Methods for identifying genes associated with diseases or specific phenotypes

Description

TECHNICAL FIELD OF THE INVENTION

[0002] The field of this invention is genetic mapping, and also genetic variation or mutation screening.

[0003] This invention presents an improved method for screening single nucleotide polymorphisms/variations/mutations (SNPs) and identical sequences between two pools of genomic or cDNA or multiple genes, or within one of these pools, each pool comprising the DNA from one individual or multiple individuals. It may be used for isolating or and mapping the genes or genetic markers related to any trait such as human complex traits, simple Mendelian diseases or somatic genetic mutation. It may also be used for profiling and/identifying SNPs for any DNA or cDNA pool or between 2 pools of them.

[0004] This invention presents an improved method for screening genomic, cDNA, or any DNA fragments in general, and capturing and/or mapping genes related to any trait. DNA is digested with restriction endonucleases and adapters with novel designs are ligated onto the ends. The adapters allow improved methods to prepare heterohybrid DNA in which a DNA double helix is formed from single strands originating from two different individuals or two different pools of DNA. The methods allow for separating heterohybrids and homohybrids that are formed in the same annealing reaction and specifically recovering either the heterohybrids or the homohybrids. Sequence differences (polymorphisms) between the two individuals are revealed as base mismatches in the heterohybrid DNA duplex. The well-established method of genome mismatch scanning (GMS) makes use of heterohybrid DNA to reveal polymorphisms between the complementary strands. An improved method of GMS is shown in which thymine glycosylases and related enzymes are used to capture DNA containing the mismatched basepairs in the heterohybrid DNA. The novel adapters allow the selective amplification or physical capture of either heterohybrid or homohybrid DNA following the separation of perfectly base paired DNA or DNA containing mismatched basepairs by glycosylases or other GMS enzymes.

[0005] This invention presents an improved method for screening DNA pool(s) in general, and isolating the single nucleotide variations (SNPs)-containing DNA sequences from identical DNA sequences between two pools of DNA or within one of these pools, each pool comprising genomic, cDNA, or any other DNA fragments from one or multiple individuals. Mapping of any trait is realized by parallel analysis of resulted mismatch fragments and perfectly matched fragments from either/both heterohybrid pool between the desired trait DNA pool and its control, or/and from the homohybrid pool from the control DNA pool, via gel display, subtractive hybridization, DNA microarray hybridization, FISH, or other approach. DNA Glycosylases or other mismatch repair enzymes with similar function are employed to separate mismatch fragments from perfectly matched fragments. A series of strategies with novel designed adapters is given for specific and selective recovery and amplification of either homohybrids or the heterohybrids from the same mixture of two resources of DNA population after restriction digestion, adapter ligation, mixing, denature and reannealing.

[0006] The invention includes two related novel techniques:

[0007] 1) An improved method for separating and recovering either homohybrids or heterohybrids (in which the two strands of a double helix are contributed by two different individuals) formed in an annealing reaction. The heterohybrid DNA is selectively amplified via novel adapters ligated to the ends of the DNA.

[0008] 2) A method to capture those heterohybrids which contain single nucleotide sequence differences (SNPs) using DNA glycosylase(s) and other enzymes with similar functions.

BACKGROUND OF THE INVENTION

[0009] Most human traits are genetically complex, oligogenic/polygenic or multifactorial traits. This is the case for most human diseases and other medically relevant phenotypes such as drug-response.

[0010] A need exists for a method to rapidly identify the relevant genes. This will lead to improved understanding, prognosis, diagnosis, treatment, prevention, and establishment of markers for individualized medicine and development of new drugs. Complex diseases (or traits) are usually affected by two or more interacting genes. Sometimes more than 100 genes are involved, each contributing a small effect on the risk/susceptibility for the disease. Many of the related alleles appear at low frequency even in the patient population (some <10%, most <30%) since a given gene may not be absolutely required for the disease to occur. However, the related alleles will tend to occur at least at some higher frequency in the DNA of diseased individuals compared to the DNA of control subjects. These allele frequency differences form the basis for strategies to identify those genes linked to the disease.

[0011] Humans are an out-bred species. Many polymorphisms or sequence variations exist, of which may have no relation to the trait of interest. In other words, the variants exist not only between the diseased and control cohorts, but also within each group. Actually most of the human polymorphism exists within each population. So, parallel comparison of the variations between the trait population with its control population is critical when the associated genetic markers or genes is desired.

[0012] The approaches currently available for gene discovery, such as Functional cloning depending on information about the protein, and Positional cloning relying on gene position information, are mainly successful for simple Mendelian diseases. A few successful efforts have been reported on limited number of genetic markers and on some knowledge for mapping human complex traits, for example using Candidate gene approach or genome scanning with limited number of micro satellites or other markers, yet many obstacles exist. New approaches are required for complex disease gene identification.

[0013] One whole genome screening approach previously proposed, as GMS (Genomic Mismatch Screening), is a two-step method for biochemical enrichment of the regions of the genome at which two individuals share identical alleles; it is designed to map all the regions of genetic IBD (identical-by-descent) between two related individuals. First, heterohybrid DNA molecules formed by a process of solution hybridization between two genomic DNA fragment pools from two individuals are purified by a procedure based on differential restriction methylation and endonuclease digestion. A DNA methylase was used to methylate the DNA of one individual but not the DNA of the second individual. The DNAs were then mixed, denatured, and reannealed to from a mixture of heterohybrid and homohybrids DNA. This will result in hemimethylated DNA for heterohybrids that is resistant to certain restriction endonucleases. Homohybrid DNA, in contrast can be eliminated by digestion. Second, mismatch-containing hybrids formed between nonidentical alleles are eliminated by treatment with the Escherichia coli mismatch repair enzymes, Mut H, MutL and Mut S, which are capable of binding and modifying the base mispair-containing hybrids in the existence of “GATC” site. The remaining mismatch-free heterohybrids, representing loci at which the two individuals share identical alleles, can then be mapped in a single genome-wide hybridization step. Researches have shown that GMS can be used to mapp the regions of IBD (Identical-by-Descent) between 2 strains of yeast, mouse, or two human individuals (Nelson S F, Nature Genetics, 4:11, 1993; Mirzayans F, Am J Hum Genet, 61:111, 1997; Cheung V G, Genomics, 47: 1-6, 1998; McAllister L, Genomics, 47: 7, 1998; Cheung V G, Nature Genetics, 18: 225, 1998; Gerton J L, PNAS, 97; 11383,1999).

[0014] The genetic analysis of this invention is based on the frequencies of single nucleotide polymorphisms (SNPs). SNPs are the most abundant, stable and evenly distributed bi-allelic polymorphisms in the human genome, and occur at the rate of {fraction (1/300)}-{fraction (1/1000)} bp between 2 genome samples (>3×106 SNPs) or {fraction (1/2000)} bp between 2 coding sequences (cSNPs). In human populations, about 2×107 SNPs are expected. As the coding regions are 2.5-5% of the genome, so the total number of cSNPs is estimated to be >2.5% (>5×105 cSNPs) of the total SNPs in the human genome, an average of about 6 per gene, with about half of them resulting in non-synonymous codon usages (Collins F S, Genome Res., 1998: 8:1229-1231; Brookes A J, Gene, 1999, 234:177-186). 90% of the sequence variants in humans are SNPs. In recently years SNPs have been considered to be the best gene-mapping marker.

[0015] Different from that of GMS, the genetic analysis of this invention is based on the frequencies of single nucleotide polymorphisms (SNPs). SNPs are considered to be the best gene mapping marker in recent years with the human genome project is coming to be accomplished. SNP marker has several advantages comparing to the other previous commonly used genetic markers such as RFLP (restriction fragment length polymorphism), STR (short tandem repeats) or IBD used in GMS. SNPs are the most abundant, stable and evenly distributed bi-allelic polymorphisms in the human genome, and occur at the rate of {fraction (1/300)}-{fraction (1/1000)} bp between 2 genome samples (>3×106 SNPs) or {fraction (1/2000)} bp between 2 coding sequences (cSNPs). In human populations, about 2×107 SNPs are expected. As the coding regions are 2.5-5% of the genome, so the total number of cSNPs is estimated to be >2.5% (>5×105 cSNPs) of the total SNPs in the human genome, an average of about 6 per gene, with about half of them resulting in non-synonymous codon usages (Collins F S, Genome Res., 1998: 8: 1229-1231; Brookes A J, Gene, 1999, 234: 177-186). 90% of the sequence variants in humans are SNPs.

[0016] Currently, the major technique for applying SNPs in gene mapping is SNP typing based on knowledge of the individual SNPs. Variants of approaches based on variants of mechanisms have been invented to accomplish SNP typing. These include overlapping genomic sequencing or minisequencing, olignucleotide ligation assays (OLA), primer extension assays, allele-specific oligonucleotide (ASO) hybridization, exonuclease assays or 5′ nuclease assay, single base chain extension, and so on. Only some of them such as microarray techniques and mass spectrometry have the necessary features for industrial-scale SNP typing. Although these techniques succeed in some applications, few excellent examples using these techniques to map a typical complex disease have been reported. The major obstacles facing gene mapping of complex traits with SNPs are two. First, the SNPs currently or in near future available in trait mapping are too few in number, so they do not satisfy the requirement of genome coverage; second, thousands of individual samples may be required according to the theoretic prediction of geneticists. In addition, the intrinsic characteristics of the genetic complexity of complex traits and the complexity of the human genome also make the difficulties.

[0017] One approach previously proposed to use SNPs in genetic analysis is based on mixing the DNAs of two individuals together, denaturing, and then reannealing the strands back together. Therefore the DNA strands of the different individuals can base pair with each other. In this case, a mixed double stranded DNA will be formed, called a heterohybrid, in which one of the strands of the double helix is contributed by one individual, and the complementary strand is contributed by a different individual. Where the individuals have different DNA sequences (polymorphism), the strands of the heterohybrid will not form correct base pairs. Therefore, a high rate of polymorphism between the individuals will result in many mispaired bases in the heterohybrid DNA and a low rate of polymorphism will result in more perfectly matched bases. This difference forms the basis for methods to rapidly measure the diseases-related identical-by-descent (IBD) sequences. (IBD refers to sequences that individuals have in common, i.e. having low polymorphism, as a result of inheriting a trait from a common ancestor. IBD is used to associate sequences of low polymorphism to the trait.)

[0018] A class of DNA repair enzymes, MutHLS, was used to identify the mispaired bases in heterohybrid DNA. These enzymes are capable of binding and modifying base mispairs. The repair enzymes were used to remove the mispaired bases and thereby reveal the IBD sequences among two or more individuals. Such a strategy using yeast as the test organism was presented by Nelson and associates in 1993, and referred to as Genomic mismatch scanning (GMS)(Nelson S. F., et al. Nature Genetics, 1993, 4:11-18 and related subsequent papers). GMS has been modified and successfully used in screening traits related by IBD from yeast and from human chromosomes in conjunction with putative disease gene localization information. A critical step in the GMS procedure is the enrichment of heterohybrid DNA away from homohybrid DNA. A relatively complex and laborious approach was used (Nelson S. F., et al. Nature Genetics, 1993, 4:11-18; Cheung, V. G., et al. (1998) Nature Genetics, vol. 18, 224-230) that required multiple steps. A DNA methylase was used to methylate the DNA of one individual while not the methylating the DNA of the second individual. The DNAs were then mixed, denatured, and reannealed to from a mixture of heterohybrid and homohybrids DNA. This will result in hemimethylated DNA for heterohybrids that is resistant to certain restriction endonucleases. Homohybrid DNA, in contrast can be eliminated by digestion.

[0019] Some repair enzymes such as MutHLS have been used to detect the existence of mismatch-containing DNA fragments. (ref. Taylor G R, Deeble J. Genet Anal. 1999;14(5-6):181-6; Marra G, Schar P. Biochem. J. 1999;338 (Pt 1):1-13). DNA glycosylases have also been tested to detect DNA damage or mutation (Dennog C, et al. Mutat Res. 1999; 17;431(2):351-9; Gualillo O, et al. Vaughan P, et al. Genet Anal. 1999; 14(5-6):169-75).

[0020] DNA glycosylases have also been tested to detect DNA damage or mutation (Dennog C, et al. Mutat Res. 1999; 17;431(2):351-9; Gualillo O, et al. Vaughan P, et al. Genet Anal. 1999; 14(5-6):169-75). Most of the researches relaying on these MRS enzymes are focused on the detection of signal to noise ratio of the test sample comparing to control sample regarding one or a limit number of known SN Ps-containing fregmants, so as to determine whether or not any of these potential SNPs really exist, or to detect the mutation in one or a limited number of known genes. No report was found to apply this kind of enzyme to separate mismatch fragment pool from perfectly matched fragment pool.

SUMMARY OF THE INVENTION

[0021] The strategy involves several categories of new techniques:

[0022] Adapter and primer designs that allow specific amplification of either heterohybrid or homohybrid DNA. Heterohybrid is defined here as double-stranded DNA formed by annealing together complementary strands from two different individuals. For example, the DNA of two patients sharing a disease in common, can be mixed together, denatured by heating, and reannealed by cooling. In some cases, the two complementary DNA strands of one individual reanneal back together (referred to as a homohybrid). There is a need to separate the useful heterohybrid DNA form the homohybrids. Novel adapters are presented here that allow the specific amplification or capture of heterohybrids. The enriched heterohybrid DNA can then be used in polymorphism analysis. The degree of sequence polymorphism between the individuals is reflected as mismatched base pairs in the heterohybrid.

[0023] Sensitive and specific capturing of heterohybrids containing SNPs with DNA glycosylases. Glycosylases are DNA repair enzymes that bind and modify DNA containing mismatched base pairs. They can be used to separate DNA fragments with high polymorphism (resulting in base mispairs in heterohybrids or in homohybrids), from fragments with low polymorphism. The use of glycosylases in capturing DNA mismatches is an improvement over methods currently in use.

[0024] This invention is based on a parallel analysis on the products of the separately enriched SNP mismatch fragment or perfectly matched (PM) from heterohybrids or either one of the two homohybrids via DNA microarray hybridization, gel display, subtractive hybridization, FISH (fluorescent in situ hybridization), or other techniques. These heterohybrids and homohybrids are formed by mixing, denaturing and reannealling the DNA pools from the trait population and its control pool that are previously separately digested, and ligated to a pair of differentially designed heterohybrid-directed adapters (HeD adapter) and their derivations. The heterohybrids and homohybrids are separately recovered from one general pot of above mixture by the triple-recovery strategy (TRS), which is usually followed the separation of PM fragments from SNP fragments by DNA Glycosylase or other enzymes with similar functions. The gene mapping of the trait, or the SNPs or/and IBDs profiling of any given DNA pool will be obtained from the output analysis. (FIG. 1)

[0025] The DNA pool or DNA sample is the starting material for the current approach, which here refers to either genomic or cDNA or a collection of a number of genes from one individual or a population of many individuals (pooling strategy). For example, the pool of genomic DNA mixture from many patients sharing a disease or any trait in common, and its control pool from a number of control individuals without this disease or trait, can be separately digested by a restriction enzyme, ligated to differentially designed adapters. Then these two constructs are mixed together, denatured and reannealed. In some cases, two complementary DNA strands of one pool, either from the same original duplex matched or its homolouous chromosome of the same or different individual of the same pool, may reanneal together as a DNA duplex, which is referred to as a homohybrid. In other cases, two complementary strands from different DNA pools may anneal together and form another kind of DNA duplex, which is referred as a Heterohybrid. Obviously, both homohybrid and heterohybrid can be either perfect matched duplex or mismatched duplex, and most of the mismatches are produced from single nucleotide polymorphisms, called as SNP segment or SNP fragment (if digested by a restriction enzyme), including SNP homohybrid and SNP heterohybrid. DNA Glycosylases and similar enzymes are used to separate and enrich SNP mismatch fragments from perfectly matched fragments. The SNP segments between a trait DNA pool and its control, when quantitatively or qualitatively excluded the neutral SNPs within the control pool, are associated with the target traits. This exclusion or parallel analysis is important is important because most of human genetic polymorphisms exists within a population between rather than between two populations, and the “net” genetic polymorphisms that may be associated with a given human traits are very limited.

[0026] The enriched and separated perfectly matched fragments (PM segment or PM fragments, including PM homohybrids and PM heterohybrids)are taken as a result of heritage from a common ancestor and are referred as IBD segment or IBD fragment when they are derived from the same pool (i.e. PM homohybrids). It is certain that single base pair mismatches will be present in sufficiently large (ex. 2 kb-20 kb) DNA hybrids formed between allelic sequences that are not IBD. In other words, the perfectly matched fragments of hybrids from two related individuals, given it is long enough, can be regarded as IBD segments; but for a short fragment, the perfectly matched fragments of hybrids may be Identical-by-State (IBS segment or IBS fragment). The IBD fragments sharing among the different individuals with the same trait can be taken as a second information for the genetic mapping of the given trait in addition the SNP information. The IBD fragments or even the IBS fragments between the pool of trait and its control can also be quantitatively or qualitatively taken as a control information for genetic mapping of the trait. When SNP information is desired, a frequently cuffing restriction enzyme such as Sau3AI or a combination of a few of frequently cut enzymes is the choice; in this case, the IBS information may come out from the same one procedure as IBD. When IBD information is desired, a rare cutting restriction enzyme such as PstI is the choice, and genomic DNA is necessary; in this case, the SNP information are unavailable or very incomplete.

[0027] This invention involves additional categories of new techniques and a new strategy for analysis of the output fragment pools:

[0028] 1. A method for separating and amplifying either one of the two homohybrids or the heterohybrids from the mixture following by mixing two pools, denature and annealing.

[0029] Triple-Recovery-Strategy based on HeD (Heterohybroid-directed) adapters and their derivations allows specific amplification of either heterohybrids or one of the two homohybrids. Several strategies are given for this purpose.

[0030] 2. A method to separate the SNP fragments from PM fragments with high specificity and sensitivity by the employment of DNA glycosylase(s) and other enzymes with similar functions, and collect these two DNA fragment pools separately.

[0031] DNA Glycosylases are DNA repair enzymes that bind and modify DNA containing any one of 4 classes of single nucleotide mismatches (i.e. G/T and C/A, G/A and C/T, GIG and C/C, and T/T and A/A) with a high with high specificity and sensitivity. They modify the mismatched base(s) and result an abasic site, which can be tightly bound by these enzymes or other chemical or physical approach.

[0032] 3. A strategy of combining employment of the two approaches said above and other necessary designs for a parallel analysis and comparison of the output fragment pools following the treatment above, and obtaining the SNP information as well as identical sequences (IBDs or IBSs) information in terms of their nature, frequency, localization and finally the genetic and possible physical mapping of the target trait. The analysis techniques include parallel DNA microarray hybridization, gel display, subtractive hybridization, FISH (fluorescent in situ hybridization), or other techniques with the output fragments as the probes.

BRIEF DESCRIPTION OF THE DRAWINGS

[0033] 1. Selective recovery of AB in one pot by employment of a 3′ Exonuclease and Smart adaptor.Selective amplification of homohybrids in one pot by employment of a N. BstNB1 site in Smart adaptor.Use of DpnI to recover heterohybrids.Selective recovery of M, BB or AB in one pot based on DpnI site of selectively methylated LSA-adapter.Selective recovery of AA, BB or AB in one pot based on AluI site of selectively methylated LSA-adapter.Selective recovery of AA, BB or AB in one pot based on MboI/Sau3AI site of selectively methylated LSA-adapter.Selective recovery of AA, BB or AB in one pot with physical-chemical tool.

[0034] 8. General outline of the procedures of the invention.

[0035] 9. Basic HeD adapters: for recovery of heterohybrids for the mixture.

[0036] 10. Schematic for isolating particular DNA duplexes employing a 3′ Exonuclease with the HeD adapter.

[0037] 11. Schematic for isolating particular DNA duplexes employing an N. BstNB1 site in HeD adapters.

[0038] 12. Schematic for isolating particular DNA duplexes employing a 3′-exonuclease with DpnI site in the HeD adapter.

[0039] 13. Schematic for isolating particular DNA duplexes employing N. BstNB1 and Dpnl sites in HeD adapters.

[0040] 14. Schematic for isolating particular DNA duplexes employing 2 selectively methylated DpnI sites of adapters.

[0041] 15. Schematic for isolating particular DNA duplexes employing 2 selectively methylated TaqI sites of adapters.

[0042] 16. Schematic for isolating particular DNA duplexes employing 2 differentially methylated MboI/Sau3AI sites of adapters.

[0043] 17. Schematic for isolating particular DNA duplexes by physical-chemical moiety.

[0044] 18. Schematic for alternative HeD adapters.

DETAILED DESCRIPTION OF THE INVENTION

[0045] The goal of this invention is to enrich DNA sequences that are related to a disease or any other phenotype of interest. Restriction endonuclease fragments of DNA from two individuals or two pools of multiple individuals are mixed together. By melting and then reannealing it is possible to form a heterohybrid DNA double helix that contains the DNA strands of two different individuals. In some cases, the strands of DNA from a single individual may reanneal back together regenerating the original homohybrids. (“Homohybrid” is defined here to mean a double helix containing DNA strands derived from a single individual in contrast to a heterohybrid).

[0046] In the present invention, we propose the use of novel adapters that allow an improved method for selectively amplifying heterohybrid DNA without the need for methylation or digestion with restriction endonucleases. The adapters also have the advantage of resulting in an amplification of the desired heterohybrids rather than a simple enrichment away from homohybrids but with no net amplification of heterohybrid DNA.

[0047] Adapters may be Y shaped adapters, which are made up of two single-stranded DNA fragments. The two single-stranded DNA fragments of the Y shaped adapter have a region of complementarity at one end that anneals the two strands, and a region of non-complementary at the opposite end that does not anneal. These complementary and non-complementary ends give the adapters their Y shape.

[0048] The complementary end allows the adapters to ligate to the ends of other DNA fragments, such as restriction enzyme digested DNA. The complementary end may be blunt ended, or have a 5′ or 3′ overhang, depending on the DNA fragments to which the adapters are ligated. The complementary end may also be of any suitable length. The length of the complementary end can be 1, 2, 5, 10, 15, 20, 25, 30, 40, or 50 nucleotides.

[0049] The complementary end may also have features that promote the selective amplification or recovery of homohybrid or heterohybrid DNA. The complementary end of the Y shaped adapters may have adjacent methylation-sensitive restriction endonuclease recognition sites. The sites on the Y shaped adapters may be selectively methylated such that restriction endonucleases that are sensitive to DNA methylation can be used to cleave and specifically remove the homohybrids or heterohybrids. An example of a restriction enzyme that requires methylation of its recognition site to cleave DNA is DpnI. Restriction enzymes that will only cleave in the absence of methylation include MboI, MboII, ScrFI, DpnII and ApaI. AluI is also sensitive to the methylation status of DNA. Those of skill in the art will readily recognize other methylation sensitive restriction endonucleases.

[0050] The complementary end of the Y shaped adapter may also promote the selective amplification or recovery of homohybrid or heterohybrid DNA by possessing the site for a restriction endonuclease that cuts only one strand of a double stranded DNA molecule. An example of a restriction endonuclease that cleaves only one strand of a double-stranded DNA molecule is N.BstNB1. The recognition site for the restriction endonuclease can be between 1 and 50, 1 and 45, 1 and 40, 1 and 36, 1 and 30, 1 and 25, 1 and 20, 1 and 15, 1 and 10, and 1 and 5 nucleotides of the 3′ terminal sequence of the non-complementary end of the Y shaped adapter and is within the complementary end of the Y shaped adapter.

[0051] The non-complementary end of the Y shaped adapters comprises the two non-annealed strands of DNA, which are the strands with a 5′ terminal sequence and the 3′ terminal sequence. The sequences may be of any length. Typically the non-complementary 5′ terminal sequence is 1, 2, 5, 10, 15, 20, 25, 30, 40, or 50 nucleotides in length. Typically the non-complementary 3′ terminal sequence is 1, 2, 5, 10, 15, 20, 25, 30, 40, or 49 nucleotides in length. The 5′ terminal sequence and the 3′ terminal sequence may also be of different lengths. Preferably, the 5′ terminal sequence of the Y shaped adapter is longer than the 3′ terminal sequence of a Y shaped adapter.

[0052] The non-complementary ends of the Y shaped adapters also have unique features that allow the selective amplification or recovery of heterohybrid or homohybrid DNA molecules. A first Y shaped adapter with a 5′ terminal end that is longer than the 3′ terminal end may be annealed to a first DNA sample and a second Y shaped adapter that has a 5′ terminal end that is longer than the 3′ terminal end may be annealed to a second DNA sample. The 5′ terminal end of the first Y shaped adapter can be complementary to the 3′ terminal end of the second Y shaped adapter and the 5′ terminal end of the second Y shaped adapter can be complementary to the 3′ terminal end of the first Y shaped adapter. Heterohybrid DNA formed from DNA pools ligated to one or the other of these Y shaped adapters can be selectively recovered by extending the 3′ terminal ends and amplifying. Amplification may be carried out by any means known in the art, such as polymerase chain reaction or rolling circle amplification.

[0053] The non-complementary ends of the Y shaped adapters may also encode restriction enzyme sites for restriction enzymes that cleave double stranded DNA. These restriction enzyme cleavage sites are preferably in the 5′ terminal sequence of the Y shaped adapter.

[0054] Selective Amplification of Heterohybrids or Homohybrids Formed in a Single Annealing Reaction.

[0055] The first step of the invention is to selectively amplify the heterohybrids while eliminating the homohybrids. Two pools of patient DNA are formed, each pool comprising DNA of a different individual or a mixture from multiple individuals. (FIG. 1, pool 1, and pool 2). Pool 1 DNA is ligated to “Y shaped” adapters that are complementary to the ends of the restriction endonuclease digested DNA.

[0056] In this example, the 3′ recessed end of the Y adapter ends with the bases CC which form a mismatch with two corresponding Cs in the other strand (FIG. 1a). Pool 2 DNA is ligated to “Y shaped” adapters that contain a recessed 3′ GG end that is mismatched with two corresponding Gs in the other strand. The two patient DNA pools are mixed together, denatured and reannealed. Any homohybrid DNA formed will still contain either the CC or GG mismatch (FIG. 1b). The base mismatches prevent the recessed 3′ arm from being filled in by a DNA polymerase that lacks a 3′-5′ exonuclease proofreading activity such as Taq DNA polymerase (FIG. 1c). For any heterohybrid formed, the recessed 3′ arm will now be correctly base paired with the complementary strand (forming either CC base paired with GG, or GG base paired with CC). The heterohybrid recessed 3′ ends can be filled in by DNA polymerase and the four dNTPs (FIG. 1c bold lines). The method is not restricted to the 3′ CC ends in this example, but can be carried out with any combination of bases that form a recessed 3′ adapter end that is mismatched in homohybrids, and matched in heterohybrids. There are numerous ways to selectively amplify the heterohybrids once their recessed ends are filled in. For example, PCR primers can be designed that are only functional if the ends are filled in (FIG. 1d, primer 1 and 2). Therefore, only heterohybrids will be amplified while homohybrids will not anneal to the PCR primers.

[0057] In each of the examples shown (FIGS. 1-7), the goal is to be able to selectively separate 3 fractions that are formed in a single annealing reaction (referred to as a “single pot”). The 3 fractions are referred to as M, BB, and AB which represent the homohybrid obtained from one individual or DNA pool, the homohybrid obtained from the second individual or DNA pool, and the heterohybrids respectively.

[0058] An alternative strategy to that shown in FIG. 1 that allows recovery of homohybrids instead of heterohybrids is to fill in the ends with a DNA polymerase that possesses a 3′-5′ exonuclease proofreading activity. This will allow the mismatched 3′ ends of the homohybrids to be repaired and filled in. Suitable polymerases include Vent DNA polymerase, Pfu DNA polymerase, Klenow fragment of DNA polymerase 1, E. coli polymerase 1, and T4 DNA polymerase. The homohybrids can be distinguished form the heterohybrids by, for example, filling in the ends with normal nucleotides using the polymerase lacking proofreading activity, and then with modified nucleotides using the proofreading polymerase. Suitable polymerases include Taq polymerase, Tth DNA polymerase, eukaryotic DNA polymerase α, and 3′-5′ exo-minus Kienow DNA polymerase. For example, if the homohybrids are filled in with biotin containing nucleotides, they can be separated from heterohybrids with streptavidin or avidin attached to a bead of column matrix. If the homohybrids are filled in with nucleotides labeled with antigen, they can be separated from heterohybrids with an antibody specific for the antigen. If the homohybrids are filled in with nucleotides labeled with protein A, they can be separated from heterohybrids with an antibody. Other acceptable capture moieties and affinity reagents are known in the art. Also, filling in the ends of the homohybrids with methylated dATP, and in the case where an appropriate restriction endonuclease site such as Dpn1 is located on the adapter would allow the specific digestion of homohybrids and subsequent capture or amplification of the remaining heterohybrids.

[0059] An alternative way to distinguish heterohybrids from homohybrids is to use a restriction endonuclease capable of nicking only one strand of the DNA such as Bst NB1 (FIG. 2). The first step is the same as in FIG. 1. The 3′ recessed end of the heterohybrids are filled in by a DNA polymerase lacking proofreading activity. The homohybrids cannot be filled in due to the mismatched 3′ recessed ends. However, in this example, the ends are then cut with the restriction endonuclease generating a nick as indicated. It will be possible to heat the DNA to a temperature that causes the 3′ ends of the homohybrids to dissociate. Because the 3′ ends of the heterohybrids have be extended in the fill in reaction, they will remain annealed at an appropriate temperature. The temperature can be at any level that allows the dissociation of the unextended 3′ ends but not the extended ends. Treatment of the DNA with a ligase will close the nick in the heterohybrids. The homohybrids can be amplified by carrying out PCR with only primer 1 to obtain one homohybrid, or only primer 2 to obtain the other possible homohybrid (FIG. 2c). This distinguishes the homohybrids form heterohybrids that require both primer 1 and 2 for PCR (see FIG. 1d).

[0060] Another method uses adapters containing the recognition sequence for restriction endonucleases such as Dpn1 that require methylated bases (FIG. 3). The first steps are identical to FIGS. 1a-c with normal dNTPs being used to fill in the Y shaped adapters in a first 3′ end extension reaction. Bst NB1, or a similar enzyme that nicks a single strand, is then used to remove the mismatched 3′ end of the adapter as in FIG. 2. With the mismatched 3′ end now removed, a DNA polymerase can be used in a second extension reaction this time using methylated dATP. In this example, both homohybrids will be cut by DpnI which recognized the methylated A. Heterohybrids will not be cut and can therefore be selectively amplified or captured by methods described for FIG. 1.

[0061]
FIG. 4 shows another example using the recognition sequence for Dpn1, however this strategy will work for any restriction endonuclease that recognizes methylated bases. Only homohybrids will be digested by DpnI. Heterohybrids will not be digested. FIGS. 5 and 6 show additional strategies to obtain either homohybrids or heterohybrids from the same annealing reaction. In these examples, heterohybrids can be selectively amplified using both PCR primers and it can be used directly in DNA sequencing reactions. To recover either one of the possible homohybrids (for example, one from one patient and the other from another patient) the single PCR primer is used that corresponds to the adapter ligated to the sample.

[0062]
FIG. 7 demonstrates another approach to separating homohybrids and heterohybrids. Adapters with one chemical adduct (for example digoxin) are ligated to one DNA pool and adapters with a different adduct (for example, biotin) are ligated to the other pool. Only heterohybrids will contain both adducts allowing them to be differentiated from homohybrids by well-known separation techniques such as use of beads or columns.

[0063] Seven strategies based on the modification of LSA adaptor, Smart adapter or regular adapter are provided for selective recovery/amplification of AA, BB or AB (either one of the homohybrid or heterohybrid) from one pot. They are independent but they can be combined with each other or with other designs. The fragment pools recovered/enriched in this way are suitable to be taken as probes for further analysis such as applying into microarrays, FISH, and other SNP typing or gene mapping techniques. The recovered heteroduplex is also applicable to direct sequencing.

[0064] Because two primers are required for the recovery of the heteroduplex while only one primer corresponding to the original restriction fragment pool is required for the recovery of one source of homohybrid, the heterohybrid will not be recovered when one source of homohybrid is recovered. But when heterohybrid is desired, 2 primers are employed and the homohybrid adapters will not be amplified. The resultant constructs have no single stranded regions that may disturb the activities of an enzyme such as glycosylase.

[0065] The methods of the invention can also be used to estimate the extent of heterohybrid enrichment resulting from mixing, denaturing, and reannealing two or more DNA pools. In this method all the homohybrids and heterohybrids of two samples are amplified and are compared to all the heterohybrids of the two samples. Comparing may be performed by gel electrophoresis, spectrophotometry, or flourometry.

[0066] The goal of this invention is to enrich DNA sequences that are related to a disease or any other phenotype of interest. Methods and compositions are described for identifying, isolating and determining the genomic or cDNA positions where the sequences of two DNA samples or pools of DNA samples differ in the relative abundance of each of the four DNA bases. In most applications the two samples to be compared will be generally similar in sequence with variations occurring at less than one position per hundred residues. Methods include designs that make it possible to recover and further analyze DNA duplexes containing single or multiple sites of mismatched bases. Novel procedures introduced here include the enrichment for intact DNA fragments that contained mismatches or that contained perfect matches in such a way that the perfect match or mismatch containing duplexes can be amplified for subsequent analysis; the use of a special class of enzymes, the DNA glycosidases or the like for this purpose; the design of adaptors that permit selective amplification of either heterohybrids or homohybrids from a mixture of reannealed DNA fragments together with the ability to sequence each fragments using a common sequencing primer. In an alternative application, DNA heterohybrids that do not contain mismatches are recovered free of mismatch fragments, and used for mapping regions of identity between the two sources, or within each source.

[0067] The pools of DNA fragments, corresponding to mismatch SNP fragments or perfectly matched fragments of heterohybrids or homohybrids, may be divided into subsets by selective PCR amplification, and the fragments identified either by separating them on acrylamide sequencing gels and isolating and sequencing single fragments, or by annealing them to chromosome spreads or to arrays of cDNA or genomic DNA fragments and determining the nature and relative abundance of probe corresponding to homohybrids and to heterohybrids at each position in the array. The fragments in the array may represent cloned fragments of various sizes, such as are present in YAC, BAC, PAC, cosmid, or phage genomic libraries, or phage or plasmid cDNA libraries, or may represent PCR amplified fragments, corresponding, for example, to non-repetitive portions of cDNA or genomic DNA.

[0068] DNA samples to be analyzed may be derived from any source, including haploid, diploid, or polyploid genomes, and may include sources from prokaryotes or eukaryotes, including humans as well as model organisms or other mammalian or plant species or microorganisms, or the DNA may be derived from total cDNA, or cDNA libraries prepared form various sources such as different tissues or cell types, or tumors and normal cells of the cell type, including paired samples form the same individual. The DNA samples may be obtained from single individuals or clones of individual organisms, or may be prepared by pooling DNA form multiple individuals, for example from pools of individuals that share a common phenotypic trait or disease tendency. The two DNA sources to be compared will generally but not necessarily always share a large degree of sequence similarity. The two DNA “sources” to be in parallel analyzed can also be two aliquots of one original resource for the convenience to use some special designs to extract the internal mismatch segments of the DNA pool, or to map heterozygosity or homozygosity or Identical-by-Descents. DNA can be pooled form related or unrelated individuals, so that the procedure may be used either for mapping inheritance within families, such as homozygosity mapping, or for mapping allelic associations in unrelated individuals. The DNA sources may be of any complexity. But for highly complex sources DNA fragments may be divided into subsets prior to denaturation and reannealing, so as to obtain more favorable reannealing kinetics. Sometime the subdividing maybe following the denature and reannealing and the separation of perfectly matched DNA from mismatch DNA, for simplified treatment and for favorable resolution of amplification, and so on.

[0069] Selective Amplification of Heterohybrids or Either One of the Two Homohybrids Formed in a Single Annealing Reaction.

[0070] The first step of the invention is to selectively amplify the heterohybrids while eliminating the homohybrids by simply employ the HeD adapters. Two pools of DNA, or two subpools of one original pool are separately digested with a frequently cut restriction endonuclease and result 2 restricted fragment (RF) pools. Each pool comprising DNA from a single individual or a mixture from multiple individuals. (FIG. 9, pool 1, and pool 2). RF Pool-1 and RF pool-2 are separately ligated to a pair of HeD adapters that are cohesive to the restricted fragment ends. Like other adapter commonly used, HeD adapters are formed from the annealing of oligonucleotides together and can be ligated onto the ends of other DNA fragments via an appropriate blunt end or recessed 3′ or 5′ end. The cohesive ends have a necessary requirement that the 5′ end must be phosphorylated so both strand both ends can be covalent ligated to the DNA fragment. The other end of the HeD adapters of this invention, which will form the end of the adapter-DNA fragment construct following ligation, contains a region of non-complementary sequence so that the ends will be single stranded of different lengths. In this invention, the 5′ terminal sequence of the adapter-DNA fragment construct, is usually longer than the non-complementary 3′ terminal section.

[0071] In one example, the 3′ recessed end HeD adapter ends with the bases CC that forms a mismatch with the corresponding CC in the other strand, which is ligated to RF pool-1. (FIG. 9a). RF Pool-2 DNA is ligated to the other HeD adapter containing a recessed 3′ GG end that is mismatched with the corresponding GG in the other strand (FIG. 9a). These two HeD adapters pair together for heterohybrid-directed recovery and amplification, and share the same sequence in the adapter stem and the cohesive end ligated to RF pool. The two DNA construct pools are mixed together, denatured and reannealed. Any homohybrid DNA formed will still contain either the CC or GG mismatch (FIG. 9b). The base mismatches prevent the recessed 3′ arm from being filled in by a DNA polymerase that lacks a 3′-5′ exonuclease proofreading activity such as Taq DNA polymerase, Tth DNA polymerase, eukaryotic DNA polymerase α, and 3′-5′ exo-minus Klenow DNA polymerase. (FIG. 9c). For any heterohybrid formed, the recessed 3′ arm will now be correctly base paired with the complementary strand (forming either CC base paired with GG, or GG base paired with CC). The heterohybrid recessed 3′ ends can be filled in by the 3′-5′ exo-minus DNA polymerase and the four dNTPs (FIG. 9c bold lines). The method is not restricted to the 3′ CC ends in this example, but can be carried out with any combination of bases that form a recessed 3′ adapter end that is mismatched in homohybrids, and perfectly matched in heterohybrids. There are numerous ways to selectively amplify the heterohybrids once their recessed ends are filled in. For example, PCR primers can be designed that are only functional if the ends are filled in (FIG. 9d, primer 1 and 2). Therefore, only heterohybrids will be amplified while homohybrids will not anneal to the PCR primers.

[0072] In each of the examples shown below, referred as Triple-Recovery Strategy (TRS, FIGS. 10-17), the goal is to be able to selectively separate 3 fractions that are formed in a single annealing reaction (referred to as a “single pot”). The 3 fractions are referred to as M, BB, and AB which represent the homohybrid obtained from one DNA pool, the homohybrid obtained from the second DNA pool, and the heterohybrids respectively. The TRS allows a selective recovery of AA, BB or AB from one pot, which significantly reduces the treatment variation to its minimum and benefit the parallel comparison and analysis in the final step. One the other hand, the homohybrids (AA, or BB) can also be separately recovered/amplified from an independent DNA pool of denaturing and reannealing.of either one original RF DNA pool alone.

[0073] Beside the recovery of the heterohybrids so said above, an alternative strategy to shown in FIG. 10 allows recovery either one of the two homohybrids instead of heterohybrids is to fill in the ends with a DNA polymerase that possesses a 3′-5′ exonuclease proofreading activity (TRS-1, FIG. 10). This will allow the mismatched 3′ ends of the homohybrids to be repaired and filled in. Suitable polymerases include Vent DNA polymerase, Pfu DNA polymerase, Klenow fragment of DNA polymerase I, E. coli polymerase 1, and T4 DNA polymerase. The target homohybrids can be selectively recovered by the primer (one primer alone) sharing the original free 5′ terminal sequence of the adapter of the desired pool from the same one pot of DNA mixture, ex. using primer-1 sharing the original free 5′ terminal sequence of the first HeD adapter ligated to RF pool-1 to recover the homohybrids originally from pool-1, and using primer-2 to recover the homohybrids from pool-2. On the other hand, the heterohybrids of the same mixture can be recovered from a aliquot of the same mixture before the treatment of 3′-5′ exonuclease activity, as discussed above; in this case no homohybrid will be recovered.

[0074] An alternative way to distinguish heterohybrids from homohybrids is to use a restriction endonuclease capable of nicking only one strand of the DNA such as N. BstNBI (TRS-2, FIG. 11). The first step is the same as in FIG. 9. The 3′ recessed end of the heterohybrids are filled in by a DNA polymerase lacking proofreading activity, which can be optionally recovered by PCR with primer-1 plus primer-2. The heterohybrids can also be recovered together with the recovery of homohybrids after all treatment so as to reduce the variation resulted from different to its minimum (see below). The homohybrids cannot be filled in due to the mismatched 3′ recessed ends and will not be recovered. However, in this example, the ends are then cut with the restriction endonuclease generating a nick as indicated. It will be possible to heat the DNA to a temperature that causes the 3′ ends of the homohybrids to dissociate. Because the 3′ ends of the heterohybrids have been extended in the fill in reaction, they may remain annealed at an appropriate temperature. The temperature can be at any level that allows the dissociation of the unextended 3′ ends but not the extended ends. Treatment of the DNA with a ligase will close every possible nick in every internal fragment. The homohybrids can be amplified by carrying out PCR with only primer-1 to obtain one homohybrid, or only primer-2 to obtain the other possible homohybrid (FIG. 11). When the distal parts of the adapters of all the homohybrids in another aliquot of the mixture are cut off with DpnI, the heterohybrids can be selectively recovered and amplified from the with primer-1 and 2 via PCR.

[0075] When an appropriate restriction endonuclease site such as Dpn1 is located on the adapter but without N. BstNBI like site or enzyme, following the same treatment of FIGS. 10a-c, the ends of the homohybrids is then filled with methylated dATP and 3′-5′ exonuclease-containing DNA polymerase. In this case, one primer can specifically and selectively amplify the corresponding homohybrids. The key point is that this design allows the specific digestion of homohybrids by DpnI and subsequent amplification of the remaining heterohybrids with primer-1 plus primer-2. (TRS-3, FIG. 12). Any restriction endonuclease sensitive to methylation on both strands can be used to replace DpnI.

[0076] Another method uses adapters containing the recognition sequence for restriction endonucleases such as Dpn1 that require methylated bases, plus N. BstNBI for nicking single strand and filling homohybrids instead of 3′-5′ exonuclease-containing DNA polymerase. (TRS-4, FIG. 13). The first steps are identical to FIGS. 10a-c with normal dNTPs being used to fill in the HeD adapters in a first 3′ end extension reaction. N. BstNBI, or a similar enzyme that nicks a single strand, is then used to remove the mismatched 3′ end of the adapter as in FIG. 11. With the mismatched 3′ end now removed, a DNA polymerase can be used in a second extension reaction this time using methylated dATP. Homohybrids can be amplified with one corresponding primer alone. Then, both homohybrids can be cut by Dpn1 which recognized the methylated A containing GATC site. Heterohybrids will not be cut and can therefore be selectively amplified by primer-1 plus primer-2. This strategi is a combination of TRS-2 and TRS-3.

[0077]
FIG. 14 shows another strategy (TRS-5) by employment of a pair of modified HeD adaptors; each adaptopr has two differentially methylated DpnI sites After denaturing and reannealing, only homohybrids will be digested by DpnI, but Heterohybrids will not be digested. FIGS. 15 and 16 (TRS-6, and TRS-7) show alternative strategies to obtain either homohybrids or heterohybrids from the same annealing reaction. In these examples, heterohybrids can be selectively amplified using both PCR primers and it can be used directly in DNA sequencing reactions. To recover either one of the two possible homohybrids, the single PCR primer is used that corresponds to the adapter ligated to the DNA RF pool. Here, a pair of regular full double strand adapters work well for these design, but only HeD adapters/derivations are suitable for direct sequencing the heterohybrids, and only HeD adapters/derivations are suitable for selective recovery of either one of the two homohybrids from the mixture. In contrast to it, regular full duplex adapters are neither suitable for direct sequencing, nor separating one class of homohybrids from another class of homohybrids. Again, these strategies will work well for many other restriction endonuclease that recognizes methylated bases.

[0078] TRS-8 demonstrates another approach of chemical/physical separating homohybrids and heterohybrids (FIG. 17). Adapters with one chemical adduct (for example digoxin) are ligated to one DNA pool and adapters with a different adduct (for example, biotin) are ligated to the other pool. Only heterohybrids will contain both adducts allowing them to be differentiated from homohybrids by well known separation techniques such as use of beads or columns. Also, a pair of standard double strand adapters work well for these design, but only HeD adapters is suitable for direct sequencing the heterohybrids, and selective and separate recovery either one of the two homohybrids.

[0079] Alternative strategy for heterohybrid recovery and amplification is based on PCR-suppression effect (Diatchenko L, et al., Proc. Natl. Acad. Sci. USA, 93: 6025-6030, 1996; Matz M, et al, Nucl. Acids Res., 27: 6-8, 1999) and the alternative HeD adapter. The principles of PCR-suppression effect is that long inverted terminal repeats flanking DNA fragments can selectively suppress the polymerase chain amplification of these DNA fragments when a primer corresponding to the distal half of the repeat are used. We introduced this principle and modified the strategy for our purpose of selective recovery/amplification of the heterohybrids (FIG. 18, TRS-9). A pairs of adapters with different sequences are ligated to a pair of RF pools to be analyzed, or two aliquots of one original RF pools when it is required to extract the internal mismatch segments of the DNA pool, or to map heterozygosity or homozygosity or Identical-by-Descents. Each adapter has a long and a short oligonucleotides, of which the 5′ end of the long strand is to the outside of the constructs, and the 5′ end of the short oligonucleotides is onto the fragment end and unphosphorylated and so no covalent ligation occur when this adapter is ligated to DNA restriction fragment. The 5′ part of the long strand remains to be free single strand and the sequence of this free single strand part is long enough for PCR primer priming, usually longer than 20 bases. After mixing, denaturing, and reannealing the constructs of RF pool1 and RF pooll2, melting off the short strand of the adapter and filling the hybrids. When a pair of primers complementary to the two different newly extended 3′ ends are used for amplification via PCR or RCA, the heterohybrids are preferentially recovered and amplified, while the amplification of the homohybrids is suppressed because of the long inverted repeats flanking the two ends.

[0080] In the step of the recovery or amplification of the target hybrids for the strategies above, rolling cycle amplification (Fire A, et al. Proc. Natl. Acad. Sci. USA, 92:4641-4645, 1991; Lizardi P M, et al. Nat Genet. 19(3):225-32, 1998) may be used to replace PCR, whereby a splint oligonucleotides with two end sequences corresponding two end of the target constructs is used to circulating the target strands.

[0081] Several strategies based on the modification HeD adapter or regular adapter are provided for selective recovery/amplification of AA, BB or AB (either one of the homohybrid or heterohybrid) from one pot. They are independent but they can be combined with each other or with other designs.

[0082] Use of DNA glycosylases to Select Mismatched DNA.

[0083] A novel approach is also described here to use DNA glycosylases to carry out polymorphism analysis and to identify SNPs with heterohybrid or homohybrid DNA.

[0084] In the examples above, heterohybrids may be formed from two individuals or two pools of DNA that are themselves formed from the mixture of many individuals' DNA. In the example below, the use of individual patients' DNA and control DNA from an individual is used, however, the same principals apply to pools of DNA. Regions of low polymorphism that two patients share in common will tend to result in perfectly matched double stranded DNA, in a heterohybrid.

[0085] To identify SNPs in heterohybrid duplex DNA, the heterohybrids may be contacted with a DNA glycosylase to nick the heterohybrid DNA. DNA glycosylases include thymine glycosylase, E. coli mut Y protein, uracil-DNA glycosylase, MBD4, thymine-DNA glycosylase, 8-oxoguanine glycosylase, or N-methylpurine-DNA glycosylase. The glycosylase nicks the heterohybrid DNA, and the heterohybrid duplexes with nicks are identified as containing a SNP. The molecules containing steps can be separated from molecules not containing SNPs, for example by immobilizing or attaching the DNA glycosylase to a solid support. Solid supports include beads and resins used in chromatography, slides or chips.

[0086] Methods for binding a DNA glycosylase to mismatched DNA are also disclosed. DNA is bound to a DNA glycosylase by contacting a DNA sample with EDTA to withdraw Mg+2 from the DNA. EDTA may be contacted with the DNA sample and Mg+2 removed by techniques such as dialysis. The DNA sample is then contacted with a DNA glycosylase.

[0087] DNA that is bound to a DNA glycosylase is also released from the DNA glycosylase by methods of the invention. The DNA bound to the DNA glycosylase may be released by contacting with Mg+2.

[0088] Regions of polymorphism result in mismatched basepairs that can be distinguished by DNA glycosylases as described below. The procedures can also be repeated for two control individuals not expressing the phenotype of interest. This will produce a sample of control heterohybrid DNA that can be compared to the patient heterohybrids. The procedure can also be carried out to obtain heterohybrids containing one strand of patient DNA hybridized to a strand of control DNA. This will be used to determine polymorphisms between the patient population and the controls.

[0089] As we discussed before, both homohybrid and heterohybrid can be either perfect matched duplex or mismatched duplex. Separating and enriching of SNP fragments or PM fragment from a mixture with high enough specificity and sensitivity is a key issue in genetic analysis. This invention declares here that DNA Glycosylase is a useful tool for this purpose.

[0090] DNA glycosylase binds mismatched DNA with high specificity in an optimized condition with the cost that some SNP fragments may be together with PM fragment not being bound. Then the bound SNP fragments can be recovered. In this case, the SNP fragment is highly specifically separated from PM fragment and efficiently enriched. Alternatively another condition makes the PM fragment free from binding by DNA Glycosylase in a high specific mode with the cost that this enzyme may bound some PM fragments together with SNP fragments. In this case, the PM fragment is highly specifically separated from SNP fragment and efficiently enriched. The key point is to separate and enrich SNP fragment or PM fragment in separate steps and under different optimized conditions. An immobilized DNA Glycosylase in a fashion of any bead/resin column such as Glutathione-Sepharose agarose, His6 resin or any other solid matrix or free magnetic bead is presented here for specifically separate and enrich SNP fragments from PM fragments. Other fashions may also useful, for example, a recognition and binding reaction in a solution of the DNA Glycosylase with DNA fragments may be used to combine with gel shift, the binding and unbinding portion of DNA fragments can be separated and recovered.

[0091] In the procedure to separate mismatch fragments from perfectly matched fragments, the DNA samples are treated with one or more DNA glycosylases, most typically with human or methanobacterial thymidine glycosylase. The treatment is firstly performed ideally in a buffer without magnesium but with EDTA, under which condition the enzyme remains bound to the sugar aldehyde formed when the base is released. The enzyme is used in the form of a fusion protein that may be captured before or after reaction with the DNA. In a typical embodiment the enzyme is used as a fusion protein with glutathione transferase, and the fusion protein is bound to glutathione coated beads before reaction with the DNA. After binding DNA the beads containing the enzyme are washed several times to release non-specifically bound DNA, then DNA containing abasic sites is released by the addition of magnesium. The precise binding and washing conditions vary depending on whether the experiment is designed to maximize the purity of the mismatch containing, or alternatively the perfect match DNA fragments.

[0092] In an alternative method, in combining with the free DNA glycosylase recognizing and modifing the mismatched fragment and leaving an abasic (AP) site and an associated aldehyde in one or two of the duplex, a specific chemical in conjunction with a GST, His6, biotin or other groups, which is used to recover the SNP fragments in later steps, is employed here to trap the fragments with AP site(s)/aldehyde. An example of this kind of chemicals is the hydroxylamine or hydrazide derivative that forms a stable oxime bond with the open-chain aldehydes generated upon AP site formation, such as (but not limited to) FARP [5-(((2-(carbohydrazino)-methyl) thio) acetyl)-aminofluorescin, aminooxyacetyl hydrazid] (ref: Makrigiorgos G M. Int J Radiat Biol. 1998;74(1):99-109, Maulik G. Nucleic Acids Res. 1999; 1;27(5):1316-22). Before applying this procedures, the RE fragment pool is required to be treated with hydroxylamine to remove the possible spontaneously produced aldehydes.

[0093] Parallel Analysis of the Output Fragment Pools

[0094] In order to obtain the SNP information as well as IBD information in terms of their nature, frequency, localization and finally the genetic and possible physical mapping of the target trait, different output fragment pools are necessary to be analyzed in a comprehensive or parallel fashion. The possible analysis techniques include microarray hybridization, gel display, subtractive hybridization, FISH (fluorescent in situ hybridization), or other techniques. Two or more of the four output fragment pools, i.e., the PM homohybrids from the control pool, the SNP homohybrids from the control pool, the PM heterohybrids, and the SNP heterohybrids, may be analyzed parallel in one procedure/test.

[0095] Different methods can be used to identify the genomic origin or sequence of fragments that do or do not contain sequence variants in each pool, or between the two pools, depending on the research requirements. If only a small number of differences are expected, for example, in comparing tumor and normal tissue, or an inbred strain of mousse and its mutated offspring, then representational difference analysis (RDA) or another nucleic acid subtractive method may be used to subtract from the heterohybrid mismatch fragment pool (SNP heterohybrids) those fragments that are also represented in either homohybrid mismatch pool (SNP homohybrids). In another embodiment, in searching for fragments that are identical by descent (IBD) or for homozygosity mapping, the perfect match fragment pool (PM fragments) from a phenotypically marked group may be subtracted with driver from the mismatch fragment pool (SNP homohybrids) of the control DNA.

[0096] In another embodiment, the amplified fragments from the mismatch preparations within pool 1, those within pool 2, and those within the heterohybrid mismatch pool may be compared and searched for differences in various ways. This can be done by amplifying the fragments with radioactive primers and running fragments form the various pools on display gels in parallel, followed by autoradiography or phosphoimager analysis. Alternatively the fragments from each pool or from any pair of pools may be differentially labeled, either with radioisotopes with distinguishable emissions, or with fluors that emit light at different wavelengths. The labeled fragments may be annealed to arrayed cDNA pools, arrayed genomic fragments, or chromosome spreads, etc. and the presence and relative intensity of signal from each pool at each position on the arty quantitatively recorded.

[0097] For example, SNP heterohybrids between the DNA pool of the trait population and its control can be labeled as one color (ex. red) of probes, the SNP homohybrids within its control pool or the PM heterohybrids can be labeled as the different color (ex. green) of probes. These two complementary probe pools are hybridized to a set of DNA arrays, then the trait associated SNPs will show their nature, localization as well as the frequency information.

[0098] This invention differs from GMS approach in many aspects. First, It screens all possible potential related SNPs of the entire DNA pool analyzed, which is supposed to be the best genetic mapping marker; if desired, it also combining uses identical sequences (IBDs, or sometimes IBSs if desired) for dual-signals mapping of the same trait, which makes it more powerful. In GMS, IBDs are the only information it may extract out for the mapping analysis. The information for the regions of heterozygosity in the given genome may be able to extract for a given individual by GMS (U.S. Pat. No. 5,376,526), but most of the real SNP information is actually unavailable because GMS uses MutHLS system to recover the PM fragment. And the MutHLS system requires a GATC site for its activity, and only long enough DNA segment (2-20 kb) may contain this site and so the non-IBD fragment almost always contains one or more SNP site that makes SNP information unavailable. Second, a DNA pool from many individuals (pooling strategy), besides a DNA pool from an individual, can be used for one procedure analysis in the invention presented here, because GMS uses the DNA from only one individual as a DNA pool but not any of the pooling strategy. Third, either one DNA pool alone or a trait DNA pool combining with its control pool can be the right starting materials for this invention. On the other hand, GMS uses two related individuals, ex. an affected-relative-pair, or two other genomic that is supposed to contain IBD sequences, and extracts the IBD information between them, so no control information is obtained. Four, cDNA, or genomic DNA or a collection of a number of genes can be used for the analysis of this invention. In GMS, only genomic DNA can be the right sample because it is target on genomic IBD sequences. Five, a simple design of HeD adapters is designed in this invention to recovery heterohybrids, and its derivations as Triple-Recovery-Strategy (TRS) can be used to recovery either one of the homohybrids besides the heterohybrids from one pot of mixture for a parallel analysis. GMS uses a combination of a restriction methylase and methylation-sensitive endonuclease to recover the heterohybrids, which is more complex and laborious and also the homohybrids is unable to be recovered from the same pot. Six, DNA glycosylase is much more specific and sensitive than MutHLS in term of the recognition discrimination between PM and SNP fragments in vitro. Seven, an immobilized enzyme approach is designed for a convenient and efficient recover the desired fragment pools outflow.

[0099] This invention differs from the current gene cloning strategies, such as functional cloning and positional cloning, or candidate gene approach, in that it does not require any prior information of the disease or trait related protein, nor the position of the gene. It directly and greatly amplifies/captures any strong or weak association/linkage of the SNPs throughout whole genome or cDNA pool.

[0100] It is also differs from the other SNPs screening or typing approaches available currently, in that this approach simultaneously screens every and any SNPs from the entire genomic DNA/cDNA pool/subpool without any prior information. It highly specifically captures/enriches the SNP-containing fragments related to a traits for further analysis such as allele nature as well as quantitative frequency analysis, while excluding the unrelated or neutral SNPs or taking them aside as the controls for further analysis.

[0101] Again, this approach can, in the same one procedure, distinguish the SNPs within one pool from the SNPs between 2 pools, so that it greatly reduces the background of the gene mapping, and makes the results easy to interpret. In addition, this invention also provides an improved IBD screening technique, and a method for selectively amplifying a given DNA resource from a mixture DNA pool.

[0102] This invention provides a highly efficient mapping approach for any complex trait disease as well as simple Mendelian diseases or somatic mutation or drug response variation, in an outbreed as well as an inbreed population. It can be used as a highly efficient approach for screening/discovery of the SNPs in a entire, or a part of, complex genome/cDNA pool, enabling a rapid establishment of a complete SNP genetic map of the whole genome/cDNA pools and fast and efficiently evaluating the candidate genes. It can also be used for highly efficient profiling/sorting of the DNA/cDNA SNPs of any sample pair such as a chemically, physically or biologically treated sample and its untreated control.

[0103] This approach has the advantages of high sensitivity, high specificity, high-throughput, high efficiency, and high-compatibility with other related techniques.

[0104] Experiment 1

[0105] Test to separate mismatched from matched DNA duplexes. A perfectly matched duplex formed with synthetic 60 nt oligonucleotides (PM1 annealed to PM2) and a 76 bp duplex containing a G/T mismatch (MM1 annealed to MM2) were mixed together as the starting template mixture. These were labeled on the 5′ end with 32P using PNK obtained from New England Biolabs using the manufacturers suggested protocol. These were loaded onto a 0.1 ml column of immobilized hTDG or mTDG. With the steps of binding, washing and eluting, the 76 bp and 60 bp fragments were separated at different optimized conditions: the perfectly matched fragment passed through the column and were further specifically washed off (4 steps of washing). The mismatch fragment was retained in the column, and then eluted with two 1 ml washes. Fractions from different steps were collected, purified, and loaded onto an 8% sequencing gel. The concentration of matched or mismatched duplex in each fraction was determined by the level of 32P radioactivity determined with a phosphoimager. The PM duplex and Mismatched duplex was separately enriched by ≧500-fold. That is, PM duplex recovered in the flow through fraction had less than one part in 500 contamination with mismatched duplex, and mismatched duplex recovered in the flow through fraction had less than one part in 500 contamination with PM duplex.

[0106] Following the selective removal or enrichment of DNA containing basepair mismatches or having no mismatches, the novel adapters described above can be used to selectively amplify or recover the heterohybrids or either possible homohybrid for further analysis. For example, the output amplicons can be analyzed by differential display, used to make DNA probes, or used in DNA sequencing reactions. The novel adapters can also be employed in many other ways to achieve separation of heterohybrids and homohybrids and are not limited to the use of PCR or physical capture of chemical moieties, such as biotin, rhodamine, or FITC. For example, a person skilled in the art could use the adapters to initiate a rolling circle amplification reaction (Lizardi, P. M., (1998) Nature Genetics, vol 19, 225).

[0107] Our method is superior to an earlier method employing MutSLH-to eliminate homohybrids and to eliminate mismatched DNA (Nelson S F et al. Nature Genetics, 1993, 4:11-18). The Mut SLH method can only be used for DNA fragments containing the MutSLH recognition sequence (GATC) and requires the extra step of DNA methylation in order for MutSLH to be functional.

[0108] Methods

[0109] Glycosylases:

[0110] hTDG (human G/T mismatch-specific Thymine-DNA glycosylase) was cloned in a modified version from the mRNA of human JY lymphoid cell line based on the sequences reported. mTDG (the archaeon Methanobacterium thermoautotropicum DNA mismatch N-glycosylase (Mig.Mth)) was cloned in a modified version from the DNA of the archaeon Methanobacterium thermoautotropicum. Both were highly expressed in E. Coli, purified to homogeneity and also separately immobilized onto Glutathione Sepharose 4B beads (Amersham Pharmacia Biotech AB).

1Oligonucleotides PM1(60): CCG TGG TAC TTA CAT CGA GAG ATC CG*C TTG GTG TGG ATTCAC GCA CGT AGA CCG ATT CCT PM2(60) AGG AAT CGG TCT ACG TGC GTG AAT CCA CAC CAA GC*G GATCTC TCG ATG TAA GTA CCA CGG MM1(76T) GATCCGTCGACCTGCACCGTGGTACTTACATCGAGAGAATAGCTTGGTGTGGAT.TCACGCACGT.AGACCGATTC.CT MM2(76G) GATC.AGGAATCGGT.CTACGTGCGT.GAATCCACAC.CGAGCTATTC.TCTCGATGTA.AGTACCACGG.TGCAGGTCGA.CG

[0111] Binding buffer:

[0112] Hepes pH7.5, KCl 50 mM, EDTA 5 mM, ZnCl 0.2 mM, DTT 1 mM, BSA 0.25 mg/ml at 37 C. for 2 hours.

[0113] Washing buffer:

[0114] Hepes pH7.5, KCl 50 mM, EDTA 5 mM, ZnCl 0.2 mM, DTT 1 mM, BSA 0.25 mg/ml with NaCl in concentrations of 50 mM, 100 mM, 150 mM, and 200 mM at 37 C. for 2 hours.

[0115] Elution buffer1

[0116] Hepes pH7.5, KCl 50 mM, MgCl2 5 mM, ZnCl 0.2 mM, DTT 1 mM, BSA 0.25 mg/ml at 37 C. for 2 hours.

[0117] Elution buffer1

[0118] Hepes pH 8.0, Glutathione 15 mM, NaCl 100 mM at 37 C. for 2 hours.

Claims

1. A method to selectively amplify or recover heterohybrid DNA molecules resulting from mixing, denaturating, and reannealing two or more DNA pools, wherein each of the pools comprises DNA from a different individual or a mixture of DNA from multiple individuals comprising: ligating first Y-shaped adapters onto a first pool of restriction endonuclease digested DNA molecules, wherein the first Y-shaped adapters comprise a complementary and a non-complementary end, said non-complementary end comprising a 5′ terminal sequence of a first DNA strand that is non-complementary to a 3′ terminal sequence of a second DNA strand, and wherein the 5′-terminal sequence is longer than said 3′-terminal sequence; ligating second Y-shaped adapters onto a second pool of restriction endonuclease digested DNA molecules, wherein the second Y-shaped adapters comprise a complementary and a non-complementary end, said non-complementary end comprising a 5′ terminal sequence of a first DNA strand that is non-complementary to a 3′ terminal sequence of a second DNA strand, wherein the 5′ terminal sequence is longer than the 3′ terminal sequence, wherein said 5′-terminal sequence of the second Y-shaped adapters is longer than said 3′-terminal sequence of the second Y-shaped adapters, and wherein the 3′ terminal sequence of the first Y-shaped adapters is complementary to said 5′ terminal sequence of the second Y-shaped adapters and the 3′ terminal sequence of said second Y-shaped adapters is complementary to said 5′ terminal sequence of said first Y-shaped adapters; mixing, denaturing, and reannealing the first pool ligated to the first adapter with the second pool ligated to the second adapter to form a mixture containing heterohybrids and homohybrids; extending complementary 3′ terminal nucleotides of the heterohybrids using a DNA polymerase that lacks a 3′-5′ exonuclease activity to form new 3′ ends on the heterohybrids; amplifying the heterohybrids in the mixture using first and second primers complementary to the new 3′ ends, whereby the heterohybrids are selectively enriched in the mixture containing heterohybrids and homohybrids.
2. The method of claim 1 wherein the step of amplifying is carried out by rolling circle amplification, wherein one or both of the new 3′ ends serve as a primer.
3. The method of claim 1 wherein the step of amplifying is carried out by polymerase chain reaction.
4. The method of claim 1 wherein the non-complementary 5′ terminal sequence of the first and second Y-shaped adapters is between 2 and 50 nucleotides and the non-complementary 3′ terminal sequence of the first and second Y-shaped adapters is between 1 and 49 nucleotides.
5. The method of claim 1 wherein the complementary end of the first and second Y-shaped adapters contain adjacent first and second methylation-sensitive restriction endonuclease recognition sites, wherein the first site is methylated on both strands on the first adapter and unmethylated on both strands on the second adapter, and wherein the second site on the first adapter is unmethylated on both strands on the first adapter but methylated on both strands on the second adapter, whereby upon formation of the heterohybrids and the homohybrids, the homohybrids are susceptible to digestion with a restriction endonuclease that recognizes the methylation-sensitive restriction endonuclease recognition sites and the heterohybrids are not susceptible to digestion with the restriction endonuclease.
6. The method of claim 5 wherein the restriction endonuclease is MboI.
7. The method of claim 5 wherein the restriction endonuclease is AluI.
8. The method of claim 5 wherein the restriction endonuclease is DpnI.
9. A method to selectively amplify or recover homohybrid or heterohybrid DNA resulting from mixing, denaturating, and reannealing two or more DNA pools, wherein each pool comprises the DNA of a different individual or a mixture of DNA from multiple individuals, comprising: ligating first Y shaped adapters onto a first pool of restriction endonuclease digested DNA molecules, wherein the first Y shaped adapters comprise a complementary and a non-complementary end, said non-complementary end comprising a 5′ terminal sequence of a first DNA strand that is non-complementary to a 3′ terminal sequence of a second DNA strand, wherein the 5′ terminal sequence is longer than the 3′ terminal sequence, wherein the first Y shaped adapters contain a recognition site for a restriction endonuclease in the non-complementary 5′ terminal sequence, which restriction endonuclease only cuts double stranded DNA; ligating second Y shaped adapters onto a second pool of restriction endonuclease digested DNA molecules, wherein the second Y shaped adapters comprise a complementary and a non-complementary end, said non-complementary end comprising a 5′ terminal sequence of a first DNA strand that is non-complementary to a 3′ terminal sequence of a second DNA strand, wherein the 5′ terminal sequence is longer than the 3′ terminal sequence, wherein the second Y shaped adapters contain the recognition site for the restriction endonuclease in the 5′ terminal sequence which only cuts double stranded DNA, wherein the 3′ terminal sequence of the first Y-shaped adapter is complementary to the 5′ terminal sequence of the second Y-shaped adapter and the 3′ terminal sequence of the second Y-shaped adapter is complementary to the 5′ terminal sequence of the first Y-shaped adapter; mixing, denaturing and reannealing the first pool ligated to the first adapter with the second pool ligated to the second adapter to form a mixture containing heterohybrids and homohybrids; extending complementary 3′ terminal nucleotides of the heterohybrids using a DNA polymerase that lacks a 3′-5′ exonuclease activity to form new 3′ ends on the heterohybrids; contacting the restriction endonuclease with the mixture of heterohybrids and homohybrids, whereby only heterohybrids are cleaved; recovering said heterohybrids or homohybrids from the mixture of homohybrids and heterohybrids.
10. A method to selectively amplify or recover heterohybrid DNA resulting from mixing, denaturing, and reannealing two or more DNA pools, wherein each of said pools comprises the DNA of a different individual or DNA from multiple individuals, comprising: ligating first Y shaped adapters onto a first pool of restriction endonuclease digested DNA molecules, wherein the first Y shaped adapters comprise a complementary and a non-complementary end, said non-complementary end comprising a 5′ terminal sequence of a first DNA strand that is non-complementary to a 3′ terminal sequence of a second DNA strand, wherein the 5′ terminal sequence is longer than the 3′ terminal sequence; ligating second Y shaped adapters onto a second pool of restriction endonuclease digested DNA molecules, wherein the second Y shaped adapters comprise a complementary and a non-complementary end, said non-complementary end comprising a 5′ terminal sequence of a first DNA strand that is non-complementary to a 3′ terminal sequence of a second DNA strand, wherein the 5′ terminal sequence is longer than the 3′ terminal sequence, wherein the 3′ terminal sequence of the first Y shaped adapter is complementary to the 5′ terminal sequence of the second Y shaped adapter and the 3′ terminal sequence of the second Y shaped adapter is complementary to the 5′ terminal sequence of the first Y shaped adapter; mixing, denaturing and reannealing the first pool ligated to the first adapter with the second pool ligated to the second adapter to form a mixture containing heterohybrids and homohybrids; extending complementary 3′ terminal nucleotides of the heterohybrids using a DNA polymerase that lacks a 3′-5′ exonuclease activity and using deoxyribonucleotide triphosphates which are labeled with a capture moiety; recovering the heterohybrids by contacting the mixture of heterohybrids and homohybrids with an affinity reagent for the capture moiety, whereby the heterohybrids are selectively captured.
11. The method of claim 10 wherein the capture moiety is biotin and the affinity reagent is avidin or strepavidin.
12. The method of claim 10 wherein the capture moiety is an antigen and the affinity reagent is an antibody specific for the antigen.
13. A method of estimating the extent of heterohybrid enrichment resulting from mixing, denaturating, and reannealing two or more DNA pools, wherein each of the pools comprises DNA of a different individual or a mixture of DNA from multiple individuals, comprising: ligating first Y shaped adapters onto a first pool of restriction endonuclease digested DNA molecules, wherein the first Y shaped adapters comprise a complementary and a non-complementary end, said non-complementary end comprising a 5′ terminal sequence of a first DNA strand that is non-complementary to a 3′ terminal sequence of a second DNA strand, wherein the 5′ terminal sequence is longer than the 3′ terminal sequence; ligating second Y shaped adapters onto a second pool of restriction endonuclease digested DNA molecules, wherein the second Y shaped adapters comprise a complementary and a non-complementary end, said non-complementary end comprising a 5′ terminal sequence of a first DNA strand that is non-complementary to a 3′ terminal sequence of a second DNA strand, wherein the 5′ terminal sequence is longer than the 3′ terminal sequence, wherein the 3′ terminal sequence of the first Y shaped adapter is complementary to the 5′ terminal sequence of the second Y shaped adapter and the 3′ terminal sequence of the second Y shaped adapter is complementary to the 5′ terminal sequence of the first Y shaped adapter; mixing, denaturing and reannealing the first pool ligated to the first adapter with said second pool ligated to the second adapter to form a mixture containing heterohybrids and homohybrids; dividing the mixture containing the heterohybrids and homohybrids into first and second sample portions; treating the homohybrids and the heterohybrids of the first sample portion with a DNA polymerase that has 3′-5′ exonuclease activity to extend the complementary 3′ terminal nucleotides of the heterohybrids and to repair and extend the 3′ terminal sequence of the non-complementary end of the homohybrids; treating the homohybrids and the heterohybrids of the second sample portion with a DNA polymerase that lacks 3′-5′ exonuclease activity to extend the complementary 3′ terminal nucleotides of the heterohybrids and form new 3′ ends; amplifying the homohybrids and the heterohybrids of the first sample portion using first and second primers complementary to each of the non-complementary 5′ ends of the first and the second Y shaped adapters; amplifying the heterohybrids of the second sample portion using the first and second primers; comparing the amplified homohybrids and heterohybrids obtained in the first sample portion to the amplified heterohybrids obtained in the second sample portion, whereby an estimate of the extent of heterohybrid enrichment can be obtained.
14. A method to selectively amplify or recover homohybrid DNA resulting from mixing, denaturating, and reannealing two or more DNA pools, wherein each of the pools comprises DNA of different individual or DNA of a mixture from multiple individuals, comprising: ligating first Y shaped adapters onto a first pool of restriction endonuclease digested DNA molecules, wherein the first Y shaped adapters comprise a complementary and a non-complementary end, said non-complementary end comprising a 5′ terminal sequence of a first DNA strand that is non-complementary to a 3′ terminal sequence of a second DNA strand, wherein the 5′ terminal sequence is longer than the 3′ terminal sequence, wherein the first Y shaped adapter further comprises a recognition site for a restriction endonuclease that cuts only one strand of a double stranded DNA molecule; ligating second Y shaped adapters onto a second pool of restriction endonuclease digested DNA molecules, wherein the second Y shaped adapters comprise a complementary and a non-complementary end, said non-complementary end comprising a 5′ terminal sequence of a first DNA strand that is non-complementary to a 3′ terminal sequence of a second DNA strand, wherein the 5′ terminal sequence is longer than the 3′ terminal sequence, wherein the second Y shaped adapter further comprises the recognition site for the restriction endonuclease, wherein the 3′ terminal sequence of the first Y shaped adapter is complementary to the 5′ terminal sequence of the second Y shaped adapter and the 3′ terminal sequence of the second Y shaped adapter is complementary to the 5′ terminal sequence of the first Y shaped adapter; mixing, denaturing and reannealing the first pool ligated to the first adapter with the second pool ligated to the second adapter to form a mixture containing heterohybrids and homohybrids; extending complementary 3′ terminal nucleotides of the heterohybrids using a DNA polymerase that lacks a 3′-5′ exonuclease activity to form new 3′ ends on the heterohybrids; contacting the restriction endonuclease with the mixture of heterohybrids and homohybrids under conditions where the restriction endoclease makes a single stranded nick; heating the mixture of heterohybrids and homohybrids to a temperature which will denature the 3′ terminal sequences on the homohybrids but which will not denature the extended 3′ terminal sequences on the heterohybrids; amplifying the homohybrid DNA by polymerase chain reaction employing a first primer with the same nucleotide sequence as the non-complementary 5′ terminal sequence of the first Y shaped adapters or a second primer with the same nucleotide sequence as the non-complementary 5′ terminal sequence of the second Y shaped adapters.
15. The method of claim 14 wherein the restriction endonuclease is N.BstNB1.
16. The method of claim 14 wherein the recognition site for the endonuclease is between 1 and 36 nucleotides of the 3′ terminal sequence of the non-complementary end of the first or second Y-shaped adapters and is within the complementary end of said first or second Y-shaped adapters.
17. The method of claim 14 wherein the recognition site for the endonuclease is between 1 and 15 nucleotides of the 3′ terminal sequence of the non-complementary end of the first or second Y-shaped adapters and is within the complementary end of said first or second Y-shaped adapters.
18. A method for identifying SNPs comprising: forming a heterohybrid duplex DNA molecule, wherein each strand of the heterohybrid duplex DNA is from a different individual or a different allele; contacting the heterohybrid duplex DNA with a DNA glycosylase to nick the heterohybrid duplex DNA, whereby heterohybrid duplexes with nicks are identified as containing a SNP.
19. A method of enriching a population of molecules for those containing SNPs comprising: forming a heterohybrid duplex DNA molecule, wherein each strand of the heterohybrid duplex DNA molecule is from a different individual or a different allele; contacting the heterohybrid duplex with a DNA glycosylase under conditions where said enzyme binds to mismatched base pairs but does not cleave them; separating heterohybrid duplexes into a first fraction which is bound to DNA glycosylase and a second fraction which is not bound, thereby enriching the first fraction for heterohybrid duplexes which contain SNPs.
20. The method of claim 18 or 19 further comprising the step of selectively amplifying the heterohybrid duplexes using Y-shaped adapters.
21. The method of claim 19 wherein said step of contacting is done in the presence of EDTA.
22. The method of claim 19 wherein said DNA glycosylase is attached to a solid matrix.
23. The method of claim 20 wherein the DNA glycosylase is attached to a solid matrix.
24. The method of of claims 18, 19, or 21 wherein the DNA glycosylase is thymine glycosylase.
25. The method of claim 20 wherein the DNA glycosylase is thymine glycosylase.
26. The method of claim 22 wherein the DNA glycosylase is thymine glycosylase.
27. The method of any of claims 18, 19, or 21 wherein the DNA glycosylase is E. coli mut Y protein.
28. The method of claim 20 wherein the DNA glycosylase is E. coli mut Y protein.
29. The method of claim 22 wherein the DNA glycosylase is E. coli mut Y protein.
30. A method for binding a DNA glycosylase to mismatched DNA, comprising: contacting a DNA sample with EDTA to withdraw Mg+2 from the DNA; contacting the DNA sample with a DNA glycosylase, whereby the DNA glycosylase binds to mismatched base pairs in the DNA sample.
31. The method of claim 30 wherein the DNA glycosylase is thymine glycosylase.
32. A method for releasing DNA bound to a DNA glycosylase comprising: contacting DNA bound to a DNA glycosylase with Mg+2, whereby the DNA is released from the DNA glycosylase.
33. The method of claim 32 wherein the DNA glycosylase is thymine glycosylase.
34. A method for quantitatively profiling the single nucleotide polymorphisms (SNPs) within a DNA pool, or between 2 DNA pools, and quantitatively profiling the perfectly matched (PM) segments within a DNA pool. Each pool comprises the DNA of an individual or a mixture from multiple individuals, and the DNA may be from genomic DNA or cDNA or a collection of multiple genes or clones. It comprising: digesting 2 DNA pools with a certain restriction endonuclease (RE) to obtain their corresponding restriction fragment (RF) pools; ligating separately the pair of RF pools to a pair of differentially designed adapters to form the first pool of RF-adapter constructs and the second pool of RF-adapter constructs; mixing together, denaturing, and reannealing the said first pool of RF-adapter constructs with said second pool of RF-adapter constructs to form a mixture containing heterohybrids and homohybrids, both of heterohybrids and homohybrids containing PM fragments and mismatch (MM) fragments as well; Separating and enriching and collecting separately the PM fragments from mismatch (MM) fragments from the mixture; Selectively recovering either heterohybrids or one of the homohybrids, and separately collect the fragment pools of MM heterohybrids, PM heterohybrids, MM homohybrids of pool-1, PM homohybrids of pool-1, MM homohybrids of pool-2 and PM homohybrids of pool-2; Parallel analyzing a combination of two or more of above fragment pool to extract the information of the sequence nature, frequencies, and localizations in the genome.
35. A method according to claim 34 where the said adapters ligated to DNA RF pools are the heterohybrid-directed (HeD) adapters or their derivations. A set of HeD adapters comprising 2 adapters, wherein each adapter comprising a pair of annealed strands of oligonucleotides, and each comprising a complementary and a non-complementary section, wherein the said the complementary section comprising an appropriate blunt end or cohesive end with recessed 3′ or 5′ end for ligation onto the ends of DNA RFs, wherein the 5′ end of the said complementary section is phosphorylated, wherein the said non-complementary section comprising a 5′ terminal sequence of a first DNA strand that is non-complementary to a 3′ terminal sequence of a second DNA strand, wherein the 5′-terminal sequence of the said non-complementary section is longer than said 3′-terminal sequence, wherein the complementary section of the said first and second adapters share their sequences, wherein the said non-complementary 5′ terminal nucleotides of the said first and second adapter are different with the sequence Alpha and sequence Beta, wherein the said non-complementary 3′ terminal nucleotides of the said first adapter complementary to the said 5′ terminal nucleotides immediately close to the complementary section of the said second adapter, and the said non-complementary 3′ terminal nucleotides of the said second adapter complementary to the said 5′ terminal nucleotides immediately close to the complementary section of the said first adapter.
36. A method wherein a pair of HeD adapters according to claim 35 is used for selective recovery of heterohybrids or either one of the homohybrids resulting from the mixing, denaturation and reannealing of two DNA pools, comprising: digesting 2 DNA pools with a certain restriction endonuclease (RE) to obtain their corresponding restriction fragment (RF) pools: RF pool 1 and RF pool 2; ligating separately the pair of RF pools to a pair of HeD adapters to form the first and the second pools of RF-adapter constructs; mixing together, denaturing, and reannealing the said first pool of RF-adapter constructs with the said second pool of RF-adapter constructs to form a mixture containing heterohybrids and homohybrids; filling in completely the 3′ end of the heterohybrid constructs in the mixture with a 3′-5′ exonuclease minus DNA polymerase, such as but not limited to Taq DNA polymerase, Tth DNA polymerase, eukaryotic DNA polymerase α, and 3′-5′ exo-minus Klenow DNA polymerase, plus 4 deoxyribonucleotide triphosphates. then dividing this mixture into 2 aliquots. recovering and amplifying selectively of the filled in constructs, i.e. the heterohybrids, from the first aliquot with a pair of primers complementary to the two newly extended 3′ ends (i.e., sharing their sequences separately with the sequence Alpha and Beta); using a 3′-5′ exonuclease plus DNA polymerase such as but not limited to Vent DNA polymerase, Pfu DNA polymerase, Klenow fragment of DNA polymerase 1, E. coli polymerase 1, and T4 DNA polymerase, plus 4 deoxyribonucleotide triphosphates, to fill in the 3′ end of the constructs of another aliquot to fill-in both heterohybrids and homohybrids; recovering selectively the target homohybrids by the corresponding one primer from the second of aliquot of DNA mixture, ex. using primer-1 complementary to the adapter sequence Alpha to recover the homohybrids originally from RF pool-1, and using primer-2 complementary to the adapter sequence Beta to recover the homohybrids originally from RF pool-2,
37. A method wherein a pair of modified HeD adapters based on claim 35 and 36 is used for selective recovery of heterohybrids or either one of the homohybrids resulting from the mixing, denaturation and reannealing of two DNA pools, comprising: digesting 2 DNA pools with a certain restriction endonuclease (RE) to obtain their corresponding restriction fragment (RF) pools: RF pool-1 and RF pool 2; ligating separately the pair of RF pools to a pair of modified HeD adapters to form the first and the second pools of RF-adapter constructs, wherein the modified HeD adapters contain a recognition site for a nicking restriction endonuclease that is capable of cutting only one strand of a double stranded DNA molecule, and the site usually is designed at the stem of adapter and to cut off the 3′ terminal section of the adapter; mixing together, denaturing, and reannealing the said first pool of RF-adapter constructs with the said second pool of RF-adapter constructs to form a mixture containing heterohybrids and homohybrids; filling in completely the 3′ end of the constructs with a 3′-5′ exonuclease minus DNA polymerase to selectively fill-in only heterohybrids. Then dividing the mixture into 2 aliquots. recovering and amplifying selectively of the filled in constructs, i.e., the heterohybrids, from the first aliquot with a pair of primers complementary to the two newly extended 3′ ends (i.e., sharing their sequences separately with the sequence Alpha and Beta); contacting the said nicking restriction endonuclease with another aliiquot of the mixture of heterohybrids and homohybrids under conditions where it makes a single stranded nick; heating this treated aliquot of mixture of heterohybrids and homohybrids to a temperature which will denature off the short 3′ terminal sequences on homohybrids but which will not denature the internal fragment between the adapters which may be introduced a nick; ligating every possible gap of the internal fragments with a Tag ligase; recovering selectively the target homohybrids by the corresponding primer from one aliquot of the DNA mixture, ex. using primer-1 complementary to the adapter sequence Alpha to recover the homohybrids originally from RF pool-1, or separately using primer-2 complementary to the adapter sequence Beta to recover the homohybrids originally from RF pool-2,
38. The method of claim 37 wherein the restriction endonucleases is N.BstNB I.
39. The method wherein a pair of different modified adapters based on claim 35-36 is used for selective recovery of heterohybrids or either one of the homohybrids resulting from the mixing, denaturation and reannealing of two DNA pools, comprising: Digesting 2 DNA pools with a certain restriction endonuclease (RE) to obtain their corresponding restriction fragment (RF) pools: RF pool-1 and RF pool 2; ligating separately the pair of RF pools to a pair of modified HeD adapters to form the first and the second pools of RF-adapter constructs, wherein the modified HeD adapters contain a recognition site sensitive to methylated nucleotide for a corresponding restriction endonuclease (such as, but not limited to “GmethylatedATC” for DpnI site) that is capable of cutting a DNA duplex that both strands are specifically methylated at certain nucoeotide(s), and this site usually is designed at the 5′ free terminal section of the RF-adapter constructs; mixing together, denaturing, and reannealing the said first pool of RF-adapter constructs with the said second pool of RF-adapter constructs to form a mixture containing heterohybrids and homohybrids; filling in completely the 3′ end of the constructs of only heterohybrids in the mixture with a 3′-5′ exonuclease minus DNA polymerase and four standard dNTPs. Purifying the resulted mixture especially removing the free dNTPs. filling in the 3′ end of the constructs of homohybrids in the mixture with a 3′-5′ exonuclease plus DNA polymerase, plus 4 dNTPs with selective dNTP(s) being methylated, ex. methylated dATP is used with standard dCTP, dTTP and dGTP when DpnI site is designed. Then, dividing the mixture into 2 aliquots. recovering selectively the target homohybrids from the first aliquot of mixture by the corresponding primer from the of DNA mixture, ex. using primer-1 complementary to the adapter sequence Alpha to recover the homohybrids originally from RF pool-1, or separately using primer-2 complementary to the adapter sequence Beta to recover the homohybrids originally from RF pool-2, contacting the said methylation sensitive restriction endonuclease with another aliquot of the mixture of heterohybrids and homohybrids under conditions where it cuts off the distal part of the adapter from two homohybrid constructs; recovering and amplifying selectively the heterohybrids with a pair of primers complementary to the two newly extended 3′ ends.
40. A method wherein a combination designs of claim 37 and 39 is used for selective recovery of heterohybrids or either one of the homohybrids resulting from the mixing, denaturation and reannealing of two DNA pools, comprising: digesting 2 DNA pools with a certain restriction endonuclease (RE) to obtain their corresponding restriction fragment (RF) pools: RF pool-1 and RF pool 2; ligating separately the pair of RF pools to a pair of modified HeD adapters to form the first and the second pools of RF-adapter constructs, wherein the modified HeD adapters contain a recognition site sensitive to methylated nucleotide for a corresponding restriction endonuclease (such as, but not limited to “G.methylatedA.T.C” for DpnI site) that is capable of cutting a DNA duplex that both strands are specifically methylated at certain nucoeotide(s), and this site usually is designed at the 5′ free terminal section of the RF-adapter constructs, wherein the modified HeD adapters here also contain a recognition site for a nicking restriction endonuclease (such as, but not limited to N. BstNBI) that is capable of cutting only one strand of a double stranded DNA molecule, and this site usually is designed at the stem of adapter and for removing the 3′ terminal section of the adapter; mixing together, denaturing, and reannealing the said first pool of RF-adapter constructs with the said second pool of RF-adapter constructs to form a mixture containing heterohybrids and homohybrids; filling in completely and selectively the 3′ end of the constructs of the heterohybrids in the mixture with a 3′-5′ exonuclease minus DNA polymerase and four standard dNTPs. Purifying the resulted mixture and completely removing the left free dNTPs; contacting the said nicking restriction endonuclease with another aliquot of the mixture of heterohybrids and homohybrids under conditions where it makes a single stranded nick; heating the DNA mixture to a appropriate temperature that causes the 3′ ends of the homohybrids that is not extended to dissociate and the short single strands to leave off the adapters, while 3′ end of the homohybrids that is extended does not dissociated, and this right temperature is high enough to prevent the dissociated free short single strands from reannealing to the adapters; treating the mixture with a ligase to close every possible nick in every internal fragment; filling in the 3′ end of the constructs of the homohybrids in the mixture with a 3′-5′ exonuclease plus DNA polymerase, in addition 4 dNTPs with selective dNTP(s) being methylated, ex. methylated dATP is used with standard dCTP, dTTP and dGTP when DpnI site is designed in the adapters. Then dividing the mixture into 2 aliquots. recovering selectively the target homohybrids from the first aliquot of mixture via by the corresponding primer from the of DNA mixture, ex. using primer-1 complementary to the adapter sequence Alpha to recover the homohybrids originally from RF pool-1, or separately using primer-2 complementary to the adapter sequence Beta to recover the homohybrids originally from RF pool-2, contacting the said methylation sensitive restriction endonuclease with another aliquot of the mixture of heterohybrids and homohybrids under conditions where it cuts off the distal part of the adapter from two homohybrid constructs; recovering and amplifying selectively the heterohybrids from the second aliquot with a pair of primers complementary to the two newly extended 3′ ends.
41. A method according to claim 36 wherein a pair of modified HeD adapters is used for selective recovery of heterohybrids or either one of the homohybrids resulting from the mixing, denaturation and reannealing of two DNA pools. digesting 2 DNA pools with a certain restriction endonuclease (RE) to obtain their corresponding restriction fragment (RF) pools: RF pool 1 and RF pool 2; ligating separately the pair of RF pools to a pair of modified HeD adapters to form the first and the second pools of RF-adapter constructs, wherein both the first and second HeD adapters contain a first and a second adjacent nucleotide-methylation-sensitive restriction endonuclease recognition sites in the stem portion of the adapters, wherein the methylation-sensitive nucleotide of the first said site is methylated on both strands on the first adapter but unmethylated on both strands on the second adapter, wherein the methylation-sensitive nucleotide of the second said on the first adapter is unmethylated on both strands on the first adapter but methylated on both strands on the second adapter, and wherein the 3′ free terminal section of the adapter is as short as zero (i.e. the entire shorter strand is completely complementary to the stem portion of longer strand of the adapter), whereby upon formation of heterohybrids and homohybrids by denaturing and reannealing, the homohybrids are susceptible to digestion of the said restriction endonuclease and the heterohybrids are not susceptible to digestion of the said restriction endonuclease. mixing together, denaturing, and reannealing the said first pool of RF-adapter constructs with the said second pool of RF-adapter constructs to form a mixture containing heterohybrids and homohybrids; filling in the 3′ end of the constructs of both heterohybrids and homohybrids with a DNA polymerase plus 4 deoxyribonucleotide triphosphates; Then dividing the mixture into 2 aliquots; contacting the said methylation sensitive restriction endonuclease with the first aliquot of the mixture under conditions where it cuts off the distal part of the adapter from two homohybrid constructs; recovering and amplifying selectively the uncut constructs, i.e. the heterohybrids, from this aliquot with a pair of primers complementary to the two newly extended 3′ ends; recovering selectively the target homohybrids by the corresponding one primer from the second of aliquot of DNA mixture, ex. using primer-1 complementary to the adapter sequence Alpha to recover the homohybrids originally from RF pool-1, and using primer-2 complementary to the adapter sequence Beta to recover the homohybrids originally from RF pool-2,
42. The method of claim 40 wherein the restriction endonuclease is DpnI.
43. The method of claim 40 wherein the restriction endonuclease is TaqI.
44. The method of claim 40 wherein the restriction endonuclease is MboI.
45. A method according to claim 36 wherein a pair of modified HeD adapters is used for selective recovery of heterohybrids or either one of the homohybrids resulting from the mixing, denaturation and reannealing of two DNA pools. digesting 2 DNA pools with a certain restriction endonuclease (RE) to obtain their corresponding restriction fragment (RF) pools: RF pool 1 and RF pool 2; ligating separately the pair of RF pools to a pair of modified HeD adapters to form the first and the second pools of RF-adapter constructs, wherein the 5′ free terminal ends of the first and second HeD adapters are separately attached to first and second capture moieties; and wherein the 3′ free terminal section of the adapter is as short as zero (i.e. the entire shorter strand is completely complementary to the stem portion of longer strand of the adapter), whereby upon formation of heterohybrids and homohybrids by denaturing and reannealing, either one of the two homohybrid constructs has a unique capture moiety, and the heterohybrid constructs have both capture moieties; mixing together, denaturing, and reannealing the said first pool of RF-adapter constructs with the said second pool of RF-adapter constructs to form a mixture containing heterohybrids and homohybrids; filling in the 3′ end of the constructs of both heterohybrids and homohybrids with a DNA polymerase plus 4 deoxyribonucleotide triphosphates; Then dividing the mixture into 2 aliquots; recovering the said heterohybrids by capturing the first moiety with the first affinity reagent, following by capturing the second moiety with the second affinity reagent; the collection construct pool is capable to be amplified with a pair of primers complementary to the two newly extended 3′ ends (i.e., sharing their sequences separately with the sequence Alpha and Beta); recovering the said first homohybrids by capturing the first moiety with the first affinity reagent, following by removing from the output pool the constructs attached with the second moiety with the second affinity reagent; the collected construct pool is capable to be amplified with a primer complementary to the newly extended 3′ ends of the first homohybrids; recovering the said second homohybrids by capturing the second moiety with the second affinity reagent, following by removing from the output pool the constructs attached with the first moiety with the first affinity reagent; the collected construct pool is capable to be amplified with a primer complementary to the newly extended 3′ ends of the second homohybrids;
46. The method of claim 45 wherein the capture moiety/affinity reagent pairs are biotin/avidin or strepavidin, digoxigenin/anti-digoxigenin antibody;
47. A method wherein one general recovery adapters is used for efficient and selectively recovery of the fully reannealed DNA fragments resulting from the denaturation and reannealing of one DNA pools, comprising: digesting a DNA pool with a certain restriction endonuclease (RE) to obtain its restriction fragment (RF) pool; denaturing, and reannealing the said pool of RF-adapter constructs; ligating the RF pool to the cohesive end of the general recovery adapter having a long and a short oligonucleotides, of which the 5′ part of the long strand is on the outside of the constructs, and the 5′ end of the short oligonucleotides is onto the fragment end and free of phosphorylation group; melting off the short strand of the adapter at a appropriate temperature, and then adding Tag polymerase to the mixture in the existence of 4 deoxyribonucleotide triphosphates and filling the end of the construct at or the temperature above (usually at 72 C.); recovering and amplifying the filled-in constructs with a pair of primers complementary to the two newly extended 3′ ends.
48. The methods according to claim 41-47 wherein the modified HeD adapters comprising a full complementary double strand adapter, wherein the 5′ end ligated to DNA fragment is phosphorylated;
49: A method wherein a pair of different adapters is used for selective recovery of heterohybrids resulting from the mixing, denaturation and reannealing of two DNA pools, comprising: digesting a pair of DNA pools with a certain restriction endonuclease (RE) to obtain their corresponding restriction fragment (RF) pools: RF pool 1 and RF pool 2; ligating separately the pair of RF pools to a pair of special HeD adapters to form the first and the second pools of RF-adapter constructs, wherein the two adapters comprising completely different sequences, wherein each adapter having a long and a short oligonucleotides, wherein the 5′ end of the long strand being to the outside of the constructs, while the 5′ end of the short oligonucleotides being onto the fragment end and free of phosphorylation group, wherein the 5′ part of the long strand remaining to be free single strand and the sequence of this free single strand part is long enough (usually but not limited to, longer than 20 mers); denaturing, and reannealing the said constructs of RF pool1 and RF pool2; melting off the short strand of the adapter at a appropriate temperature, and then adding Tag polymerase to the mixture in the existence of 4 deoxyribonucleotide triphosphates and filling the end of the construct at or above the temperature above (usually at 72 C.); recovering and amplifying the filled-in constructs with a pair of primers complementary to the two different newly extended 3′ ends.
50. Methods according to claim 34 and claims 36 to 49 where the said step of recovery/amplification of the target fragments is realized by polymerase chain reaction (PCR) with the said primer(s);
51. Methods according to claim 34 and claims 36 to 50 wherein the recovery/amplification of the target fragments is accomplished by rolling cycle amplification (RCA) whereby a splint oligonucleotides is used to circularizing the target fragment constructs, this splint oligonucleotides having a sequence with one end complementary an end sequence of and the other end sequence complementary of target fragment construct, which are corresponding to the sequence(s) of the said primer(s).
52. A method according to claim 34 wherein the separation and enrichment of and mismatch fragments (MM) or perfectly matched fragments (PM) from the mixture is accomplished by one or more DNA Glycosylase(s), or and any other mismatch repair enzymes with similar function, comprising: contacting the contacting this class of enzyme(s) with a DNA mixture of PM fragments or constructs and MM fragments or constructs under a optimized condition wherein the said enzyme(s) removes the mismatched nucleotide of from the MM fragment, leaving an abasic site and an associated aldehyde in one or two of the duplex, and nicking the basic site; recovering the DNA fragments containing basic site as MM fragments.
53. A method according to claim 34 wherein the separation and enrichment of and mismatch fragments (MM) or perfectly matched fragments (PM) from the mixture is accomplished by one or more DNA Glycosylase(s), or and any other mismatch repair enzymes with similar function. contacting this class of enzyme(s) with a DNA mixture of PM fragments or constructs and MM fragments or constructs under a optimized condition wherein the said enzyme(s) removes the mismatched nucleotide of from the MM fragment and leaving an abasic site and an associated aldehyde in one or two of the duplex but not cleave them. separating the said DNA mixture into a first fraction which is bound to the enzyme(s) and a second fraction which is not bound, releasing and enriching the first fraction for MM fragments.
54. A method according to claim 53 wherein the DNA Glycosylase itself, after recognizing the MM fragment and leaving an abssic site, binds to the abasic sites of the MM fragments under a condition with enough EDTA to withdraw Mg+2 from the reaction buffer, the MM fragments with abasic site tightly binding to Glycosylase.
55. A methods according to claim 53 wherein the Glycosylase bound MM fragments and unbound PM fragments are separate by washing off the PM fragment and retain MM fragments when the DNA glycosylase is previous immobilized to any solid matrix such as but not limit to resin, agarose or any other beads including magnetic beads via GST, His6, or other molecules, wherein the bound MM fragments are eluted under a condition with enough Mg+2, or the complex of Glycosylase with DNA fragments are eluted from the solid matrix corresponding to the immobilization fashion of Glycosylase, such as Gluththione corresponding to GST-immobilized TDG, wherein the free PM fragments are collected separately.
56. A method according to claim 53 wherein the separation and enrichment of and mismatch fragments (MM) or perfectly matched fragments (PM) from the mixture is accomplished by one or more DNA Glycosylase(s), or and any other mismatch repair enzymes with similar function, comprising: contacting this class of enzyme(s) with a DNA mixture of PM fragments or constructs and MM fragments or constructs under a optimized condition wherein the said enzyme(s) removes the mismatched nucleotide of from the MM fragment and leaving an abasic site and an associated aldehyde in one or two of the duplex but not cleave or bound them; applying an immobilized-chemical molecules to bound specifically and tightly to the abasic site-associated aldehyde in MM fragments and separating the MM fragment from the free PM fragments.
57. A method of claim 58 wherein the chemical molecule is hydroxylamine or hydrazide derivative, which forms a stable oxime bond with the open-chain aldehydes generated upon absic site formation, such as (but not limited to) FARP [5-(((2-(carbohydrazino)-methyl) thio) acetyl)-aminofluorescin, aminooxyacetyl hydrazid].
58. Methods according to claims 53 to 56 wherein the DNA glycosylase is thymine DNA glycosylase, such as but not limit to, hTDG (human G/T mismatch-specific Thymine-DNA glycosylase), mTDG (the archaeon Methanobacterium thermoautotropicum DNA mismatch N-glycosylase (Mig.Mth)), and E. coli MutY protein.
59. A method according to claim 58 wherein E. coli MutY is used for recognizing and binding of the MM fragments at the mismatch nucleotide in the presence of sodium borohydride (NaBH4) to produce covalent protein-DNA complex, which is then separated from PM fragments;
60. A method according to claim 34 wherein the output fragment pools, i.e. the PM homohybrids from the control pool, the MM homohybrids from the control pool, the PM heterohybrids, and the MM heterohybrids, are analyzed via one or a combination of the following techniques: microarray hybridization, gel display, subtractive hybridization or representative differential analysis, FISH (fluorescent in situ hybridization), or other techniques.
61. The methods according to claim 35-60 wherein any technique of one claim is combined with the technique(s) of another one or more claims.

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of U.S. provisional application serial No. 60/243,407 filed Oct. 27, 2000, the disclosure of which is expressly incorporated herein.

Provisional Applications (1)

	Number	Date	Country
	60243407	Oct 2000	US

Methods for identifying genes associated with diseases or specific phenotypes

Information

Publication Number

Date Filed

Date Published

Inventors

CPC

US Classifications

International Classifications

Abstract

Description

Claims

CROSS-REFERENCE TO RELATED APPLICATIONS

Provisional Applications (1)