WO2003020968A2

WO2003020968A2 - Method for analyzing nucleic acid sequences and gene expression

Info

Publication number: WO2003020968A2
Application number: PCT/EP2002/009614
Authority: WO
Inventors: Dimitri Tcherkassov
Original assignee: Genovoxx Gmbh
Priority date: 2001-08-29
Filing date: 2002-08-28
Publication date: 2003-03-13
Also published as: DE10239504A1; AU2002337030A1; WO2003020968A3

Abstract

The invention relates to a method for analyzing nucleic acid sequences and gene expression. The method is based on the detection of fluorescence signals of nucleotide molecules inserted into growing nucleic acid chains by means of a polymerase. The reaction occurs on a plane surface, wherein a plurality of individual nucleic acid molecules are immobilized on said surface. All said nucleic acid molecules are exposed to the same conditions so that a build-up reaction can simultaneously take place in all nucleic acid molecules.

Description

Method for the analysis of nucleic acid chain sequences and gene expression

The invention relates to a method for analyzing nucleic acid chains and gene expression. The basis of the method is the detection of fluorescence signals of individual nucleotide molecules labeled with dyes, which are incorporated into a growing nucleic acid chain by a polymerase. The reaction takes place on a flat surface. Many individual nucleic acid molecules are bound to this surface. All of these nucleic acid molecules are exposed to the same conditions, so that a build-up reaction can take place simultaneously on all nucleic acid molecules.

The process essentially comprises the following steps:

1) Binding of the nucleic acid chain fragments (NSKFs) on a flat surface with subsequent hybridization of primers, alternatively binding of primers with subsequent hybridization of NSKFs, so that NSKF-primer complexes are formed.

2) Carrying out a cyclic build-up reaction, each cycle consisting of the following steps: a) adding a solution with labeled nucleotides (NTs ^* ) and polymerase to the bound NSKF-primer complexes, b) incubating the bound NSKF-primer complexes with them Solution under conditions suitable for extending the complementary strands by one NT, c) washing, d) detection of signals from individual molecules, e) removal of the label from the built-in nucleotides, f) washing.

If necessary, the cycle is repeated several times.

3) Analysis of the detected signals of the individual molecules.

4) Reconstruction of the sequences from the individual data. 1. Abbreviations and explanations of terms

DNA - deoxyribonucleic acid of different origins and different lengths (genomic DNA, cDNA, ssDNA, dsDNA)

RNA - ribonucleic acid (mostly mRNA)

Polymerases - enzymes that can incorporate complementary nucleotides into a growing DNA or RNA strand (e.g. DNA polymerases, reverse transcriptases, RNA polymerases)

dNTP - 2 'deoxy nucleoside triphosphates, substrates for DNA polymerases and reverse transcriptases

NTP - nucleoside triphosphates, substrates for RNA polymerases

NT - natural nucleotide, usually dNTP, unless expressly stated otherwise.

Abbreviation "NT" is also used when specifying the length of a nucleic acid sequence, e.g. NT 1,000. In this case "NT" stands for nucleoside monophosphates.

The majority of abbreviations in the text are formed by using the suffix "s", ^» NT ^M stands for" nucleotide "," NTs "stands for several nucleotides.

NT ^* - modified nucleotide, usually dNTP, unless expressly stated otherwise. NTs ^* means: modified nucleotides

NSK - nucleic acid chain. DNA or RNA in their original length

NSKF - Nucleic acid chain fragment (DNA or RNA) that corresponds to part of the total sequence, NSKFs Nucleic acid chain fragments. The sum of the NSKFs is equivalent to the overall sequence. The NSKFs can, for example, be fragments of the entire DNA or RNA sequence that arise after a fragmentation step.

5

Total sequence - the sequence or sequences used in the sequencing reaction, mostly converted into NSKFs. It can originally consist of one or more NSKs. The total sequence can be parts or equivalents

Represent 0 of another sequence or of sequence populations (e.g. mRNA, cDNA, plasmid DNA with insert, BAC, YAC) and originate from one or different species.

Primer binding site (PBS) - part of the sequence in the NSK or 5 NSKF to which the primer binds.

Reference sequence - a sequence already known, to which the deviations in the sequence to be examined or in the sequences to be examined (overall sequence) are determined. D Sequences accessible in databases can be used as reference sequences, e.g. from the NCBI database.

Tm - melting temperature

5 Plane surface - surface which preferably has the following features: 1) It allows several individual molecules, preferably more than 100, more preferably more than 1000, to be detected simultaneously with the given objective-surface distance at one objective position. 2) The

The immobilized individual molecules are in the same focal plane, which can be set reproducibly.

Wide-field optical detection system - detection system that can simultaneously detect fluorescence signals from individual molecules distributed over a surface, the surface being approximately 100 μm ² and larger. An example of wide-field detection optics is provided by the Axiovert 200 or fluorescence microscope Axioplan 2e (Zeiss) with a planofluar objective lOOx NA 1.4 oil immersion (Zeiss), or a planapochromatic objective lOOx NA 1.4 oil immersion (Zeiss); The fluorescence can be excited using a lamp, for example a mercury vapor lamp, or a laser or diode. Both epifluorescence mode and total internal reflection fluorescence microscopy (TIRF microscopy) mode or laser scanning microscopy mode can be used. This wide field detection optics is used in this application

Definition of termination: In this application, termination is the reversible stop in the installation of the modified, uncleaved NTs ^* .

This term is to be separated from the usual use of the word "termination" by dideoxy-NTP in conventional sequencing.

The termination comes after the installation of a modified NT ^* . A substituent leading to the termination or a modification of the 3 "-OH position on the deoxyribose of a nucleotide, which leads to the termination. The substituent can be split off under mild conditions, so that the 3" - OH function again for the incorporation of an NT * is available. A fluorescent dye is coupled to these substituents.

Gene products - The gene products are the primary gene products of the genes. Essentially, these are RNA transcripts of the genes mentioned, which are also referred to as target sequences (or target nucleic acid sequences). In addition to mRNA, these target sequences also include single-stranded and double-stranded cDNA derived therefrom, RNA derived from cDNA or DNA amplified from cDNA.

Single nucleotide poly orphisms (single nucleotide polymorphisms, SNPs) - changes in the sequences that can occur as a substitution (transition or transversion) or as a deletion or insertion of individual NT.

2. State of the art

Nucleic acid chain sequence analysis and

Gene expression analysis has become an important tool in many areas of science, medicine and industry. Several methods have been developed for analysis.

The best known methods are chain termination sequencing according to Sanger (F. Sanger et al. PNAS 1977 v.74 p. 5463), which is based on the installation of chain terminators, and the Maxam-Gilbert method, which is based on bases specific modification and cleavage of nucleic acid chains is based (AM Maxam and W. Gilbert PNAS 1977, v.74 p.560). Both methods deliver a number of nucleic acid chain fragments of different lengths. These fragments are separated lengthwise in a gel. All disadvantages of electrophoresis (such as long running times, relatively short stretches of sequences that can be determined in one approach, limited number of parallel approaches and relatively large amounts of DNA) have to be accepted. These methods are very labor intensive and slow.

Another sequencing method is based on the hybridization of nucleic acid chains with short oligonucleotides. Mathematical methods are used to calculate how many oligonucleotides of a certain length are required to determine a complete sequence (ZT Strezoska et al. PNAS 1991 v.88 p.10089, RSDrmanac et al. Science 1993 v.260 p.1649 ). This method also has problems: only one sequence can be determined in one approach, secondary structures disrupt it Hybridization and repeats prevent correct analysis.

Working groups have another possibility for sequencing, for example, of (Dower US Patent 5,547,839, Canard et al. US Patent 5,798,210, Rasolonjatovo Nucleosides & Nucleotides 1999, v.18 p.1021, Metzker et al. NAR 1994, v.22 , P.4259, Welch et al. Nucleosides & Nucleotides 1999, v.18, p.197). This method is abbreviated as BASS (Base Addition Sequencing Scheme) or SBS (Sequecing by Synthesis). A large number of identical single-stranded DNA pieces are fixed at a defined location on a surface and the signal from all of these many identical DNA pieces is analyzed. A solution with polymerase and nucleotides is added to this fixed DNA so that a complementary strand can be synthesized. The polymerase should work step by step: only one nucleotide is incorporated in each step. This is detected, whereupon the polymerase incorporates the next nucleotide in a next cycle.

Despite the success of some of the method's individual steps, it has not been developed into a functional process. This can be based, for example, on the following facts: When the complementary strands are built up, the synthesis is desynchronized very quickly, so that the errors accumulate with each step. Therefore only very short fragments can be sequenced. It should be emphasized that all BASS methods described are not based on the detection of individual molecules. Instead, the signal is registered by a large number of identical molecules immobilized at a defined location. The customary use of the terms “individual molecules” and “molecules” in these methods does not aim at individual, separate molecules, but at a population consisting of many identical molecules. In this case, identical means that the molecules have the same sequence. Analysis of the gene expression spectrum has become an important tool in science. The comparison of the gene expression spectra between different cell lines, tissues or developmental stages allows conclusions to be drawn about the specific biological processes taking place in them. For example, one can expect that the comparison between tumor cells and healthy cells of the same origin will provide information about the genes involved in the tumor process. It is important that the activity of as many or all genes as possible is analyzed at the same time.

The analysis of gene expression is a complex task: the number of genes active in a cell type can be several thousand. However, the analysis should consider all genes contained in the genome of the species in question (approximately 32,000 in humans). In addition, the genes active in the respective cell type are first of all mostly not yet completely known and secondly they are expressed to different extents.

Many methods for gene expression analysis have already been developed, e.g. Differential Display (Nature 1984, v.308 p.149, Science 1992 v.257 p.967), Expressed Sequence Tags (EST) (Science 1991 v.252, p.1656, Nature Genetics 1992, v.2 p.173 ), Northern blotting or RT-PCR (PNAS 1977, v.74, p.5350, Cell 1983 v.34 p.865, "The PCR Technique, RT-PCR" 1998, Ed. Paul Suebert, Eaton Publishing). All of these methods can only analyze a very limited number of genes per reaction and are sometimes very labor-intensive.

The most widespread method for parallel analysis of the gene expression patterns is the hybridization of a mixture of cDNA molecules to be analyzed with oligonucleotides bound to a surface, which are fixed in a certain arrangement, usually as a "microarray"("Microarray Biochip Technology" 2000, Ed M.Schena, Eaton Publishing, Zhao et al. Gene 1995, v. 156, p.207, Schena et al. Science 1995 v.270, p.467, Lockhart et al - US Patent 6,040,138, Wang U.S. Patent 6,004,755, Arlinghaus et al. U.S. Patent 5,821,060, Southern U.S. Patent 5,700,637, Fodor et al. U.S. Patent 5,871,928).

The major disadvantages of the hybridization method include: The production of the oligonucleotides bound to the surface is expensive. The analysis is limited to genes whose sequences are already known. Multiple mismatch controls increase the number of oligonucleotides that need to be immobilized.

The object of the present invention is therefore to provide a method for the sequence analysis of nucleic acid chains and the analysis of gene expression which does not have the disadvantages of the above-mentioned methods and, above all, enables a cheaper, faster and more efficient analysis of nucleic acid sequences. In particular, the method should be able to determine many sequences in parallel. It can then be used, for example, for the analysis of very long nucleic acid chains (several Mb) or for variant analysis on many short chains (mutation analysis, SNP analysis) in one approach.

3. Brief description

The present invention relates to a method for parallel sequence analysis of nucleic acid sequences (nucleic acid chains, NSKs), in which

Fragments (NSKFs) of single-stranded NSKs with a length of about 50 to 1000 nucleotides are generated, which represent overlapping partial sequences of the total sequences

the NSKFs are bound on a reaction surface in a random arrangement using one or more different primers in the form of NSKF-primer complexes cyclically build the complementary strand of the NSKFs using one or more polymerases by:

a) a solution is added to the NSKF primer complexes bound to the surface, which contains one or more polymerases and one to four modified nucleotides (NTs ^* ) which are labeled with fluorescent dyes, the simultaneous use of at least two NTs ^* Fluorescent dyes located on the NTs ^* are selected such that the NTs ^{* used} can be distinguished from one another by measuring different fluorescence signals, the NTs ^{* being} structurally modified such that the polymerase does not form after incorporating such an NT ^* into a growing complementary strand is able to incorporate another NT ^* in the same strand, the terminating substituent being cleavable with the fluorescent dye

b) the stationary phase obtained in stage a) is incubated under conditions which are suitable for the extension of the complementary strands, the complementary strands being extended in each case by one NT ^* , man

c) the stationary phase obtained in stage b) is washed under conditions which are not suitable for the removal of NTs ^* which are not incorporated into a complementary strand

d) the individual NTs ^* built into complementary strands are detected by measuring the signal which is characteristic of the respective fluorescent dye, the relative position at the same time of the individual fluorescence signals on the reaction surface is determined

e) to generate unlabeled (NTs or) NSKFs, the substituents leading to the termination and the fluorescent dyes are cleaved off from the NTs ^* attached to the complementary strand, man

f) the stationary phase obtained in step e) is washed under conditions which are suitable for removing the fluorescent dyes and the ligands

stages a) to f) are repeated several times, if necessary,

the relative position of individual NSKF-primer complexes on the reaction surface and the sequence of these NSKFs being determined by specific assignment of the fluorescence signals detected in step d) in successive cycles at the respective positions to the NTs.

The overall sequence of the NSKs can be determined, for example, from the partial sequences determined. In this context, a parallel sequence analysis is understood to mean the simultaneous sequence analysis of many NSKFs (for example 1,000,000 to 10,000,000), these NSKFs being derived from a uniform NSK population or from several different NSK populations.

The resulting population of overlapping partial sequences can be combined, for example, in de novo sequencing with commercially available programs to form the overall sequence of the NSK (Huang et al. Genom Res. 1999 v.9 p.868, Huang Genomics 1996 v.33 p.21, Bonfield et al. NAR 1995 v.23 p.4992, Miller et al. J. Comput. Biol. 1994 vl p.257).

When analyzing variants of a known reference sequence, mutations or single nucleotide polymorphisms can be identified by comparing the obtained overlapping partial sequences with the reference sequence.

According to a particular embodiment of the invention, the process can be carried out by repeating steps a) to f) of the cyclic build-up reaction several times, with

a) only one marked NT ^* in each cycle, b) two differently marked NTs ^* in each cycle or c) four differently marked NTs ^* in each cycle

starts.

If the NSKs are variants of a known reference sequence, the method can also be carried out by repeating steps a) to f) of the cyclic build-up reaction several times, alternately using two differently marked NTs ^* and two unmarked NTs in the cycles and one Total sequences determined by comparison with the reference sequence.

The present invention also relates to a method for highly parallel analysis of gene expression, in which

provides single-stranded gene products, one

the gene products are bound on a reaction surface in a random arrangement using one or more different primers in the form of gene product-primer complexes, one

perform a cyclic assembly reaction of the complementary strand of the gene products using one or more polymerases by a) a solution is added to the gene product-primer complexes bound to the surface which contains one or more polymerases and one to four modified nucleotides (NTs ^* ) which are labeled with fluorescent dyes, the simultaneous use of at least two NTs ^* each ^* located at the NTs fluorescent dyes are selected so located that can differ from each other, the NTs used ^* by measuring different fluorescent signals, the NTs ^* are structurally modified ^'so that the polymerase after incorporation of such NT ^* not in a growing complementary strand is able to incorporate another NT ^* in the same strand, the terminating substituent being cleavable with the fluorescent dye, man

d) the individual NTs ^* built into complementary strands are detected by measuring the signal characteristic of the respective fluorescent dye, at the same time determining the relative position of the individual fluorescent signals on the reaction surface

e) for the generation of unlabelled (NTs or) NSKFs, the substituents leading to the termination with the Splits off fluorescent dyes from the NTs ^* attached to the complementary strand, man

stages a) to f) are repeated several times, if necessary,

whereby the relative position of individual gene product-primer complexes on the reaction surface and the sequence of these gene products are determined by specific assignment of the fluorescence signals detected in step d) in successive cycles at the respective positions to the NTs and the identity of the gene products is determined from the partial sequences determined certainly.

The gene products are the primary gene products of the genes whose expression is to be analyzed. Essentially, these are RNA transcripts of the genes mentioned, which are also referred to as target sequences (or target nucleic acid sequences). In addition to RNA, these target sequences also include single-stranded and double-stranded cDNA derived therefrom, RNA derived from cDNA or DNA amplified from cDNA.

The gene products or target sequences can either be isolated as mRNAs directly from a biological sample (e.g. cell extract, tissue extract or extract from whole organisms) or obtained as cDNAs by reverse transcription of the mRNAs.

In this context, a highly parallel analysis is understood to mean the simultaneous sequence analysis of many gene product molecules (for example 1,000,000 to 10,000,000), these gene product molecules being a complex one represent a heterogeneous population that corresponds, for example, to a complete expression profile or an expression spectrum of a tissue.

starts.

The method can also be carried out by repeating steps a) to f) of the cyclic build-up reaction several times, alternately using two differently labeled NTs ^* and two unlabeled NTs in the cycles and comparing the identity of the gene products with the reference sequences determined.

The invention further relates to a kit for carrying out the method which contains a reaction surface, reaction solutions required for carrying out the method, one or more polymerases, and nucleotides (NTs), one to four of which are labeled with fluorescent dyes, the NTs ^* on the 3 'position are structurally modified so that after installing such an NT ^* in a growing complementary strand, the polymerase is not able to incorporate another NT ^* in the same strand, the terminating substituent being cleavable with the fluorescent dye. According to a particular embodiment of the invention, the kit also contains for the production of Single strands of double strands required reagents, single stranded nucleic acid molecules, which are introduced as PBS in the NSKFs, oligonucleotide primers, for cleaving off the fluorescent dyes and the reagents and / or washing solutions leading to termination.

The method according to the invention is used to determine the nucleic acid sequences and can be used in various areas of genetics. This includes in particular the determination of unknown, long sequences, analysis of sequence polymorphisms and point mutations as well as the parallel analysis of a large number of gene sequences and the analysis of gene expression.

When analyzing long nucleic acid chains (e.g. 100 Kb and longer), the preparation of the material to be analyzed (single and double-stranded nucleic acid sequences) depends on the task at hand and the aim is to create a population of relatively small, single-stranded nucleic acid chain fragments (NSKFs) from a long nucleic acid chain. to form, to provide these fragments with a primer suitable for starting the sequencing reaction (NSKF-primer complexes) and to fix them on a flat surface.

Individual NSKFs are fixed on a flat surface in such a way that an enzymatic reaction can take place on these molecules. In principle, different types of immobilization are possible, which depend on the objective, the type of NSK and the polymerase used for the reaction. The NSKFs are randomly distributed on the surface during immobilization or binding, that is, it is not necessary to ensure that the individual chains are positioned exactly. NSKF primer complexes can be bound to the surface via the NSKFs or primers. The NSKF primer complexes must be fixed to the surface in such a density ensure that the later-detected signals are clearly assigned by the installed NT * s to individual NSKFs.

After the preparation of the NSKFs, the sequencing reaction is started with all NSKF primer complex molecules immobilized on the surface. The sequencing is based on the synthesis of the complementary strand to each individual bound NSKF. Labeled NTs ^{* are} installed in the newly synthesized strand. The polymerase incorporates only one labeled NT ^* into the growing chain in one cycle.

The sequencing reaction proceeds in several cycles. A cycle comprises the following steps:

a) adding a solution with labeled nucleotides (NTs ^* ) and polymerase to the bound NSKF-primer complexes, b) incubating the bound NSKF-primer complexes with this solution under conditions that are suitable for extending the complementary strands by one NT , c) washing, d) detection of signals from individual molecules, e) removal of the label from the built-in nucleotides, f) washing.

If necessary, the cycle is repeated several times (a-f).

The reaction conditions of step (b) in a cycle are selected so that the polymerases can incorporate a labeled NT ^* on more than 50% of the NSKFs (NSKF primer complexes capable of expansion) which are involved in the sequencing reaction, preferably on more than 90%.

The number of cycles to be carried out depends on the task at hand, and is not theoretically limited and is preferably between 20 and 5000.

Then, for each fixed NSKF, its specific sequence is determined from the order of the installed NTs ^* .

In one embodiment, the original NSK sequence can be reconstructed from the overlapping NSKF sequences ("Automated DNA sequencing and analysis" p. 231 ff. 1994 M. Adams et al. Academic Press, Huang et al. Genom Res. 1999 v. 9 p.868, Huang Genomics 1996 v.33 p.21, Bonfield et al. NAR 1995 v.23 p.4992, Miller et al. J.Comput. Biol. 1994 vl p.257). The entire population of NSKF sequences is searched for matches / overlaps in the sequences of NSKFs. Through these matches / overlaps, the NSKF can be aligned, e.g .:

ACTGTGCGTCCGTATGATGGTCATTCCATG

CATTCCATGGTACGTTAGCTCCTAG

TCCTAGTAAAATCGTACC.

In practice, when sequencing unknown sequences, it has proven useful to achieve a length of the sequenced pieces of more than 300 bp. This allows the sequencing of genomes from eukaryotes in the shotgun process.

The errors in the method can be recorded and corrected using various means. All steps of the process can be largely automated.

Working with individual molecules offers great advantages over the BASS method described earlier:

1. Since the molecules are detected individually, there is no risk that the signal will become incorrect due to the desynchronization in the population. For each fixed NSKF created its own sequence. Therefore, it does not matter whether the synthesis of a neighboring molecule has already progressed or lagged behind. This makes highly parallel sequencing of long NSKF possible.

It is not necessary to fix molecules in a defined arrangement on the surface, since the signal comes from individual molecules and not from a spatially defined population (which is necessary with BASS methods). It is not absolutely necessary to produce multiple copies of the nucleic acid chains to be analyzed, so that PCR and / or cloning steps can be omitted. This leads to an enormous acceleration of the analysis compared to existing methods.

The simultaneous sequencing of individual gene product molecules gives the method according to the invention for analyzing gene expression several advantages over known methods of analyzing gene expression:

1) The gene products can bind to the surface in any arrangement. A previous complex synthesis of different oligonucleotides at certain positions (such as in the hybridization method) is therefore not necessary.

2) The material can be analyzed on a standardized surface.

3) The expression of still unknown genes can also be determined because all gene products contained in the mixture are analyzed.

4). The large number of molecules analyzed also allows the detection of poorly expressed genes.

5) Smallest amounts of starting material can be used: mRNA from a single cell can be sufficient for the analysis. 6) All steps of the process can be largely automated.

The method is based on several principles:

1. Short nucleotide sequences (10-50 NTs) contain enough information to identify the corresponding gene if the gene sequence itself is already contained in a database.

A sequence of, for example, 10 NTs can form more than 10 ^δ different combinations. For example, this is sufficient for most genes in the human genome, which according to current estimates contains 32,000 genes. The sequence can be even shorter for organisms with fewer genes.

2. The method is based on a new method for sequencing individual nucleic acid chain molecules.

3. Mixtures of nucleic acid chains can be examined.

4. The sequencing reaction takes place simultaneously on many molecules, the sequence of each individual bound nucleic acid chain being analyzed.

It is known that mRNAs or nucleic acid chains derived from the mRNA (e.g. single-stranded cDNAs, double-stranded cDNAs, cDNA-derived RNA or cDNA-amplified DNA) can be used to investigate gene expression. Regardless of the exact composition, they are referred to below as gene products. Partial sequences of these gene products are also referred to below as gene products.

These gene products are a mixture of different nucleic acid chains.

The synthesis is based on the synthesis of a strand that is complementary to the gene product. The aim of the preparation is to provide gene product-primer complexes which are bound in a random manner on a flat surface and on which the incorporation of NT * s by the polymerase can take place (gene product-primer complexes which can be extended).

The sequencing reaction is carried out with these bound gene product-primer complexes.

It runs in several cycles. Only one marked NT ^* per cycle is installed in the growing strand. A cycle comprises the following steps: a) addition of a solution with labeled nucleotides (NTs ^* ) and polymerase to bound gene product-primer complexes, b) incubation of the bound gene product-primer complexes with this solution under conditions which extend the complementary strands are suitable for an NT, c) washing, d) detection of the signals from individual modified NTs ^* molecules incorporated into the newly synthesized strands, e) removal of the label from the built-in nucleotides, f) washing.

This cycle can be repeated several times, so that preferably 10 to 50 NTs are determined for each gene product-primer complex participating in the sequencing reaction. The nucleic acid sequences are then reconstructed from the detected signals. The determined sequences of the bound gene products are compared with one another to determine the abundances and are assigned to specific genes by comparison with gene sequences in databases.

4. Detailed description

General principles of reaction, material selection and material preparation (generation of short NSKFs, introduction of a PBS, single strand preparation, primer selection, fixation of NSKFs), as well as the detection apparatus and detection are shown using the example of the procedure for sequencing long NSKs. The method for analyzing gene expression is then described in Example 3.

4.1 General principles of the reaction

In the following, the general principles of the reaction are to be illustrated by way of example using the sequencing of a piece of DNA of several Mb length (FIG. 1). The sequencing and reconstruction of nucleic acid sequences is based on the shotgun principle ("Automated DNA sequencing and analysis" p. 231 ff. 1994 M. Adams et al. Academic Press, Huang et al. Genom Res. 1999 v.9 p. 868 , Huang Genomics 1996 v.33 p.21, Bonfield et al. NAR 1995 v.23 p.4992, Miller et al. J.Comput. Biol. 1994 vl p.257). The sequence of a long piece of DNA is determined by sequencing small DNA fragments and subsequent reconstruction. The material to be analyzed (1) is prepared for the sequencing reaction by breaking it down into fragments of preferably 50 to 1000 bp in length (2). Each fragment is then provided with a primer binding site and a primer (3). This mixture of different DNA fragments is now fixed on a flat surface (4). The unbound DNA fragments are removed by a washing step. The sequencing reaction is then carried out on the entire reaction surface. This reaction is cyclical. In the first step of the cycle, an NT ^* labeled with a fluorescent dye is incorporated into the growing strand: the reaction is controlled so that only one labeled NT ^* from a polymerase can be incorporated into the growing strand in each cycle. This is achieved by using NTs ^* which have a reversibly coupled substituent leading to the termination at the 3 "position of the deoxyribose. The incorporation of a further labeled NT ^* is thereby made impossible. The polymerase and the labeled NTs ^* become simultaneous (5). The reaction mixture is then removed and washed the surface in a suitable manner (6). Now there is a detection step (7): The surface is scanned with a device suitable for single molecule detection (consisting of light source, microscope, camera, scanning table, computer with control and image recognition or image processing software) and the signals of the individual, built-in marked NTs ^* identified. After the detection step, the marker and the substituent leading to the termination are removed from all built-in NTs ^* (8). After a subsequent washing step, a new cycle can begin. For the reconstruction of a larger original DNA sequence (eg several Mb long piece of DNA), the DNA fragments should be several hundred NT long if the reconstruction is carried out according to the shotgun principle ("Automated DNA sequencing and analysis" p. 231 ff 1994 M. Adams et al. Academic Press, Huang et al. Genom Res. 1999 v.9 p.868, Huang Genomics 1996 v.33 p.21, Bonfield et al. NAR 1995 v.23 p.4992, Miller et al. J. Comput. Biol. 1994 vl p.257). Since only one marked NT * is installed per cycle, at least 300 cycles are required for sequencing.

4.2 Selection of the material

With the aid of the method according to the invention, it is possible to preselect DNA sequences (for example in YAC, PAC or BAC vectors (R. Anand et al. NAR 1989 v.17 p. 3425, H. Shizuya et al. PNAS 1992 v.89 p.8794, "Construction of bacterial artificial chromosome libraries using the modified PAC System" in "Current Protocols in Human genetics" 1996 John Wiley & Sons Inc.) cloned sections of a genome) as well as non-preselected DNA (e.g. genomic DNA, cDNA mixtures) to analyze. By preselection, it is possible to filter out relevant information in advance, such as sequence sections from a genome or populations of gene products, from the large amount of genetic information and thus to limit the amount of the sequences to be analyzed. They are particularly noteworthy Embodiments in which the methods according to the invention are used without preselection and without duplicating the material. Dispensing with PCR and cloning brings about a decisive acceleration in the highly parallel analysis of nucleic acid sequences, which has not been possible with other methods until now.

4.3 Preparation of the material

The aim of the material preparation is to obtain bound single-stranded NSKFs with a length of preferably 50-1000 NTs, a single primer binding site and a hybridized primer (bound NSKF-primer complexes). These NSKF primer complexes have, for example, the structure shown in FIG. 2. In particular, very variable constructions can be derived from this general structure. To improve the clarity, here are some examples, whereby the methods listed can be used individually or in combination.

4.3.1 Generation of Short Nucleic Acid Chain Fragments (50-1000 NTs) (Fragmentation Step)

It is important that the NSKs are fragmented in such a way that fragments are obtained which represent overlapping partial sequences of the total sequences. This is achieved by methods in which fragments of different lengths are formed as fission products in a random distribution.

According to the invention, the nucleic acid chain fragments (NSKFs) can be produced by several methods, for example by fragmentation of the starting material with ultrasound or by endonucleases ("Molecular cloning". 1989 j.Sambrook et al. Cold Spring Harbor Laborotary Press), such as, for example, by non-specific endonuclease mixtures , According to the invention, ultrasound fragmentation is preferred. The conditions can be set so that fragments with an average Length of 100 bp to 1 kb arise. These fragments are then filled in at their ends by the Klenow fragment (E. coli polymerase I) or by the T4 DNA polymerase ("Molecular cloning" 1989 J.Sambrook et al. Cold Spring Harbor Laborotary Press).

In addition, complementary short NSKFs can be synthesized from a long NSK using randomized primers. This method is particularly preferred when analyzing the gene sequences. Single-stranded DNA fragments with randomized primers and a reverse transcriptase are formed on the mRNA (Zhang-J et al. Biochem. J. 1999 v.337 p.231, Ledbetter et al. J.Biol.Chem. 1994 v.269 P.31544, Rolls et al. Anal.Biochem. 1993 v.208 p.264, Decraene et al. Biotechniques 1999 v.27 p.962).

4.3.2 Introduction of a primer binding site in the NSKF.

The primer binding site (PBS) is a sequence section which is intended to enable selective binding of the primer to the NSKF.

In one embodiment, the primer binding sites can be different, so that several different primers must be used. In this case, certain sequence segments of the overall sequence can serve as natural PBSs for specific primers. This embodiment is particularly suitable for the investigation of already known SNP sites, see. Example 4 "SNP analysis with sequence-specific primers".

In another embodiment, for reasons of simplification of the analysis, it is favorable if there is a uniform primer binding site in all NSKFs. According to a preferred embodiment of the invention, the primer binding sites are therefore introduced separately into the NSKFs. In this way, primers with a uniform structure can be used for the reaction. This embodiment will be described in detail below.

The composition of the primer binding site is not restricted. Their length is preferably between 20 and 50 NTs. The primer binding site can carry a functional group for immobilizing the NSKF. This functional group can e.g. be a biotin group.

As an example of the introduction of a uniform primer binding site, the ligation and the nucleotide tailing on DNA fragments are described below.

a) Ligation:

A double-stranded oligonucleotide complex with a primer binding site is used (FIG. 3a). This is ligated to the DNA fragments using commercially available ligases ("Molecular cloning" 1989 J.Sambrook et al. Cold Spring Harbor Laborotary Press). It is important that only a single primer binding site be ligated to the DNA fragment. This is achieved e.g. by modification of one side of the oligonucleotide complex on both strands (FIG. 3b). The results after the ligation or after subsequent denaturation are shown in FIGS. 3c and 3d. The modifying groups on the oligonucleotide complex can be used for immobilization. The synthesis and modification of such an oligonucleotide complex can be carried out according to standardized regulations. For synthesis, e.g. the DNA synthesizer 380 A Applied Biosystems can be used. Oligonucleotides with a certain composition with or without modifications are also commercially available as custom synthesis, e.g. from MWG-Biotech GmbH, Germany.

b) Nucleotide tailing:

Instead of ligation with an oligonucleotide, one terminal deoxynucleotidyl transferase can be used for several (e.g. between 10 and 20) attach nucleoside monophosphates to the 3 'end of an SS DNA fragment ("molecular cloning" 1989 J.Sambrook et al. Cold Spring Harbor Laborotary Press, "Method in Enzymology" 1999 v.303, p. 37-38) (FIG. 4), for example several guanosine monophosphates (called (G) n-tailing). The resulting fragment is used to bind the primer, in this example a (C) n primer.

4.3.3 Single strand preparation

Single-stranded NSKFs are required for the sequencing reaction. If the starting material is in double-stranded form, there are several possibilities for producing a single-stranded form from double-stranded DNA (e.g. heat denaturation or alkali denaturation) ("Molecular cloning" 1989 J.Sambrook et al. Cold Spring Harbor Laborotary Press).

4.3.4 Primers for the sequencing reaction

This has the function of enabling the start at a single point of the NSKF. It binds to the primer binding site in the NSKF. The composition and the length of the primer are not restricted. In addition to the start function, the primer can also perform other functions, such as to create a connection to the reaction surface. Primers should be adapted to the length and composition of the primer binding site so that the primer enables the sequencing reaction to be started with the respective polymerase.

When using different primer binding sites, for example naturally occurring in the original overall sequence, the sequence-specific primers for the respective primer binding site are used. In this case, a primer mixture is used for sequencing.

With a uniform, for example through the ligation A single primer is used at the primer binding site coupled to NSKFs.

The length of the primer is preferably between 6 and 100 NTs, optimally between 15 and 30 NTs. The primer can carry a functional group which serves to immobilize the NSKF, for example such a functional group is a biotin group (see section Immobilization). It should not interfere with sequencing. The synthesis of such a primer can e.g. can be carried out with the DNA synthesizer 380 A Applied Biosystems or as a custom synthesis from a commercial provider, e.g. MWG-Biotech GmbH, Germany).

Prior to hybridization to the NSKFs to be analyzed, the primer can be fixed on the surface using various techniques or synthesized directly on the surface, for example according to (McGall et al. US Patent 5412087, Barrett et al. US Patent 5482867, Mirzabekov et al. US Patent 5981734, "Microarray biochip technology" 2000 M. Schena Eaton Publishing, "DNA Microarrays" 1999 M. Schena Oxford University Press, Fodor et al. Science 1991 v.285 p.767, Timofeev et al. Nucleic Acid Research (NAR) 1996, v.24 p.3142, Ghosh et al. NAR 1987 v.15 p.5353, Gingeras et al. NAR 1987 v.15 p.5373, Maskos et al. NAR 1992 v.20 p.1679).

The primers are on the surface of microns, for example, in a density of between 10 to 100 microns per 100 ^2, 100 to 10,000 per 100 ^2, or 10,000 to 1,000,000 per lOOμm ² bound.

The primer or mixture of primers is incubated with NSKFs under hybridization conditions, which allow it to bind selectively to the primer binding site of the NSKF. This primer hybridization (annealing) can take place before (1), during (2) or after (3) the binding of the NSKFs to the surface. The optimization of the hybridization conditions depends on the exact structure of the primer binding site and the primer and can be according to Rychlik et al. Calculate NAR 1990 v.18 p.6409. In the following, these hybridization conditions are referred to as standardized hybridization conditions.

If a primer binding site with a known structure that is common to all NSKFs is introduced, for example by ligation, primers with a uniform structure can be used. The primer binding site can carry a functional group at its 3 'end which e.g. is used for immobilization. For example, this group is a biotin group. The primer has a structure that is complementary to the primer binding site.

An example of a primer binding site and a primer is shown below.

5 'TAATACGACTCACTATAGG3' primer (T7-19 primer) Biotin-3 ΑTTATGCTGAGTGATATCC5 'primer binding site

4.3.5 Fixation of NSKF primer complexes on the surface (binding or immobilization of NSKFs).

The aim of the fixation (immobilization) is to fix NSKF primer complexes on a suitable flat surface in such a way that a cyclic enzymatic sequencing reaction can take place. This can be done, for example, by binding the primer (see above) or the NSKF to the surface.

The order of the steps in the fixation of NSKF-Pri er complexes can be variable:

1) The NSKF primer complexes can first be formed in a solution by hybridization (annealing) and then bound to the surface.

2) Primers can first be bound to a surface and NSKFs can then be hybridized to the bound primers, whereby NSKF-primer complexes are formed (NSKFs indirectly bound to the surface) 3) The NSKFs can first be bound to the surface (NSKFs bound directly to the surface) and in the subsequent step the primers can be hybridized to the bound NSKFs, resulting in NSKF-primer complexes.

The NSKFs can therefore be immobilized on the surface by direct or indirect binding.

In the present case, the surface and the reaction surface are to be understood as equivalent terms, unless explicitly referred to another meaning. The surface of a solid phase of any material serves as the reaction surface. This material is preferably inert towards enzymatic reactions and does not cause any interference with the detection. Silicon, glass, ceramics, plastics (e.g. polycarbonates or polystyrenes), metal (gold, silver, or aluminum) or any other material that meets these functional requirements can be used. The surface is preferably not deformable, since otherwise the signals are likely to be distorted upon repeated detection.

If a gel-like solid phase (surface of a gel) is used, this gel can e.g. be an agarose or polyacrylamide gel. The gel is preferably freely passable for molecules with a molecular mass below 5000 Da

(For example, 1 to 2% agarose gel or 10 to 15% polyacrylamide gel can be used). Such a gel surface has the advantage over other solid surfaces that there is a significantly lower non-specific binding of NT * s to the surface. By binding the NSKF primer complexes on the surface, the detection of the fluorescence signals from built-in NTs ^{* is} possible. The signals from free NTs ^* are not detected because they do not bind to the material of the gel and are therefore not immobilized. The gel is preferably attached to a solid surface

(Fig. 5a). This firm base can be silicone, glass, ceramic, Plastic (e.g. polycarbonates or polystyrenes), metal (gold, silver, or aluminum) or any other material.

The thickness of the gel is preferably not more than 0.1 mm. The gel thickness is preferably greater than the simple depth of field of the lens, so that NTs ^* bound non-specifically to the solid base do not reach the focal plane and are therefore detected. If the depth of focus is 0.3 μm, for example, the gel thickness is preferably between 1 μm and 100 μm. The surface can be produced as a continuous surface or as a discontinuous surface composed of individual small components (eg agarose beads) (FIG. 5b). The reaction surface must be large enough to be able to immobilize the necessary number of NSKFs with the appropriate density. The reaction surface should preferably not be larger than 20 cm ² .

The different cycle steps require an exchange of the different reaction solutions above the surface. The reaction surface is preferably part of a reaction vessel. The reaction vessel is in turn preferably part of a reaction apparatus with a flow device. The flow device enables the solutions in the reaction vessel to be exchanged. The exchange can take place with a pump device controlled by a computer or manually. It is important that the surface does not dry out. The volume of the reaction vessel is preferably less than 50 μl. Ideally, its volume is less than 1 μl. An example of such a flow system is given in Fig.6.

If the NSKF primer complexes are fixed on the surface via the NSKFs, this can be done, for example, by binding the NSKFs to one of the two chain ends. This can be achieved by appropriate covalent, affine or other bonds. There are many examples of that Immobilization of nucleic acids is known (McGall et al. US Patent 5412087, Nikiforov et al. US Patent 5610287, Barrett et al. US Patent 5482867, Mirzabekov et al. US Patent 5981734, "Microarray biochip technology" 2000 M. Schena Eaton Publishing, " DNA Microarrays "1999 M. Schena Oxford University Press, Rasmussen et al. Analytical Biochemistry v.198, p.138, Allemand et al. Biophysical Journal 1997, v.73, p.2064, Trabesinger et al. Analytical Chemistry 1999, v .71, p.279, Osborne et al. Analytical Chemistry 2000, v.72, p.3678, Timofeev et al. Nucleic Acid Research (NAR) 1996, v.24 p.3142, Ghosh et al. NAR 1987 v. 15 p.5353, Gingeras et al. NAR 1987 v.15 p.5373, Maskos et al. NAR 1992 v.20 p.1679). The fixation can also be achieved by a non-specific binding, such as by drying out the sample containing NSKFs on the flat surface.

The NACFs be on the surface of microns, for example, at a density between 10 and 100 microns NACFs per 100 ^2, 100 to 10,000 per 100 ^2, bound 10,000 to 1,000,000 per lOOμrn. ²

The density of NSKF primer complexes which can be extended is necessary for the detection to be approximately 10 to 100 per 100 μm ² . It can be achieved before, during or after hybridization of the primers to the gene products.

Some methods for binding NSKF-primer complexes are shown in more detail below by way of example: In one embodiment, the NSKFs are immobilized via biotin-avidin or biotin-streptavidin binding. Avidin or streptavidin is covalently bound on the surface, the 5 'end of the primer contains biotin. After the labeled primers have hybridized with the NSKFs (in solution), they are fixed on the surface coated with avidin / streptavidin. The concentration of the hybridization products labeled with biotin and the time of incubation of this solution with the surface is chosen such that a density suitable for sequencing is already in this step is achieved.

In another preferred embodiment, the primers suitable for the sequencing reaction are fixed on the surface using suitable methods before the sequencing reaction (see above). The single-stranded NSKFs, each with one primer binding site per NSKF, are thus incubated under hybridization conditions (annealing). They bind to the fixed primers and are thereby bound (indirect binding), resulting in primer-NSKF complexes. The concentration of the single-stranded NSKFs and the hybridization conditions are chosen such that an immobilization density of 10 to 100 NSKF primer complexes capable of extension per 100 μm ² suitable for sequencing is achieved. After hybridization, unbound NSKFs are removed by a washing step. In this embodiment, a surface with a high primer density is preferred, for example approximately 1,000,000 primers per 100 μm ² or even higher, since the desired density of NSKF-primer complexes is achieved more quickly, the NSKFs only binding to a part of the primers ,

In another embodiment, the NSKFs are bound directly to the surface (see above) and then incubated with primers under hybridization conditions. With a density of approx. 10 to 100 NSKFs per 100 μm ² , attempts will be made to prime all available NSKFs and to make them available for the sequencing reaction. This can be achieved, for example, by high primer concentration, for example 1 to 100 mmol / 1. With a higher density of the fixed NSKFs on the surface, for example 10,000 to 1,000,000 per 100 μm ² , the density of the NSKF-primer complexes required for optical detection can be achieved during the primer hybridization. The hybridization conditions (eg temperature, time, buffer, primer concentration) should be selected so that the primers only bind to a part of the immobilized NSKFs. If the surface of a solid phase (eg silicone or glass) is used for immobilization, a blocking solution is preferably applied to the surface before step (a) in each cycle, which serves to avoid non-specific adsorption of NTs ^* on the surface. For example, an albumin solution (BSA) with a pH between 8 and 10 fulfills these conditions for a blocking solution.

4. Choice of polymerase

In principle, all DNA-dependent DNA polymerases without 3 '-5' exonuclease activity (DNA replication "1992 Ed. A. Kornberg, Freeman and Company NY), for example modified T7 polymerase of the type" Sequenase Version 2, are suitable as polymerases "

(Amersham Pharmacia Biotech), 3 '-5' exonuclease-free Klenow fragment of DNA polymerase I (Amersham Pharmacia Biotech), polymerase beta of various origins (Animal Cell DNA Polymerases "1983, Fry M., CRC Press Inc., commercially available from Chimerx) thermostable polymerases such as Taq polymerase (GibcoBRL), proHATM polymerase

(Eurogentech).

Polymerases with 3 '-5' exonuclease activity can be used (eg Klenow fragment of E. coli polymerase I), provided reaction conditions are selected, suppress the existing 3 '-5' exonuclease activity, such as a low pH Value (pH 6.5) for the Klenow fragment (Lehman and Richardson, J. Biol. Chem. 1964 v.239 p.233) or addition of NaF for the installation reaction. Another possibility is to use NTs ^* with a phosphorothioate compound (Kunkel et al. PNAS 1981, v.78 p.6734). Built-in NTs ^{* are} not attacked by the 3 '-5' exonuclease activity of the polymerase. All these types of polymerases are referred to below as "polymerase". 4th 5 chemistry

4.5.1 General NT structure

Different NT * s (preferably 2'-deoxy nucleotide triphosphates) which carry a substituent at their 3 'position of the ribose ring can be used in the methods according to the invention. This substituent, alone or together with the fluorescent dye, can lead to the termination of the incorporation reaction and can be split off from the nucleotide under mild conditions. A fluorescent dye which is characteristic of the respective NT * is coupled to these substituents, so that the substituent also assumes the role of a linker between the nucleotide and the fluorescent dye. The fluorescent dye is preferably coupled to this linker by a bond which can be cleaved under mild conditions.

"Mild conditions" are understood to mean cleavage conditions which neither lead to denaturation of the primer-nucleic acid complex, nor to the cleavage of its individual components.

Formulas (1-3) are examples of the reversible cleavable terminators:

1) NT-3 '-O-S (l) -F

2) NT-3'-0-S (2) -N-F

3) NT-3 '-0-S (2) -N-L-F

NT-3'-0 - represents the 2 'deoxy nucleoside triphosphate residue.

S (l) - represents a substituent (Formula 1) that can be split off from the NT * under mild conditions. A fluorescent dye (F) is coupled to these substituents.

S (2) -N - represents another substituent (formula 2 and 3), which can be split off from the NT * under mild conditions. This substituent is linked to the fluorescent dye (F) by a group (N) which is cleavable under mild conditions. The fluorescent dye can be coupled directly to the cleavable group (formula 2) or by a further linker (L) (formula 3).

Examples of NT * structures, NT * synthesis, polymerase selection for the incorporation reaction, reaction conditions of the NT * incorporation reaction and cleavage reaction are described in (Kwiatkoxski WO patent 01/25247, Kwiatkowski US patent 6,255,475, Conard et al U.S. Patent 6,001,566, Dower (U.S. Patent 5,547,839), Canard et al. (U.S. Patent 5,798,210), Rasolonjatovo

(Nucleosides & Nucleotides 1999, v.18 p.1021), Metzker et al.

(NAR 1994, v.22, p.4259), Welch et al. (Nucleosides & Nucleotides 1999, v.18, p.197).

4.5.2 Markers, fluorophores

Each nucleotide is marked with a characteristic marker (F). The marker is a fluorescent dye. The choice is not restricted if the dye meets the following requirements:

a) The detection apparatus used must have this marker as the only molecule bound to DNA under mild conditions

(preferably reaction conditions) can identify. The dyes preferably have great photostability. Their fluorescence is preferably not quenched by the DNA or only to a minor extent.

b) The dye bound to the NT must not cause an irreversible disturbance of the enzymatic reaction.

c) NTs ^* marked with the dye must be incorporated into the nucleic acid chain by the polymerase. d) When labeled with different dyes, these dyes should not have any significant overlaps in their emission spectra.

For example, some fluorophores which can be used in the context of the present invention are compiled with structural formulas in "Handbook of Fluorescent Probes and Research Chemicals" 6th ed. 1996, R.Haugland, Molecular Probes. According to the invention, the following classes of dyes are preferably used as markers: cyanine dyes and their derivatives (for example Cy2, Cy3, Cy5, Cy7 Amersham Pharmacia Biotech, Waggoner US Pat. No. 5,268,486), rhodamines and their derivatives (for example TAMRA, TRITC, RG6, R110 , ROX, Molecular Probes, see manual), xanthene derivatives (e.g. Alexa 568, Alexa 594, Molecular Probes, Mao et al. US Pat. No. 6,130,101) and porphyrins (Porphyrin-Systems, Germany). These dyes are commercially available.

Depending on the spectral properties and the equipment available, appropriate dyes can be selected. The dyes are bound to the NT * via a cleavable linker. The dyes can be attached to the linker e.g. via thiocyanate or ester linkage ("Handbook of Fluorescent Probes and Research Chemicals" 6th ed. 1996, R.Haugland, Molecular Probes, Jameson et al. Methods in Enzymology 1997 v.278 p.363, Waggoner Methods in Enzymology 1995 v.246 p.362)

4.5.3 Cleavable bond between the nucleotide and the substituent, cleavage.

The substituent leading to the termination is coupled to the NT by a bond which can be split under mild conditions.

Examples of these compounds are esters and acetals.

The esters are preferably cleaved in the basic pH Range (e.g. 9 to 11). Acetals are split in the acidic range (for example between 3 and 4).

Esters can also be eliminated enzymatically by polymerases or esterases.

In a preferred embodiment of the invention, the substituent is split off together with the fluorescent dye in one step.

4.5.4 Cleavable bond between the substituent and the fluorescent dye, cleavage.

In another preferred embodiment of the invention, the fluorescent dye is coupled to the substituents by a group which is cleavable under mild conditions.

The group mentioned preferably belongs to chemically or enzymatically cleavable or photolabile compounds. Ester, thioester, disulfide compounds and photolabile compounds are particularly suitable as a cleavable connection between the substituent and the fluorescent dye.

As examples of chemically cleavable groups, ester, thioester and disulfide compounds are preferred (^ Chemistry of protein conjugation and crosslinking "" Shan S. Wong 1993 CRC Press Inc., Herman et al. Method in Enzymology 1990 v. 184 S. 584, Lomant et al. J. Mol. Biol. 1976 v.104 243, "Chemistry of carboxylic acid and esters" S.Patai 1969 Interscience Publ.). Examples of photolabile compounds can be found in the following references: "Protective groups in organic synthesis" 1991 John Willey & Sons, Inc., V. Pillai Synthesis 1980 Sl, V. Pillai Org. Photochem. 1987 v.9 p.225, dissertation i ^ New photolabile protective groups for light-controlled oligonucleotide synthesis "" H. Giegrich, 1996, Konstanz, dissertation ξξNew photolabile protective groups for light-controlled oligonucleotide synthesis "SM Bühler, 1999, Constance).

The cleavage step is present in every cycle and must take place under mild conditions so that the nucleic acids are not damaged or modified.

The cleavage preferably proceeds chemically (for example in a mild acidic or basic environment for an ester compound or by adding a reducing agent, for example dithiothreitol or mercaptoethanol (Sigma) when cleaving a disulfide compound), or physically (for example by illuminating the surface with light a certain wavelength for the cleavage of a photolabile group, dissertation ξ§New photolabile protecting groups for the light-controlled oligonucleotide synthesis ""^" H. Giegrich, 1996, Constance).

In this embodiment, after the detection, the fluorescent dye is first split off and only then is the substituent which is coupled to the 3 'position and leads to the termination.

4.5.5 Color coding scheme, number of dyes Each NT * must be clearly marked with a characteristic dye. A cycle can preferably be carried out with:

a) four differently marked NT * s b) two differently marked NT * s c) one marked NT * d) two differently marked NT * s and two unmarked NTs,

(other combinations should also be obvious to a specialist) i.e.

a) You can mark all 4 NTs with different dyes and use all 4 at the same time in the reaction. The sequencing of a nucleic acid chain is achieved with a minimum number of cycles. However, this variant of the invention places high demands on the detection system: 4 different dyes have to be identified in each cycle.

b) To simplify the detection, a label with two dyes can be selected. Two pairs of NTs ^{* are} formed, each marked differently, eg A and G are marked "X", C and U are marked "Y". Two differently labeled NTs ^* are used simultaneously in the reaction in one cycle (s), for example C ^* in combination with A ^* , and U ^* and G ^* are then added in the subsequent cycle (n + 1).

c) You can also use only a single dye to mark all 4 NTs ^* and use only one NT ^* per cycle.

d) In a technically simplified embodiment, two differently marked NT ^* s are used per cycle and two unmarked NTs (so-called 2NT * s / 2NTs method). This embodiment can be used to determine variants (eg mutations, or alternatively spliced genes) of an already known sequence.

Under reaction conditions, the incorporation of NT ^* s into the NSKFs preferably takes place in such a way that a labeled NT ^{* is} incorporated in more than 50% of the NSKFs involved in the sequencing reaction, preferably in more than 90%. This is due to the fact that the reaction takes place very slowly on some nucleic acid chains. An installation of the NTs ^* at every complementary position in each cycle is aimed for, but is not necessary because only the successful installation reactions are detected and evaluated; a delayed reaction in the subsequent cycle does not lead to a sequencing error.

The same polymerase is preferred for all NTs ^* used. However, different polymerases can also be used for different NTs ^* .

4.6 Detection equipment

Individual molecules on a surface can be examined using various methods. Several methods are known: e.g. AtomForce microscopy, electron microscopy, near-field fluorescence microscopy, wide-field fluorescence

Microscopy, TIR microscopy etc. (Science 1999 v.283 1667, Unger et al. BioTechniques 1999 v.27 p.1008, Ishijaima et al. Cell 1998 v.92 p.161, Dickson et al. Science 1996 v.274 P.966, Xie et al. Science 1994 v.265 p.361, Nie et al. Science 1994 v.266 p.1018, Betzig et al. Science 1993 v.262 p.1422).

According to the invention, fluorescence signals of individual NTs ^* built into the nucleic acid chain are preferably measured using a wide-field fluorescence microscope (epifluorescence) or a laser scanning microscope (epifluorescence) or a TIRF microscope (Total Internal Reflection Fluorescence Microscope).

Various variants of the construction of such an apparatus are possible (Weston et al. J. Chem. Phys. 1998 v.109 p.7474, Trabesinger et al. Anal. Chem. 1999 v.71 p.279, Adachi et al. Journal of Microscopy 1999 v.195 p.125, Unger et al. BioTechniques 1999 v.27 p.1008, Ishijaima et al. Cell 1998 v.92 p.161, Dickson et al. Science 1996 v.274 p.966, Tokunaga et al. Bichem.Biophys.Res.Com. 1997 v.235 p.47, "Confocal Laser Scanning Microscopy" 1997 Ed. Sheppard, BIOS Scientific Publishers, "New Techniques of optical microscopy and microspectroscopy" 1991 Ed. R.Cherry CRC Press, Inc., "Fluorescence microscopy" 1998 2nd ed. Herman BIOS Scientific Publishers, "Handbook of biological confocal microscopy" 1995 J. Pavley Plenum Press). Differences in their concrete structure result from the variation of their individual parts. The device for the excitation light can function, for example, on the basis of a laser, a lamp or diodes. For the detection device can serve both CCD cameras and PMT. For other examples of technical details see ("Confocal Laser Scanning Microscopy" 1997 Ed. Sheppard, BIOS Scientific Publishers, "New Techniques of optical microscopy and microspectroscopy" 1991 Ed. R.Cherry CRC Press, Inc., "Fluorescence microscopy" 1998 2. ed. Herman BIOS Scientific Publishers, "Handbook of biological confocal microscopy" 1995 J. Pavley Plenum Press). It is not the object of this invention to list all possible technical variants of a detection device. The basic structure of a suitable apparatus is explained in a diagram in FIG. 8. It consists of the following elements:

Light source to excite fluorescence (1)

Light guiding part (2)

Scan table (3)

Spectra selection device (4)

Detection device (5)

Computers with control and analysis functions (6)

These elements of the apparatus can be purchased commercially (microscope companies: Zeiss, Leica, Nikon. Olympus).

In the following, for example, a combination of these elements suitable for the detection of individual molecules will be presented:

Wide field fluorescence microscope Axioplan 2 (Zeiss) with

Mercury vapor lamp

Objective Planeofluar lOOx, NA 1.4 (Zeiss)

Camera Photometrix or AxioCam (Zeiss)

Computer with software for control and analysis

The procedure for detection will be explained below. Please note the general rules of fluorescence microscopy ("Confocal Laser Scanning Microscopy") 1997 Ed. Sheppard, BIOS ^' Scientific Publishers, "New Techniques of optical microscopy and microspectroscopy" 1991 Ed. R.Cherry CRC Press, Inc., "Fluorescence microscopy" 1998 2nd ed. Herman BIOS Scientific Publishers, "Handbook of biological confocal microscopy" 1995 J. Pavley Plenum Press).

The detection comprises the following phases:

1) Preparation for detection

2) Execution of a detection step in each cycle, each detection step running as a scanning process and comprising the following operations: a) adjusting the position of the lens (X, Y axis), b) adjusting the focal plane (Z axis), c) detecting the Signals of individual molecules, assignment of the signal to NT ^* and assignment of the signal to the respective NSKF, d) shift to the next position on the surface.

The signals from NTs ^* built into the NSKFs are registered by scanning the surface. The scanning process can be carried out in various ways ("Confocal Laser Scanning Microscopy" 1997 Ed. Sheppard, BIOS Scientific Publishers, "New Techniques of optical microscopy and microspectroscopy" 1991 Ed. R.Cherry CRC Press, Inc., "Fluorescence microscopy" 1998 2. ed. Herman BIOS Scientific Publishers, "Handbook of biological confocal microscopy" 1995 J. Pavley Plenum Press). For example, a discontinuous scanning process is selected. The objective is moved step by step over the surface (FIG. 8a), so that a two-dimensional image (2D image) is created from each surface position (FIG. 8b, c).

This 2D image can be created using various methods: for example by laser scanning a position of the microscope field (laser scanning microscopy) or by Camera recording at one position (see manuals of microscopy). The detection of individual molecules with a CCD camera is described as an example.

The detection is explained schematically using the example of sequencing a 1Mb piece of DNA:

1) Preparation for detection:

At the beginning it is determined how many NSKF sequences have to be analyzed in order to reconstruct the original sequence. In the case of a reconstruction according to the shotgun method ("Automated DNA sequencing and analysis" p. 231 ff. 1994 M. Adams et al. Academic Press, Huang et al. Genom Res. 1999 v.9 p. 868, Huang Genomics 1996 v.33 p.21, Bonfield et al. NAR 1995 v.23 p.4992, Miller et al. J. Comput. Biol. 1994 vl p.257) the following factors play a role: 1) Each NSKF is used in the Sequencing determined a sequence of approximately 300-500 NTs. 2) The total length of the sequence to be analyzed is important. 3) A certain degree of redundancy must be achieved in the sequencing in order to increase the accuracy and to correct any errors. Overall, approximately 10 to 100 times the amount of raw sequences is required to reconstruct most of the original sequence, i.e. in this one Mb example, 10 to 100 Mb raw sequence data is needed. With an average sequence length of 400 bp per NSKF, 25,000 to 250,000 DNA fragments are required.

2) Performing a detection step in every cycle

The positions of the NSKFs must be determined for sequencing so that one has a basis for the assignment of the signals. Knowing these positions allows a statement to be made as to whether the signals of individual molecules come from built-in NTs ^* or from randomly bound NTs ^* . These positions can be identified using various methods. In a preferred embodiment, the positions of bound NSKF primer complexes are identified during sequencing. This makes use of the fact that the signals from the NTs ^* built into the nucleic acid chain always have the same coordinates. This is ensured by fixing the nucleic acid chains. The non-specifically bound NTs ^* bind randomly at different points on the surface.

To identify the positions of fixed NSKFs, the signals are checked for agreement of their coordinates from several successive cycles. This can e.g. at the beginning of the sequencing. The matching coordinates are evaluated as coordinates of the DNA fragments and saved.

The scan system must be able to scan the surface reproducibly over several cycles. X, Y and Z axis settings at each surface position can be controlled by a computer. The stability and reproducibility of the setting of lens positions in each scanning process determine the quality of the detection and thus the identification of the signals of individual molecules.

a) Setting the position of the lens (X, Y axis)

The mechanical instability of the commercially available scanning tables and the low reproducibility of repeated adjustment of the same X, Y positions make it difficult to precisely analyze the signals of individual molecules over several cycles. There are many ways to improve the coincidence of the coordinates with repeated settings or to check possible deviations. A control option is given as an example. After a rough mechanical adjustment of the lens position, a control image is made of a pattern firmly attached to the surface. Even if the mechanical adjustment is not has exactly the same coordinates (deviations of up to 10 μm are quite possible), you can make a correction using an optical control. The control image from the sample serves as a coordinate system for the image with signals from built-in NTs ^* . A prerequisite for such a correction is that no further surface movements are made between these two images. Signals from individual molecules are placed in relation to the pattern, so that an X, Y deviation in the pattern position means the same X, Y deviation in the position of the signals of individual molecules. The control image of the pattern can be taken before, during or after the detection of individual molecules. Such a control picture must be made accordingly with each setting on a new surface position.

b) Setting the focus plane (Z axis)

The surface is not absolutely flat and has various unevenness. This changes the surface-lens distance when scanning neighboring locations. These differences in distance can lead to individual molecules leaving the focal plane and thus avoiding detection.

For this reason, it is important that when scanning the surface, a reproducible adjustment of the focal plane is achieved at every lens position.

There are various ways to reproducibly adjust the focus level. For example, the following method can be used: Since the excitation of individual molecules can lead to the extinction of their fluorescence, a marker is applied to the surface, which serves to adjust the focal plane. The signals of individual molecules are then detected. The marker can be of any nature (eg dye or pattern), but must not interfere with the detection and the reaction. c) Detection of the signals of individual molecules, assignment of the signal to NT ^* and assignment of the signal to the respective NSKF.

The two-dimensional image of the reaction surface generated with the aid of the detection system contains the signal information from NT ^* s built into the NSKFs. Before further processing, these must be extracted from the total amount of image information using suitable methods. The algorithms required for scaling, transforming and filtering the image information are part of the standard repertoire of digital image processing and pattern recognition (Haberäcker P. "Practice of digital image processing and pattern recognition". Hanser-Verlag, Munich, Vienna, 1995; Galbiati LJ "Machine vision and digital image processing fundamentals ". Prentice Hall, Englewood Cliffs, New Jersey, 1990). The signal extraction is preferably carried out via a gray value image, which depicts the brightness distribution of the reaction surface for the respective fluorescence channel. If several nucleotides with different fluorescent dyes are used in the sequencing reaction, a separate gray value image can first be generated for each fluorescence-labeled nucleotide (A, T, C, G or U). In principle, two methods can be used for this:

1. A gray value image is generated for each fluorescence channel by using suitable filters (Zeiss filter sets).

2. The relevant color channels are extracted from a recorded multichannel color image with the aid of a suitable algorithm by an image processing program and are individually processed further as a gray-scale image. A channel-specific color threshold algorithm is used for channel extraction. Individual gray value images 1 to N are thus initially created from a multi-channel color image. These images are defined as follows: GB _N = (s (x, y)) single-channel gray value image N = {l, ..., number of fluorescence channels}.

M = {θ, 1, ..., 255} gray value set

S = (s (x, y)) image matrix of the gray value image x = 0, 1, ..., L-1 image lines y = 0, 1, ..., R-1 image columns

(x, y) location coordinates of a pixel s (x, y) e M gray value of the pixel.

The relevant image information is then extracted from this amount of data by a suitable program. Such a program should implement the following steps:

For GB _X to GB _N :

I. Preprocessing of the image, for example, if necessary, reducing the image noise caused by the digitization of the image information, for example by gray value smoothing.

II. Checking each pixel (x, y) of the gray-scale image as to whether this point fulfills the properties of a fluorescence point in connection with the immediate and further distant neighboring pixels surrounding it. These properties depend, among other things, on the detection equipment used and the resolution of the gray-scale image. For example, you can display a typical distribution pattern of brightness intensity values over a matrix surrounding the pixel. The image segmentation methods that can be used for this range from simple threshold value methods to the use of neural networks.

If a pixel (x, y) fulfills these requirements, then a comparison with the coordinates of NSKFs identified in previous sequencing cycles follows. If there is a match, the signal is assigned to the nucleotide resulting from the respective fluorescence channel NACF. Signals with mismatched coordinates are evaluated as background signals and rejected. The signals can be analyzed in parallel with the scanning process.

In an exemplary embodiment, an 8-bit gray value image with a resolution of 1317 x 1035 pixels was used. In order to reduce the changes in the image caused by digitization, the overall image was first preprocessed: the average value of the brightness of its 8 neighbors was assigned to each pixel. With the selected resolution, this results in a typical pattern for a fluorescence dot of a central pixel with the greatest brightness value and neighboring pixels with brightness decreasing on all sides. If a pixel met these criteria and the centrifugal drop in brightness exceeded a certain threshold value (to exclude fluorescence spots that were too weak), this central pixel was evaluated as the coordinate of a fluorescence spot.

d) Moving the lens to the next position on the surface. After detecting the signals of individual molecules, the lens is positioned over another position on the surface.

Overall, for example, a sequence of recordings can be made with the control of the X, Y position, the adjustment of the focal plane and with the detection of individual molecules at each new lens position. These steps can be controlled by a computer.

4.7 Timing of the procedural steps

The scanning process and the biochemical reaction take a certain amount of time. If you switch these processes one after the other, you can achieve an optimal performance of the equipment. In a preferred embodiment, the response to performed two separate surfaces.

As an example, a surface with bound NSKF primer complexes can be separated into two spatially isolated parts, so that reactions on these two parts can take place independently of one another. In another example, NSKFs can also be immobilized on two separate surfaces from the outset.

The reaction is then started. The principle is that while the reaction and washing steps take place on part of the surface, the second part is scanned. This enables the analysis to run continuously and the speed of sequencing to be increased.

The number of surfaces on which the reaction takes place can also be greater than 2. This makes sense if the reaction occurs as a time-limiting step, i.e. the detection of the signals on the surface is faster than the reaction and washing steps. In order to adapt the total duration of the reaction to the detection duration, each individual step of the reaction can take place on a single surface with a time delay compared to the next surface.

The invention is illustrated below using examples.

Examples

Example 1:

Sequence analysis with 4 marked NTs ^*

In a preferred embodiment of the invention, all four NTs ^* used in the reaction are labeled with fluorescent dyes.

1A. Reconstruction of the original sequences according to the shotgun principle ("Automated DNA sequencing and analysis" p. 231 ff. 1994 M. Adams et al. Academic Press, Huang et al. Genom Res. 1999 v.9 p. 868, Huang Genomics 1996 v.33 p.21, Bonfield et al. NAR 1995 v.23 p.4992, Miller et al. J. Comput. Biol. 1994 vl p.257). (This principle is particularly useful when analyzing new, unknown sequences.)

1A-1 sequencing of a long piece of DNA

In the following, the sequencing of long nucleic acid chains will be shown schematically on the basis of the sequencing of a 1Mb piece of DNA (FIG. 1). The sequencing is based on the shotgun principle ("Automated DNA sequencing and analysis" p. 231 ff. 1994 M. Adams et al. Academic Press, Huang et al. Genom Res. 1999 v.9 p.868, Huang Genomics 1996 v .33 p.21, Bonfield et al. NAR 1995 v.23 p.4992, Miller et al. J.Comput. Biol. 1994 vl p.257). The material to be analyzed is prepared for the sequencing reaction by breaking it down into fragments of preferably 50 to 1000 bp in length. Each fragment is then provided with a primer binding site and a primer. This mixture of different DNA fragments is now fixed on a flat surface. The unbound DNA fragments are removed by a washing step. The sequencing reaction is then carried out on the entire reaction Surface performed. To reconstruct a 1 Mb long DNA sequence, the sequences of NSKFs should preferably be longer than 300 NTs, on average approx. 400 bp. Since only one marked NT ^{* is} installed per cycle, at least 400 cycles are required for sequencing.

Overall, approximately 10 to 100 times the amount of raw sequences is required to reconstruct the original sequence, i.e. 10 to 100 Mb. With an average sequence length of approx. 400 bp per NSKF, 25,000 to 250,000 DNA fragments are required in order to cover more than 99.995% of the total sequence.

The NSKF sequences determined represent a population of overlapping partial sequences which can be combined with commercially available programs to form the overall sequence of the NSK ("Automated DNA sequencing and analysis" p. 231 ff. 1994 M. Adams et al. Academic Press, Huang et al. Genom Res. 1999 v.9 p.868, Huang Genomics 1996 v.33 p.21, Bonfield et al. NAR 1995 v.23 p.4992, Miller et al. J. Comput. Biol. 1994 vl S. 257).

A-2 sequencing of gene products using the example of cDNA sequencing

In a preferred embodiment, several sequences can be analyzed in one approach instead of one sequence. The original sequences can be extracted from the raw data e.g. be reconstructed according to the shotgun principle.

First, NSKFs are created. For example, one can convert mRNA into a double-stranded cDNA and fragment this cDNA with ultrasound. These NSKFs are then provided with a primer binding site, denatured, immobilized and hybridized with a primer. It should be noted with this variant of the sample preparation that the cDNA molecules can represent incomplete mRNA sequences (Method in Enzymology 1999, v.303, p.19 and other articles in this volume, "cDNA library protocols" 1997 Humana Press).

Another possibility for the generation of single-stranded NSKFs from mRNA is the reverse transcription of the mRNA with randomized primers. Many relatively short antisense DNA fragments are formed (Zhang-J et al. Biochem. J. 1999 v.337 p.231, Ledbetter et al. J.Biol.Chem. 1994 v.269 p.31544, Rolls et al Anal.Biochem. 1993 v.208 p.264, Decraene et al. Biotechniques 1999 v.27 p.962). These fragments can then be provided with a primer binding site (see above). Further steps correspond to the processes described above. With this method, complete mRNA sequences (from the 5 'to the 3' end) can be analyzed, since the randomized primers bind over the entire length of the mRNA.

Immobilized NSKFs are analyzed using one of the sequencing embodiments listed above. Since mRNA sequences have significantly fewer repetitive sequences than, for example, genomic DNA, the number of signals detected by the built-in NTs ^* from an NSKF can be less than 300 and is preferably between 20 and 1000. The number of NSKFs that need to be calculated is calculated following the same principles as a shot-shot reconstruction of a long sequence.

The original gene sequences are reconstructed from NSKF sequences according to the principles of the shotgun method.

This method allows the sequencing of many mRNAs without prior cloning.

B. Analysis of sequence variants

Confirming a previously known sequence or detecting variants of this sequence places far less demands on the length and redundancy of the NSKF sequences determined. Sequence editing is also easier in this case. The full sequence does not need to be reconstructed. Rather, the NSKF sequences are assigned to the full sequence using a commercially available program and any deviations are detected. Such a program can e.g. BLAST or FASTA algorithm are based ("Introduction to computational Biology" 1995 M.S. Waterman Chapman & Hall).

The sequence to be analyzed is converted into NSKFs using one of the methods mentioned above. These NSKFs are sequenced using the method according to the invention, wherein both a uniform primer and a uniform primer binding site as well as different, sequence-specific primers and natural ones which occur in the overall sequence to be investigated

Can use primer binding sites. The determined sequences of NSKFs are then not assembled using the shotgun method, but are compared with the reference sequence and in this way assigned to their positions in the full sequence. This can be genomic or cDNA sequences.

In contrast to a shotgun reconstruction, the analysis of a sequence variant requires considerably less raw sequence data. For example, 5 to 10 times the raw sequence quantity can be sufficient to restore a new variant of a full sequence. With the shotgun process is for a Recovery ^' requires a 10 to 100-fold amount of raw sequences ("Automated DNA sequencing and analysis" p. 231 ff. 1994 M. Adams et al. Academic Press, Huang et al. Genom Res. 1999 v.9 p. 868 , Huang Genomics 1996 v.33 p.21, Bonfield et al. NAR 1995 v.23 p.4992, Miller et al. J. Comput. Biol. 1994 vl p.257).

The length of the NSKF sequences determined should be sufficient for an unambiguous assignment to a specific position in the reference sequence, for example sequences with a length of 20 NTs (e.g. from non-repetitive sections in the human genome) can be clearly identified. Longer sequences are required for the comparative analysis of the repetitive sections. The exact length of the sequences depends on the task. The length of the NSKF sequences determined is preferably more than 20 NTs when analyzing non-repetitive sections. For the analysis of the repetitive sections, it is preferably over 500 NTs.

The objectives when sequencing new variants of an already known full sequence can be very different. A comparison of the newly determined sequence with the known full sequence / reference sequence is usually sought. The two sequences can originate from species that are evolutionarily different from one another. Different parameters of the composition of these two sequences can be compared. Examples of such an analysis are: mutation or polymorphism analyzes and the analysis of alternative spliced gene products.

In the following, a comparison of the sequence to be examined with a reference sequence without prior reconstruction of the sequence to be analyzed is to be considered schematically and as an example. Such a Comparison can e.g. ^' B. serve for mutation or SNP analysis.

B-1

A long sequence to be analyzed, e.g. 1 Mb, is shared in NSKFs using one of the above methods. These NSKFs are sequenced using uniform primers using the method according to the invention. The sequences determined from each individual NSKF are compared directly with the reference sequence. The reference sequence serves as the basis for the assignment of determined NSKF sequences, so that the time-consuming reconstruction using the shotgun method is not necessary. The length of the NSKF sequences determined is preferably more than 20 NTs when analyzing non-repetitive sections. For the analysis of the repetitive sections, it is preferably over 500 NTs. The number of NSKFs to be analyzed depends on the total length of the sequence to be examined, the average length of the NSKF sequences and the necessary accuracy of the sequencing. With an average length of the determined NSKF sequence of 100 NTs, a total length of the sequence to be examined of 1 Mb and an accuracy that corresponds to the raw sequence determination (i.e. each location should be sequenced only once if possible), e.g. approx. 5 times the amount of raw sequences, i.e. 5 Mb because the NSKFs are randomly distributed over the entire sequence. A total of 50,000 NSKFs must be analyzed to cover more than 99% of the total route.

The determined NSKF sequences are then assigned to the full sequence using a commercially available program and any deviations are detected. Such a program can be based on, for example, BLAST or FASTA algorithm ("Introduction to computational Biolögy "1995 MS Waterman Chapman & Hall)

Example 2:

Sequence analysis with 2 marked NTs ^* and 2 unmarked NTs (2NTs ^* / 2NTs method).

In another embodiment, 2 modified NTs ^* and 2 unmodified NTs are used for the analysis of the sequences.

This method is particularly suitable for analyzing the sequence variants (e.g. SNP or mutation analysis) and requires knowledge of a reference sequence. The full sequence is not reconstructed, but the determined sequences are assigned to the reference sequence using a program and any deviations are registered. Such a program can e.g. based on the BLAST or FASTA algorithm ("Introduction to computational Biology" 1995 M.S. Waterman Chapman & Hall).

This embodiment is based on the principle that a sequence of 2 signals (marked NT ^* s) can contain enough information to identify a sequence. The determined sequence is compared with the reference sequence and assigned to a specific position, for example:

ACCAAAACACCC - determined sequence (dCTP ^* and dATP ^* are marked)

ATCATCGTTCGAAATATCGATCGCCTGATGCC - reference sequence

A-C C-AAA-A-C-A-C-CC (assigned determined sequence)

ATCATCGTTCGAAATATCGATCGCCTGATGCC (reference sequence) The unknown variant of the reference sequence to be analyzed is prepared for sequencing as described above (NSK is converted into NSKFs, these are ligated with PBS, then hybridized with a primer and immobilized on the reaction surface). NSKFs prepared in this way are sequenced using the 2NTs ^* / 2NTs method. NSKF sequences are obtained, each NSKF sequence being a sequence of 2NTs ^* . In order to enable the determined sequence to be clearly assigned to a known reference sequence, this sequence must be long enough. The length of the NSKF sequences determined is preferably more than 40 NT ^* s. Since 2 marked NTs ^{* represent} only part of the sequence, the total length of the complementary strand synthesized is approximately twice as long as the sequence of the detected NTs ^* (with 40 detected NTs ^* , the total length is z, e.g. 80 NTs on average).

4 nucleotides are required to synthesize a complementary strand. Since the NTs ^* marked with a fluorescent dye appear as semiterminators in the present invention, ie termination occurs only when modified NTs ^{* are available} , unmodified NTs must be added to the reaction in an additional step in each cycle. The exact position of this step in the cycle can vary. It is important that the marked NTs ^* and the unmodified NTs are used separately.

A cycle in this embodiment may look as follows, for example:

a) adding a solution with modified NTs ^* and polymerases to the surface with the NSKFs provided b) incubating the immobilized nucleic acid chains with this solution under conditions which are suitable for extending the complementary strands by an NT c) washing d) Detection of the signals from individual, modified NTs ^* molecules built into the newly synthesized strands complementary to the NSKFs e) removal of the label and the terminating group in the built-in nucleotides f) washing g) addition of 2 unmodified NTs and polymerases h) washing ,

This 2NT ^* s / 2NTs method is suitable, for example, for the SNP analysis of a genomic stretch of a gene or for double-stranded cDNA analysis. It is based on the following principles:

1) The genetic information in each of the two complementary DNA strands is identical, so that missing information in one strand can be completed by the information from the other strand.

2) By combining certain pairs of labeled NTs ^* , you can obtain complete information from a double-stranded DNA with only 2 NTs ^* . Permissible combinations of NT ^* s in this embodiment are: A ^* C ^* ; A ^* G ^* ; C ^* T ^* / C ^* U ^* ; G ^* T ^* / G ^* U ^* . The combination C ^* and U ^* is preferred.

3) An already known reference sequence serves as the basis for the analysis.

4) The NSKFs come from both strands of the NSK to be analyzed and • the NSKF sequences determined cover the entire length of the sequence to be analyzed.

The following example explains how the information is obtained from a double-stranded DNA fragment with only 2 labeled NTs ^* and how the differences from the original or non-mutated sequence (reference sequence / comparison sequence) can be determined. Sequences under (1) and (2) are identical except for one position (underlined). A ^* and C ^* are marked. 1) Sequence to be checked:

The sequence to be checked is sequenced using the 2NT ^* s / 2NTs method, so that a population of NSKF sequences (determined NSKF sequences (n)) is formed. The NSKF sequences determined contain information from each strand:

5'A-C C-AAA-A-C-A-C-CC3 '- NSKF sequence determined (i)

5 'ATCGTTCGAAATATCGATCGCCTG3' 3 'TAGCAAGCTTTATAGCTAGCGGAC5' 3 'A-CAA-C --- A-A-C-A-C --- C5' - determined NSKF sequence (i + 1)

2) Comparison sequence:

A comparison sequence (reference sequence) is required for analysis:

5 'ATTGTTCGAAATATCGATCGCCTG3' 3 'TAACAAGCTTTATAGCTAGCGGAC5'

3) Comparison sequence with adapted NSKF sequences determined:

With the help of a program, determined NSKF sequences are assigned to specific points in the comparison sequence and possible deviations are detected:

5'A-C C-AAA-A-C-A-C-CC3 '- NSKF sequence determined (i)

5 'ATTGTTCGAAATATCGATCGCCTG3' 3 'TAACAAGCTTTATAGCTAGCGGAC5' 3 'A-CAA-C A-A-C-A-C --- C5' - NSKF sequence determined (i + 1)

1t (single nucleotide mutation)

With this embodiment, a double-stranded nucleic acid can be examined for SNP or mutations. The NSKF sequences determined are compared with a reference sequence compared. The basic rules of comparing a partial sequence and a complete sequence in the analysis with only 2 marked NTs do not differ fundamentally from those that apply when comparing the sequences using all 4 marked NTs ^* . For more information, see Sequence comparison in mutation analysis and SNP analysis with 4NTs ^* (Example 1B).

Example 3: Analysis of gene expression

The basic principles of the sequencing reaction in gene expression analysis correspond to those of the sequencing reaction of long NSKs (Fig. 7). The basic principles for carrying out a reaction cycle (the choice of the NT * structure, the polymerase, the reaction conditions for the NT * incorporation reaction and the cleavage reaction) and for the detection of the signals from built-in NT * correspond to those in the process for sequencing long NSKs. The main differences between the two methods lie in the selection and preparation of materials and the processing of the data obtained.

Choice of material:

Gene products can come from various biological objects, e.g. of individual cells, cell populations, a tissue or of entire organisms. Biological fluids such as blood, sputum or cerebrospinal fluid can also serve as a source of the gene products. The methods for obtaining the gene products from the various biological objects can be found, for example, in the following literature sources: "Molecular cloning" 1989, Ed. Maniatis, Cold Spring Harbor Laboratory, "Method in Enzymology" 1999, v303, "cDNA library protocols" 1997, Ed. I. G. Cowell, Humana Press Inc.

Both the entirety of the gene products isolated and a part thereof selected by preselection can be used in the sequencing reaction. By preselection one can determine the amount of the to be analyzed Reduce gene products ^' . The preselection can be carried out, for example, by means of molecular biological methods such as, for example, PCR amplification, gel separation or hybridization with other nucleic acid chains ("Molecular cloning" 1989, Ed. Maniatis, Cold Spring Harbor Laboratory, "Method in Enzymology" 1999, v303, "cDNA library protocols "1997, Ed. IG Cowell, Humana Press Inc.)

The entirety of the gene products is preferably selected as the starting material.

Preparation of the material:

The aim of the preparation of the material is to form extensible gene product-primer complexes from the starting material that are bound to the surface. Whereby only a maximum of one primer should bind per gene product.

Priming site (PBS): Each gene product preferably has only one priming site.

A primer binding site is a sequence section which is intended to enable selective binding of the primer to the gene product.

Sections in the nucleic acid sequence that naturally occur in the sequences to be analyzed can serve as primer binding sites (e.g. polyA stretches in mRNA). A primer binding site can also be introduced into the gene product (Molecular cloning "1989, Ed. Maniatis, Cold Spring Harbor Laboratory," Method in Enzymology "1999, v303," cDNA library protocols "1997, Ed. IG Cowell, Humana Press Inc. ).

In order to simplify the analysis, it may be important to have a primer binding site that is as uniform as possible in all gene products. Then primers with a uniform structure can be used in the reaction. The composition of the primer binding site is not limits. Their length is preferably between 10 and 100 NTs. The primer binding site can carry a functional group, for example to bind the gene product to the surface. This functional group can be, for example, a biotin or digoxigenin group.

The nucleotide tailing of antisense cDNA fragments is described as an example of the introduction of a primer binding site into the gene products.

First, single-stranded cDNAs are synthesized from mRNAs. The result is a population of cDNA molecules that represent a copy of the mRNA population, so-called antisense cDNA. (Molecular cloning "1989, Ed. Maniatis, Cold Spring Harbor Laboratory," Method in Enzymology "1999, v303," cDNA library protocols "1997, Ed. IG Cowell, Humana Press Inc.). With a terminal deoxynucleotide transferase, one can do several ( for example between 10 and 20) attach nucleoside monophosphates to the 3 'end of this antisense cDNA, for example several adenosine monophosphates (called ((dA) n-tail). The resulting fragment is used to bind the primer, in this example one (dT ) n-primers used ("Molecular cloning" 1989 J. Sambrook et al. Cold Spring Harbor Laborotary Press, "Method in Enzymology" 1999 v.303, pp. 37-38).

Primer for the secretion reaction: This has the function of enabling starting at a single point in the gene product. It preferably binds to the primer binding site in the gene product. The composition and the length of the primer are not restricted. In addition to the start function, the primer can also perform other functions, such as creating a connection between the gene product-primer complexes and the reaction surface. Primers should be adapted to the length and composition of the primer binding site so that the primer enables the sequencing reaction to be started with the respective polymerase. The length of the primer is preferably between 6 and 100 NTs, optimally between 15 and 30 NTs. The primer can carry a functional group which is used, for example, to bind the primer to the surface, for example such a functional group is a biotin group (see section Immobilization). It should not interfere with sequencing. The synthesis of such a primer can be carried out, for example, with the 380 A Applied Biosystems DNA synthesizer or can be created as a custom synthesis from a commercial provider, for example MWG-Biotech GmbH, Germany.

Different primers can also be used, a defined primer set, or a primer mixture.

Before hybridization, the primer can be fixed to the fragments to be analyzed on the surface using various techniques or synthesized directly on the surface, for example according to (McGall et al. US Patent 5412087, Barrett et al. US Patent 5482867, Mirzabekov et al. US Patent 5981734, "Microarray biochip technology" 2000 M. Schena Eaton Publishing, "DNA Microarrays" 1999 M. Schena Oxford University Press, Fodor et al. Science 1991 v.285 p.767, Timofeev et al. Nucleic Acid Research (NAR) 1996 , v.24 p.3142, Ghosh et al. NAR 1987 v.15 p.5353, Gingeras et al. NAR 1987 v.15 p.5373, Maskos et al. NAR 1992 v.20 p.1679).

The primers are on the surface at a density between 10 to 100 microns per 100 ^2, 100 to 10,000 per 100 micron ^2, bound 10,000 to 1,000,000 per lOOμm ² or greater than 1,000,000 per 100 microns. ²

The primer or mixture of primers is incubated with gene products under hybridization conditions that selectively bind it to the primer binding site of each gene product. This primer hybridization (annealing) can take place before (1), during (2) or after (3) the binding of the gene products to the surface respectively. If gene products are present as double-stranded nucleic acids, they are denatured by heat before hybridization ("Molecular cloning" 1989 J.Sambrook et al. Cold Spring Harbor Laborotary Press). The optimization of the hybridization conditions depends on the exact structure of the primer binding site and the primer and can be done according to Rychlik et al. (NAR 1990 v.18 p.6409). In the following, these hybridization conditions are referred to as standardized hybridization conditions.

If the starting material has a poly-A stretch or a poly-dA stretch (for example mRNA, sense cDNA or antisense cDNA with (dA) _n -Tail), an oligo-dT primer can be used. However, a primer mixture consisting of 12 different primers with the following general structure 5 '(K) _n MN3' can also be used. Where (n) is between 10 and 50, preferably between 20 and 30. "K" stands for dT or dU, "M" and "N" each stand for dA, dT or dU, dC, dG (e.g. .5 " ^* -dTdTdTdTdTdTdTdTdTdT ₁₀ dTdTdTdTdTdTdTdTdTdT _2Q dAdG-3 ^{" *} ). Such a primer mixture enables an exact start of the sequencing reaction at the end of the polyA segment or the poly-dA segment (anchored primer).

Fixation of gene product-primer complexes to the surface (binding or immobilization of gene products):

The aim of the fixation (binding, immobilization) is to fix gene product-primer complexes on a suitable flat surface in such a way that a cyclic enzymatic sequencing reaction can take place. This can be done, for example, by binding the primer (see above) or the gene product to the surface.

The order of the steps in binding gene product-primer complexes can be variable:

4) The gene product-primer complexes can first be formed in a solution by hybridization (annealing) and then be bound to the surface.

5) Primers can first be bound to a surface and gene products can then be hybridized to the bound primers, producing gene product-primer complexes (gene products indirectly bound to the surface)

6) The gene products can first be bound to the surface (gene products bound directly to the surface) and in the subsequent step the primers can be hybridized to the bound gene products, producing gene product-primer complexes.

The gene products can therefore be immobilized on the surface by direct or indirect binding.

In this application, the surface and the reaction surface are to be understood as equivalent terms, unless explicitly referred to another meaning. The surface of a solid phase of any material serves as the reaction surface. This material is preferably inert towards enzymatic reactions and does not cause any interference with the detection. Silicon, glass, ceramics, plastics (e.g. polycarbonates or polystyrenes), metal (gold, silver, or aluminum) or any other material that meets these functional requirements can be used. The surface is preferably not deformable, since otherwise the signals are likely to be distorted upon repeated detection.

If a gel-like solid phase (surface of a gel) is used, this gel can be, for example, an agarose or polyacrylamide gel. The gel is preferably freely passable for molecules with a molecular mass below 5000 Da (for example, a 1 to 2% agarose gel or 5 to 15% polyacrylamide gel can be used). Such a gel surface has the advantage over other solid surfaces that there is a significantly lower non-specific binding of NT * s to the surface. By binding the gene product Primer complexes on the surface make it possible to detect the fluorescence signals from built-in NTs ^* . The signals from free NTs ^* are not detected because they do not bind to the material of the gel and are therefore not immobilized. The gel is preferably attached to a solid surface. This solid base can be silicone, glass, ceramic, plastic (for example polycarbonate or polystyrene), metal (gold, silver or aluminum) or any other material. The thickness of the gel is preferably not more than 0.1 mm. However, the gel thickness is preferably greater than the simple depth of field of the lens, so that NTs ^* bound non-specifically to the solid base do not get into the focal plane and are therefore detected. If the depth of focus is 0.3 μm, for example, the gel thickness is preferably between 1 μm and 100 μm. The surface can be produced as a continuous surface or as a discontinuous surface composed of individual small components (eg agarose beads). The reaction surface must be large enough to be able to bind the necessary number of gene products with the appropriate density. The reaction surface should preferably not be larger than 20 cm ² .

The different cycle steps require an exchange of the different reaction solutions above the surface. The reaction surface is preferably part of a reaction vessel. The reaction vessel is in turn preferably part of a reaction apparatus with a flow device. The flow device enables the solutions in the reaction vessel to be exchanged. The exchange can take place with a pump device controlled by a computer or manually. It is important that the surface does not dry out. The volume of the reaction vessel is preferably less than 50 μl. Ideally, its volume is less than 1 μl.

If the gene product-primer complexes are fixed on the surface via the gene products, this can for example by binding the gene products at one of the two chain ends. This can be achieved by appropriate covalent, affine or other bonds. Many examples of immobilization of nucleic acids are known (McGall et al. US Patent 5412087, Nikiforov et al. US Patent 5610287, Barrett et al. US Patent 5482867, Mirzabekov et al. US Patent 5981734, "Microarray biochip technology" 2000 M. Schena Eaton Publishing, "DNA Microarrays" 1999 M. Schena Oxford University Press, Rasmussen et al. Analytical Biochemistry v.198, p.138, Allemand et al. Biophysical Journal 1997, v.73, p.2064, Trabesinger et al. Analytical Chemistry 1999, v.71, p.279, Osborne et al. Analytical Chemistry 2000, v.72, p.3678, Timofeev et al. Nucleic Acid Research (NAR) 1996, v.24 p.3142, Ghosh et al NAR 1987 v.15 p.5353, Gingeras et al. NAR 1987 v.15 p.5373, Maskos et al. NAR 1992 v.20 p.1679). The fixation can also be achieved by a non-specific binding, for example by drying out the sample containing gene products on the flat surface. The gene products are bound to the surface at a density between 10 and 100 microns per 100 ^2, 100 to 10,000 per 100 micron ^2, 10,000 to 1000,000 per lOOμm. ²

The density of extensible gene product-primer complexes required for the detection is approx. 10 to 100 per 100 μm ² . It can be achieved before, during or after hybridization of the primers to the gene products.

Some binding methods are described in more detail below:

In one embodiment, the gene product-primer complexes are bound via biotin-avidin or biotin-streptavidin binding. Avidin or streptavidin is covalently bound on the surface, the 5 'end of the primer is modified with biotin. After hybridization of the modified primers with the gene products (in solution), these are fixed on the surface coated with avidin / streptavidin. The concentration of the gene product primer labeled with biotin Complexes as well as the time of incubation of this solution with the surface are chosen in such a way that a density suitable for sequencing is achieved.

In another embodiment, the primers suitable for the sequencing reaction are fixed on the surface using suitable methods before the sequencing reaction (see above). The single-stranded gene products, each with one primer binding site per gene product molecule, are thus incubated under hybridization conditions (annealing). They bind to the fixed primers and are thus bound to the surface (indirect binding). The concentration of the single-stranded gene products and the hybridization parameters (eg temperature, time, buffer) are selected so that a density of approximately 10 to 100 gene product-primer complexes capable of extension per 100 μm ^{2 is} achieved which is suitable for sequencing. After hybridization, unbound gene products are removed by a washing step. In this embodiment, a surface with a high primer density is preferred, for example approximately 1,000,000 primers per 100 μm ² or even higher, since the desired density of gene product-primer complexes is achieved more quickly, the gene products binding only to a part of the primers ,

In another embodiment, the gene products are bound directly to the surface (see above) and then incubated with primers under hybridization conditions. At a density of approx. 10 to 100 gene products per 100 μm ² , attempts will be made to provide all available gene products with a primer and to make them available for the sequencing reaction. This can be achieved by high primer concentration, for example 1 to 100 mmol / 1. With a higher density of the fixed gene products on the surface, for example 10,000 to 1,000,000 per 100 μm ² , the density of the gene product-primer complexes required for optical detection can be achieved during the primer hybridization. The hybridization conditions (e.g. temperature, time, buffer, Primer concentrate) ^' so that the primers bind only to a part of the immobilized gene products.

If the surface of a solid phase (eg silicone or glass) is used for immobilization, a blocking solution is preferably applied to the surface before step (a) in each cycle, which serves to avoid non-specific adsorption of NTs ^* on the surface. For example, an albumin solution (BSA) with a pH between 8 and 10 fulfills these conditions for a blocking solution.

Choice of polymerase:

The type of immobilized nucleic acid (RNA or DNA) used plays a decisive role in the choice of polymerase:

If RNA is used as a gene product (e.g. mRNA) in the sequencing reaction, commercially available RNA-dependent DNA polymerases can be used, e.g. AMV reverse transcriptase (Sigma), M-MLV reverse transcriptase (Sigma), HIV reverse transcriptase without RNAse activity. All reverse transcriptases must be largely free of RNAse activity ("Molecular cloning" 1989, Ed. Maniatis, Cold Spring Harbor Laboratory).

If DNA is used as a gene product (eg cDNA), all DNA-dependent DNA polyases without 3 "-5" exonuclease activity (DNA replication "1992 Ed. A. Kornberg, Freeman and Company NY) are suitable in principle as polymerases. , eg modified T7 polymerase of the "Sequenase Version 2" type (Amersham Pharmacia Biotech), Klenow fragment of DNA polymerase I without 3 '-5' exonuclease activity (Amersham Pharmacia Biotech), polymerase beta of various origins (animal cell DNA polymerases "1983, Fry M., CRC Press Inc., commercially available from Chimerx) thermostable polymerases such as Taq polymerase (GibcoBRL), proHA DNA polymerase (Eurogentec). detection:

As with the sequencing of long NSKs, the detection comprises the following phases:

1) Preparation for detection

Carrying out a detection step in each cycle, each detection step running as a scanning process and comprising the following operations: a) adjusting the position of the objective (X, Y axis), b) adjusting the focal plane (Z axis), c) detecting the signals of individual people Molecules, assignment of the signal to NT ^* and assignment of the signal to the respective gene product, d) shift to the next position on the surface.

The signals from NTs ^* built into the strands complementary to the gene products are registered by scanning the surface. The scanning process is carried out as with the sequencing of long NSKs. The lens is gradually moved over the surface, so that a two-dimensional image (2D image) is created from each surface position.

Preparation for detection:

At the beginning it is determined how many copies of the gene products are necessary for the expression analysis. Several factors play a role in this. The exact number depends e.g. on the relative presence of the gene products in the approach and on the desired accuracy of the analysis. The number of gene products analyzed is preferably between 1000 and 10,000,000. For highly expressed genes, the number of gene products analyzed can be low, e.g. 1000 to 10,000. When analyzing poorly expressed genes, it must be increased, e.g. to 100,000 or even further.

For example, 100,000 individual gene products are analyzed simultaneously. In this way, weakly expressed genes (e.g. with approx. 100 mRNA molecules / cell, what approx. 0.02% total mRNA corresponds) in the reaction with an average of 20 identified gene products.

2) Performing a detection step in each cycle is the same as in the sequencing of long NSKs. Genetic products are used instead of NSKFs.

Analysis:

The data obtained (short sequences) are compared using a program with known gene sequences. Such a program can e.g. based on a BLAST or FASTA algorithm ("Introduction to computational Biology" 1995 M.S. Waterman Chapman & Hall).

The choice of the method for material preparation determines, among other things, in which sections of the gene products the sequences are determined and to which strand (sense or antisense) they belong. For example, are determined when using the polyA stretches as primer binding sites in mRNA sequences from NTRs (non-translating regions). When using the method with antisense cDNA as a template, the sequences determined come, inter alia, from the protein-coding regions of the gene products.

In a preferred simple variant of the invention, the gene expression is only determined qualitatively. Only the fact that certain genes are expressed is important.

In another preferred embodiment, a quantitative determination of the relationships between individual gene products in the batch is of interest. It is known that the activity of a gene in a cell is represented by a population of identical mRNA molecules. Many genes are active in a cell at the same time and are expressed to different extents, which leads to the presence of many different mRNA populations with different strengths. The quantitative analysis of gene expression is discussed in more detail below:

For quantitative analysis of gene expression ^abundances' individual gene products are determined in the sequencing reaction. The products of highly expressed genes are more frequently represented in the sequencing reaction than the poorly expressed genes. After assigning the sequences to specific genes, the proportion of the sequences determined is determined for each individual gene. Genes with strong expression have a higher proportion of the total population of gene products than genes with weak expression.

The number of gene products analyzed is preferably between 1000 and 10,000,000. The exact number of gene products to be analyzed depends on the task. It can be low for highly expressed genes, e.g. 1000 to 10,000. When analyzing poorly expressed genes, it must be increased, e.g. to 100,000 or higher.

If, for example, 100,000 individual gene products are analyzed at the same time, weakly expressed genes such as e.g. approx. 100 mRNA molecules / cell (which corresponds to approx. 0.02% total mRNA), represented in the reaction with an average of 20 identified gene products,

The following method can be used as internal control of the hybridization, the immobilization and the sequencing reaction:

One or more nucleic acid chains with known sequences can be used as a control. The composition of these control sequences is not restricted, provided they do not interfere with the identification of the gene products. RNA control samples are used in the sequence analysis of the mRNA samples, and corresponding DNA control samples are used in the analysis of the cDNA samples. These samples are preferably carried along in all steps. she can be added after mRNA isolation, for example. In general, the control samples are prepared for sequence analysis in the same way as the gene products to be analyzed.

The control sequences are added to the gene products to be analyzed in known, fixed concentrations. Concentrations of the control samples can be different, these concentrations are preferably between 0.01% and 10% of the total concentration of the sample to be analyzed (100%). For example, if the concentration of the mRNA is 10ng / μl, the concentrations of control samples are between 1pg / μl and 1ng / μl.

In the quantitative analysis of gene expression, the general metabolic activity of the cells must also be taken into account, in particular when a comparison of the expression of certain genes under different external conditions is sought.

The change in the level of expression of a particular gene can occur as a result of the change in the transcription rate of that gene or as a result of a global change in gene expression in the cell. The expression of the so-called "house-keeping genes" can be analyzed to observe the metabolic states in the cell. In the absence of important metabolites, for example, the general level of expression in the cell is low, so that constitutively expressed genes also have a low level of expression.

In principle, all constitutively expressed genes can serve as "house-keeping genes". Examples include the transferrin receptor gene or the beta actin gene. The expression of these house-keeping genes thus serves as

Reference value for the analysis of the expression of other genes. The

Sequence determination and quantification of the expression of the

House-keeping genes are preferably part of the analysis program for gene expression.

As with the sequencing of long NSKs, one can carry out a sequencing reaction with 4 marked or 2 marked and 2 unmarked NT *.

Sequence analysis with 4 marked NTs ^* Sequence analysis with - 4 marked NTs ^*

One of the color coding schemes mentioned above is used. The number of NTs determined for each sequence from a gene product is between 5 and 100, ideally between 20 and 50. These sequences are compared with a sequence with known sequences in gene databases and assigned to corresponding genes. Such a program can e.g. based on the BLAST or FASTA algorithm ("Introduction to computational Biology" 1995 M.S. Waterman Chapman & Hall).

A cycle has the following steps: a) adding a solution with labeled nucleotides (NTs ^* ) and polymerase to immobilized nucleic acid chains, b) incubating the immobilized nucleic acid chains with this solution under conditions which are suitable for extending the complementary strands by one NT, c) Washing d) Detection of signals from individual molecules e) Removal of the label from the built-in nucleotides and that leading to the termination Substituents, f) washing.

ACCAAAACACCC - determined sequence (dCTP ^* and dATP ^* are marked)

ATCATCGTTCGAAATATCGATCGCCTGATGCC - reference sequence

A-C C-AAA-A-C-A-C-CC (assigned determined sequence)

ATCATCGTTCGAAATATCGATCGCCTGATGCC (reference sequence)

The ascertained sequences are preferably assigned to the reference sequence with the aid of a program. Such a program can e.g. based on the BLAST or FASTA algorithm ("Introduction to computational Biology" 1995 M.S. Waterman Chapman & Hall).

The gene products are prepared for sequencing as described above and sequenced using the 2NTs ^* / 2NTs method. Sequence sections are obtained from gene products, each sequence representing a sequence of 2NTs ^* . Known gene sequences serve as reference sequences. To be clear To enable the determined sequence to be assigned to a known reference sequence, this sequence must be long enough. The length of the sequences determined is preferably more than 20 NT ^* s. Since 2 marked NTs ^{* represent} only part of the sequence, the total length of the complementary strand synthesized is approximately twice as long as the sequence of the detected NTs ^* (with 20 detected NTs ^* , the total length is, for example, an average of 40 NTs).

A cycle in this embodiment may look as follows, for example:

a) adding a solution with modified NTs ^* and polymerases to the surface with the gene products provided b) incubating the immobilized nucleic acid chains with this solution under conditions which are suitable for extending the complementary strands by one NT, c) washing d) detecting the signals from individual, modified NTs ^* molecules built into the newly synthesized strands complementary to the gene products e) removal of the label and the terminating group in the built-in nucleotides f) washing g) addition of 2 unmodified NTs and polymerases h) washing.

The concentration of the NTs is preferably below 1 M, ideally below 10 μM.

Example 4:

A special embodiment of the method is the analysis of single nucleotide polymorphisms with sequence-specific primers.

In addition to Section 1 "Abbreviations and Definitions", the following terms are defined for this example:

Primer - In order to clarify the Indian idea, the following terms are distinguished in this example: a) In the present case, a "primer" is generally understood to mean a population of primer molecules with a uniform structure Understand primer molecules that have different structures. C) A "primer molecule" means a single oligonucleotide molecule. D) ^ Several primer moleculesξjξ mean several individual oligonucleotide molecules; they can have a uniform or different structure.

SNP position - a position in NSK that is checked for the presence or absence of SNP.

Target Sequence - Part of an overall sequence that is sequenced / determined using a specific primer in the sequencing reaction. A Entire sequence can contain several target sequences. A target sequence is long enough to ensure that it is very likely to be positioned within the overall sequence. Target sequences can contain one or more SNP sites, for example.

Detection Sequence - Part of the target sequence that is used to assign this target sequence to the overall sequence.

In this embodiment for SNP analysis, several potential SNP positions in the reference sequence are selected which are examined in an NSK to be analyzed. Accordingly, different, sequence-specific primers are provided for these positions. These primers can form a standardized primer set for SNP analysis for a specific question and can be used as a kit for the relevant analyzes.

The preparation of the material to be analyzed (single and double-stranded nucleic acid chains to be examined for SNP) has the aim according to the invention of creating a population of relatively small, between 30 and 2000 NT long, single-stranded nucleic acid chain fragments (NSKFs) from one or more long nucleic acid chains (overall sequence) form.

These NSKF molecules happen to be random on a flat surface with a density between 10 and 1,000,000 per 100 μm ² , preferably 10 and 100 NSKFs per 100 μm ² , 100 to 10,000 per 100 μm ² or 10,000 to 1,000,000 per 100 μm ² immobilized. Primers are hybridized to the NSKFs bound on the surface, so that the density of the NSKF primer complexes capable of extension is approximately 10-100 per 100 μm ² . After hybridization, unbound primers are removed and the sequencing reaction started.

By selecting the target sequences and the sequence-specific primers are only the relevant ones

Sections of the overall sequence examine what reduces the amount of irrelevant information and shortens the analysis time.

This embodiment of the method for SNP analysis is based on the following principles:

Locations in a reference sequence are selected which are to be checked for single nucleotide polymorphisms (SNPs) in the NSKs to be examined (overall sequence).

1) Specific primers are provided for the analysis of each selected SNP site, so that each SNP site to be examined either occupies the next position in the 3 g direction from the primer or within 2 to 100, preferably 2 to 50, ideally 2 to 20 positions in the 3§ direction from the primer. The SNP site is thus within the target sequence that is determined during the sequencing reaction. Several SNP sites are preferably analyzed simultaneously, so that several specific primers have to be used. The primers are preferably selected so that they have annealing temperatures that are as uniform as possible, i.e. Differences between melting temperatures of individual primer populations are, for example, within a range of approximately 4 degrees, better within 2 degrees, even better within 1 degree.

2) Short nucleic acid chain fragments (NSKFs) are derived from the total sequence, these fragments being single-stranded and having a length of 20 to 2000 NT, preferably 30 to 500 NT.

3) NSKF molecules are arranged in a random order immobilized on the surface.

4) After the hybridization (annealing) of sequence-specific primers to the NSKFs immobilized on the surface, a cyclic sequencing reaction is carried out, a target sequence being determined for each NSKF molecule involved in the reaction. The sequencing reaction takes place on many molecules simultaneously.

5) The determined target sequences contain information about the affiliation to a certain section in the overall sequence and about the SNP in this section for the sample to be examined. The length of the target sequences and thus the number of cycles should be selected so that an identification of the sequences can be guaranteed.

In an advantageous embodiment, the determined target sequences are compared with the reference sequence and assigned by sequence matching. If the target sequence is sufficiently long, it is very likely that it can be assigned to a specific position in the reference sequence. For example, a sequence of 10 NTs can form more than 10 ^ε different combinations and can therefore be identified with a high probability in an NSC of only 100,000 NT. After assigning the determined target sequence to the specific position within the reference sequence, differences in the sequences, the SNPs, become visible.

In another advantageous embodiment, the already known number of primers, their composition and an already known sequence section of the reference sequence which adjoins the primer binding site are used to identify the target sequences used. The target sequences determined are analyzed according to their affiliation with the primers, only the sequences close to the primer binding site having to be taken into account. If, for example, only 1000 primers are used, fewer than 10 NTs of the determined target sequences are sufficient to enable an assignment to the corresponding primers.

The sample to be analyzed usually contains several identical total sequence molecules, e.g. multiple copies of genomic DNA from cells of a tissue or multiple identical mRNA populations from cells of a tissue.

a) Choice of the SNP position

The method according to the invention can be used to analyze known SNP positions as well as to determine new SNP positions. Any position in the NSK can appear as a potential SNP position. The selection depends on the question, e.g. SNP analysis in genes whose products are associated with certain diseases, or SNP analysis in conserved, coding sections of the genes that code for membrane receptors, or checking known SNP sites in regulatory sequences of genes that are important for cell division.

An SNP site to be analyzed lies within a target sequence that is determined during the sequencing reaction. You can identify multiple SNP locations within a target sequence. On the other hand, one can also select several target sequences, for example within one gene. It is important that the target sequences are at a sufficient distance from one another in the overall sequence. This distance is necessary so that only one sequence-specific primer hybridizes per NSKF, and it depends on the average NSKF length: the shorter the NSKFs, the closer the target sequences can be. With an appropriate choice of primers, the SNP sites can do both Strands of a double-stranded nucleic acid chain are analyzed.

The method also offers the possibility, for example, of controlling several SNP positions from many individuals (as a sample of a population) at the same time. This can e.g. the SNP profile of a population are examined.

b) primers for the sequencing reaction

Sequencing reaction on a single NSKF molecule is made possible by a primer molecule. According to the invention, a sequence-specific primer is necessary in order to be able to carry out the sequencing reaction in each case on a specific / specific target sequence within the overall sequence. The sequence-specific primer to be used for the analysis of an SNP site or a target sequence represents a population of primer molecules with an identical structure. Several different primer populations are necessary for the analysis of several different target sequences.

By using sequence-specific primers, only the relevant sequence segments, the target sequences, are analyzed. In the method according to the invention, the length of the sequences to be sequenced is kept as short as possible so that the speed of the analysis increases.

A sequence-specific primer binds to a specific primer binding site in the sequence to be analyzed, PBS. The composition and length of the primers are optimized for each potential SNP site or target sequence. Examples of optimization steps are in Rychlik et al. NAR 1990 v.18 p.6409 shown. When choosing a primer or choosing a PBS (priming agent), the following aspects must be taken into account:

1) The SNP position to be analyzed should either be the same after the 3 'end of the primer or within the next 2 to 50 NTs, preferably 2 to 20 NTs.

2) The positioning (the choice of the sequence length and the composition) of the PBS to the SNP site should take place in such a way that the different PBS sequences and the corresponding primer sequences have as similar as possible "annealing temperatures" in order to achieve the most uniform hybridization conditions possible tie. This can be done, for example, by changing the PBS position in relation to the respective SNP site to be analyzed or by changing the length of the primer sequence (Rychlik et al. NAR 1990 v.18 p.6409).

3) The minimum distance between primers that bind to the same strand in the overall sequence should not be less than the average NSKF length.

Primers can be used for both strands of a double-strand NSK. This makes it possible, for example, to record SNP points that are close to one another, or to check an SNP point in both lines.

The length of the primer is preferably between 6 and 100 NTs, optimally between 10-30 or 30-40 or 40-50. Primers with different lengths can be used for different SNP sites or target sequences.

For the SNP analysis with sequence-specific primers, primers are hybridized according to the invention in a hybridization solution to the NSKFs immobilized on the reaction surface (annealing reaction).

c) Immobilization of NSKFs

In this embodiment, the NSKF primer complexes are bound to the surface exclusively via the NSKFs (direct binding of NSKFs to the Surface), the NSKF molecules provided being bound to the flat surface in a random arrangement.

The NSKFs are preferably immobilized at one of the two chain ends (see above). The immobilization can also be achieved by a non-specific binding, for example by drying out the sample containing NSKFs on the flat surface. The density of the immobilization can be between 10 and 100, 100 and 10,000, 10,000 and 1,000,000 NSKFs per 100 μm ² .

d) hybridization

The bound NSKFs and the primers are incubated under stringent hybridization conditions which allow the primers to be linked as selectively as possible (annealing) to the corresponding primer binding sites of the NSKFs. Optimal hybridization conditions depend on the exact structure of the primer binding sites and the respective primer structures and can be determined, for example, according to Rychlik et al. Calculate NAR 1990 v.18 p.6409.

The primers are preferably a mixture of primers. The concentrations of individual sequence-specific primers (individual concentrations of primer populations) are, for example, between 10 pmol / l and 1 mmol / 1, preferably between 0.1 μmol / l and 10 μmol / l. The total concentration of primers in the primer mixture is preferably between 1 mol / 1 and 10 mmol / l. The ratio between individual primer populations can vary. Primers can be added in a significant excess over the immobilized NSKFs, so that the hybridization time is short.

The density of NSKF primer complexes which can be extended is necessary for the detection to be approximately 10 to 100 per 100 μm ² . It can be achieved before, during or after hybridization of the primers. In the case of a known NSKF concentration, in one embodiment the immobilization conditions can be selected such that the NSKFs are bound in a density of approx. 10 to 1000 molecules per 100 μm ² . NSKFs thus determine the density of the NSKF primer complexes.

In another embodiment, the density of the immobilized NSKFs can be substantially higher than 1000 NSKFs per 100 μm ² , for example 1,000,000 per 100 μm ² . The density of the NSKF primer complexes required for optical detection is achieved during the primer hybridization. The hybridization conditions (eg temperature, time, buffer) should be selected so that the primers only bind to a part of the immobilized NSKFs.

If the NSKF concentration is unknown and the immobilization density is accordingly unknown, the hybridization (annealing) of primers to the NSKFs can lead to a higher than optimal density of NSKF-primer complexes.

For this reason, in an advantageous embodiment, part of the sample containing NSKFs is used to determine the optimal density. This part is immobilized on a reaction surface, the primers are hybridized to the NSKFs and the resulting NSKF-primer complexes are marked by the incorporation of fluorescent dye-bearing NT * s (eg Cy3-dCTP, Amersham Pharmacia Biotech). On the one hand, the density determined can be used to calculate the dilution or concentration of the original sample that may be necessary for the final sequencing approach (the hybridization conditions are maintained). On the other hand, necessary changes in the hybridization conditions can be calculated from this, for example a shortening of the hybridization time, the NSKF immobilization density remaining constant. The quantity ratio between primer populations can be different or the same size. A higher primer concentration makes it possible to bind certain, for example rarer, sequences with a higher probability in a certain period of time.

The great advantage of the described process arrangement compared to a process arrangement with sequence-specific primers immobilized on a surface and a subsequent hybridization of samples to these primers is the significant reduction in the time for hybridization (annealing) between the sequence-specific primers and the samples to be analyzed on the reaction surface.

Legends for Figures 1 to 8

Legend to Fig. 1

Schematic representation of the sequencing of a long nucleic acid chain

The sequencing and reconstruction of long nucleic acid sequences (NSKs) is based on the shotgun principle. The sequence of a long piece of DNA is determined by sequencing small fragments (NSKFs) and a subsequent reconstruction.

1) Starting material - the long nucleic acid sequence to be analyzed, total sequence

2) Fragments of 50-1000 bp - the NSKFs generated from the total sequence in the fragmentation step

3) Fragments with one primer each - NSKF-primer complexes

4) Immobilized fragments - on the flat surface bound NSKF primer complexes, in this embodiment the binding takes place at the 3 end of the NSKFs

5) Add a solution with polymerases and NT * s - the first step in a cycle of the sequencing reaction

6) Washing step - after the installation step, the surface is washed

7) Detection - the signals from individual built-in NT * s are detected

8) Remove the marker and the group leading to the termination

Legend for Fig. 2

Examples of general structure of NSKF primer complexes

In this embodiment, a uniform primer binding site (PBS) is coupled to the 3 'end of the NSKFs and a uniform primer binds to this PBS.

Legend for Fig. 3

An example of the coupling of a uniform primer binding site (PBS), which carries a functional group for binding to the surface.

In this case, a double-stranded oligonucleotide complex (3a), which has, for example, a modification on both strands (3b), is located on the double-stranded NSKFs (3c). After denaturation, single-strand NSKFs with uniform PBS (3d) are created. Legend for Fig. 4

Another example of the creation of a uniform primer binding site (PBS).

In this case, NTs are coupled to the 3 'end of the single-strand NSKFs (a so-called "tailing"). By using a uniform NT, a uniform PBS is created.

Legend for Fig. 5

Example of the binding of NSKFs to a gel-like reaction surface.

A gel layer (2) adheres to a solid base (1), e.g. a polyacrylamide gel (Fig. 5a), or many gel beads (5), e.g. Agarose beads (Fig. 5b). NSKFs (4) are bound to the surface of the gel. The NSKFs have a functional group, e.g. Biotin, and are bound to the gel via streptavidin or avidin (3).

Legend for Fig. 6

Example of a flow device

A gel-like reaction surface (1) is attached to a solid base (2) which is permeable to the excitation and fluorescent light. Together they form the lid of the flow cell. The liquids in the flow cell can be exchanged in a controlled manner, the flow cell forming a flow device together with the reservoir (3), pump (4) and valve (5). NSKF primer complexes are bound to the reaction surface (not shown here). The signals of the installed NT * s are detected with the detection apparatus (6).

Legend for Fig. 7

Schematic representation of the analysis of mRNA population

The analysis is based on the sequencing of short sections of mRNA.

1) mRNA - the mRNA population to be analyzed, in this example consisting of two different mRNA molecule populations (thin and thick strips represent mRNA molecules)

2) Immobilized mRNA - mRNA-primer complexes bound to the flat surface, in this example the binding is carried out by the oligo-dT primer

3) Add a solution with polymerases and NT * s - the first step in a cycle of the sequencing reaction

4) Washing step - after the installation step, the surface is washed

5) Detection - the signals from individual built-in NT * s are detected

6) Remove the marker and the group leading to the termination

Legend for Fig. 8

Example of a detection system

A wide-field optical detection system is shown. After installing marked NT * s, the surface (7) scanned, the fluorescence signals being detected by individual dye molecules coupled to the NTs.

Fig. 8a Schematic representation of a portion of the reaction surface (gray) that is scanned. The circles each correspond to the recording of a 2D image and represent the areas from which the fluorescence signals are detected. Several signals (for example 100 to 10,000) of individual molecules are registered simultaneously per exposure. The reaction surface is scanned in each cycle, several images being taken from different locations on the surface during the scanning process. Up to several million signals can be recorded by built-in NT * s. The high degree of parallelism is the basis for the speed of the process.

Fig. 8b A recording (a 2D image) with signals from individual, built-in NT * s.

Fig. 8c detail from Figure 8b. The section shows signals from four built-in NTs. Each signal has characteristic properties of the single molecule signals (see description) and can be identified on the basis of these (preferably with the aid of a computer program). The corresponding X, Y coordinates are assigned to each of the identified signals.

Claims

claims;

1. Method for parallel sequence analysis of nucleic acid sequences (nucleic acid chains, NSKs), in which one

Generated fragments (NSKFs) of single-stranded NSKs with a length of about 50 to 1000 nucleotides, which can represent overlapping partial sequences of an entire sequence

the NSKFs are bound on a reaction surface in a random arrangement using one or more different primers in the form of NSKF-primer complexes

cyclically build the complementary strand of the NSKFs using one or more polymerases by

a) a solution is added to the NSKF primer complexes bound to the surface, which contains one or more polymerases and one to four modified nucleotides (NTs ^* ) which are labeled with fluorescent dyes, the simultaneous use of at least two NTs ^* Fluorescent dyes located on each of the NTs ^* are selected such that the NTs ^{* used} can be distinguished from one another by measuring different fluorescence signals, the NTs ^* at the 3 'position being structurally modified such that the polymerase after incorporation of such an NT ^* in a growing complementary strand is unable to incorporate a further NT ^* into the same strand, the terminating substituent and the fluorescent dye being removable, man b) the stationary phase obtained in stage a) is incubated under conditions which are suitable for the extension of the complementary strands, the complementary strands being extended in each case by one NT ^* , man

stages a) to f) are repeated several times, if necessary,

2. The method according to claim 1, characterized in that steps a) to f) of the cyclic reaction are repeated several times, only one marked NT ^* being used in each cycle.

3. The method according to claim 1, characterized in that the steps a) to f) of the cyclic reaction are repeated several times, using two differently labeled NTs ^* in each cycle.

4. The method according to claim 1, characterized in that the steps a) to f) of the cyclic reaction are repeated several times, using four differently marked NTs ^* in each cycle.

5. The method according to claim 1, characterized in that the NSKs are variants of a known reference sequence and steps a) to f) of the cyclic reaction are repeated several times, alternately using two differently marked NTs ^* and two unmarked NTs in the cycles and the total sequences are determined by comparison with the reference sequence.

6. The method according to claims 1 to 5, characterized in that in each case a primer binding site (PBS) is introduced into the NSKFs, wherein in the case of double-stranded NSKs one PBS is introduced on each of two complementary single strands and the primer binding sites are the same or different for all NSKFs Have sequences.

7. The method according to claims 1 to 6, characterized in that the NSKFs are brought into contact with primers in a solution under conditions which lead to hybridization. tion of the primers to the primer binding sites (PBSs) of the NSKFs are suitable, the primers having identical or different sequences to one another, and the NSKF-primer complexes formed are then bound to the reaction surface.

8. The method according to claims 1 to 6, characterized in that the NSKFs are first immobilized on the reaction surface and only then brought into contact with primers under conditions which are suitable for hybridizing the primers to the primer binding sites (PBSs) of the NSKFs, wherein NSKF primer complexes are formed, the primers having the same or different sequences.

9. The method according to claims 1 to 6, characterized in that the primers are first immobilized on the reaction surface and only then brought into contact with NSKFs under conditions which hybridize the primers to the primer binding sites

(PBSs) of the NSKFs are suitable, as a result of which NSKFs are bound to the surface and NSKF-primer complexes are formed, the primers having identical or different sequences to one another.

10. The method according to claims 1 to 9, characterized in that the density of the expandable

NSKF primer complexes is between about 10 and 100 per 100 μm ² .

11. The method according to claim 1, characterized in that the nucleic acid sequences (NSKs) are sequence sections of an entire sequence and the primers are sequence-specific primers, wherein one provides single-stranded NSKFs with a length of about 30 to 1000 nucleotides, which correspond to overlapping partial sequences of the entire sequence

the NSKF molecules bind directly to a flat surface in a random arrangement, one

hybridization (annealing) to the immobilized NSKFs is carried out with one or more sequence-specific primer populations, the density of the individual NSKF-primer complexes capable of extension being between 10 and 100 per 100 μm ²

performs a cyclic build-up reaction of the strands complementary to NSKFs by

a) a solution is added to the bound NSKF-primer complexes, which contains one or more polymerases and one to four modified nucleotides (NTs ^* ), which are labeled with fluorescent dyes, with the simultaneous use of at least two NTs ^* each on the Fluorescent dyes located in NTs ^* are selected such that the NTs ^{* used} can be distinguished from one another by measuring different fluorescence signals, the NTs ^* at the 3 'position being structurally modified so that the polymerase after incorporation of such an NT ^* into a growing complementary one Strand is unable to incorporate another NT ^* into the same strand, the terminating substituent and the fluorescent dye being cleavable

b) the stationary phase obtained in stage a) is incubated under conditions which are suitable for the extension of the complementary strands, the complementary strands being extended in each case by one NT ^* , man c) the stationary phase obtained in stage b) is washed under conditions which are not suitable for the removal of NTs ^* which are not incorporated in a complementary strand

e) the individual NT ^* molecules built into complementary strands are detected by measuring the signal characteristic of the respective fluorescent dye, at the same time determining the relative position of the individual fluorescent signals on the reaction surface

e) the substituents leading to the termination and the fluorescent dyes are cleaved from the NTs ^* attached to the complementary strand to produce unlabelled (NTs or) NSKFs, man

f) the stationary phase obtained in step e) is washed under conditions which are suitable for removing the groups leading to termination with the fluorescent dyes

stages a) to f) are repeated several times, if necessary,

12. Method for highly parallel analysis of gene expression in which one provides single-stranded gene products, one

perform a cyclic assembly reaction of the complementary strand of the gene products using one or more polymerases by

a) a solution is added to the gene product-primer complexes bound to the surface which contains one or more polymerases and one to four modified nucleotides (NTs ^* ) which are labeled with fluorescent dyes, the simultaneous use of at least two NTs ^* Fluorescent dyes located on each of the NTs ^* are selected such that the NTs ^{* used} can be distinguished from one another by measuring different fluorescence signals, the NTs ^* at the 3 'position being structurally modified such that the polymerase after incorporation of such an NT ^* in a growing complementary strand is unable to incorporate a further NT ^* into the same strand, the terminating substituent and the fluorescent dye being removable, man

c) the stationary phase obtained in stage b) below Conditions that are not suitable for the removal of NTs ^* built into a complementary strand, one

d) the individual NTs ^* built into complementary strands are detected by measuring the signal which is characteristic of the respective fluorescent dye, at the same time determining the relative position of the individual fluorescent signals on the reaction surface

e) to generate unlabelled (NTs or) gene products, the substituents leading to the termination and the fluorescent dyes are cleaved from the NTs ^* attached to the complementary strand, man

stages a) to f) are repeated several times, if necessary,

13. The method according to claim 12, characterized in that steps a) to f) of the cyclic reaction are repeated several times, only one marked NT ^* being used in each cycle.

14. The method according to claim 12, characterized in that stages a) to f) of the cyclic reaction are repeated several times, with two differently marked NTs ^* being used in each cycle.

15. The method according to claim 12, characterized in that steps a) to f) of the cyclic reaction are repeated several times, using four differently marked NTs ^* in each cycle.

16. The method according to claim 12, characterized in that already known genes serve as reference sequences and steps a) to f) of the cyclic reaction are repeated several times, alternately using two differently labeled NTs ^* and two unlabeled NTs in the cycles and the identity of the gene products is determined by comparing the sequences obtained with those of the reference sequences.

17. The method according to claims 12 to 16, characterized in that in each case a primer binding site (PBS) is introduced into the gene products, the primer binding sites each having the same or different sequences for all gene products.

18. The method according to claims 12 to 17, characterized in that the gene products are brought into contact with primers in a solution under conditions which hybridize the primers to the primer binding sites

(PBSs) of the gene products are suitable, the primers having identical or different sequences to one another, and the gene product-primer complexes formed are then bound to the reaction surface.

19. The method according to claims 12 to 17, characterized in that the gene products are first immobilized on the reaction surface and only then brought into contact with primers under conditions which lead to Hybridization of the primers to the primer binding sites (PBSs) of the gene products are suitable, gene product-primer complexes being formed, the primers having identical or different sequences to one another.

20. The method according to claims 12 to 17, characterized in that the primers are first immobilized on the reaction surface and only then brought into contact with gene products under conditions which hybridize the primers to the primer binding sites

(PBSs) of the gene products are suitable, as a result of which gene products are bound to the surface and gene product-primer complexes are formed, the primers having identical or different sequences to one another.

21. The method according to claims 12 to 20, characterized in that the density of the extensible gene product-primer complexes is between about 10 and 100 per lOOμm ² .

22. The method according to claims 1 to 21, characterized in that the fluorescent dye is split off together with the substituent leading to the termination.

23. The method according to claims 1 to 21, characterized in that the fluorescent dye is first split off and only then the substituent leading to the termination is split off.

24. The method according to claims 1 to 21, characterized in that the following types of detection are used in detection step (d): wide-field epifluorescence microscopy, laser scanning Fluorescence microscopy, TIRF microscopy.

25. The method according to claims 1 or 11, characterized in that it is a method for SNP analysis and one uses a sequence-specific primer to identify each SNP site in the total sequence.

26. The method according to claim 25, characterized in that the number of SNP sites to be analyzed in parallel is greater than 2 and a sequence-specific primer is used for each SNP site.

27. The method according to claims 1 to 26, characterized in that the reaction surface is selected from the group consisting of silicone, glass, ceramic, plastics, gels.

28. The method according to claim 27, characterized in that the plastics are polycarbonates or polystyrenes or derivatives thereof.

29. The method according to claim 27, characterized in that the gels are agarose or polyacrylamide gels or derivatives thereof.

30. The method according to claim 29, characterized in that the gels are 1 to 2% agarose gels or 10 to 15% polyacrylamide gels.

31. The method according to claims 1 to 30, characterized in that the polymerase is a DNA polymerase without 3 '-5' exonuclease activity.

32. The method according to claim 31, characterized in that the polymerase from the group consisting of viral DNA polymerases of the sequenase type, bacterial thermolabile and thermostable DNA polymerases, DNA polymerases beta from eukaryotes and reverse transcriptases is selected.

33. The method according to claims 1 to 32, characterized in that the fluorescent dyes are selected from the group consisting of cyanine dyes, rhodamines, xanthenes, porphyrins and their derivatives.

34. Carrier for carrying out the method according to claims 1 to 33, characterized in that the nucleic acid chains or their fragments are immobilized on their surface in a random arrangement, the density of the immobilized nucleic acid chain molecules or their fragments between 10 and 100 per 100 μm ² lies.

35. Carrier for carrying out the method according to claims 1 to 33, characterized in that the nucleic acid chains or their fragments are immobilized in a random arrangement on its surface, the density of the immobilized nucleic acid chain molecules or their fragments between 100 and 1,000,000 per lOOμm ² .

36. Kit for performing the method according to claims 1 to 35, characterized in that it contains a reaction surface (a solid support), reaction solutions required to carry out the method, one or more polymerases, and nucleotides (NTs), one of which is four are marked with fluorescent dyes, the marked NTs also being structurally modified (NT ^* or NTs ^* ) in such a way that the polymerase is incapable of incorporating such an NT ^* into a growing complementary strand is to install another NT ^* in the same strand, the terminating substituent being cleavable with the fluorescent dye

37. Kit according to claim 36, characterized in that it further contains components:

a) reagents required for the production of single strands from double strands, b) nucleic acid molecules which are introduced as PBS into the NSKFs, c) oligonucleotide primers, d) reagents required to split off the substituents with the fluorescent dyes, e) washing solutions.

38. Kit according to claim 36 or 37, characterized in that the reaction surface is selected from the group consisting of silicone, glass, ceramic, plastics, gels.

39. Kit according to claim 38, characterized in that the gels are polyacrylamide gels.

40. Kit according to claim 39, characterized in that the gels are 5 to 30% polyacrylamide gels.

41. Kit according to claims 36 to 40, characterized in that the DNA polymerase is a DNA polymerase without 3'- 5 'endonuclease activity.

42. Kit according to claims 36 to 41, characterized in that the fluorescent dyes coupled to the nucleotides are selected from the group consisting of cyanine dyes, rhodamines, xanthenes, porphyrins and their derivatives.