selecting the biotincontaining fragments with streptavidin beads.

[0169] After purification, non-labeled fragments are discarded, and a sequencing library is constructed using the pulled-down (biotin-containing) DNA. The labeled fragments may be further flanked with sequencing adapters by ligation or by PCR. Said sequencing adapters may be those suitable for sequencing by Illumina technology. For example, the sequencing adapters suitable for ligation may have universal sequence and may typically comprise, from 5’ to 3’, a sequence required to bind to a support surface for sequencing, the index sequence, and the sequencing primer binding site. The adapters may also comprise, from 5’ to 3’ , a part of universal sequence as described above and a part of target-specific sequence. Said sequencing adapters are suitable for addition by PCR. After the addition of sequencing adapters, the labeled fragments may be released from streptavidin conjugated surface upon DNA amplification by any DNA polymerase. This library may be directly submitted to sequencing. Because the annealing oligonucleotides originally differed from the endogenous sequence, e.g. by comprising mutation(s), the distinct ligation junction fragments, e.g. comprising said mutation(s), may be detected by sequencing and e.g. distinguished from ligation junction fragments obtained from dsDNA interactions which do not comprise said mutation(s).

[0170] Enrichment by capture oligonucleotides

[0171] The library obtained in step (d) may be submitted to one or more further enrichments of the ligation junction fragments.

[0172] In some embodiments, the method of the invention further comprises at least a step d’) of submitting the library obtained in step (d) to a further enrichment of the ligation junction fragments.

[0173] Said enrichment may be obtained with a set of capture oligonucleotides.

[0174] The capture oligonucleotides are designed so that the sequence of each capture oligonucleotide overlaps or is partially identical to the sequence of an annealing oligonucleotide used in step (a). Said capture oligonucleotides are substantially complementary to the sequence of the endogenous ssDNA at a target genomic DNA to which said annealing oligonucleotide anneals. The sequence of said capture oligonucleotide may comprise one or more mutations compared to the sequence of said ssDNA. These capture oligonucleotides can anneal both with their complementary or substantial complementary oligonucleotides which derive from ssDNA contacts and dsDNA interactions.

[0175] Said capture oligonucleotides may be an RNA, e.g. a locked nucleic acid (LNA), a DNA, a peptide nucleic acid (PNA), or a hybrid RNA-DNA.

[0176] In a particular embodiment, each capture oligonucleotide has the same length as its targeting digested annealing oligonucleotide. In another particular embodiment, each capture oligonucleotide is shorter than its targeting digested annealing oligonucleotide. In a particular embodiment, each capture oligonucleotide comprises from 50 to 70 nucleotides. In a more particular embodiment, each capture oligonucleotide comprises from 55 to 65 nucleotides.

[0177] The appropriate number of capture oligonucleotides in a set may be determined by the number of annealing oligonucleotides used in step (a). In a particular embodiment, the method of the invention is conducted with a same number of capture oligonucleotides than of annealing oligonucleotides. In a particular embodiment, the method of the invention is conducted with a greater number of capture oligonucleotides than of annealing oligonucleotides, the additional capture oligonucleotides targeting other junction fragments.

[0178] In a particular embodiment, said further enrichment comprises at least one cycle of denaturation and annealing with a set of capture oligonucleotides.

[0179] During denaturation cycle, the fragments obtained in step (d) unwind. Both single-stranded DNA comprising digested annealing oligonucleotides and single- stranded DNA of the junctions derived from dsDNA interactions are released. Under annealing condition, both these singlestranded DNA are hybridized by capture oligonucleotides and form dsDNA with the last one.

[0180] Said dsDNA may be purified by several means. In some embodiments, the enrichment is conducted with selective purification using a marker. In a particular embodiment, each capture oligonucleotide is labeled with a marker to facilitate selective purification. Said marker may be, for example, a biotin, a streptavidin, a FLAG tag, an antibody or any other bioconjugate ligands. The marker may be located at 5’ or 3’ end of, or be internal to, a capture oligonucleotide. In a preferable embodiment, the marker is located at 5’ end of a capture oligonucleotide.

[0181] In a more particular embodiment, each capture oligonucleotide is labeled with biotin and the captured ligation junction fragment is purified by streptavidin pulldown.

[0182] In some other embodiments, the enrichment is conducted with PCR amplification. Each dsDNA is amplified with a pair of primers that specifically bind to sequencing adapters. In some other embodiments, the ligation junction fragments obtained from ssDNA contacts are further separated from ligation junctions obtained from dsDNA interactions. The separation is conducted with a second restriction enzyme digestion. Said method is usable when the annealing oligonucleotides comprise at least one mutation which inactivates a second restriction site which may be present in a dsDNA interaction. A dsDNA comprising such a mutation is resistant to an enzyme which specifically recognizes said second restriction site. Therefore, ligation junctions obtained from dsDNA interactions are digested by said second restriction site, while ligation junctions obtained from ssDNA contacts are maintained.

[0183] In a particular embodiment, each capture oligonucleotide comprising at least one mutation compared to the sequences of a ssDNA at a target genomic site wherein said mutation inactivates the restriction site of Mfel or Sspl.

[0184] In another particular embodiment, step d’) consists of submitting the library obtained in step (d) to a further enrichment, comprising at least one cycle of denaturation and annealing with capture oligonucleotides comprising the mutation which inactivates the restriction site of Mfel or Sspl, and a further step of submitting the enriched fragments to a digestion with an enzyme that specifically recognizes Mfel or Sspl.

[0185] In a particular embodiment, more than one enrichment process is conducted successively. In a more particular embodiment of step d’), the further enrichment successively comprises:

[0186] (i) at least one cycle of denaturation and annealing with capture oligonucleotides,

[0187] (ii) biotin pulldown of captured oligonucleotides,

[0188] (iii) PCR amplification of captured oligonucleotides,

[0189] (iv) treatment with a second restriction enzyme.

[0190] In said embodiment, steps (i), (ii) and (iii) may be repeated one or more times.

[0191] In another embodiment, the further enrichment of the ligation junction fragments successively comprises:

[0192] (i) treatment with a second restriction enzyme,

[0193] (ii) at least one cycle of denaturation and annealing with capture oligonucleotides,

[0194] (iii)biotin pulldown of captured oligonucleotides,

[0195] (iv)PCR amplification of captured oligonucleotides.

[0196] In said embodiment, steps (ii), (iii) and (iv) may be repeated one or more times.

[0197] After one or more purification process as described in steps (d) and optionally (d’), a library enriched with desired ligation junction fragments is created. Said library is suitable for analyzing the ssDNAs contacts.

[0198] In one embodiment, the library is analyzed by sequencing the ligation junction fragments.

[0199] Sequencing allows to directly identify the sequences of ligation junction fragments and the mutation(s) present in the labeled ligation junction fragments obtained from ssDNA contacts. Based on sequence information, ligation junction fragments from ssDNA contacts can be directly distinguished from ligation junction fragments from dsDNA interactions. The frequency and / or the number of ssDNA contacts may then be determined.

[0200] The term “DNA sequencing” refers to the process of determining the nucleotide order of a given DNA molecule. Generally, the sequencing can be performed using automated Sanger sequencing (AB 13730x1 genome analyzer), pyrosequencing on a solid support (454 sequencing, Roche), sequencing-by- synthesis with reversible terminations (ILLUMINA® Genome Analyzer), sequencing-by-ligation (ABI SOLiD®) or sequencing-by-synthesis with virtual terminators (HELISCOPE®). Other next generation sequencing techniques for use with the disclosed methods include, Massively parallel signature sequencing (MPSS), Polony sequencing, Ion Torrent semiconductor sequencing, DNA nanoball sequencing, Heliscope single molecule sequencing, Single molecule real time (SMRT) sequencing, and Nanopore DNA sequencing.

[0201] In some embodiments, a library (e.g., for next generation sequencing applications, such as Illumina® sequencing (Illumina® Inc., San Diego, CA)) is constructed on magnetic particles.

[0202] In a particular embodiment, the library is analyzed by high throughput sequencing.

[0203] The term “high throughput sequencing” refers to a process of sequencing through a combination of robotics, data processing and control software, liquid handling devices, and detectors, high throughput techniques allows the rapid screening of potential reagents, conditions, or targets in a short period of time, for example in less than 24, less than 12, less than 6 hours, or even less than 1 hour.

[0204] Sequences of ligation junction fragments may be analyzed with any available sequence analysis software, e.g. Bowtie2, HiCpro, Hicstuff, pairtools. The frequencies and / or the absolute numbers of ssDNA contacts may be calculated with any available data analysis software, e.g. R studio®, Python®. In some embodiments, the information from one or more ligation junction fragments is used to simultaneously map protein-DNA interactions and DNA-DNA interactions. In some embodiments, the information from one or more ligation junction fragments is used to simultaneously map methylation and three-dimensional structure. In some embodiments, the information from more than one ligation junction fragments is used to assemble whole genomes or parts of genomes

[0205] Kit

[0206] The present invention also provides kits for conducting an in vitro method for analyzing ssDNA contacts as described above. The term “kit” refers to any necessary reagents, and optionally means, for conducting the method as described above.

[0207] The kit of the invention comprises a set of annealing oligonucleotides as described above. In some embodiment, kit further comprises a set of capture oligonucleotides as described above that are specific to ssDNA at target genomic sites.

[0208] The kit may further comprise any reagent for conducting the method as described above, such as one or more crosslinking agent, one or more restriction enzymes, a junction marker, e.g. labeled nucleotides, reagents, e.g. beads, particles or surface covered with an affinity agent, for purifying labeled ligation junction fragments or enriching captured oligonucleotides.

[0209] In a particular embodiment, the kit comprises:

[0210] (i) a set of annealing oligonucleotides,

[0211] (ii) optionally a set of capture oligonucleotides,

[0212] (iii) optionally a crosslinking agent capable of fixing genomic DNA;

[0213] (iv) optionally a first restriction enzyme,

[0214] (v) optionally a junction marker and a DNA polymerase,

[0215] (vi) optionally a DNA ligase,

[0216] (vii) optionally beads or a surface covered with affinity agent, and

[0217] (viii) optionally a second restriction enzyme.

[0218] In a more particular embodiment, the kit comprises a set of annealing oligonucleotides as described above, a set of capture oligonucleotides as described above, DpnII, biotinylated nucleotides, beads or a surface covered with streptavidin, and optionally Mfel or Sspl. In another particular embodiment, the kit further comprises PCR primers for amplifying the ligation junction fragments.

[0219] In another particular embodiment, the kit further comprises any necessary probes for sequencing the ssDNA contacts. In a more particular embodiment, said probes are probes adapted for massive parallel sequencing (next-generation sequencing). Said probes may contain probes for sequencing and universal sequencing adapters which may comprise different index sequences for multiplex sequencing.

[0220] In one embodiment, the kit further comprises buffer and reagents capable of supporting the genomic DNA fixing. In one embodiment, the kit further comprises buffers and reagents capable of supporting the enzyme nucleic acid fragmentation. In one embodiment, the kit further comprises buffers and reagents capable of supporting selective purification using the junction marker.

[0221] In some embodiments, the kit of the invention further comprises capture oligonucleotides that are specific to at least a reference double-stranded genomic site.

[0222] The term “reference double- stranded genomic site”, as used herein, refers to a double-stranded genomic site wherein its frequency and / or number are known. A reference double-stranded genomic site may be used as a control to normalize the efficiency of capture oligonucleotides that are specific to ligation junction fragments. Examples of reference genomic sites that may be cited include any housekeeping gene, e.g. beta-actin.

[0223] The capture oligonucleotides that are specific to a reference double- stranded genomic site are designed so that the sequence of each said capture oligonucleotide is complementary or identical to the sequence of one strand of the reference double- stranded genomic site.

[0224] The kit of the invention may further comprise any delivery system for delivering materials. Such delivery systems include systems for the storage, transport, or delivery of reaction reagents from one location to another, and / or supporting materials (e.g., reaction medium, written instructions for performing the assay etc.). For example, kits may include one or more enclosures (e.g., boxes) containing the relevant reaction reagents and / or supporting materials.

[0225] Applications of the method of the invention As described above and demonstrated in the examples below, the methods of the invention successfully detect ssDNA contacts which are so far undetectable by using conventional “Chromosome Conformation Capture” technologies, e.g. a Hi-C method.

[0226] The method of the invention finds wide applications in the study of DNA damage repair, DNA replication, meiosis, adenovirus and retrovirus life cycles, transposons and retrotransposons life cycles. In addition, the method of the invention may be used for screening of compounds that modifies DNA interactions.

[0227] The present invention is illustrated in more detail in the following examples. These examples should not be construed as a limitation to the present invention.

[0228] Examples

[0229] The project leading to this application has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation program (grant agreement No 851006).

[0230] Materials and methods

[0231] • Saccharomyces cerevisiae strains

[0232] Genotypes of Saccharomyces cerevisiae (W303 RAD5+ background) strains are the following: 47'a-inc, ura3::LY-HOcs, lys2::LY, trpl::GAL-HO-hphMX, his3-ll,15, canl-100, leu2- 3,112, ade2-l, RAD5. The inducible HO expression construct pGALl-HO at trpl (chr. IV), the mutagenesis of the endogenous HO cut-site at MAT (chr. Ill), the DSB-inducible construct replacing URA3 (chr. V) and the donor sequence replacing LYS2 (chr. II) have been described previously (Piazza et al., 2019). The DSB-inducible construct contains 327 bp of a fragment of the PhiX phage genome flanked by multiple restriction sites (including EcoRI) making a unique 453 bp sequence pad for annealing of an oligonucleotide in ssHi-C, a 2,086 bp-long sequence corresponding to the first half of the LYS2 gene (+4 to +2,090), and a 117 bp HO cut-site. It is positioned at position 116,167 (S288c coordinates), replacing the URA3 coding sequence. A 2,086 bp-long ectopic donor (+4 to +2,090 of the LYS2 gene) is present at its endogenous locus on chr. II. The rad51::kanMX mutation were previously described (Piazza et al., 2021). Gene deletions and point mutations were verified by PCR and bam files generated upon alignment of Hi-C reads. Protein tagging was verified by Western blotting.

[0233] • Culture media

[0234] Liquid YPD (1% yeast extract, 2% peptone, 2% dextrose), YEP-lactate (1% yeast extract, 2% peptone, 2% lactate) and Lactate-MET-TRP (0.17% yeast nitrogen base, 0.5% ammonium sulfate, 0.2% synthetic dropout lacking methionine and tryptophan, 2% lactate) were prepared accordingly.

[0235] • Induction of DSB formation in cells synchronously released in S-phase

[0236] Exponentially growing cultures in YEP-lactate (30°C) were synchronized at the Gl / S transition by addition of 1 pg / ml alpha-factor (ProteoGenix, cat. 559627) every 30 minutes for 3-4 hours. Cells were washed 3 times with 50 mL of pre-warmed YEP-lactate and released in S-phase in YEP-lactate supplemented with 2% galactose to simultaneously induce HO expression. 2% galactose was added to induced DSB, while cells were maintained in G1 by alpha-factor addition for 4 more hours and then collected for Hi-C. The cell-cycle stage was verified by flow cytometry.

[0237] • Flow cytometry

[0238] Approximately 107cells were collected by centrifugation, re-suspended in 70% ethanol and fixed at 4°C for at least 24 hours. Cells were pelleted, washed in 1 mL of 50 mM Sodium citrate pH 7.0, and resuspended in 1 mL 50 mM Sodium citrate pH 7.0. 200 pL of the resuspension was treated with 200pg Rnase A (Euromedex, cat. 9707-C). Following incubation at 50°C for 1 hour or 37°C overnight, cells were pelleted and resuspended in 1.8 mL of 50 mM Sodium citrate pH 7.0 with 2 pM of SYTOX™ Green (Thermo Scientific, S7020), incubated for 1 hour at room temperature in the dark, and sonicated for 10 seconds on a Bioruptor. Flow cytometry profiles were obtained on a MACSQuant machine and analyzed using Flowing Software 2.5.1 or Flowjo software 10.8.0.

[0239] Annealing and capture oligonucleotide design

[0240] Annealing oligonucleotides have been designed as follows: a length of 80 nucleotides;

[0241] Overlap a DpnII restriction site and a Mfel or SspI restriction site located at less than 40 bp from the DpnII site.

[0242] Asymmetrically positioned relative to the DpnII site so has to ensure sufficient basepairing near the DpnII site (> 4 bp) and maximize the length of annealed dsDNA produced on the Mfel / Ss pl-containing side.

[0243] Contain 4-6 single nucleotide substitutions introduced as evenly as possible within the \1feI / SspI-containg side. One of these substitutions inactivates the Mfel / SspI site. Other substitutions are chosen preferentially to inactivate cryptic DpnII (e.g. AATC instead of GATC) and Hinfl e.g. AANTC instead of GANTC). Substitutions are not introduced within 4 bp of the DpnII site to be restored. Finally, they must be as evenly distributed as possible.

[0244] Oriented so as to anneal to the ssDNA present in the cell.

[0245] Be HPLC- or PAGE-purified.

[0246] Capture oligonucleotides target the dsDNA part restored by the annealing oligonucleotides, apart from control dsDNA capture oligonucleotides that target single dsDNA fragments without specific constrains. They contain a 5’ Biotin. Annealing and capture oligonucleotides were synthesized by Integrated DNA technologies and provided in Table 1 and Table 2, respectively. For an example of the design of the annealing and capture oligonucleotide for the DpnII site L2599 (corresponding to the restriction fragments 18613 (SEQ ID NO: 71) and 18614 (SEQ ID NO: 72)), see Figure 4B.

[0247] Table 1: Annealing oligonucleotides

[0248] Table 2: Capture oligonucleotides

[0249] Cell fixation with formaldehyde and EGS

[0250] Cell fixation is adapted from Micro-C-XL (Hsieh et al., 2016). Approximately 1.5xl09haploid cells were fixed with 3% formaldehyde (Sigma-Aldrich, cat. F8775) for 30 minutes at RT at 120 rpm. Formaldehyde was quenched with 330 mM glycine for 20 minutes at RT at 120 rpm.

[0251] Cells were washed twice with cold water at 3,000 g for 10 minutes. Pellets were split in two tubes and frozen at -80° C.

[0252] Cell pellets were thawed in ice, resuspended in 10 mL of Zymolyase solution (0.25 mg / mE Zymolyaze 100T (Carl Roth, cat. 9329), 1 M D-sorbitol (Sigma-Aldrich, cat. S6021), 50 mM Tris-HCl pH 7.5, 10 mM 2-mercaptoethanol (Sigma- Aldrich, cat. M3148)) and spheroplasted at 210 rpm for 40 minutes at 30° C (cells spheroplasting is checked under the microscope and the incubation time adjusted accordingly). Spheroplasts are pelleted at 3,000 g for 10 minutes at 4° C, washed with 4 mL cold IX PBS, and resuspended in a fresh EGS crosslinking solution (3 mM ethylene glycol bis(succinimidyl succinate) (Fisher Scientific, cat. 10350924) in IX PBS), and incubated at 30° C for 40 minutes at 100 rpm. EGS was quenched with 400 mM glycine at RT for 5 minutes at 100 rpm. The crosslinked cellular material was pelleted at 3,000 g for 10 minutes at 4° C and washed twice in cold IX PBS supplemented with anti-protease (Sigma-Aldrich, cat. 11836170001). Pellets were snap-freezed in liquid nitrogen and stored at -80°C.

[0253] • Hi-C procedure with restriction site restoration in ssDNA

[0254] The crosslinked pellets were thawed in ice and resuspended in cold H2O supplemented with anti-protease. Between 2 and 5.107cells were used for the remainder of the procedure. The ssHi-C procedure uses the Arima Hi-C kit (Arima Genomics, cat. A410079), which employs a dual restriction digestion (DpnII and Hinfl) yielding a median fragment length of 108 bp in S. cerevisiae. The protocol follows manufacturer’s instructions with the following modification: after the lysis step, the pool of HPLC- or PAGE-purified annealing oligonucleotides (stock concentration 100 nM each) is added at a final concentration of 7 nM each. Annealing occur during the subsequent incubation at 65° C for 10 minutes.

[0255] • Sequencing library preparation

[0256] DNA was fragmented into 300-400 bp fragments using Covaris M220 sonicator. Preparation of the libraries for paired-end sequencing on an Illumina platform was performed using the Thermofisher Collibri ES DNA Library Prep Kit for Illumina Systems with UD indexes (cat. A38606024) following manufacturer’s instructions. The library was amplified in triplicate PCR reactions using oligonucleotides corresponding to the Illumina sequencing adaptors (5’- AATGATACGGCGACCACCGAGATCTACAC-3’ (SEQ ID NO: 69) and 5’- CAAGCAGAAGACGGCATACGAGAT-3’ (SEQ ID NO: 70)) and Phusion DNA polymerase (New England Biolabs, cat. M0531) for 11 cycles. PCR products were purified with AMPure XP beads (Beckman-Coulter, cat. A63881) and resuspended in pure H2O. The Hi-C library is quantified using the Qubit DNA high sensitivity kit (Thermo Scientific, cat. Q32851) on a Qubit 2 fluorometer (Thermo Scientific, cat. Q32866). ssHi-C library Capture

[0257] The ssHi-C library consists in an enrichment of ssDNA contacts from the Hi-C library using a panel of biotinylated “capture” oligonucleotides. 500 ng of PCR products were saturated with 2 (ig of unlabeled Cotl human DNA (Invitrogen, cat. 15279-011) and 0.2 nmol of oligonucleotides corresponding to the universal Illumina sequencing adaptors (see above) and dehydrated at 60° C for 1 hour in a Speedvac (Thermo Scientific, cat. DNA130-230). DNA was resuspended in IX xGen Hybridization buffer and 0.15X of xGen Hybridization Enhancer (Integrated DNA Technologies, cat. 10010352), and the pool of biotinylated capture oligonucleotides (stock 1.8 pm each) at a final concentration of 0.6 pM. DNA is denatured at 95° C for 5 minutes, and annealing occurs at 65° C for 4 hours. Biotin pulldown on streptavidin- coated dynabeads Cl (Invitrogen) was performed according to manufacturer’s instructions, and DNA was recovered in pure H2O. PCR amplification (10 cycles) and DNA purification on AMPure XP beads was carried out as before. The capture procedure was repeated once and the final library was amplified for only 8 cycles.

[0258] • Elimination of the dsDNA-bome contacts

[0259] This step eliminates the chimeric molecules involving the fragments of interest (i.e. targeted by annealing and capture oligonucleotides) that were in a dsDNA form in the cell population at the time of crosslink, upon digestion by restriction enzymes whose site was eliminated in the annealing oligonucleotide (rational Figs. 4A-B) 300 ng of PCR products were digested at 37° C for 1 hour with 10 units of Mfel-HF and SspI-HF in IX rCutsmart buffer (New England Biolabs, cat. R3589, R3132 and B7204, respectively) and the enzymes heat-inactivated at 65° C for 20 minutes. DNA was purified with AMPure XP beads, resuspended in pure H2O, and quantified on Qubit.

[0260] • Paired-end illumine sequencing ssHi-C libraries were paired-end sequenced (2x150 bp) on an Illumina Novaseq6000 platform by Novogene UK.

[0261] Data analysis

[0262] Generation of contact maps Paired-end reads are aligned and processed using Hicstuff (Matthey-Doret et al., 2020). Briefly, pairs of reads are aligned iteratively and independently using Bowtie2 in its most sensitive mode to the S. cerevisiae reference genome deleted for the URA3 and LYS2 sequences. Each uniquely mapped read was assigned to a restriction fragment. Uncuts, loops and circularization events were filtered as described (Cournac et al., 2012) and PCR duplicates discarded. Contact maps were generated and normalized using Hicstuff (view function). Contact maps were generated, loglO transformed, and rendered. For 80 kb DSB-donor interaction windows, bins were set at 4 kb. No transformation was performed before rendering.

[0263] Ratio maps were generated from sparse matrices using Hicstuff view function, normalized, log- transformed and binned at either 1 or 2 kb. Other ratio maps were generated with Serpentine (Baudry et al., 2020). Briefly, 1 kb matrices were subsampled to contain the same number of contacts. For intra-chromosomal interactions, only chromosomes of interest were loaded into Serpentine. Comparison between pairs of contact maps was performed using the default threshold parameters (50 and 5) and the detrending constant determined automatically by Serpentine for each comparison.

[0264] • Computation of the contact probability as a function of genomic distance

[0265] Computation of the contact probability as a function of genomic distance Pc(s) and its derivative have been determined using Hicstuff distance law function with default parameters, averaging the contact data of entire chromosome arms.

[0266] • Quantification of coverage from Hi-C data

[0267] Quantification of coverage from Hi-C data was determined with custom scripts from sparse contact matrices generated with Hicstuff pipeline function. Briefly, the sum of total contact frequencies within 1 kb bins was computed along the broken chromosome for each sample, and normalized onto the average coverage of undamaged cells obtained from a biological replicate.

[0268] • Statistics & Reproducibility

[0269] The proportion of intra-chromosomal contacts were compared using a non-parametric Mann- Whitney Wilcoxon rank sum test, two-tailed. Parallel DLC values were compared using a nonparametric Mann- Whitney Wilcoxon rank sum test, paired, two-tailed. Statistical cut-off was set at 0.05. P-values below this cut-off are denoted by a star (*) in figures. Pearson correlation coefficients (r) between DLC and Hi-C contact data were computed with Graphpad Prism 9.1.2. Other statistical tests were performed under R 3.6.2. No statistical method was used to predetermine sample size. No data were excluded from the analyses. The experiments were not randomized. The investigators were not blinded to allocation during experiments and outcome assessment.

[0270] Results ssHi-C: a Hi-C-based methodology for mapping ssDNA contacts

[0271] A single cut-site for the galactose-inducible HO endonuclease (HOcs) was introduced 35 kb away from the centromere of chr. V, with a 2 kb-long region of homology to the left DSB end located on chr. II (Fig. 1A; Methods). HO expression is induced upon S-release of Gl- synchronized MATa-inc cells, yielding a highly homogeneous population of checkpoint- arrested cells attempting DSB repair by HR in G2 / M at 2 and 4 hours post-DSB induction23(Fig. IB). The efficient DSB formation in this system (>90% cut in Ih) authorizes kinetics study of HR up to D-loop extension (Piazza et al., 2019). Hi-C is performed on spheroplasted cells crosslinked with FA and EGS upon chromatin digestion with DpnII and Hinfl (median fragment length = 108 bp; Methods), which yields high-quality stereotypic contacts maps at sub-kilobase resolution (see below, Fig. IH).

[0272] DSB resection produces long 3’-protruding ssDNA that eliminates restriction sites and the Hi- C readout (Fig. 1C). We devised ssHi-C, a procedure to restore, isolate and enrich genomewide contacts made by ssDNA (Fig. ID and Methods). First, 80mer oligonucleotides are annealed to the resected ssDNA at the chromatin solubilization step so as to overlap specific DpnII sites (Fig. ID and Fig. 4A-B). These “annealing oligos” have their 5’ and 3’ ends chemically blocked to prevent artefactual biotinylation and ligation in the absence of annealing to, and digestion of, their DpnII target site. They also contain single-nucleotide polymorphisms (SNPs), some of which inactivate a secondary restriction site (Mfel or SspI; Fig. 4A-B). These SNPs will allow distinguishing contacts made at that site in the ssDNA or dsDNA form (Fig. 4A). The Hi-C library is subjected to two rounds of targeted enrichment for the ssDNA sites of interest, as well as 7 dsDNA control sites, using biotinylated “capture oligos”. Isolation of ssDNA contacts is achieved by combining Mfel / SspI digestion of the ssHi-C library prior to sequencing with a SNP-based post-sequencing read filtering step (Fig. 4A-C, Methods). We designed 21 annealing oligos targeting 19 DpnII restriction fragments on both sides of, and up to 17 kb away from, the DSB site, and 26 capture oligos targeting 17 out of the 19 ssDNA fragments as well as 7 control dsDNA fragments (Fig. IE). One control dsDNA site (ARG4) is at the same distance from a centromere as the DSB site (Fig. 4C). Annealing oligos partially restored Hi-C coverage, which could be specifically enriched using capture oligos (Fig. 1F-H and Fig. 5A-B). No recovery and enrichment were observed at these sites in the absence of a DSB-inducible construct (Fig. 1F-H and Fig. 5A-B), indicating that ssHi-C contacts are not produced by random ligation of annealing oligos. Site-to-site differences in ssDNA fragment enrichment partially depended on the fraction of resected DNA: the greater the resected fraction, the greater the ssDNA enrichment (Fig. II). Some other sites remained poorly restored, despite being fully resected (e.g. L1560 vs L2599 Fig. IF), suggesting differences in annealing efficiency. Various annealing oligo dropouts indicated that they (i) are each required, (ii) act independently, and (iii) do not interfere with one another for ssDNA site recovery (Fig. 1J, Fig. 5B). In conclusion, the ssHi-C procedure recovers ssDNA-specific contacts for 16 out of the 17 fragments, with varying site-to-site efficiency.

[0273] Despite the strong enrichment for the 16 ssDNA and 7 dsDNA fragments targeted by the capture oligos, 87% of the total contacts originated from the 74,884 other, non-targeted fragments. The resulting contact data were highly similar to that of conventional Hi-C (Fig. 1H, Fig. 5C-D). Conversely, the 7 control dsDNA sites recapitulate conventional Hi-C contacts (Fig. 5E and Fig. 2A,D). Consequently, the capture procedure does not introduce bias at target and non-target sites. ssHi-C library can thus be used to study both ssDNA-specific contacts at the enriched sites as well as genome-wide pairwise contacts (Fig. IK). ssDNA predominantly contacts chromatin in trans

[0274] Discrete stripes of non-uniform contacts emanate from the DSB region and stretch across chr. V and the genome (Fig. 1H). No such stripes are observed at the 7 captured dsDNA regions, indicating that this contact behavior is specific to the DSB region and not an enrichment artefact (Fig. 1H). A finer 4C-like representation reveals a remarkable propensity of ssDNA on both sides of the DSB to contact inter-chromosomal sites (hereafter trans contacts, 64% at 4 hours), with local variations in contact frequencies (e.g. enriched at centromeres, the donor site, and chr. Ill; Fig. 2A-B). It contrasts with both a control dsDNA site located at the same distance from a centromere as the DSB site (ARG4, 22%) and to a likely over-estimation of DSB- flanking trans dsDNA contacts detected by conventional Hi-C (33.7%; Fig. 2A-B; rationale for the over-estimation Fig. 6A-B). ssDNA trans contacts increase significantly from 2 to 4 hours, opposite to the genome-wide chromatin condensation (Fig. 2B). Notably, trans contacts increase at a distance from the resection front, with -40% of DSB -distal contacts made with other chromosomes vs. -74% for DSB-proximal sites (Fig. 2C). Consequently, ssDNA predominantly contacts chromatin in trans with a DSB-proximal-to-distal gradient in WT cells.

[0275] DSB-proximal ssDNA contacts with cis chromatin are broadly distributed

[0276] Although conversely depleted in cis, ssDNA fragments still engage the immediate flanking sequences -2 orders of magnitude more frequently than trans sites on average (—5.10sper kb, Fig. 2A, D). Again, the distribution of cis contacts departs from that of dsDNA regions detected with Hi-C (Fig. 2D). First, ssDNA contacts are not asymmetrically distributed across the DSB region: e.g. the left DSB end contacts regions on both sides with similarly frequencies. Second, contact enrichment with DSB-flanking chromatin decreases from 2 to 4 hours, coincident with an increase in trans contact. Third, the decay of ssDNA contacts with genomic distance does not follow a power law (e.g. left-end side): instead, it appears monotonous at 4 hours (Fig. 2D). Per-site decomposition of ssDNA contacts reveals that the peak of immediate cis interaction can be attributed to DSB-distal sites while DSB-proximal sites make long-range contacts, enriched 10-fold at the highest over the genome average (Fig. 2E). More specifically, the proximity to the resection correlates with frequent interaction of ssDNA sites with chromatin in cis, as well as across the resected region (e.g. the R8073 site, at the resection edge at 2 hours and -8 kb away at 4 hours post-DSB induction, Fig. 2C, E). Accordingly, DSB-distal sites follow the genome- averaged probability of contact Pcas a function of genome distance s, while other ssDNA sites contact cis chromatin at long-range (Fig. 2G). These observations suggest that ssDNA sites are progressively relieved from constraints confining most dsDNA contacts in cis over time. This relief correlates with the distance to the resection front.

[0277] DSB-proximal and DSB-distal ssDNA regions are differently emancipated from the cisacting spatial constraint imposed by the nearby centromere

[0278] The Rabi organization of yeast chromosomes is defined by anchoring of centromeres at the spindle-pole body (SPB) and of telomeres at the nuclear envelope, whose distance from the SPB scales with chromosome arm length (Berger et al., 2008; Bystricky et al., 2005; Duan et al., 2010). This organization spatially constrains chromatin, promoting clustering between -100 kb centromeric regions and inhibiting centromere-telomere contacts (Duan et aL, 2010; Jin et al., 2000; Therizols et al., 2010; Verdaasdonk et al., 2013) (Fig. 1H). Positioning the DSB site 35 kb left of CEN5 (Fig. 1A) enabled us to address the influence a spatially constrained loci in cis exerts on the neighboring ssDNA contacts. Centromeres are preferred ssDNA interaction sites (Fig. 2A, G). This preference is more pronounced for the centromere-tethered right DSB end, especially for sites close to the resection front (Fig. 2G-H). The fraction of trans contacts made with centromeres was nonetheless lower for DSB ends than for the undamaged dsDNA ARG4 locus located at the same distance from a centromere on chr. VIII (Fig. 2G). Centromere de-clustering upon Nocodazole treatment was associated with a reduced engagement of centromeres in trans by the DSB ends (Fig. 6C-D). Finally, trans contacts with centromeres decreased from 2 to 4 hours, in a manner that correlated with the distance to the resection front. Likewise, inhibition of trans contacts with telomeres was partly relieved for ssDNA sites relative to dsDNA sites, for both short and long chromosomal arms (Fig. 21). These observations show that a m-acting spatial tether differently constrains the two DSB ends. Furthermore, this tether exerts an overall reduced influence on the contacts made by DSB ends compared to intact dsDNA loci.

[0279] Altogether, these results show that resected DSB ends are partly emancipated from constraints exerted in cis, allowing them to reach initially distant genomic sites. This emancipation increases along the ssDNA in the 5 ’-3’ direction and over time, inversely correlating with the distance from the resection front.

[0280] Rad51 promotes genome-wide ssDNA-chromatin contacts

[0281] We evaluated the contribution Rad51 in regulating genome-wide ssDNA contacts at 2 and 4 hours post-DSB induction in a rad51A mutant. Absence of Rad51 did not affect resection, DSB end-tethering, metaphase-arrest, cohesin-mediated loop folding, chromosome individualization, or the Rabi chromosome organization (Piazza et al., 2021) (Fig. 3A and Fig. 7A-E). The amount of captured ssDNA fragments decreased ~2-fold compared to WT cells, while dsDNA fragments remained constant (Fig. 3A). This observation is consistent with the stabilization by protein-DNA crosslinking of Rad51 / RecA-ssDNA NPF transient association with heterologous dsDNA in vitro (Tsang et al., 1985; Pinsince and Griffith, 1992) which leads to the biotinprotecting ligation events reported by Hi-C (Lieberman-Aiden et al., 2009) .

[0282] Calibration of the contact distribution on the capture efficiency of each ssDNA fragment enabled determining the absolute gain and loss of contacts relative to the WT context (Methods). Strikingly, elevated trans contacts were abolished in the absence of Rad51, down to levels observed for dsDNA sites (Fig. 3A-B, Fig. 7F). Likewise, long-range cis contacts decreased while short-range contacts increased modestly or remained constant, both in cis and across the resected DSB region (Fig. 3C). Unlike in WT cells, the contact profile in cis remained similar for ssDNA sites ~10 kb apart (Fig. 3D) and followed the genome-averaged Pc(s), irrespective of their distance to the DSB end (Fig. 3E). It suggests that distant ssDNA sites are spatially associated in the absence of Rad51, and behave as DSB-distal sites. The trans contacts were enriched at the spatially tethered centromeres and depleted from the more distant telomeres compared to WT cells, similar to an intact dsDNA site (Fig. 3F-H). These observations show that reduced spatial constraints of ssDNA sites observed in WT cells are not simply due to the rupture of the dsDNA polymer at the DSB, but instead reflects ongoing homology search by the Rad51-ssDNA NPF.

[0283] References:

[0284] Belton JM, McCord RP, Gibcus JH, Naumova N, Zhan Y, Dekker J. (2012). Hi-C: a comprehensive technique to capture the conformation of genomes. Methods. 58(3):268-276.

[0285] Rao SS, Huntley MH, Durand NC, Stamenova EK, Bochkov ID, Robinson JT, Sanborn AL, Machol I, Omer AD, Lander ES, Aiden EL. (2014). A 3D map of the human genome at kilobase resolution reveals principles of chromatin looping. Cell.l59(7):1665-1680.

[0286] Gyarmati P, Song Y, Hallman J, Kaller M. (2013) Chemical fragmentation for massively parallel sequencing library preparation. J Biotechnol. 168(1 ):95- 100.

[0287] Cullen KE, Kladde MP, Seyfred MA. (1993). Interaction between transcription regulatory regions of prolactin chromatin. Science. 61(5118):203-206

[0288] Piazza, A., Shah, S.S., Wright, W.D., Gore, S.K., Koszul, R., and Heyer, W.-D. (2019). Dynamic processing of displacement loops during recombinational DNA repair. Mol. Cell 73, 1255-1266.

[0289] Piazza, A., Bordelet, H., Dumont, A., Thierry, A., Savocco, J., Girard, F., and Koszul, R. (2021). Cohesin regulates homology search during recombinational DNA repair. Nat. Cell Biol. 23, 1176-1186.

[0290] Dauban L, Montagne R, Thierry A, Lazar-Stefanita L, Bastie N, Gadal O, Cournac A, Koszul R, Beckouet F. (2020). Regulation of Cohesin-Mediated Chromosome Folding by Ecol and Other Partners. Mol Cell. 77(6):1279-1293.

[0291] Hsieh TS, Fudenberg G, Goloborodko A, Rando OJ. (2016). Micro-C XL: assaying chromosome conformation from the nucleosome to the entire genome. Nat Methods. 13(12):1009-1011.

[0292] Matthey-Doret, C., baudrly, Amaury, axelcoumac, Remi-Montagne, Guiglielmoni, N., Foutel- Rodier, T, and Scolari, V. F. (2020). koszullab / hicstuff: Use miniconda layer for docker and improved P(s) normalisation (v2.3.1).10.528 l / zenodo.4066363

[0293] Cournac, A., Marie-Nelly, H., Marbouty, M. et al. (2012). Normalization of a chromosomal contact map. BMC Genomics 13, 436. Baudry L, Millot GA, Thierry A, Koszul R, Scolari VF. Serpentine: a flexible 2D binning method for differential Hi-C analysis. Bioinformatics. 2020 Jun 1 ;36( 12):3645 -3651.

[0294] Berger, A.B., Cabal, G.G., Fabre, E., Duong, T., Buc, H., Nehrbass, U., Olivo-Marin, J.-C., Gadal, O., and Zimmer, C. (2008). High-resolution statistical mapping reveals gene territories in live yeast. Nat. Methods 5, 1031-1037.

[0295] Bystricky, K., Laroche, T., van Houwe, G., Blaszczyk, M., and Gasser, S.M. (2005). Chromosome looping in yeast: telomere pairing and coordinated movement reflect anchoring efficiency and territorial organization. J. Cell Biol. 168, 375-387.

[0296] Duan, Z., Andronescu, M., Schutz, K., McIlwain, S., Kim, Y.J., Lee, C., Shendure, J., Fields, S., Blau, C.A., and Noble, W.S. (2010). A three-dimensional model of the yeast genome. Nature 465, 363-367.

[0297] Qw, Jin., J, F., and J, L. (2000). Centromere clustering is a major determinant of yeast interphase nuclear organization. J. Cell Sci. 113 ( Pt 11).

[0298] Therizols, P., Duong, T., Dujon, B., Zimmer, C., and Fabre, E. (2010). Chromosome arm length and nuclear constraints determine the dynamic relationship of yeast subtelomeres. Proc. Natl. Acad. Sci. U. S. A. 107, 2025-2030.

[0299] Verdaasdonk, J.S., Vasquez, P. A., Barry, R.M., Barry, T., Goodwin, S., Forest, M. G., and Bloom, K. (2013). Centromere Tethering Confines Chromosome Domains. Molecular Cell, 52(6), 819-831. Tsang, S.S., Chow, S.A., and Radding, C.M. (1985). Networks of DNA and RecA protein are intermediates in homologous pairing. Biochemistry 24, 3226-3232.

[0300] Pinsince, J.M., and Griffith, J.D. (1992). Early stages in RecA protein-catalyzed pairing. Analysis of coaggregate formation and non-homologous DNA contacts. J. Mol. Biol. 228, 409- 420.

[0301] Lieberman- Aiden E, van Berkum NL, Williams L, Imakaev M, Ragoczy T, Telling A, Amit I, Lajoie BR, Sabo PJ, Dorschner MO, Sandstrom R, Bernstein B, Bender MA, Groudine M, Gnirke A, Stamatoyannopoulos J, Mirny LA, Lander ES, Dekker J. (2009). Comprehensive mapping of long-range interactions reveals folding principles of the human genome. Science. 326(5950):289-293.

Claims

Claims1. An in vitro method for analyzing interactions between i) single-strand DNA (ssDNA) and ii) double-strand or single-strand DNA (dsDNA or ssDNA), also designated as ssDNA contacts, preferably in chromatin, which method comprises a) mixing genomic DNA with a set of annealing oligonucleotides, wherein each annealing oligonucleotide is designed to anneal to a ssDNA at a target genomic site, while differing from an endogenous sequence that is complementary to said target genomic site, whereby the annealing of the oligonucleotide with the target genomic site restores a double-stranded restriction site; b) digesting the restored genomic DNA, by means of an enzyme that specifically recognizes said first double- stranded restriction site, to produce digested genomic DNA; c) labelling the ends of the digested genomic DNA and ligating the spatially proximal labelled digested ends, whereby labelled ligation junction fragments are provided; d) purifying the labelled ligation junction fragments, and preparing a library from the purified fragments; e) sequencing the library; wherein sequences that differ from said target genomic sites are identified and marked as a proximal interaction between ssDNA and dsDNA or ssDNA in the genomic DNA.

2. The method of claim 1, wherein said set of annealing oligonucleotides consists of identical oligonucleotides or includes oligonucleotides of different sequences.

3. The method of claim 1 or 2, wherein the annealing oligonucleotides comprise DNA nucleotides or modified nucleotides, optionally LNA.

4. The method of any of claims 1 to 3, wherein the set of annealing oligonucleotides is designed so that all oligonucleotides anneal to the target genomic site(s) under the same hybridization conditions, in particular the same temperature and salt concentration, and / or wherein the set of annealing oligonucleotides is designed so that the digested annealing oligonucleotides remain annealed to the target genomic sites at thedenaturation temperature of the enzyme that specifically recognizes the first doublestranded restriction site.

5. The method of any of claims 1 to 4, wherein each annealing oligonucleotide restores the first double- stranded restriction site located at about 1-15 nucleotides, preferably 4-10 nucleotides from the 3’ or the 5’ end of the oligonucleotide.

6. The method of any of claims 1 to 5, wherein the first double-stranded restriction site is DpnII.

7. The method of any of claims 1 to 6, wherein each annealing oligonucleotide comprises at least one, preferably at least 2 or 3, mutations compared to the endogenous sequence that is complementary to said target genomic site, optionally wherein the annealing oligonucleotide comprises a single point mutation every 10 to 30 nucleotides.

8. The method of claim 7, wherein at least one of said mutations inactivates a second double- stranded restriction site, optionally Mfel or Sspl.

9. The method of any of claims 1 to 8, wherein step c) comprises labelling the ends of the digested genomic DNAby filling in the overhanging ends with biotinylated nucleotides.

10. The method of any of claims 1 to 9, which method further comprises at least a step d’) of submitting the library to a further enrichment of the labelled ligation junction fragments obtained from ssDNA contacts, comprising at least one cycle of denaturation and annealing with capture oligonucleotides, optionally wherein the capture oligonucleotides are biotinylated and the annealed capture oligonucleotides are recovered by streptavidin pulldown.

11. The method of claim 10, wherein at least one of said mutations within the annealing oligonucleotide inactivates a second double- stranded restriction site, optionally Mfel or Sspl, and, before sequencing, the method further comprises submitting the fragments enriched by step d’) to a digestion with an enzyme that specifically recognizes said second double- stranded restriction site.

12. The method of claim 10, wherein at least one of said mutations within the annealing oligonucleotide inactivates a second double- stranded restriction site, optionally Mfel or SspI, and the method further comprises a step of digestion with an enzyme that specifically recognizes said second double- stranded restriction site, before the step d’) of further enrichment.

13. The method of any of claims 1 to 12, which method comprises a preliminary step of providing a sample of one or more cells, cellular nuclei, or an acellular nucleic acid system comprising DNA strands, wherein the DNA strands are held in a fixed position relative to one another, optionally by a crosslinking treatment, preferably with a chemical crosslinker that comprises an aldehyde, preferably with formaldehyde.

14. The method of any of claims 1 to 13, for use in identifying interaction frequencies.

15. A kit for conducting the method of any of claims 1 to 14, wherein the kit comprises: a set of annealing oligonucleotides as defined in any of claims 1 to 14, optionally capture oligonucleotides that are specific to ssDNA at target genomic sites, and optionally capture oligonucleotides that are specific to at least one reference doublestranded target genomic site.

Images